From 391b3a473294afe427a1ca836c3bfad2c444ae0f Mon Sep 17 00:00:00 2001
From: wuyang <5700876+banisherwy@user.noreply.gitee.com>
Date: Wed, 8 Jul 2026 12:25:30 +0800
Subject: [PATCH] Expand arXiv paper corpus

---
 .../2026-07-08-arxiv-expanded-sweep.md        |    55 +
 ...t-candidates-2025-07-08-to-2026-07-08.json | 53788 ++++++++++++++++
 ...agent-papers-2025-07-08-to-2026-07-08.json | 35331 ++++++++++
 data/index.json                               | 39106 +++++++++++
 data/summary.json                             |   106 +-
 papers/README.md                              |     1 +
 papers/corpus-summary-2026-07-08.md           |   143 +
 ...tial-decision-making-in-language-models.md |    61 +
 ...-calling-for-relative-evaluation-of-ano.md |    63 +
 ...-context-protocol-mcp-tool-use-benchmar.md |    60 +
 ...agentic-recovery-from-external-failures.md |    61 +
 ...tion-for-agentic-multi-turn-interaction.md |    61 +
 ...-for-automating-distribution-grid-analy.md |    63 +
 ...long-horizon-and-efficient-mobile-agent.md |    66 +
 ...or-power-system-analysis-and-operations.md |    60 +
 ...ture-for-geospatial-analysis-function-c.md |    62 +
 ...aluate-agent-architectures-in-enterpris.md |    64 +
 ...ic-intelligence-via-environment-scaling.md |    59 +
 ...ngual-and-regionalized-agent-evaluation.md |    60 +
 ...uation-of-llm-agents-beyond-final-state.md |    61 +
 ...tion-free-controllable-evaluation-frame.md |    61 +
 ...-a-survey-of-architectures-capabilities.md |    65 +
 ...rning-with-a-multi-turn-multi-task-fram.md |    59 +
 ...eyond-utility-an-open-ended-perspective.md |    61 +
 ...ework-for-llm-based-multi-agent-applica.md |    61 +
 ...gal-behavior-of-llm-agents-under-eu-law.md |    61 +
 ...guage-model-susceptibility-in-agent-to-.md |    62 +
 ...ta-synthesis-solution-for-real-world-mu.md |    61 +
 ...owards-agentic-tool-use-reward-modeling.md |    60 +
 ...-llm-agents-via-environment-interaction.md |    63 +
 ...-through-multi-agent-interaction-in-vqa.md |    63 +
 ...pi-centric-llm-agent-defense-frameworks.md |    62 +
 ...all-language-models-for-agentic-tasks-o.md |    60 +
 ...r-general-ai-agents-a-technical-white-p.md |    65 +
 ...entic-reasoning-in-the-neurips-cure-ben.md |    65 +
 ...l-use-data-via-multi-agent-role-playing.md |    62 +
 ...earning-for-agentic-information-seeking.md |    65 +
 ...t-small-language-coder-model-mify-coder.md |    64 +
 ...ation-of-llm-agents-under-real-world-ap.md |    60 +
 ...and-execution-of-llm-generated-programs.md |    64 +
 ...ompt-caching-for-long-horizon-agentic-t.md |    61 +
 ...xt-engineering-for-agentic-data-science.md |    60 +
 ...del-paper-reading-agents-more-efficient.md |    62 +
 ...-multi-agent-reasoning-through-holistic.md |    63 +
 ...-deployable-in-real-world-dynamic-envir.md |    61 +
 ...-agent-memory-with-systematic-benchmark.md |    63 +
 ...gent-creation-for-agentic-orchestration.md |    63 +
 ...-language-agents-under-communication-ba.md |    61 +
 ...gents-in-realistic-negotiation-scenario.md |    61 +
 ...fficient-agent-defense-with-hierarchica.md |    62 +
 ...lutionary-security-evaluation-of-agents.md |    60 +
 ...erabilities-across-deep-research-agents.md |    64 +
 ...nder-controllable-and-extreme-context-g.md |    61 +
 ...d-detecting-tool-use-hallucinations-via.md |    62 +
 ...g-and-benchmarking-attacks-on-openclaw-.md |    63 +
 ...ramework-for-agent-system-observability.md |    62 +
 ...-agent-safety-through-incident-response.md |    61 +
 ...fending-multi-turn-safety-risks-in-tool.md |    65 +
 ...-with-episodic-memory-in-language-agent.md |    62 +
 ...lls-for-agentic-ai-through-hybrid-model.md |    59 +
 ...-framework-for-long-horizon-search-agen.md |    62 +
 ...adaptive-trust-calibration-in-model-con.md |    62 +
 ...ety-alignment-in-vision-language-agents.md |    59 +
 ...ond-single-channel-agentic-benchmarking.md |    61 +
 ...iation-as-a-causal-mechanism-of-agent-f.md |    62 +
 ...perception-vulnerability-in-llm-driven-.md |    63 +
 ...gents-with-parametric-reflective-memory.md |    61 +
 ...iminal-prompting-in-multi-agent-systems.md |    61 +
 ...ctured-analysis-and-reporting-of-agenti.md |    63 +
 ...ini-internets-for-diagnosing-epistemic-.md |    62 +
 ...ime-via-dynamic-importance-estimation-f.md |    64 +
 ...us-llm-fine-tuning-with-language-agents.md |    61 +
 ...pproach-to-study-affective-polarization.md |    60 +
 ...rnance-framework-for-military-ai-agents.md |    63 +
 ...uage-agents-toward-strategic-exploratio.md |    61 +
 ...tform-for-function-calling-data-synthes.md |    63 +
 ...benchmark-for-self-evolving-language-ag.md |    61 +
 ...rchical-autonomy-evolution-of-ai-agents.md |    64 +
 ...ion-of-data-over-exposure-in-llm-agents.md |    62 +
 ...-are-language-agents-from-human-experts.md |    63 +
 ...e-to-metal-kernel-generation-on-emergin.md |    62 +
 ...-considerations-for-multi-agent-systems.md |    66 +
 ...sis-with-evidence-integrated-language-a.md |    63 +
 ...pe-of-agentic-ai-a-comprehensive-survey.md |    61 +
 ...-through-multi-agent-dialectical-negoti.md |    61 +
 ...-for-tool-use-under-complex-constraints.md |    61 +
 ...more-precise-instructions-for-language-.md |    59 +
 ...zed-llm-agents-the-curious-case-of-ment.md |    62 +
 ...tomated-alzheimer-s-disease-detection-w.md |    61 +
 ...ion-and-coverage-audit-of-llm-agent-too.md |    63 +
 ...work-for-formalizing-llm-agent-security.md |    61 +
 ...a-geometry-aware-vision-language-agents.md |    62 +
 ...lay-for-llm-agent-trajectory-relabeling.md |    62 +
 ...hierarchical-memory-for-language-agents.md |    64 +
 ...or-persistent-and-semantically-consiste.md |    60 +
 ...system-for-autonomous-industrial-safety.md |    62 +
 ...-cross-the-line-before-they-actually-do.md |    60 +
 ...e-usage-of-agents-with-real-world-tools.md |    63 +
 ...eral-purpose-agent-for-open-agentic-web.md |    64 +
 ...-for-small-uas-separation-assurance-und.md |    64 +
 ...jectory-benchmark-for-safety-evaluation.md |    62 +
 ...ought-budget-effects-in-function-callin.md |    61 +
 ...and-internal-reward-for-language-agents.md |    63 +
 ...upled-latent-reasoning-for-agent-safety.md |    62 +
 ...lexity-for-tool-augmented-language-agen.md |    62 +
 ...t-emerging-supply-chain-injections-in-a.md |    60 +
 ...ulecon-agentic-security-rule-conversion.md |    60 +
 ...optimization-for-safe-multi-agent-navig.md |    65 +
 ...c-agentic-data-reactivates-general-tool.md |    59 +
 ...-user-instructions-expose-critical-vuln.md |    63 +
 ...tion-data-and-evaluation-for-llm-agents.md |    60 +
 ...why-ai-agents-that-think-must-never-act.md |    63 +
 ...ntic-llm-driven-obfuscation-for-ip-prot.md |    61 +
 ...ing-information-and-control-to-filesyst.md |    61 +
 ...trinsic-non-attack-trajectory-benchmark.md |    62 +
 ...odied-vision-language-agent-framework-f.md |    63 +
 ...do-harmful-skills-weaponize-your-agents.md |    62 +
 ...ty-symbolic-guardrails-for-domain-speci.md |    63 +
 ...ge-reliability-propagation-cascades-and.md |    61 +
 ...ection-architecture-for-agentic-systems.md |    65 +
 ...issing-threat-model-for-ai-agent-safety.md |    64 +
 ...hitectures-for-offensive-security-tasks.md |    62 +
 ...d-harm-recovery-for-computer-use-agents.md |    65 +
 ...imization-framework-for-language-agents.md |    62 +
 ...ot-be-it-mitigating-trust-boundary-conf.md |    62 +
 ...cks-novel-threats-for-function-calling-.md |    62 +
 ...f-vision-language-agents-for-spatial-re.md |    61 +
 ...context-fragmented-violations-in-multi-.md |    67 +
 ...pecifications-from-1-bit-danger-signals.md |    64 +
 ...nformation-flow-tracking-for-llm-agents.md |    62 +
 ...ecture-matters-for-multi-agent-security.md |    65 +
 ...-agents-with-efficient-dynamic-analysis.md |    63 +
 ...luation-of-ai-agent-security-guardrails.md |    61 +
 ...-for-open-source-llms-in-interactive-to.md |    59 +
 ...rk-for-automated-3d-cutscene-generation.md |    64 +
 ...idation-and-zero-trust-security-for-sem.md |    63 +
 ...l-firewall-for-structured-workflow-ai-a.md |    63 +
 ...rdrails-for-clinical-safety-hallucinati.md |    63 +
 ...ic-discovery-on-a-real-optical-platform.md |    63 +
 ...r-autonomous-agent-frameworks-a-layered.md |    63 +
 ...al-framework-for-proactive-embodied-age.md |    64 +
 ...ement-learning-in-large-language-models.md |    62 +
 ...y-efficient-red-teaming-for-prompt-inje.md |    62 +
 ...-contracts-for-agentic-security-systems.md |    63 +
 ...urity-pattern-selection-for-iot-systems.md |    62 +
 ...constraint-guided-large-language-agents.md |    63 +
 ...he-loop-ai-speech-therapy-agent-for-per.md |    68 +
 ...-framework-for-agent-safety-measurement.md |    61 +
 ...m-agents-in-real-world-ehr-environments.md |    63 +
 ...d-benchmark-rewriting-and-analogical-re.md |    64 +
 ...tration-for-small-language-model-agents.md |    63 +
 ...ork-for-pre-print-anomaly-detection-in-.md |    61 +
 ...or-root-cause-analysis-in-microservices.md |    61 +
 ...compilation-for-agentic-llm-deployments.md |    62 +
 ...ollable-and-interactive-red-teaming-pla.md |    64 +
 ...rieval-for-agentic-search-via-direct-co.md |    63 +
 ...ugmented-guardrail-for-llm-agent-safety.md |    63 +
 ...t-interference-in-llm-agent-scaffolding.md |    64 +
 ...arning-for-long-horizon-language-agents.md |    63 +
 ...yber-offense-forecast-consequences-and-.md |    64 +
 ...e-evolution-of-llm-agent-memory-mechani.md |    63 +
 ...or-reliable-llm-based-autonomous-agents.md |    64 +
 ...m-agents-a-unified-graph-representation.md |    64 +
 ...neral-sequential-decision-making-agents.md |    64 +
 ...interpretability-of-agentic-ai-tool-use.md |    63 +
 ...s-for-hierarchical-generalized-planning.md |    60 +
 ...-safety-fail-to-generalize-across-tasks.md |    60 +
 ...i-model-router-for-agentic-tool-calling.md |    61 +
 ...luating-llms-on-chemical-cost-reasoning.md |    62 +
 ...n-llm-agents-for-cyber-attack-scenarios.md |    60 +
 ...lving-memory-agents-over-provenance-dag.md |    64 +
 ...l-layers-a-mechanistic-evaluation-of-pe.md |    66 +
 ...ramework-for-automated-cyber-intrusions.md |    64 +
 ...-reasoning-level-denial-of-service-in-l.md |    60 +
 ...nd-accountability-from-models-to-agents.md |    63 +
 ...ention-verification-for-language-agents.md |    62 +
 ...dicts-action-control-in-language-agents.md |    62 +
 ...e-benchmark-for-evaluating-agent-values.md |    60 +
 ...-agentic-ai-systems-openclaw-case-study.md |    62 +
 ...s-of-llm-agents-in-real-os-environments.md |    62 +
 ...-a-rate-distortion-framework-for-agent-.md |    60 +
 ...y-argument-level-provenance-solves-enfo.md |    65 +
 ...in-llm-agents-via-trajectory-refinement.md |    64 +
 ...se-agents-via-structured-meta-cognition.md |    63 +
 ...tent-in-simulated-embodied-environments.md |    64 +
 ...marking-heterogeneous-geospatial-reason.md |    63 +
 ...ajectories-for-agentic-safety-alignment.md |    60 +
 ...ark-and-domain-randomized-rl-recipe-for.md |    60 +
 ...tual-trace-auditing-of-llm-agent-skills.md |    61 +
 ...fety-under-skill-facing-attack-surfaces.md |    68 +
 ...-engine-for-structure-aware-associative.md |    62 +
 ...ent-aware-structured-memory-for-long-ho.md |    62 +
 ...traversal-retrieval-with-planning-mecha.md |    62 +
 ...text-icu-data-a-benchmark-beyond-behavi.md |    66 +
 ...work-for-distributed-materials-informat.md |    64 +
 ...es-as-self-maintaining-software-ecosyst.md |    62 +
 ...gents-in-fast-healthcare-interoperabili.md |    63 +
 ...ld-adopt-the-plan-then-execute-paradigm.md |    65 +
 ...enchmark-for-real-world-teaching-workfl.md |    60 +
 ...guided-enforcement-for-llm-agent-memory.md |    62 +
 ...t-supply-chains-via-payload-less-skills.md |    62 +
 ...ent-memory-in-multi-party-conversations.md |    62 +
 ...learning-interatomic-potential-developm.md |    64 +
 ...collaboration-failure-attribution-and-s.md |    63 +
 ...-memory-in-large-vision-language-models.md |    62 +
 ...curing-ai-agents-like-operating-systems.md |    61 +
 ...-open-source-agentic-modeling-framework.md |    64 +
 ...n-framework-for-multimodal-agent-memory.md |    63 +
 ...arly-to-save-energy-in-consumer-devices.md |    65 +
 ...icle-monte-carlo-workflows-for-colloida.md |    64 +
 ...ng-and-retrieving-agent-memory-via-a-hy.md |    61 +
 ...rce-distributed-multimodal-agent-memory.md |    62 +
 ...ng-for-efficient-long-term-agent-memory.md |    61 +
 ...weight-updates-via-population-broadcast.md |    65 +
 ...g-video-understanding-via-online-indexi.md |    63 +
 ...ice-a-systematic-analysis-of-generator-.md |    63 +
 ...bustness-of-ai-enabled-security-orchest.md |    63 +
 ...-framework-for-resilient-multi-agent-ev.md |    61 +
 ...lm-agents-under-untrusted-tool-feedback.md |    61 +
 ...ture-for-long-horizon-scientific-agents.md |    64 +
 ...tric-memory-layer-for-software-reposito.md |    64 +
 ...esign-for-multi-agent-systems-in-port-h.md |    62 +
 ...mory-control-for-long-horizon-gui-agent.md |    63 +
 ...me-guarantee-architecture-is-structural.md |    60 +
 ...a-locally-correct-but-non-transferable-.md |    63 +
 ...s-for-efficient-and-accurate-llm-agents.md |    61 +
 ...omic-facts-in-lifelong-llm-agent-memory.md |    62 +
 ...rounding-benchmark-for-vision-language-.md |    63 +
 ...lling-precise-decoding-for-agentic-llms.md |    64 +
 ...emory-consolidation-for-language-agents.md |    61 +
 ...rizon-memory-environment-for-llm-agents.md |    66 +
 ...e-by-construction-for-generalist-agents.md |    64 +
 ...xploration-for-self-evolving-llm-agents.md |    63 +
 ...-world-small-molecule-drug-design-tasks.md |    62 +
 ...via-speculative-planning-for-llm-agents.md |    63 +
 ...-temporal-spatial-and-semantic-evasions.md |    64 +
 ...multi-turn-benchmark-for-agentic-safety.md |    63 +
 ...ts-an-empirical-study-of-curriculum-eff.md |    60 +
 ...uantitative-goal-persistence-in-long-ho.md |    64 +
 ...struments-with-natural-language-underst.md |    63 +
 ...ent-memory-via-causal-attribution-and-s.md |    63 +
 ...a-systematic-study-of-model-generated-a.md |    62 +
 ...tem-with-hierarchical-temporal-indexing.md |    62 +
 ...k-to-evaluate-mcp-poisoning-attacks-for.md |    62 +
 ...llm-agents-via-theory-of-mind-reasoning.md |    67 +
 ...level-hallucinations-in-multi-agent-ind.md |    61 +
 ...olar-agentic-rl-on-any-harness-at-scale.md |    61 +
 ...y-as-an-agent-human-interaction-problem.md |    60 +
 ...nst-llm-agents-via-feedback-guided-iter.md |    62 +
 ...nforcement-learning-for-code-generation.md |    64 +
 ...ing-using-edge-and-iot-data-a-review-of.md |    65 +
 ...-agents-on-multi-person-travel-planning.md |    61 +
 ...y-decodable-in-llm-agent-residual-strea.md |    60 +
 ...ce-aware-language-model-for-autonomous-.md |    64 +
 ...undamentals-attacks-and-countermeasures.md |    63 +
 ...onsistency-in-legal-agentic-search-thro.md |    61 +
 ...c-rag-under-constrained-context-budgets.md |    62 +
 ...a-foundations-for-long-term-ai-agent-me.md |    62 +
 ...5-experiments-in-agentic-ai-for-science.md |    63 +
 ...ion-a-dual-graph-defense-for-llm-agents.md |    60 +
 ...lf-evolving-llm-agents-in-cuda-kernel-g.md |    61 +
 ...ic-retrieval-augmented-generation-frame.md |    61 +
 ...ven-logical-retrieval-beyond-embeddings.md |    60 +
 ...ion-language-agents-for-mobile-gui-navi.md |    63 +
 ...-retrieval-for-emotional-support-agents.md |    62 +
 ...e-safety-harness-for-finance-llm-agents.md |    62 +
 ...kill-creation-memory-management-and-eva.md |    62 +
 ...i-turn-llm-agents-via-trajectory-state-.md |    63 +
 ...ough-contrastive-internalization-of-exp.md |    64 +
 ...erence-attacks-on-memory-in-chat-agents.md |    63 +
 ...stigation-of-layer-wise-dynamics-in-seq.md |    60 +
 ...y-as-cognition-in-conversational-agents.md |    64 +
 ...-augmented-generation-for-reliable-lega.md |    64 +
 ...val-augmented-generation-for-multi-agen.md |    61 +
 ...utilization-for-out-of-distribution-gen.md |    59 +
 ...mework-for-automatic-workflow-execution.md |    65 +
 ...-safe-agents-as-recursive-program-holes.md |    63 +
 ...arative-study-in-agentic-data-retrieval.md |    62 +
 ...or-accurate-and-generalizable-function-.md |    62 +
 ...eedback-alignment-in-llm-trading-agents.md |    65 +
 ...-memory-through-action-world-interactio.md |    63 +
 ...r-attributing-retrieval-lift-in-agent-m.md |    63 +
 ...tem-for-stateful-llm-based-applications.md |    61 +
 ...agents-master-pok-mon-trading-card-game.md |    58 +
 ...en-optimized-formats-in-agentic-ai-syst.md |    60 +
 ...ution-for-llm-based-multi-agent-systems.md |    61 +
 ...lignment-framework-for-ai-agent-safety-.md |    63 +
 ...ch-a-multi-agent-harness-for-interleave.md |    66 +
 ...tacks-through-conversational-interactio.md |    62 +
 ...lm-agents-exhibit-human-like-psychology.md |    63 +
 ...generation-with-personalized-multi-agen.md |    64 +
 ...ta-engineering-for-model-specialization.md |    64 +
 ...architecture-for-regulated-cybersecurit.md |    67 +
 ...-as-a-learnable-resource-for-llm-agents.md |    63 +
 ...icient-memory-evolution-in-agentic-llms.md |    65 +
 ...forecasting-with-adaptive-factor-memory.md |    63 +
 ...tive-self-evolving-agentic-jailbreaking.md |    65 +
 ...ng-llm-agents-on-financial-spreadsheets.md |    62 +
 ...es-a-multi-agent-framework-for-evidence.md |    62 +
 ...used-memorization-for-multimodal-agents.md |    61 +
 ...026-2605-31268-mellum2-technical-report.md |    62 +
 ...nce-the-glide-library-for-reliable-gena.md |    61 +
 ...-diagnosing-and-improving-agent-traject.md |    62 +
 ...-tree-for-time-sensitive-news-retrieval.md |    63 +
 ...00198-bagen-are-llm-agents-budget-aware.md |    63 +
 ...vior-arising-from-ordinary-computer-use.md |    63 +
 ...em-for-graph-retrieval-augmented-genera.md |    65 +
 ...mpression-for-long-horizon-agent-safety.md |    60 +
 ...ic-memory-systems-as-evolvable-programs.md |    63 +
 ...or-forward-looking-ai-research-judgment.md |    59 +
 ...irculation-for-long-horizon-llm-agents-.md |    64 +
 ...-agent-decisions-against-their-defaults.md |    63 +
 ...autonomous-agentic-design-for-photonics.md |    61 +
 ...-agent-for-generalizable-radiotherapy-t.md |    63 +
 ...mo-with-disagree-or-commit-deliberation.md |    63 +
 ...ts-learn-from-experience-via-latent-rag.md |    63 +
 ...wire-format-for-agent-memory-operations.md |    63 +
 ...ld-threats-to-safer-computer-use-agents.md |    63 +
 ...entric-optimization-of-lakehouse-agents.md |    63 +
 ...in-long-horizon-organizational-dynamics.md |    64 +
 ...-task-decomposition-for-beyond-5g-auto-.md |    62 +
 ...lti-agent-orchestration-with-external-k.md |    65 +
 ...liable-tool-augmented-large-language-mo.md |    65 +
 ...xploration-learning-via-novelty-signals.md |    62 +
 ...-rag-for-technical-literature-reasoning.md |    67 +
 ...mplex-task-dependencies-and-human-align.md |    60 +
 ...cal-autoresearch-with-agentic-ai-models.md |    61 +
 ...-evaluation-for-generative-enterprise-r.md |    63 +
 ...thesis-for-evaluating-autonomous-agents.md |    64 +
 ...odels-and-agent-policies-for-llm-agents.md |    64 +
 ...gic-deception-in-agents-via-plan-action.md |    63 +
 ...odeling-co-training-for-language-agents.md |    61 +
 ...t-benchmark-grounded-in-korean-contexts.md |    59 +
 ...f-continual-learning-in-language-agents.md |    66 +
 ...time-series-forecasting-with-llm-agents.md |    62 +
 ...ystem-for-patient-trajectory-modeling-i.md |    64 +
 ...r-evaluating-abstention-competence-in-a.md |    62 +
 ...raining-harnesses-for-autonomous-agenti.md |    61 +
 ...ion-in-llm-agents-with-information-gain.md |    61 +
 ...linical-decision-making-with-large-lang.md |    60 +
 ...self-supervised-context-memory-training.md |    63 +
 ...ts-with-answer-conditioned-information-.md |    61 +
 ...poral-memory-system-for-embodied-agents.md |    63 +
 ...ocialized-evolution-in-agent-ecosystems.md |    62 +
 ...-an-agentic-benchmark-for-novel-api-acq.md |    61 +
 ...ility-controlled-self-evolving-llm-agen.md |    64 +
 ...reinforcement-learning-for-agent-safety.md |    62 +
 ...nitive-memory-for-conversational-agents.md |    63 +
 ...of-intervention-timing-why-affect-based.md |    63 +
 ...entic-memory-systems-diagnostics-and-a-.md |    62 +
 ...y-segment-trees-for-long-horizon-agents.md |    62 +
 ...-inspired-agentic-system-for-industrial.md |    63 +
 ...h-priority-aware-runtime-transformation.md |    59 +
 ...-for-person-understanding-in-llm-agents.md |    63 +
 ...mework-for-planning-capabilities-in-llm.md |    61 +
 ...idence-tracing-and-execution-provenance.md |    65 +
 ...h-agents-measuring-performance-inflatio.md |    61 +
 ...for-verifiable-reinforcement-learning-o.md |    63 +
 ...-early-failure-alerting-in-dialogs-and-.md |    65 +
 ...l-intelligence-for-clinical-literature-.md |    63 +
 ...nchmark-for-evaluating-llms-in-patient-.md |    62 +
 ...development-kits-via-llm-as-a-developer.md |    61 +
 ...for-off-policy-evaluation-of-llm-agents.md |    62 +
 ...-in-large-language-model-agents-under-w.md |    61 +
 ...tive-study-on-structured-and-multi-hop-.md |    61 +
 ...ime-adaptive-memory-for-language-agents.md |    63 +
 ...ent-communication-in-llm-based-multi-ag.md |    63 +
 ...emediation-a-guardrail-feedback-driven-.md |    63 +
 ...hy-memory-search-for-personal-ai-agents.md |    60 +
 ...ecution-state-management-for-long-horiz.md |    62 +
 ...set-of-action-level-mental-model-annota.md |    64 +
 ...-investigating-collaborative-competence.md |    64 +
 ...implications-of-stateful-long-horizon-w.md |    63 +
 ...hmark-everything-everywhere-all-at-once.md |    61 +
 ...tomated-machine-learning-algorithm-disc.md |    64 +
 ...for-llm-based-quantum-software-debuggin.md |    64 +
 ...-preventing-cheating-via-capped-evaluat.md |    63 +
 ...y-for-realistic-user-agent-interactions.md |    62 +
 ...d-to-end-autonomous-scientific-research.md |    62 +
 ...ary-propagation-failures-in-vision-lang.md |    64 +
 ...mplete-ultra-long-horizon-software-work.md |    65 +
 ...ry-adaptive-memory-for-cross-llm-agents.md |    62 +
 ...-quasiparticle-and-excitonic-properties.md |    66 +
 ...the-cold-start-safety-gap-in-llm-agents.md |    60 +
 ...ntropy-principle-and-the-inevitable-dis.md |    59 +
 ...afety-gating-civility-steering-and-affe.md |    65 +
 ...-integrating-cognition-culture-values-a.md |    64 +
 ...i-agent-coordination-in-language-agents.md |    66 +
 ...on-and-safety-evaluation-framework-for-.md |    62 +
 ...iteria-rubrics-across-the-evolving-llm-.md |    62 +
 ...on-native-clearing-for-agentic-commerce.md |    62 +
 ...rks-with-adversarial-hacker-fixer-loops.md |    62 +
 ...imization-via-an-fea-ai-hybrid-approach.md |    64 +
 ...ibution-for-silent-failures-in-llm-agen.md |    61 +
 ...with-memory-augmented-social-simulation.md |    63 +
 ...owledge-into-reusable-skills-for-agents.md |    64 +
 ...ous-web-navigation-grounded-in-human-br.md |    64 +
 ...hmark-for-computer-use-agents-with-hybr.md |    61 +
 ...-real-world-cloud-environments-via-dist.md |    61 +
 ...tive-memory-system-for-self-evolving-ll.md |    63 +
 ...claw-clawing-back-control-of-llm-agents.md |    62 +
 ...age-for-structured-3d-indoor-scene-gene.md |    62 +
 ...for-personally-intelligent-phone-agents.md |    62 +
 ...-with-lightweight-coding-agent-adapters.md |    65 +
 ...characterizing-false-success-in-llm-age.md |    60 +
 ...ised-policy-optimization-for-llm-agents.md |    61 +
 ...ext-engineering-for-long-horizon-tool-u.md |    64 +
 ...lipping-encoding-subspace-in-llm-agents.md |    63 +
 ...agent-for-spreadsheet-manipulation-and-.md |    63 +
 ...vidence-grounded-muon-collider-analysis.md |    66 +
 ...nt-benchmarking-for-realistic-scenarios.md |    59 +
 ...able-and-efficient-generalist-web-agent.md |    63 +
 ...on-folding-for-long-horizon-llm-agent-l.md |    62 +
 ...e-memory-for-long-horizon-llm-reasoning.md |    62 +
 ...ge-navigation-as-a-tool-calling-harness.md |    62 +
 ...afe-memory-retention-via-constrained-op.md |    63 +
 ...ocuments-for-long-term-llm-agent-memory.md |    62 +
 ...i-agent-llm-training-with-cross-agent-l.md |    62 +
 ...ng-of-multimodal-memories-in-web-agents.md |    64 +
 ...urfaces-attacks-defenses-and-evaluation.md |    65 +
 ...on-demand-hypergraph-memory-for-long-do.md |    64 +
 ...g-to-adapt-to-unfamiliar-programming-la.md |    61 +
 ...ion-of-computer-use-agentic-tasks-in-re.md |    62 +
 ...grounded-critic-for-computer-use-agents.md |    64 +
 ...simulation-toolkit-for-agent-evaluation.md |    61 +
 ...-framework-for-efficient-agentic-reinfo.md |    64 +
 ...data-into-verifiable-multimodal-stories.md |    66 +
 ...cation-for-hierarchical-language-agents.md |    62 +
 ...an-building-interaction-via-programmati.md |    64 +
 ...-memory-navigation-for-efficient-agents.md |    66 +
 ...ion-firewall-for-unattended-long-horizo.md |    62 +
 ...ta-a-benchmark-for-clinical-tool-agents.md |    63 +
 ...-building-custom-ai-agents-from-substra.md |    62 +
 ...ls-with-multimodal-contextual-reasoning.md |    63 +
 ...untime-governance-of-production-ai-agen.md |    67 +
 ...dgets-for-privacy-preserving-llm-agents.md |    61 +
 ...-openclaw-style-agent-harnesses-on-codi.md |    61 +
 ...-agentic-procedural-policy-optimization.md |    61 +
 ...a-cognition-layer-for-autonomous-agents.md |    61 +
 ...ger-leakage-in-vision-language-agentic-.md |    62 +
 ...ided-credit-distillation-for-long-horiz.md |    64 +
 ...or-human-mobility-trajectory-generation.md |    65 +
 ...table-tool-workflows-for-compact-agents.md |    62 +
 ...mory-poisoning-in-persistent-llm-agent-.md |    63 +
 ...rld-models-for-self-evolving-llm-agents.md |    64 +
 ...ch-agents-beyond-the-human-difficulty-c.md |    61 +
 ...rounded-multi-factor-value-model-for-ag.md |    64 +
 ...on-over-heterogeneous-earth-system-data.md |    64 +
 ...-compression-for-long-term-agent-memory.md |    64 +
 ...-multimodal-llms-task-benchmark-and-app.md |    62 +
 ...ogy-aware-skill-self-evolution-for-llm-.md |    60 +
 ...ompt-injection-benchmarking-for-real-wo.md |    65 +
 ...on-of-ai-agents-on-epigenomics-analysis.md |    59 +
 ...or-openness-standardization-and-reprodu.md |    63 +
 ...26-2606-13643-recursive-agent-harnesses.md |    63 +
 ...se-tool-calls-for-tool-augmented-agents.md |    61 +
 ...y-under-e-commerce-deceptive-interfaces.md |    61 +
 ...s-for-qa-agents-over-massive-data-lakes.md |    64 +
 ...safety-against-decomposition-attacks-wi.md |    64 +
 ...ough-a-failure-mode-study-of-gui-agents.md |    63 +
 ...g-and-agent-memory-you-can-replay-diff-.md |    64 +
 ...adigm-shift-toward-persistent-autonomou.md |    62 +
 ...e-attacks-on-llm-based-agent-guardrails.md |    63 +
 ...ent-memory-for-future-oriented-assistan.md |    60 +
 ...m-executable-planning-with-a-world-mode.md |    64 +
 ...system-for-reliable-multi-agent-workflo.md |    65 +
 ...lay-debugging-of-multi-agent-llm-traces.md |    61 +
 ...s-worth-their-tokens-a-budget-constrain.md |    62 +
 ...hmark-for-safety-in-computer-use-agents.md |    61 +
 ...ent-and-instant-agentic-intelligence-at.md |    64 +
 ...ual-social-intelligence-in-multimodal-s.md |    61 +
 ...n-security-risks-in-agent-skill-ecosyst.md |    62 +
 ...urrency-control-for-multi-agent-systems.md |    63 +
 ...ed-equation-chains-a-controlled-generat.md |    62 +
 ...e-language-model-agents-via-memory-base.md |    62 +
 ...complementary-collaboration-in-minecraf.md |    62 +
 ...twork-management-with-proof-of-concept-.md |    64 +
 ...soning-and-coherent-decision-making-of-.md |    62 +
 ...e-agentic-programming-for-agent-harness.md |    60 +
 ...-an-architectural-study-of-agent-memory.md |    62 +
 ...dence-for-agentic-multimodal-rag-in-lon.md |    64 +
 ...tem-for-therapeutic-reasoning-over-hist.md |    63 +
 ...kload-migration-via-in-context-learning.md |    61 +
 ...ents-with-pareto-ranking-policy-optimiz.md |    62 +
 ...rsonalized-agent-for-the-physical-world.md |    63 +
 ...playbooks-for-agentic-security-auditing.md |    64 +
 ...ontextual-grounding-for-language-agents.md |    62 +
 ...py-controllable-narrative-script-genera.md |    60 +
 ...entic-ai-generated-julia-code-on-superc.md |    61 +
 ...evidence-from-agentic-automata-learning.md |    62 +
 ...ble-active-tool-discovery-in-llm-agents.md |    62 +
 ...-agents-in-heterogeneous-multi-agent-ec.md |    62 +
 ...-language-models-for-sms-to-webpage-fra.md |    61 +
 ...sonally-intelligent-computer-use-agents.md |    62 +
 ...earch-for-agentic-large-language-models.md |    63 +
 ...gents-for-scientific-instrument-control.md |    62 +
 ...al-minimal-tool-filtering-in-llm-agents.md |    60 +
 ...-software-tool-discovery-case-for-log-a.md |    62 +
 ...ridge-scalable-evaluation-for-ai-agents.md |    62 +
 ...analysis-articles-from-nature-portfolio.md |    63 +
 ...es-computes-and-self-reviews-climate-sc.md |    64 +
 ...ol-using-llm-agents-in-realistic-scenar.md |    63 +
 ...nts-for-operational-disaster-geo-intell.md |    65 +
 ...s-architecture-key-mechanisms-and-proto.md |    63 +
 ...pomdp-based-framework-for-belief-state-.md |    63 +
 ...nergy-based-retrieval-augmented-generat.md |    67 +
 ...-aware-map-agents-through-behavior-grou.md |    59 +
 ...esource-reallocation-with-multi-role-ag.md |    65 +
 ...-transactions-for-tool-using-llm-agents.md |    63 +
 ...amics-in-agentic-reinforcement-learning.md |    62 +
 ...efficient-test-time-computation-scaling.md |    63 +
 ...y-verification-for-mcp-based-llm-agents.md |    64 +
 ...m-agents-decompose-retrieve-and-compose.md |    61 +
 ...-premature-diagnostic-handoff-and-silen.md |    61 +
 ...lfight-an-agentic-benchmark-for-implici.md |    61 +
 ...ents-for-energy-efficient-6g-autonomous.md |    63 +
 ...vidence-and-sandbox-harm-in-tool-using-.md |    63 +
 ...ersal-harness-for-embodied-manipulation.md |    64 +
 ...uided-distillation-for-long-term-memory.md |    63 +
 ...agentic-ai-under-retrieval-and-tool-use.md |    62 +
 ...ment-of-multi-agent-systems-for-enterpr.md |    62 +
 ...y-detection-via-specification-inference.md |    63 +
 ...ent-trajectories-for-interactive-verifi.md |    61 +
 ...c-ai-in-power-system-steady-state-studi.md |    64 +
 ...in-multi-principal-shared-memory-agents.md |    63 +
 ...gic-reasoning-by-vision-language-models.md |    64 +
 ...-via-suspicious-api-knowledge-and-agent.md |    62 +
 ...e-compliance-verification-for-ai-agents.md |    59 +
 ...-on-small-molecule-preclinical-pharmaco.md |    64 +
 ...entered-runtime-state-for-agent-systems.md |    64 +
 ...untime-governance-of-agentic-ai-systems.md |    63 +
 ...oding-agents-over-100-interaction-turns.md |    61 +
 ...lidity-for-the-evaluation-of-llm-agents.md |    61 +
 ...ging-operations-research-tasks-end-to-e.md |    60 +
 ...tration-for-ai-assisted-legal-discovery.md |    65 +
 ...s-workflows-for-lung-pathology-extracti.md |    62 +
 ...cal-capabilities-and-risks-of-ai-agents.md |    64 +
 ...obile-gui-agent-with-proactive-context-.md |    61 +
 ...r-mobile-gui-agents-with-hierarchical-f.md |    60 +
 ...licy-self-improvement-in-the-real-world.md |    62 +
 ...ng-over-privileged-tool-selection-in-ll.md |    62 +
 ...or-model-grounded-economic-analysis-wit.md |    63 +
 ...on-as-a-pluggable-engine-for-llm-agents.md |    61 +
 ...b-issue-resolution-via-multi-agent-llms.md |    64 +
 ...ntic-ai-in-power-system-dynamic-studies.md |    66 +
 ...model-guided-automated-attacks-on-agent.md |    62 +
 ...lures-in-vision-language-agents-via-tra.md |    62 +
 ...robabilistic-verification-for-ai-agents.md |    62 +
 ...f-repository-guidance-for-coding-agents.md |    64 +
 ...cits-reasoning-for-spatial-intelligence.md |    62 +
 ...kflow-design-for-global-agentic-collabo.md |    61 +
 ...the-cost-accuracy-frontier-of-llm-agent.md |    65 +
 ...or-vulnerability-detection-in-web-agent.md |    65 +
 ...ng-environments-for-computer-use-agents.md |    63 +
 ...-agents-against-tool-description-poison.md |    61 +
 ...evaluation-of-ai-agents-in-electric-pow.md |    61 +
 ...agent-memory-from-a-few-kilobytes-of-le.md |    62 +
 ...astructure-for-future-event-forecasting.md |    63 +
 ...akes-reasoning-evaluation-and-interpret.md |    62 +
 ...ting-system-architecture-for-autonomous.md |    61 +
 ...2606-21228-sakana-fugu-technical-report.md |    61 +
 ...eduling-for-low-latency-agentic-systems.md |    62 +
 ...e-feedback-breaks-tool-using-llm-agents.md |    61 +
 ...-systems-with-primitive-representations.md |    62 +
 ...on-for-multi-hop-qa-with-a-local-7b-mod.md |    62 +
 ...a-building-blocks-towards-design-time-v.md |    62 +
 ...ta-evaluation-dataset-for-agentic-tasks.md |    62 +
 ...r-long-context-retrieval-and-agentic-me.md |    62 +
 ...extual-privacy-alignment-for-llm-agents.md |    63 +
 ...-the-compression-boundary-of-llm-agents.md |    62 +
 ...proach-to-end-to-end-pddl-planning-with.md |    62 +
 ...-architectural-design-space-exploration.md |    62 +
 ...l-attacks-on-non-prefix-kv-cache-in-rag.md |    62 +
 ...ill-of-materials-for-agentic-ai-systems.md |    66 +
 ...ixed-language-mobile-crashes-at-industr.md |    63 +
 ...-world-model-for-long-term-agent-memory.md |    64 +
 ...work-for-repository-level-code-generati.md |    63 +
 ...-of-agentic-program-repair-trajectories.md |    64 +
 ...-research-contributions-through-structu.md |    62 +
 ...ety-vulnerability-detection-for-reposit.md |    64 +
 ...riven-skill-optimization-for-llm-agents.md |    64 +
 ...ning-of-llm-tool-use-agents-in-large-sc.md |    62 +
 ...al-codebase-index-inside-a-coding-agent.md |    61 +
 ...d-human-oversight-for-agentic-code-gene.md |    62 +
 ...tic-ai-needs-deterministic-environments.md |    62 +
 ...g-ai-agents-on-real-world-macos-desktop.md |    65 +
 ...s-research-and-human-in-the-loop-refine.md |    61 +
 ...-rag-for-automated-vulnerability-repair.md |    64 +
 ...ia-mechanistic-subspaces-for-multi-turn.md |    62 +
 ...ss-discipline-in-autonomous-ai-coding-a.md |    63 +
 ...r-llm-as-judge-in-stateful-agent-evalua.md |    62 +
 ...n-of-llm-agent-dependency-and-execution.md |    60 +
 ...nstatement-for-long-term-agentic-memory.md |    62 +
 ...evaluation-protocol-for-hidden-state-pr.md |    60 +
 ...fied-search-for-long-horizon-gui-agents.md |    63 +
 ...nagement-is-load-bearing-for-llm-agents.md |    61 +
 ...ial-analysts-beyond-finance-agent-v2-wi.md |    62 +
 ...-in-security-of-vibe-coded-applications.md |    64 +
 ...tion-of-evaluator-bias-via-agent-memory.md |    63 +
 ...hancing-implicit-logical-memory-retriev.md |    60 +
 ...ork-for-video-understanding-and-editing.md |    63 +
 ...al-prompts-in-populations-of-llm-agents.md |    61 +
 ...-agent-framework-with-3d-spatial-memory.md |    66 +
 ...ization-improve-multi-agent-llm-systems.md |    61 +
 ...ry-layer-for-continuity-handoff-and-cur.md |    60 +
 ...cieties-from-collective-affect-to-autho.md |    62 +
 ...amic-red-teaming-for-agentic-ai-systems.md |    61 +
 ...2026-2606-23991-critique-of-agent-model.md |    67 +
 ...d-multi-agent-drl-framework-for-low-alt.md |    63 +
 ...g-agent-for-spatial-proteomics-analysis.md |    63 +
 ...st-poisoning-non-malleable-origin-bound.md |    60 +
 ...-poisoning-effects-on-ai-security-agent.md |    62 +
 ...tion-of-policy-driven-physical-layer-sy.md |    61 +
 ...mory-sustains-mixture-of-agents-scaling.md |    61 +
 ...4453-bayesian-control-for-coding-agents.md |    62 +
 ...r-use-agents-with-autonomous-evaluation.md |    61 +
 ...arison-as-process-reward-for-gui-agents.md |    62 +
 ...ared-memory-for-multi-agent-llm-systems.md |    62 +
 ...n-only-and-skill-mediated-computer-use-.md |    64 +
 ...t-memory-via-hidden-user-state-recovery.md |    61 +
 ...anguage-world-models-for-general-agents.md |    63 +
 ...mantic-rewriting-achieving-confidential.md |    63 +
 ...lt-attribution-via-active-investigation.md |    63 +
 ...ognition-for-zero-shot-3d-understanding.md |    62 +
 ...ments-an-llm-based-multi-agent-approach.md |    61 +
 ...earning-in-supply-chain-via-contextual-.md |    61 +
 ...ready-for-an-agent-native-memory-system.md |    64 +
 ...riant-prioritization-and-diagnosis-of-g.md |    62 +
 ...tic-localization-for-code-repair-agents.md |    63 +
 ...luating-an-agentic-data-analysis-system.md |    62 +
 ...s-agent-data-recipes-for-agentic-models.md |    59 +
 ...-agentic-ai-from-foundations-to-systems.md |    68 +
 ...lures-in-agentic-persuasion-via-taxonom.md |    63 +
 ...tinual-learning-via-budget-curated-memo.md |    61 +
 ...-benchmarking-agentic-ai-skills-in-buil.md |    60 +
 ...olidation-for-llm-agents-with-long-term.md |    63 +
 ...-policy-enforcement-for-agent-harnesses.md |    62 +
 ...essment-for-cost-efficient-multi-agent-.md |    62 +
 ...ion-progress-pitfalls-and-paths-forward.md |    63 +
 ...ion-with-a-visuo-spatio-temporal-memory.md |    65 +
 ...le-multi-agent-framework-for-safe-and-c.md |    62 +
 ...lm-architecture-for-stealth-assessment-.md |    63 +
 ...w-different-memory-roles-shape-conversa.md |    63 +
 ...multi-agent-framework-for-autonomous-br.md |    62 +
 ...n-agentic-ai-framework-for-e-scooter-mo.md |    63 +
 ...ve-multi-agent-scaffolding-for-efficien.md |    64 +
 ...lation-as-a-hidden-cost-of-low-bit-reas.md |    63 +
 ...amework-for-cross-library-test-migratio.md |    64 +
 ...its-evaluating-multi-agent-systems-for-.md |    65 +
 ...-medical-error-detection-and-correction.md |    62 +
 ...h-agentic-solutions-with-context-optimi.md |    63 +
 ...d-exploration-of-user-sensitive-screens.md |    60 +
 ...se-agents-a-benchmark-across-vision-lan.md |    64 +
 ...-using-agents-under-tool-environment-un.md |    61 +
 ...is-multi-environment-evaluation-of-fron.md |    61 +
 ...me-ai-alignment-for-ai-agents-and-other.md |    64 +
 ...re-an-llm-powered-pipeline-for-comparat.md |    64 +
 ...l-health-medication-information-seeking.md |    60 +
 ...t-contracts-against-real-world-on-chain.md |    61 +
 ...-silver-bullet-for-coding-agent-rewards.md |    63 +
 ...rm-on-real-world-energy-analytics-tasks.md |    63 +
 ...ence-in-prompt-composed-agentic-systems.md |    61 +
 ...substrate-for-privacy-memory-and-tool-u.md |    60 +
 ...ing-tools-as-expert-surrogates-for-llm-.md |    63 +
 ...es-against-prompt-injection-in-llm-agen.md |    64 +
 ...minating-stale-fact-errors-for-ai-agent.md |    65 +
 ...ioral-specifications-in-ai-agent-skills.md |    60 +
 ...n-the-loop-agentic-system-for-scientifi.md |    67 +
 ...centric-survey-of-privacy-in-llm-agents.md |    64 +
 ...-agent-instructions-into-policy-as-code.md |    62 +
 ...orkflow-for-agent-mediated-knowledge-co.md |    67 +
 ...d-agent-framework-for-kernel-generation.md |    62 +
 ...guided-mcts-red-teaming-for-agentic-rag.md |    65 +
 ...parametric-consolidation-for-long-runni.md |    61 +
 ...socio-economic-systems-powered-by-llm-a.md |    60 +
 ...g-task-insensitivity-in-language-agents.md |    61 +
 ...tic-control-plane-for-llm-coding-agents.md |    64 +
 ...ng-system-administration-with-ai-agents.md |    61 +
 ...-stopping-for-iterative-llm-agent-loops.md |    63 +
 ...me-labels-to-causal-process-supervision.md |    61 +
 ...or-architecture-evolution-in-industrial.md |    64 +
 ...rience-exploration-and-hindsight-experi.md |    65 +
 ...led-agentic-ai-driven-hardware-software.md |    61 +
 ...nts-in-open-ended-positive-sum-bargaini.md |    64 +
 ...it-software-world-models-in-coding-llms.md |    63 +
 ...esearch-with-parallel-llm-coding-agents.md |    61 +
 ...ing-the-memory-update-gap-in-llm-agents.md |    64 +
 ...c-training-paradigm-for-world-model-pla.md |    62 +
 ...tion-topologies-for-token-efficient-llm.md |    61 +
 ...dal-agents-visual-memory-with-incidenta.md |    65 +
 ...anguage-model-for-content-and-ai-safety.md |    62 +
 ...d-planning-for-reliable-language-agents.md |    64 +
 ...towards-embodied-collective-intelligenc.md |    62 +
 ...ti-agent-llm-honeypot-for-ssh-deception.md |    61 +
 ...g-llm-agents-for-fault-tolerant-control.md |    66 +
 ...-bound-privacy-in-tool-using-llm-agents.md |    62 +
 ...-modeling-embodied-multi-agent-behavior.md |    66 +
 ...ions-for-optimizing-multi-agent-systems.md |    60 +
 ...m-architecture-taxonomy-and-engineering.md |    65 +
 ...sign-as-repository-level-code-evolution.md |    60 +
 ...emory-system-for-long-context-reasoning.md |    64 +
 ...ith-institutional-guardrails-for-academ.md |    63 +
 ...-evolving-agents-via-held-out-selection.md |    61 +
 ...hesizable-c-conversion-and-verification.md |    63 +
 ...teganography-in-multi-agent-llm-systems.md |    65 +
 ...r-what-you-check-not-what-you-requested.md |    60 +
 ...nagement-for-long-horizon-coding-agents.md |    63 +
 ...free-program-verifier-for-coding-agents.md |    61 +
 ...ve-survey-of-self-security-and-empowere.md |    61 +
 ...idence-from-gaslighting-ai-agents-in-a-.md |    61 +
 ...l-energy-anomaly-detection-and-llm-driv.md |    65 +
 ...for-general-purpose-terminal-use-agents.md |    63 +
 ...ic-framework-for-holistic-athlete-profi.md |    64 +
 ...ains-from-trism-guided-agentic-workflow.md |    64 +
 ...nfused-deputy-failures-in-llm-agent-fra.md |    60 +
 ...asoning-over-a-biomedical-tool-universe.md |    61 +
 ...agents-know-when-to-stop-instead-of-act.md |    60 +
 ...-28739-agent-safety-is-action-alignment.md |    60 +
 ...owledge-topology-for-agent-first-memory.md |    63 +
 ...software-engineering-and-the-evolution-.md |    65 +
 ...uring-output-distribution-coupling-in-m.md |    62 +
 ...tic-framework-with-mcp-and-proof-repair.md |    64 +
 ...agent-framework-for-sar-data-generation.md |    63 +
 ...ion-a-wildchat-benchmark-and-cost-aware.md |    65 +
 ...egrity-in-multi-agent-llm-collaboration.md |    61 +
 ...ortation-engineering-practice-a-develop.md |    63 +
 ...lti-agent-ai-through-runtime-monitoring.md |    62 +
 ...-a-study-on-multiple-choice-question-an.md |    60 +
 ...ntic-workflows-a-study-on-n8n-ecosystem.md |    62 +
 ...-practitioner-systematization-of-autono.md |    62 +
 ...y-retention-for-long-horizon-llm-agents.md |    62 +
 ...llm-agents-in-microservice-failure-diag.md |    62 +
 ...-verifier-for-policy-adherence-in-llm-a.md |    62 +
 ...ority-voting-in-multi-agent-llm-debates.md |    61 +
 ...315-hierarchical-experimentalist-agents.md |    64 +
 ...unication-for-efficient-multi-agent-rea.md |    61 +
 ...-tasks-via-generalized-keyframe-extract.md |    62 +
 ...ocess-level-social-influence-evaluation.md |    61 +
 ...agents-on-long-horizon-real-world-tasks.md |    67 +
 ...iberation-with-local-reliability-bounds.md |    64 +
 ...r-audit-of-evaluator-driven-preference-.md |    60 +
 ...framework-for-automatic-microservice-de.md |    63 +
 ...-language-agents-with-turn-level-credit.md |    61 +
 ...tform-for-medical-deep-research-with-in.md |    63 +
 ...ers-are-llm-agents-a-case-study-on-molt.md |    59 +
 ...nsistent-benchmark-for-diagnostic-evalu.md |    64 +
 ...emory-for-agentic-embodied-manipulation.md |    64 +
 ...mory-system-for-long-term-conversations.md |    63 +
 ...mation-leaks-in-multimodal-agent-memory.md |    59 +
 ...gents-with-implicit-activation-steering.md |    64 +
 ...ation-retrieval-evaluation-in-mathemati.md |    62 +
 ...en-confounds-in-agent-memory-evaluation.md |    61 +
 ...-long-horizon-civrealm-strategy-plannin.md |    63 +
 ...ing-agents-in-interactive-user-sessions.md |    61 +
 ...mory-agents-via-dual-space-distillation.md |    61 +
 ...iciting-self-managed-context-via-a-prop.md |    60 +
 ...-design-of-embodied-agent-architectures.md |    66 +
 ...m-bot-unmasking-web-agents-with-multi-l.md |    63 +
 ...ol-evolution-for-vision-language-agents.md |    60 +
 ...redit-optimization-for-agentic-tool-use.md |    61 +
 ...ce-llms-to-mitigate-disinformation-thre.md |    60 +
 ...lora-variants-for-incremental-motion-un.md |    59 +
 ...trations-with-real-time-voice-question-.md |    66 +
 ...i-party-principal-loyalty-in-llm-agents.md |    60 +
 ...thout-individual-fidelity-in-llm-agents.md |    61 +
 ...dme-md-generation-evaluating-single-age.md |    63 +
 ...-framework-for-reliable-multi-agent-sys.md |    62 +
 ...-defense-in-multi-agent-systems-routing.md |    64 +
 ...-coding-agent-workloads-for-llm-serving.md |    62 +
 ...es-for-agent-memory-poisoning-detection.md |    63 +
 ...s-user-driven-long-horizon-coding-sessi.md |    63 +
 ...n-channels-for-securing-multi-agent-sys.md |    62 +
 ...aching-trillion-parameter-performance-w.md |    63 +
 ...ing-world-models-for-llm-agent-planning.md |    65 +
 ...er-for-accessibility-grounded-ai-agents.md |    63 +
 ...ent-security-through-a-computer-systems.md |    61 +
 ...ction-for-iterative-prompt-optimization.md |    62 +
 ...rom-advanced-regulatory-control-theory-.md |    62 +
 ...nt-systems-for-human-aligned-mental-hea.md |    63 +
 ...igating-multi-agent-deliberation-in-law.md |    61 +
 ...w-for-hls-compatibility-and-performance.md |    64 +
 ...mous-ai-agents-the-agentbound-framework.md |    61 +
 ...-collective-intelligence-in-human-agent.md |    65 +
 ...ificial-life-with-autonomous-llm-agents.md |    60 +
 ...nchmark-and-framework-for-multi-uav-col.md |    67 +
 ...low-for-drug-drug-interaction-predictio.md |    63 +
 ...or-autoformalizing-research-mathematics.md |    62 +
 ...estration-and-dynamic-workflows-in-lang.md |    61 +
 ...e-of-realistic-agentic-healthcare-envir.md |    63 +
 ...l-augmented-generation-with-self-reflec.md |    62 +
 ...-via-structured-autoregressive-modeling.md |    66 +
 ...ework-for-multi-layer-agent-red-teaming.md |    59 +
 ...-trajectories-synthesis-for-scientific-.md |    63 +
 ...-for-parametric-b-rep-assembly-modeling.md |    62 +
 ...-dynamic-model-discovery-in-power-syste.md |    60 +
 ...governance-for-intelligent-industrial-m.md |    64 +
 ...606-31410-xiaomi-gui-0-technical-report.md |    65 +
 ...anguage-agents-for-incremental-3d-scene.md |    59 +
 ...m-passive-records-to-active-task-drivin.md |    64 +
 ...ontrol-using-knowledge-grounded-llm-age.md |    63 +
 ...of-large-language-model-vulnerabilities.md |    69 +
 ...nt-adaptation-of-multilingual-tool-usin.md |    63 +
 ...ith-selective-turn-memory-in-agentic-rl.md |    62 +
 ...t-agent-search-system-for-geopolitical-.md |    61 +
 ...to-item-fulfillment-in-agentic-shopping.md |    64 +
 ...sics-based-household-digital-twins-for-.md |    62 +
 ...xecution-time-improvement-patches-in-ja.md |    60 +
 ...ientific-discovery-in-plant-phenotyping.md |    60 +
 ...ersation-assessing-the-capacity-of-llms.md |    62 +
 ...gaps-in-human-and-agentic-computer-use-.md |    63 +
 ...rative-skill-composition-for-llm-agents.md |    62 +
 ...ion-signals-for-long-horizon-llm-agents.md |    63 +
 ...eering-the-loops-that-replace-step-by-s.md |    63 +
 ...ssion-for-multi-agent-code-co-synthesis.md |    64 +
 ...itecture-drives-language-emergence-in-l.md |    63 +
 ...uav-enabled-wpt-systems-in-low-altitude.md |    60 +
 ...-evaluator-preference-dynamics-in-llm-a.md |    62 +
 ...fety-and-governance-for-single-and-mult.md |    63 +
 ...ing-eddops-with-evaluation-drivenregist.md |    60 +
 ...ing-latent-design-intents-for-agentic-s.md |    60 +
 ...g-reasoning-in-agentic-retrieval-augmen.md |    60 +
 ...r-tool-augmented-scientific-simulator-a.md |    61 +
 ...rk-for-provenance-based-backward-tracki.md |    63 +
 ...-llm-for-context-aware-agricultural-adv.md |    67 +
 ...tion-for-long-horizon-mobile-gui-agents.md |    63 +
 ...alysis-for-deep-learning-framework-bugs.md |    65 +
 ...age-models-an-overview-and-perspectives.md |    63 +
 ...ark-framework-for-world-modeling-agents.md |    61 +
 ...ing-context-for-long-horizon-llm-agents.md |    60 +
 ...skills-are-written-adapted-and-maintain.md |    60 +
 ...multi-agent-story-generation-for-long-f.md |    62 +
 ...enerate-quantum-applications-for-test-o.md |    63 +
 ...tic-rag-pipelines-a-proof-of-concept-st.md |    62 +
 ...gents-with-runtime-diagnosis-from-multi.md |    62 +
 ...collectives-as-interpretable-substrates.md |    59 +
 ...r-deterministic-self-expanding-reaction.md |    62 +
 ...benchmarking-sycophancy-in-agent-memory.md |    62 +
 ...nveiling-the-fragility-of-static-traini.md |    61 +
 ...rning-systems-enable-self-evolving-agen.md |    61 +
 ...hmarks-reliably-measuring-coding-agents.md |    61 +
 ...ts-on-whole-repository-compatibility-re.md |    61 +
 ...earch-for-federated-learning-algorithms.md |    60 +
 ...ng-teams-an-organizational-framework-fo.md |    64 +
 ...-involved-agentic-permission-management.md |    61 +
 ...1523-multi-head-recurrent-memory-agents.md |    62 +
 ...ith-ontology-error-prioritized-interact.md |    63 +
 ...uced-representational-coupling-in-multi.md |    60 +
 ...s-for-static-analysis-of-agent-programs.md |    63 +
 ...ng-infinite-agentic-loops-in-llm-agents.md |    62 +
 ...gent-deliberation-under-information-asy.md |    60 +
 ...istant-for-hardware-security-verificati.md |    65 +
 ...arnesses-for-image-generation-workflows.md |    64 +
 ...nt-system-for-dynamic-3d-scene-creation.md |    64 +
 ...le-world-model-correction-for-agent-rol.md |    62 +
 ...tem-in-hyper-scale-microservice-systems.md |    63 +
 ...isk-discovery-to-evidence-grounded-veri.md |    63 +
 ...ork-for-automated-topology-optimization.md |    62 +
 ...luating-and-enhancing-agentic-skill-use.md |    64 +
 ...code-memory-for-repository-level-progra.md |    62 +
 ...l-modal-structural-reasoning-for-agenti.md |    64 +
 ...mory-failures-in-long-term-agent-memory.md |    60 +
 ...proxy-for-agentic-capability-evaluation.md |    61 +
 ...collaboration-for-reliable-software-dev.md |    62 +
 ...or-ai-agent-decisions-in-autonomous-tel.md |    63 +
 ...e-for-equitable-mental-wellness-support.md |    62 +
 ...ory-testbed-for-long-horizon-llm-agents.md |    64 +
 ...on-boundary-violations-in-underspecifie.md |    61 +
 ...gent-simplification-for-spanish-easy-to.md |    62 +
 ...ing-of-fdm-parts-via-multi-agent-llm-re.md |    61 +
 ...nal-analysis-of-open-source-multi-agent.md |    59 +
 ...ng-social-structure-and-latent-objectiv.md |    63 +
 ...lidity-audit-of-tool-calling-evaluation.md |    60 +
 ...-promotion-from-correlated-agent-traces.md |    63 +
 ...or-measuring-enforcing-and-training-pro.md |    63 +
 ...gents-on-multi-bug-software-maintenance.md |    60 +
 ...er-optimizations-evaluating-llm-agents-.md |    60 +
 ...r-streaming-egocentric-memory-retrieval.md |    63 +
 ...-ai-for-scientific-software-development.md |    65 +
 ...sion-making-in-agent-based-urban-mobili.md |    64 +
 ...-coding-agents-for-open-ended-discovery.md |    60 +
 ...-environment-modeling-for-agentic-tasks.md |    61 +
 ...composition-attack-in-llm-coding-agents.md |    63 +
 ...ex-medical-calculations-with-llm-agents.md |    59 +
 ...agentic-workflow-via-symbolic-inference.md |    61 +
 ...servation-compression-for-coding-agents.md |    61 +
 ...rch-with-multi-tool-agentic-reasoning-v.md |    62 +
 ...-serving-layer-for-agentic-applications.md |    61 +
 ...ous-agents-in-scientific-quantum-progra.md |    60 +
 ...-ability-of-large-language-model-agents.md |    61 +
 ...configurations-of-personalizable-agents.md |    64 +
 ...-intelligence-and-cyber-investigations-.md |    65 +
 ...ve-security-framework-for-wireless-pbft.md |    63 +
 ...elopers-feedback-to-coderabbit-reviews-.md |    61 +
 ...ing-to-accelerate-agentic-llm-inference.md |    64 +
 ...dynamic-real-time-compositional-policie.md |    62 +
 ...entic-test-time-training-for-llm-agents.md |    61 +
 ...ce-evaluation-for-enterprise-agentic-ai.md |    66 +
 ...g-agents-on-real-c-runtime-environments.md |    64 +
 ...suring-ai-agents-as-computer-architects.md |    62 +
 ...agent-reasoning-for-smart-city-security.md |    62 +
 ...scaffolding-evolution-shapes-coding-age.md |    63 +
 ...607-03695-social-networks-of-llm-agents.md |    59 +
 ...g-via-pivotal-aware-self-feedback-retry.md |    62 +
 ...em-self-optimizing-memory-for-ai-agents.md |    63 +
 ...-prompt-injection-in-personal-ai-agents.md |    61 +
 ...-framework-for-radiology-report-generat.md |    61 +
 ...-ai-agents-with-natural-language-tools-.md |    63 +
 ...level-jailbreak-construction-in-ide-cod.md |    62 +
 ...work-for-discovering-turbulence-physics.md |    61 +
 ...agentic-reliability-in-function-calling.md |    61 +
 ...-aware-memory-plane-for-lifelong-agents.md |    61 +
 ...scene-reasoning-with-benchmarking-and-q.md |    63 +
 ...lation-via-zero-shot-workflow-reasoning.md |    66 +
 ...ode-generation-on-repository-scale-prob.md |    60 +
 ...d-challenges-toward-the-internet-of-age.md |    63 +
 ...0-biological-motifs-for-agentic-control.md |    62 +
 ...-causal-thinking-of-llm-agents-in-games.md |    64 +
 ...ng-state-belief-reliance-on-pixels-vers.md |    61 +
 ...an-auditable-agentic-memory-architectur.md |    61 +
 ...driven-agents-for-mathematical-research.md |    63 +
 ...-agentic-tool-use-for-neuron-kernel-gen.md |    61 +
 ...ndational-model-for-physical-agentic-ai.md |    64 +
 ...roadmap-for-agentic-recommender-systems.md |    66 +
 ...ugmented-cooperative-multi-agent-reinfo.md |    63 +
 ...ief-divergence-in-multi-step-llm-agents.md |    61 +
 ...llms-for-agentic-home-energy-management.md |    61 +
 ...mory-substrate-for-long-lived-ai-agents.md |    62 +
 ...ures-a-recovery-aware-evaluation-of-dia.md |    63 +
 ...gnosing-tool-use-failures-in-llm-agents.md |    62 +
 ...ency-structure-and-merge-conflict-rates.md |    60 +
 ...-optimization-for-multi-turn-llm-agents.md |    62 +
 ...icy-optimization-for-llm-agent-training.md |    60 +
 ...ber-threat-intelligence-knowledge-graph.md |    64 +
 ...ged-reasoning-attacks-on-llm-agent-memo.md |    62 +
 ...rge-language-model-agents-in-healthcare.md |    62 +
 ...acks-are-realistic-threats-to-ai-agents.md |    63 +
 ...ence-and-exploitation-in-repeated-games.md |    62 +
 ...el-agents-in-de-idealized-real-world-en.md |    61 +
 ...t-programming-horizons-in-coding-agents.md |    61 +
 ...ent-self-evolution-via-ability-transfer.md |    63 +
 ...nt-of-llm-agents-via-two-timescale-meta.md |    61 +
 ...integrity-in-multi-user-agentic-systems.md |    62 +
 ...ersonal-agents-under-evolving-intent-pl.md |    63 +
 ...context-compaction-for-long-horizon-age.md |    60 +
 ...-general-purpose-verification-framework.md |    65 +
 ...al-augmented-generation-system-for-evid.md |    66 +
 ...er-agentic-ai-system-for-bioinformatics.md |    64 +
 ...ses-with-offline-reinforcement-learning.md |    63 +
 ...ntity-bound-authorization-for-ai-agents.md |    61 +
 ...d-emotion-in-multi-agent-software-teams.md |    61 +
 ...rical-taxonomy-of-mutation-patterns-in-.md |    59 +
 ...erizing-coding-agent-in-open-source-sof.md |    64 +
 ...s-extendedworking-memory-for-language-a.md |    62 +
 ...esearch-for-ai-coding-agents-isolation-.md |    62 +
 ...via-multi-stage-reasoning-with-llm-base.md |    63 +
 ...ion-environments-for-scalable-agentic-r.md |    61 +
 ...l-use-planning-and-reasoning-failures-i.md |    65 +
 ...avigation-learning-to-use-memory-as-a-s.md |    63 +
 ...simulator-for-cryogenic-fault-diagnosis.md |    61 +
 ...r-engine-grounded-pcb-design-automation.md |    60 +
 ...t-to-execution-integrity-for-llm-agents.md |    60 +
 ...-in-economies-of-frontier-llm-agents-a-.md |    62 +
 ...ng-multilingual-long-horizon-llm-agents.md |    64 +
 ...plying-putnam-s-social-capital-theory-w.md |    64 +
 ...ental-learning-back-into-ai-assisted-so.md |    68 +
 ...benchmark-for-efficient-web-agent-evalu.md |    65 +
 ...tion-evolving-for-agentic-post-training.md |    60 +
 ...-a-study-on-joint-decision-making-under.md |    62 +
 ...agent-frameworks-with-an-agentic-oracle.md |    60 +
 ...imization-an-adaptive-tree-structured-r.md |    61 +
 ...ntime-intervention-for-reliable-llm-age.md |    62 +
 ...nts-for-automatic-software-verification.md |    64 +
 ...-benchmark-with-natively-authored-russi.md |    62 +
 ...ting-agentic-ai-s-autonomous-model-disc.md |    61 +
 ...-type-aware-llm-pipelines-for-bioasq-14.md |    63 +
 papers/paper-insights.md                      |    23 +-
 papers/source-registry.md                     |     2 +-
 tools/collection/README.md                    |    39 +
 tools/collection/collect_arxiv.py             |   425 +
 tools/collection/promote_arxiv_manifest.py    |   101 +
 980 files changed, 189464 insertions(+), 58 deletions(-)
 create mode 100644 collection-runs/2026-07-08-arxiv-expanded-sweep.md
 create mode 100644 data/arxiv-agent-candidates-2025-07-08-to-2026-07-08.json
 create mode 100644 data/arxiv-agent-papers-2025-07-08-to-2026-07-08.json
 create mode 100644 papers/corpus-summary-2026-07-08.md
 create mode 100644 papers/items/2025-2507-20395-mazeeval-a-benchmark-for-testing-sequential-decision-making-in-language-models.md
 create mode 100644 papers/items/2025-2507-20666-mimii-agent-leveraging-llms-with-function-calling-for-relative-evaluation-of-ano.md
 create mode 100644 papers/items/2025-2508-07575-mcptoolbench-a-large-scale-ai-agent-model-context-protocol-mcp-tool-use-benchmar.md
 create mode 100644 papers/items/2025-2508-11027-hell-or-high-water-evaluating-agentic-recovery-from-external-failures.md
 create mode 100644 papers/items/2025-2508-12685-toolace-mt-non-autoregressive-generation-for-agentic-multi-turn-interaction.md
 create mode 100644 papers/items/2025-2508-17094-powerchain-a-verifiable-agentic-ai-system-for-automating-distribution-grid-analy.md
 create mode 100644 papers/items/2025-2509-02444-appcopilot-toward-general-accurate-long-horizon-and-efficient-mobile-agent.md
 create mode 100644 papers/items/2025-2509-02494-gridmind-llms-powered-agents-for-power-system-analysis-and-operations.md
 create mode 100644 papers/items/2025-2509-08863-geojson-agents-a-multi-agent-llm-architecture-for-geospatial-analysis-function-c.md
 create mode 100644 papers/items/2025-2509-10769-agentarch-a-comprehensive-benchmark-to-evaluate-agent-architectures-in-enterpris.md
 create mode 100644 papers/items/2025-2509-13311-towards-general-agentic-intelligence-via-environment-scaling.md
 create mode 100644 papers/items/2025-2509-14477-ticket-bench-a-kickoff-for-multilingual-and-regionalized-agent-evaluation.md
 create mode 100644 papers/items/2025-2509-20998-core-full-path-evaluation-of-llm-agents-beyond-final-state.md
 create mode 100644 papers/items/2025-2509-26553-towards-reliable-benchmarking-a-contamination-free-controllable-evaluation-frame.md
 create mode 100644 papers/items/2025-2510-03847-small-language-models-for-agentic-systems-a-survey-of-architectures-capabilities.md
 create mode 100644 papers/items/2025-2510-04206-agentrl-scaling-agentic-reinforcement-learning-with-a-multi-turn-multi-task-fram.md
 create mode 100644 papers/items/2025-2510-14548-llm-agents-beyond-utility-an-open-ended-perspective.md
 create mode 100644 papers/items/2025-2510-18586-tokencake-a-kv-cache-centric-serving-framework-for-llm-based-multi-agent-applica.md
 create mode 100644 papers/items/2025-2510-21524-eu-agent-bench-measuring-illegal-behavior-of-llm-agents-under-eu-law.md
 create mode 100644 papers/items/2025-2510-22768-seeing-is-believing-evaluating-vision-language-model-susceptibility-in-agent-to-.md
 create mode 100644 papers/items/2025-2510-24645-funreason-mt-technical-report-advanced-data-synthesis-solution-for-real-world-mu.md
 create mode 100644 papers/items/2025-2510-26167-toolrm-towards-agentic-tool-use-reward-modeling.md
 create mode 100644 papers/items/2025-2511-04847-test-time-adaptation-for-llm-agents-via-environment-interaction.md
 create mode 100644 papers/items/2025-2511-11169-refine-and-align-confidence-calibration-through-multi-agent-interaction-in-vqa.md
 create mode 100644 papers/items/2025-2511-15203-taxonomy-evaluation-and-exploitation-of-ipi-centric-llm-agent-defense-frameworks.md
 create mode 100644 papers/items/2025-2511-22138-tinyllm-evaluation-and-optimization-of-small-language-models-for-agentic-tasks-o.md
 create mode 100644 papers/items/2025-2512-02605-iact-a-self-organizing-recursive-model-for-general-ai-agents-a-technical-white-p.md
 create mode 100644 papers/items/2025-2512-11682-medai-evaluating-txagent-s-therapeutic-agentic-reasoning-in-the-neurips-cure-ben.md
 create mode 100644 papers/items/2025-2512-23611-close-the-loop-synthesizing-infinite-tool-use-data-via-multi-agent-role-playing.md
 create mode 100644 papers/items/2025-2512-23647-nested-browser-use-learning-for-agentic-information-seeking.md
 create mode 100644 papers/items/2025-2512-23747-state-of-the-art-small-language-coder-model-mify-coder.md
 create mode 100644 papers/items/2026-2601-00268-beyond-perfect-apis-a-comprehensive-evaluation-of-llm-agents-under-real-world-ap.md
 create mode 100644 papers/items/2026-2601-05467-stelp-secure-transpilation-and-execution-of-llm-generated-programs.md
 create mode 100644 papers/items/2026-2601-06007-don-t-break-the-cache-an-evaluation-of-prompt-caching-for-long-horizon-agentic-t.md
 create mode 100644 papers/items/2026-2601-06606-cedar-context-engineering-for-agentic-data-science.md
 create mode 100644 papers/items/2026-2601-12988-paperguide-making-small-language-model-paper-reading-agents-more-efficient.md
 create mode 100644 papers/items/2026-2601-14652-mas-orchestra-understanding-and-improving-multi-agent-reasoning-through-holistic.md
 create mode 100644 papers/items/2026-2602-03117-agentdyn-are-your-agent-security-defenses-deployable-in-real-world-dynamic-envir.md
 create mode 100644 papers/items/2026-2602-03224-tame-a-trustworthy-test-time-evolution-of-agent-memory-with-systematic-benchmark.md
 create mode 100644 papers/items/2026-2602-03786-aorchestra-automating-sub-agent-creation-for-agentic-orchestration.md
 create mode 100644 papers/items/2026-2602-05115-socialveil-probing-social-intelligence-of-language-agents-under-communication-ba.md
 create mode 100644 papers/items/2026-2602-05302-piearena-ranking-and-profiling-language-agents-in-realistic-negotiation-scenario.md
 create mode 100644 papers/items/2026-2602-05386-spider-sense-intrinsic-risk-sensing-for-efficient-agent-defense-with-hierarchica.md
 create mode 100644 papers/items/2026-2602-07391-naamse-framework-for-evolutionary-security-evaluation-of-agents.md
 create mode 100644 papers/items/2026-2602-07652-agent-fence-mapping-security-vulnerabilities-across-deep-research-agents.md
 create mode 100644 papers/items/2026-2602-07962-loca-bench-benchmarking-language-agents-under-controllable-and-extreme-context-g.md
 create mode 100644 papers/items/2026-2602-08082-spectral-guardrails-for-agents-in-the-wild-detecting-tool-use-hallucinations-via.md
 create mode 100644 papers/items/2026-2602-08412-from-assistant-to-double-agent-formalizing-and-benchmarking-attacks-on-openclaw-.md
 create mode 100644 papers/items/2026-2602-10133-agenttrace-a-structured-logging-framework-for-agent-system-observability.md
 create mode 100644 papers/items/2026-2602-11749-air-improving-agent-safety-through-incident-response.md
 create mode 100644 papers/items/2026-2602-13379-unsafer-in-many-turns-benchmarking-and-defending-multi-turn-safety-risks-in-tool.md
 create mode 100644 papers/items/2026-2602-13530-remem-reasoning-with-episodic-memory-in-language-agent.md
 create mode 100644 papers/items/2026-2602-13665-hyfunc-accelerating-llm-based-function-calls-for-agentic-ai-through-hybrid-model.md
 create mode 100644 papers/items/2026-2602-14234-redsearcher-a-scalable-and-cost-efficient-framework-for-long-horizon-search-agen.md
 create mode 100644 papers/items/2026-2602-14281-mcpshield-a-security-cognition-layer-for-adaptive-trust-calibration-in-model-con.md
 create mode 100644 papers/items/2026-2602-16931-narrow-fine-tuning-erodes-safety-alignment-in-vision-language-agents.md
 create mode 100644 papers/items/2026-2602-18456-beyond-single-channel-agentic-benchmarking.md
 create mode 100644 papers/items/2026-2602-19008-capable-but-unreliable-canonical-path-deviation-as-a-causal-mechanism-of-agent-f.md
 create mode 100644 papers/items/2026-2602-21127-are-you-sure-an-empirical-study-of-human-perception-vulnerability-in-llm-driven-.md
 create mode 100644 papers/items/2026-2602-23320-parammem-augmenting-language-agents-with-parametric-reflective-memory.md
 create mode 100644 papers/items/2026-2603-00131-thought-virus-viral-misalignment-via-subliminal-prompting-in-multi-agent-systems.md
 create mode 100644 papers/items/2026-2603-00623-tracesir-a-multi-agent-framework-for-structured-analysis-and-reporting-of-agenti.md
 create mode 100644 papers/items/2026-2603-00801-the-synthetic-web-adversarially-curated-mini-internets-for-diagnosing-epistemic-.md
 create mode 100644 papers/items/2026-2603-01438-enhancing-persona-following-at-decoding-time-via-dynamic-importance-estimation-f.md
 create mode 100644 papers/items/2026-2603-01712-ft-dojo-towards-autonomous-llm-fine-tuning-with-language-agents.md
 create mode 100644 papers/items/2026-2603-02711-a-natural-language-agentic-approach-to-study-affective-polarization.md
 create mode 100644 papers/items/2026-2603-03515-the-controllability-trap-a-governance-framework-for-military-ai-agents.md
 create mode 100644 papers/items/2026-2603-03680-mage-meta-reinforcement-learning-for-language-agents-toward-strategic-exploratio.md
 create mode 100644 papers/items/2026-2603-05553-eigendata-a-self-evolving-multi-agent-platform-for-function-calling-data-synthes.md
 create mode 100644 papers/items/2026-2603-05578-tool-genesis-a-task-driven-tool-creation-benchmark-for-self-evolving-language-ag.md
 create mode 100644 papers/items/2026-2603-07496-from-thinker-to-society-security-in-hierarchical-autonomy-evolution-of-ai-agents.md
 create mode 100644 papers/items/2026-2603-07557-agentraft-automated-detection-of-data-over-exposure-in-llm-agents.md
 create mode 100644 papers/items/2026-2603-07980-onemillion-bench-how-far-are-language-agents-from-human-experts.md
 create mode 100644 papers/items/2026-2603-08721-kernelcraft-benchmarking-for-agentic-close-to-metal-kernel-generation-on-emergin.md
 create mode 100644 papers/items/2026-2603-09002-security-considerations-for-multi-agent-systems.md
 create mode 100644 papers/items/2026-2603-10492-human-ai-co-reasoning-for-clinical-diagnosis-with-evidence-integrated-language-a.md
 create mode 100644 papers/items/2026-2603-11088-the-attack-and-defense-landscape-of-agentic-ai-a-comprehensive-survey.md
 create mode 100644 papers/items/2026-2603-11890-quare-quality-aware-requirements-analysis-through-multi-agent-dialectical-negoti.md
 create mode 100644 papers/items/2026-2603-15309-cctu-a-benchmark-for-tool-use-under-complex-constraints.md
 create mode 100644 papers/items/2026-2603-15666-compiled-memory-not-more-information-but-more-precise-instructions-for-language-.md
 create mode 100644 papers/items/2026-2603-16734-differential-harm-propensity-in-personalized-llm-agents-the-curious-case-of-ment.md
 create mode 100644 papers/items/2026-2603-17392-agentic-cognitive-profiling-realigning-automated-alzheimer-s-disease-detection-w.md
 create mode 100644 papers/items/2026-2603-18245-who-tests-the-testers-systematic-enumeration-and-coverage-audit-of-llm-agent-too.md
 create mode 100644 papers/items/2026-2603-19469-a-framework-for-formalizing-llm-agent-security.md
 create mode 100644 papers/items/2026-2603-19684-tsegagent-zero-shot-tooth-segmentation-via-geometry-aware-vision-language-agents.md
 create mode 100644 papers/items/2026-2603-21357-agenther-hindsight-experience-replay-for-llm-agent-trajectory-relabeling.md
 create mode 100644 papers/items/2026-2603-21564-toward-a-theory-of-hierarchical-memory-for-language-agents.md
 create mode 100644 papers/items/2026-2603-24257-memory-augmented-vision-language-agents-for-persistent-and-semantically-consiste.md
 create mode 100644 papers/items/2026-2603-25353-safeguard-asf-sr-agentic-humanoid-robot-system-for-autonomous-industrial-safety.md
 create mode 100644 papers/items/2026-2603-27148-safetydrift-predicting-when-ai-agents-cross-the-line-before-they-actually-do.md
 create mode 100644 papers/items/2026-2603-28166-evaluating-privilege-usage-of-agents-with-real-world-tools.md
 create mode 100644 papers/items/2026-2603-28428-synergy-a-next-generation-general-purpose-agent-for-open-agentic-web.md
 create mode 100644 papers/items/2026-2603-28900-robust-multi-agent-reinforcement-learning-for-small-uas-separation-assurance-und.md
 create mode 100644 papers/items/2026-2604-02022-atbench-a-diverse-and-realistic-agent-trajectory-benchmark-for-safety-evaluation.md
 create mode 100644 papers/items/2026-2604-02155-brief-is-better-non-monotonic-chain-of-thought-budget-effects-in-function-callin.md
 create mode 100644 papers/items/2026-2604-03098-co-evolution-of-policy-and-internal-reward-for-language-agents.md
 create mode 100644 papers/items/2026-2604-03242-draft-task-decoupled-latent-reasoning-for-agent-safety.md
 create mode 100644 papers/items/2026-2604-04131-profile-then-reason-bounded-semantic-complexity-for-tool-augmented-language-agen.md
 create mode 100644 papers/items/2026-2604-04426-shieldnet-network-level-guardrails-against-emerging-supply-chain-injections-in-a.md
 create mode 100644 papers/items/2026-2604-06762-arulecon-agentic-security-rule-conversion.md
 create mode 100644 papers/items/2026-2604-06972-differentiable-environment-trajectory-co-optimization-for-safe-multi-agent-navig.md
 create mode 100644 papers/items/2026-2604-08388-awakening-the-sleeping-agent-lean-specific-agentic-data-reactivates-general-tool.md
 create mode 100644 papers/items/2026-2604-10577-the-blind-spot-of-agent-safety-how-benign-user-instructions-expose-critical-vuln.md
 create mode 100644 papers/items/2026-2604-11557-unitoolcall-unifying-tool-use-representation-data-and-evaluation-for-llm-agents.md
 create mode 100644 papers/items/2026-2604-12986-parallax-why-ai-agents-that-think-must-never-act.md
 create mode 100644 papers/items/2026-2604-13298-can-agents-secure-hardware-evaluating-agentic-llm-driven-obfuscation-for-ip-prot.md
 create mode 100644 papers/items/2026-2604-13536-don-t-let-ai-agents-yolo-your-files-shifting-information-and-control-to-filesyst.md
 create mode 100644 papers/items/2026-2604-13954-hintbench-horizon-agent-intrinsic-non-attack-trajectory-benchmark.md
 create mode 100644 papers/items/2026-2604-14399-spacemind-a-modular-and-self-evolving-embodied-vision-language-agent-framework-f.md
 create mode 100644 papers/items/2026-2604-15415-harmfulskillbench-how-do-harmful-skills-weaponize-your-agents.md
 create mode 100644 papers/items/2026-2604-15579-don-t-make-models-guess-security-and-safety-symbolic-guardrails-for-domain-speci.md
 create mode 100644 papers/items/2026-2604-16706-evaluating-tool-using-language-agents-judge-reliability-propagation-cascades-and.md
 create mode 100644 papers/items/2026-2604-17562-safeagent-a-runtime-protection-architecture-for-agentic-systems.md
 create mode 100644 papers/items/2026-2604-18658-owner-harm-a-missing-threat-model-for-ai-agent-safety.md
 create mode 100644 papers/items/2026-2604-18718-towards-optimal-agentic-architectures-for-offensive-security-tasks.md
 create mode 100644 papers/items/2026-2604-18847-human-guided-harm-recovery-for-computer-use-agents.md
 create mode 100644 papers/items/2026-2604-19821-jtpro-a-joint-tool-prompt-reflective-optimization-framework-for-language-agents.md
 create mode 100644 papers/items/2026-2604-19844-if-you-re-waiting-for-a-sign-that-might-not-be-it-mitigating-trust-boundary-conf.md
 create mode 100644 papers/items/2026-2604-20994-breaking-mcp-with-function-hijacking-attacks-novel-threats-for-function-calling-.md
 create mode 100644 papers/items/2026-2604-21190-spatio-adaptive-test-time-orchestration-of-vision-language-agents-for-spatial-re.md
 create mode 100644 papers/items/2026-2604-22879-beyond-single-agent-alignment-preventing-context-fragmented-violations-in-multi-.md
 create mode 100644 papers/items/2026-2604-23210-discovering-agentic-safety-specifications-from-1-bit-danger-signals.md
 create mode 100644 papers/items/2026-2604-23374-ghost-in-the-agent-redefining-information-flow-tracking-for-llm-agents.md
 create mode 100644 papers/items/2026-2604-23459-architecture-matters-for-multi-agent-security.md
 create mode 100644 papers/items/2026-2604-24212-empowering-autonomous-debugging-agents-with-efficient-dynamic-analysis.md
 create mode 100644 papers/items/2026-2604-24826-a-comparative-evaluation-of-ai-agent-security-guardrails.md
 create mode 100644 papers/items/2026-2604-25135-fama-failure-aware-meta-agentic-framework-for-open-source-llms-in-interactive-to.md
 create mode 100644 papers/items/2026-2604-25318-cutscene-agent-an-llm-agent-framework-for-automated-3d-cutscene-generation.md
 create mode 100644 papers/items/2026-2604-25555-from-crud-to-autonomous-agents-formal-validation-and-zero-trust-security-for-sem.md
 create mode 100644 papers/items/2026-2604-26274-enforcing-benign-trajectories-a-behavioral-firewall-for-structured-workflow-ai-a.md
 create mode 100644 papers/items/2026-2604-26959-careguardai-context-aware-multi-agent-guardrails-for-clinical-safety-hallucinati.md
 create mode 100644 papers/items/2026-2604-27092-end-to-end-autonomous-scientific-discovery-on-a-real-optical-platform.md
 create mode 100644 papers/items/2026-2604-27464-security-attack-and-defense-strategies-for-autonomous-agent-frameworks-a-layered.md
 create mode 100644 papers/items/2026-2604-27699-bridging-values-and-behavior-a-hierarchical-framework-for-proactive-embodied-age.md
 create mode 100644 papers/items/2026-2604-27859-rethinking-agentic-reinforcement-learning-in-large-language-models.md
 create mode 100644 papers/items/2026-2604-28157-flashrt-towards-computationally-and-memory-efficient-red-teaming-for-prompt-inje.md
 create mode 100644 papers/items/2026-2605-00081-alignment-contracts-for-agentic-security-systems.md
 create mode 100644 papers/items/2026-2605-00741-self-adaptive-multi-agent-llm-based-security-pattern-selection-for-iot-systems.md
 create mode 100644 papers/items/2026-2605-00845-graph-query-generation-with-constraint-guided-large-language-agents.md
 create mode 100644 papers/items/2026-2605-01101-virtual-speech-therapist-a-clinician-in-the-loop-ai-speech-therapy-agent-for-per.md
 create mode 100644 papers/items/2026-2605-01644-toward-a-principled-framework-for-agent-safety-measurement.md
 create mode 100644 papers/items/2026-2605-02240-physicianbench-evaluating-llm-agents-in-real-world-ehr-environments.md
 create mode 100644 papers/items/2026-2605-03242-enhancing-agent-safety-judgment-controlled-benchmark-rewriting-and-analogical-re.md
 create mode 100644 papers/items/2026-2605-03312-memflow-intent-driven-memory-orchestration-for-small-language-model-agents.md
 create mode 100644 papers/items/2026-2605-03328-llm-adam-a-generalizable-llm-agent-framework-for-pre-print-anomaly-detection-in-.md
 create mode 100644 papers/items/2026-2605-03505-lats-rca-language-agent-tree-search-for-root-cause-analysis-in-microservices.md
 create mode 100644 papers/items/2026-2605-04107-tscg-deterministic-tool-schema-compilation-for-agentic-llm-deployments.md
 create mode 100644 papers/items/2026-2605-04808-decodingtrust-agent-platform-dtap-a-controllable-and-interactive-red-teaming-pla.md
 create mode 100644 papers/items/2026-2605-05242-beyond-semantic-similarity-rethinking-retrieval-for-agentic-search-via-direct-co.md
 create mode 100644 papers/items/2026-2605-05704-safeharbor-hierarchical-memory-augmented-guardrail-for-llm-agent-safety.md
 create mode 100644 papers/items/2026-2605-05716-more-is-not-always-better-cross-component-interference-in-llm-agent-scaffolding.md
 create mode 100644 papers/items/2026-2605-06078-milestone-guided-policy-learning-for-long-horizon-language-agents.md
 create mode 100644 papers/items/2026-2605-06713-agentic-ai-and-the-industrialization-of-cyber-offense-forecast-consequences-and-.md
 create mode 100644 papers/items/2026-2605-06716-from-storage-to-experience-a-survey-on-the-evolution-of-llm-agent-memory-mechani.md
 create mode 100644 papers/items/2026-2605-06737-a-self-healing-framework-for-reliable-llm-based-autonomous-agents.md
 create mode 100644 papers/items/2026-2605-06812-towards-security-auditable-llm-agents-a-unified-graph-representation.md
 create mode 100644 papers/items/2026-2605-06869-agentick-a-unified-benchmark-for-general-sequential-decision-making-agents.md
 create mode 100644 papers/items/2026-2605-06890-beyond-the-black-box-interpretability-of-agentic-ai-tool-use.md
 create mode 100644 papers/items/2026-2605-06957-learning-and-reusing-policy-decompositions-for-hierarchical-generalized-planning.md
 create mode 100644 papers/items/2026-2605-06992-why-does-agentic-safety-fail-to-generalize-across-tasks.md
 create mode 100644 papers/items/2026-2605-07112-switchcraft-ai-model-router-for-agentic-tool-calling.md
 create mode 100644 papers/items/2026-2605-07251-can-agents-price-a-reaction-evaluating-llms-on-chemical-cost-reasoning.md
 create mode 100644 papers/items/2026-2605-07830-cybiasbench-benchmarking-bias-in-llm-agents-for-cyber-attack-scenarios.md
 create mode 100644 papers/items/2026-2605-08374-memq-integrating-q-learning-into-self-evolving-memory-agents-over-provenance-dag.md
 create mode 100644 papers/items/2026-2605-08442-defense-effectiveness-across-architectural-layers-a-mechanistic-evaluation-of-pe.md
 create mode 100644 papers/items/2026-2605-08763-when-llms-team-up-a-coordinated-attack-framework-for-automated-cyber-intrusions.md
 create mode 100644 papers/items/2026-2605-08876-otora-a-unified-red-teaming-framework-for-reasoning-level-denial-of-service-in-l.md
 create mode 100644 papers/items/2026-2605-08964-trustworthy-ai-ensuring-reliability-and-accountability-from-models-to-agents.md
 create mode 100644 papers/items/2026-2605-09168-civex-causal-intervention-verification-for-language-agents.md
 create mode 100644 papers/items/2026-2605-09692-causal-state-binding-predicts-action-control-in-language-agents.md
 create mode 100644 papers/items/2026-2605-10365-agent-valuebench-a-comprehensive-benchmark-for-evaluating-agent-values.md
 create mode 100644 papers/items/2026-2605-10763-matra-modeling-the-attack-surface-of-agentic-ai-systems-openclaw-case-study.md
 create mode 100644 papers/items/2026-2605-10779-litmus-benchmarking-behavioral-jailbreaks-of-llm-agents-in-real-os-environments.md
 create mode 100644 papers/items/2026-2605-10870-remember-the-decision-not-the-description-a-rate-distortion-framework-for-agent-.md
 create mode 100644 papers/items/2026-2605-11039-the-granularity-mismatch-in-agent-security-argument-level-provenance-solves-enfo.md
 create mode 100644 papers/items/2026-2605-11225-pivot-bridging-planning-and-execution-in-llm-agents-via-trajectory-refinement.md
 create mode 100644 papers/items/2026-2605-11388-deep-reasoning-in-general-purpose-agents-via-structured-meta-cognition.md
 create mode 100644 papers/items/2026-2605-11534-prism-planning-and-reasoning-with-intent-in-simulated-embodied-environments.md
 create mode 100644 papers/items/2026-2605-11633-can-llm-agents-respond-to-disasters-benchmarking-heterogeneous-geospatial-reason.md
 create mode 100644 papers/items/2026-2605-11882-on-policy-self-evolution-via-failure-trajectories-for-agentic-safety-alignment.md
 create mode 100644 papers/items/2026-2605-11928-when-simulation-lies-a-sim-to-real-benchmark-and-domain-randomized-rl-recipe-for.md
 create mode 100644 papers/items/2026-2605-11946-counterfactual-trace-auditing-of-llm-agent-skills.md
 create mode 100644 papers/items/2026-2605-12015-skillsafetybench-evaluating-agent-safety-under-skill-facing-attack-surfaces.md
 create mode 100644 papers/items/2026-2605-12061-sage-a-self-evolving-agentic-graph-memory-engine-for-structure-aware-associative.md
 create mode 100644 papers/items/2026-2605-12260-prism-pareto-efficient-retrieval-over-intent-aware-structured-memory-for-long-ho.md
 create mode 100644 papers/items/2026-2605-13481-personalai-2-0-enhancing-knowledge-graph-traversal-retrieval-with-planning-mecha.md
 create mode 100644 papers/items/2026-2605-13542-realicu-do-llm-agents-understand-long-context-icu-data-a-benchmark-beyond-behavi.md
 create mode 100644 papers/items/2026-2605-13618-openaaas-an-open-agent-as-a-service-framework-for-distributed-materials-informat.md
 create mode 100644 papers/items/2026-2605-13716-skillops-managing-llm-agent-skill-libraries-as-self-maintaining-software-ecosyst.md
 create mode 100644 papers/items/2026-2605-14126-reinforcement-learning-for-tool-calling-agents-in-fast-healthcare-interoperabili.md
 create mode 100644 papers/items/2026-2605-14290-web-agents-should-adopt-the-plan-then-execute-paradigm.md
 create mode 100644 papers/items/2026-2605-14322-are-agents-ready-to-teach-a-multi-stage-benchmark-for-real-world-teaching-workfl.md
 create mode 100644 papers/items/2026-2605-14421-memlineage-lineage-guided-enforcement-for-llm-agent-memory.md
 create mode 100644 papers/items/2026-2605-14460-exploiting-llm-agent-supply-chains-via-payload-less-skills.md
 create mode 100644 papers/items/2026-2605-14498-groupmembench-benchmarking-llm-agent-memory-in-multi-party-conversations.md
 create mode 100644 papers/items/2026-2605-14527-lang2mlip-end-to-end-language-to-machine-learning-interatomic-potential-developm.md
 create mode 100644 papers/items/2026-2605-14892-beyond-individual-intelligence-surveying-collaboration-failure-attribution-and-s.md
 create mode 100644 papers/items/2026-2605-14906-memlens-benchmarking-multimodal-long-term-memory-in-large-vision-language-models.md
 create mode 100644 papers/items/2026-2605-14932-toward-securing-ai-agents-like-operating-systems.md
 create mode 100644 papers/items/2026-2605-15040-orchard-an-open-source-agentic-modeling-framework.md
 create mode 100644 papers/items/2026-2605-15128-memeye-a-visual-centric-evaluation-framework-for-multimodal-agent-memory.md
 create mode 100644 papers/items/2026-2605-15206-agentstop-terminating-local-ai-agents-early-to-save-energy-in-consumer-devices.md
 create mode 100644 papers/items/2026-2605-15625-colpackagent-agent-skill-guided-hard-particle-monte-carlo-workflows-for-colloida.md
 create mode 100644 papers/items/2026-2605-15701-h-mem-a-novel-memory-mechanism-for-evolving-and-retrieving-agent-memory-via-a-hy.md
 create mode 100644 papers/items/2026-2605-15710-smmbench-a-benchmark-for-source-distributed-multimodal-agent-memory.md
 create mode 100644 papers/items/2026-2605-15759-dimmem-dimensional-structuring-for-efficient-long-term-agent-memory.md
 create mode 100644 papers/items/2026-2605-16233-forge-self-evolving-agent-memory-with-no-weight-updates-via-population-broadcast.md
 create mode 100644 papers/items/2026-2605-16481-visual-agentic-memory-enabling-online-long-video-understanding-via-online-indexi.md
 create mode 100644 papers/items/2026-2605-16821-multi-paradigm-agent-interaction-in-practice-a-systematic-analysis-of-generator-.md
 create mode 100644 papers/items/2026-2605-17075-a-red-teaming-framework-for-evaluating-robustness-of-ai-enabled-security-orchest.md
 create mode 100644 papers/items/2026-2605-17348-taming-zombie-agents-a-markov-state-aware-framework-for-resilient-multi-agent-ev.md
 create mode 100644 papers/items/2026-2605-17453-trust-no-tool-evaluating-and-defending-llm-agents-under-untrusted-tool-feedback.md
 create mode 100644 papers/items/2026-2605-17625-episodic-semantic-memory-architecture-for-long-horizon-scientific-agents.md
 create mode 100644 papers/items/2026-2605-18284-commitdistill-a-lightweight-knowledge-centric-memory-layer-for-software-reposito.md
 create mode 100644 papers/items/2026-2605-18502-the-distance-based-formation-controller-design-for-multi-agent-systems-in-port-h.md
 create mode 100644 papers/items/2026-2605-18652-mementogui-learning-agentic-multimodal-memory-control-for-long-horizon-gui-agent.md
 create mode 100644 papers/items/2026-2605-18672-position-a-three-layer-probabilistic-assume-guarantee-architecture-is-structural.md
 create mode 100644 papers/items/2026-2605-18930-oep-poisoning-self-evolving-llm-agents-via-locally-correct-but-non-transferable-.md
 create mode 100644 papers/items/2026-2605-19604-formal-skill-programmable-runtime-skills-for-efficient-and-accurate-llm-agents.md
 create mode 100644 papers/items/2026-2605-19952-rethinking-how-to-remember-beyond-atomic-facts-in-lifelong-llm-agent-memory.md
 create mode 100644 papers/items/2026-2605-20306-wildroadbench-a-wild-aerial-road-damage-grounding-benchmark-for-vision-language-.md
 create mode 100644 papers/items/2026-2605-20315-mix-quant-quantized-prefilling-precise-decoding-for-agentic-llms.md
 create mode 100644 papers/items/2026-2605-20616-auto-dreamer-learning-offline-memory-consolidation-for-language-agents.md
 create mode 100644 papers/items/2026-2605-20833-memgym-a-long-horizon-memory-environment-for-llm-agents.md
 create mode 100644 papers/items/2026-2605-20874-governance-by-construction-for-generalist-agents.md
 create mode 100644 papers/items/2026-2605-21240-apex-autonomous-policy-exploration-for-self-evolving-llm-agents.md
 create mode 100644 papers/items/2026-2605-21740-smdd-bench-can-llms-solve-real-world-small-molecule-drug-design-tasks.md
 create mode 100644 papers/items/2026-2605-22154-idlespec-exploiting-idle-time-via-speculative-planning-for-llm-agents.md
 create mode 100644 papers/items/2026-2605-22321-benchmarking-autonomous-agents-against-temporal-spatial-and-semantic-evasions.md
 create mode 100644 papers/items/2026-2605-22643-boiling-the-frog-a-multi-turn-benchmark-for-agentic-safety.md
 create mode 100644 papers/items/2026-2605-23067-what-training-data-teaches-rl-memory-agents-an-empirical-study-of-curriculum-eff.md
 create mode 100644 papers/items/2026-2605-23574-push-your-agent-measuring-and-enforcing-quantitative-goal-persistence-in-long-ho.md
 create mode 100644 papers/items/2026-2605-23636-rf-instrument-agent-rfia-empowering-rf-instruments-with-natural-language-underst.md
 create mode 100644 papers/items/2026-2605-23723-memaudit-post-hoc-auditing-of-poisoned-agent-memory-via-causal-attribution-and-s.md
 create mode 100644 papers/items/2026-2605-23899-from-raw-experience-to-skill-consumption-a-systematic-study-of-model-generated-a.md
 create mode 100644 papers/items/2026-2605-23986-memforest-an-efficient-agent-memory-system-with-hierarchical-temporal-indexing.md
 create mode 100644 papers/items/2026-2605-24069-when-the-manual-lies-a-realistic-benchmark-to-evaluate-mcp-poisoning-attacks-for.md
 create mode 100644 papers/items/2026-2605-24216-agent-tom-learning-to-monitor-autonomous-llm-agents-via-theory-of-mind-reasoning.md
 create mode 100644 papers/items/2026-2605-24219-beyond-final-answers-auditing-trajectory-level-hallucinations-in-multi-agent-ind.md
 create mode 100644 papers/items/2026-2605-24220-polar-agentic-rl-on-any-harness-at-scale.md
 create mode 100644 papers/items/2026-2605-24309-reframing-llm-agent-security-as-an-agent-human-interaction-problem.md
 create mode 100644 papers/items/2026-2605-24659-iterinject-indirect-prompt-injection-against-llm-agents-via-feedback-guided-iter.md
 create mode 100644 papers/items/2026-2605-24812-core-code-collaborative-reinforcement-learning-for-code-generation.md
 create mode 100644 papers/items/2026-2605-25141-llm-agent-based-renewable-energy-forecasting-using-edge-and-iot-data-a-review-of.md
 create mode 100644 papers/items/2026-2605-25200-grouptravelbench-benchmarking-llm-agents-on-multi-person-travel-planning.md
 create mode 100644 papers/items/2026-2605-25310-tool-call-dependency-structure-is-linearly-decodable-in-llm-agent-residual-strea.md
 create mode 100644 papers/items/2026-2605-25393-decision-making-with-lightweight-confidence-aware-language-model-for-autonomous-.md
 create mode 100644 papers/items/2026-2605-25435-security-of-openclaw-agents-fundamentals-attacks-and-countermeasures.md
 create mode 100644 papers/items/2026-2605-25920-can-llms-time-travel-enhancing-temporal-consistency-in-legal-agentic-search-thro.md
 create mode 100644 papers/items/2026-2605-26165-tool-schema-compression-enables-agentic-rag-under-constrained-context-budgets.md
 create mode 100644 papers/items/2026-2605-26252-is-agent-memory-a-database-rethinking-data-foundations-for-long-term-ai-agent-me.md
 create mode 100644 papers/items/2026-2605-26305-experiments-in-agentic-ai-for-science.md
 create mode 100644 papers/items/2026-2605-26497-aligning-provenance-with-authorization-a-dual-graph-defense-for-llm-agents.md
 create mode 100644 papers/items/2026-2605-26720-towards-feedback-to-plan-decisions-for-self-evolving-llm-agents-in-cuda-kernel-g.md
 create mode 100644 papers/items/2026-2605-26926-from-norms-to-indicators-n2i-rag-an-agentic-retrieval-augmented-generation-frame.md
 create mode 100644 papers/items/2026-2605-27123-rethinking-agentic-rag-toward-llm-driven-logical-retrieval-beyond-embeddings.md
 create mode 100644 papers/items/2026-2605-27134-scaling-benchmarking-and-reasoning-of-vision-language-agents-for-mobile-gui-navi.md
 create mode 100644 papers/items/2026-2605-27240-enpmr-bench-benchmarking-proactive-memory-retrieval-for-emotional-support-agents.md
 create mode 100644 papers/items/2026-2605-27333-finharness-an-inline-lifecycle-safety-harness-for-finance-llm-agents.md
 create mode 100644 papers/items/2026-2605-27366-muse-autoskill-self-evolving-agents-via-skill-creation-memory-management-and-eva.md
 create mode 100644 papers/items/2026-2605-27690-traces-proactive-safety-auditing-for-multi-turn-llm-agents-via-trajectory-state-.md
 create mode 100644 papers/items/2026-2605-27762-peam-parametric-embodied-agent-memory-through-contrastive-internalization-of-exp.md
 create mode 100644 papers/items/2026-2605-27825-mrmmia-membership-inference-attacks-on-memory-in-chat-agents.md
 create mode 100644 papers/items/2026-2605-27935-do-agents-think-deeper-a-mechanistic-investigation-of-layer-wise-dynamics-in-seq.md
 create mode 100644 papers/items/2026-2605-28046-memcog-from-memory-as-tool-to-memory-as-cognition-in-conversational-agents.md
 create mode 100644 papers/items/2026-2605-28120-legalgraphrag-multi-agent-graph-retrieval-augmented-generation-for-reliable-lega.md
 create mode 100644 papers/items/2026-2605-28175-mixture-of-experts-knowledge-graph-retrieval-augmented-generation-for-multi-agen.md
 create mode 100644 papers/items/2026-2605-28424-skill0-5-joint-skill-internalization-and-utilization-for-out-of-distribution-gen.md
 create mode 100644 papers/items/2026-2605-28607-adaptive-multimodal-agents-based-framework-for-automatic-workflow-execution.md
 create mode 100644 papers/items/2026-2605-28617-lacuna-safe-agents-as-recursive-program-holes.md
 create mode 100644 papers/items/2026-2605-28787-do-agents-need-semantic-metadata-a-comparative-study-in-agentic-data-retrieval.md
 create mode 100644 papers/items/2026-2605-28835-genesisfunc-multi-agent-data-generation-for-accurate-and-generalizable-function-.md
 create mode 100644 papers/items/2026-2605-28850-representation-signatures-and-risk-feedback-alignment-in-llm-trading-agents.md
 create mode 100644 papers/items/2026-2605-29341-worldmemarena-evaluating-multimodal-agent-memory-through-action-world-interactio.md
 create mode 100644 papers/items/2026-2605-29630-entity-collision-a-stratified-protocol-for-attributing-retrieval-lift-in-agent-m.md
 create mode 100644 papers/items/2026-2605-29640-vikingmem-a-memory-base-management-system-for-stateful-llm-based-applications.md
 create mode 100644 papers/items/2026-2605-29653-ptcg-bench-can-llm-agents-master-pok-mon-trading-card-game.md
 create mode 100644 papers/items/2026-2605-29676-notation-matters-a-benchmark-study-of-token-optimized-formats-in-agentic-ai-syst.md
 create mode 100644 papers/items/2026-2605-29790-evolve-as-a-team-collaborative-self-evolution-for-llm-based-multi-agent-systems.md
 create mode 100644 papers/items/2026-2605-29801-agentdog-1-5-a-lightweight-and-scalable-alignment-framework-for-ai-agent-safety-.md
 create mode 100644 papers/items/2026-2605-29861-towards-verifiable-multimodal-deep-research-a-multi-agent-harness-for-interleave.md
 create mode 100644 papers/items/2026-2605-29960-hijacking-agent-memory-stealthy-trojan-attacks-through-conversational-interactio.md
 create mode 100644 papers/items/2026-2605-30058-heart-bench-do-llm-agents-exhibit-human-like-psychology.md
 create mode 100644 papers/items/2026-2605-30090-directorbench-diagnosing-long-form-video-generation-with-personalized-multi-agen.md
 create mode 100644 papers/items/2026-2605-30407-exploring-autonomous-agentic-data-engineering-for-model-specialization.md
 create mode 100644 papers/items/2026-2605-30604-an-organization-scoped-llm-agent-runtime-architecture-for-regulated-cybersecurit.md
 create mode 100644 papers/items/2026-2605-30690-elasticmem-latent-memory-as-a-learnable-resource-for-llm-agents.md
 create mode 100644 papers/items/2026-2605-30711-sage-a-novelty-gate-for-efficient-memory-evolution-in-agentic-llms.md
 create mode 100644 papers/items/2026-2605-30858-forecastcompass-guiding-agentic-forecasting-with-adaptive-factor-memory.md
 create mode 100644 papers/items/2026-2605-30883-trace-task-aware-adaptive-self-evolving-agentic-jailbreaking.md
 create mode 100644 papers/items/2026-2605-30907-bluefin-benchmarking-llm-agents-on-financial-spreadsheets.md
 create mode 100644 papers/items/2026-2605-30947-extending-ai-for-research-to-the-humanities-a-multi-agent-framework-for-evidence.md
 create mode 100644 papers/items/2026-2605-31075-task-focused-memorization-for-multimodal-agents.md
 create mode 100644 papers/items/2026-2605-31268-mellum2-technical-report.md
 create mode 100644 papers/items/2026-2605-31278-industrializing-prediction-powered-inference-the-glide-library-for-reliable-gena.md
 create mode 100644 papers/items/2026-2605-31308-tracegraph-shared-decision-landscapes-for-diagnosing-and-improving-agent-traject.md
 create mode 100644 papers/items/2026-2605-31377-dynatree-dynamic-agentic-retrieval-tree-for-time-sensitive-news-retrieval.md
 create mode 100644 papers/items/2026-2606-00198-bagen-are-llm-agents-budget-aware.md
 create mode 100644 papers/items/2026-2606-00341-rogue-misaligned-agent-behavior-arising-from-ordinary-computer-use.md
 create mode 100644 papers/items/2026-2606-00610-memgraphrag-memory-based-multi-agent-system-for-graph-retrieval-augmented-genera.md
 create mode 100644 papers/items/2026-2606-00611-trace-trajectory-risk-aware-compression-for-long-horizon-agent-safety.md
 create mode 100644 papers/items/2026-2606-00619-mempro-agentic-memory-systems-as-evolvable-programs.md
 create mode 100644 papers/items/2026-2606-00644-foresci-evaluating-llm-agents-for-forward-looking-ai-research-judgment.md
 create mode 100644 papers/items/2026-2606-00756-comic-collaborative-memory-and-insights-circulation-for-long-horizon-llm-agents-.md
 create mode 100644 papers/items/2026-2606-00914-adversarial-feeds-steer-llm-agent-decisions-against-their-defaults.md
 create mode 100644 papers/items/2026-2606-00915-autonomous-agentic-design-for-photonics.md
 create mode 100644 papers/items/2026-2606-00922-a-machine-to-machine-knowledge-guided-llm-agent-for-generalizable-radiotherapy-t.md
 create mode 100644 papers/items/2026-2606-00939-fincom-a-financial-multi-agent-demo-with-disagree-or-commit-deliberation.md
 create mode 100644 papers/items/2026-2606-01041-expweaver-llm-agents-learn-from-experience-via-latent-rag.md
 create mode 100644 papers/items/2026-2606-01138-memorywire-a-vendor-neutral-wire-format-for-agent-memory-operations.md
 create mode 100644 papers/items/2026-2606-01166-braveguard-from-open-world-threats-to-safer-computer-use-agents.md
 create mode 100644 papers/items/2026-2606-01185-skill-issues-data-centric-optimization-of-lakehouse-agents.md
 create mode 100644 papers/items/2026-2606-01199-can-llm-agents-sustain-long-horizon-organizational-dynamics.md
 create mode 100644 papers/items/2026-2606-01222-rag-driven-multi-agent-llm-framework-with-task-decomposition-for-beyond-5g-auto-.md
 create mode 100644 papers/items/2026-2606-01385-bridging-requirements-and-architecture-multi-agent-orchestration-with-external-k.md
 create mode 100644 papers/items/2026-2606-01416-self-healing-agentic-orchestrators-for-reliable-tool-augmented-large-language-mo.md
 create mode 100644 papers/items/2026-2606-01528-joint-agent-memory-and-exploration-learning-via-novelty-signals.md
 create mode 100644 papers/items/2026-2606-01613-techrag-evidence-gated-multimodal-agentic-rag-for-technical-literature-reasoning.md
 create mode 100644 papers/items/2026-2606-01815-crab-bench-evaluating-llm-agents-under-complex-task-dependencies-and-human-align.md
 create mode 100644 papers/items/2026-2606-01961-automedbench-towards-medical-autoresearch-with-agentic-ai-models.md
 create mode 100644 papers/items/2026-2606-02109-badger-bridging-agentic-and-deterministic-evaluation-for-generative-enterprise-r.md
 create mode 100644 papers/items/2026-2606-02302-seclaw-spec-driven-security-task-synthesis-for-evaluating-autonomous-agents.md
 create mode 100644 papers/items/2026-2606-02372-comap-co-evolving-world-models-and-agent-policies-for-llm-agents.md
 create mode 100644 papers/items/2026-2606-02380-spade-bench-evaluating-spontaneous-strategic-deception-in-agents-via-plan-action.md
 create mode 100644 papers/items/2026-2606-02388-policy-and-world-modeling-co-training-for-language-agents.md
 create mode 100644 papers/items/2026-2606-02404-k-browsecomp-a-web-browsing-agent-benchmark-grounded-in-korean-contexts.md
 create mode 100644 papers/items/2026-2606-02461-agentcl-toward-rigorous-evaluation-of-continual-learning-in-language-agents.md
 create mode 100644 papers/items/2026-2606-02497-bridging-the-last-mile-of-time-series-forecasting-with-llm-agents.md
 create mode 100644 papers/items/2026-2606-02812-traj-evolve-a-self-evolving-multi-agent-system-for-patient-trajectory-modeling-i.md
 create mode 100644 papers/items/2026-2606-02965-what-benchmarks-don-t-measure-the-case-for-evaluating-abstention-competence-in-a.md
 create mode 100644 papers/items/2026-2606-03108-evotrainer-co-evolving-llm-policies-and-training-harnesses-for-autonomous-agenti.md
 create mode 100644 papers/items/2026-2606-03135-uncertainty-aware-clarification-in-llm-agents-with-information-gain.md
 create mode 100644 papers/items/2026-2606-03157-clinicalmc-a-benchmark-for-multi-course-clinical-decision-making-with-large-lang.md
 create mode 100644 papers/items/2026-2606-03197-memtrain-self-supervised-context-memory-training.md
 create mode 100644 papers/items/2026-2606-03329-infomem-training-long-context-memory-agents-with-answer-conditioned-information-.md
 create mode 100644 papers/items/2026-2606-03374-emem-a-hybrid-spatio-temporal-memory-system-for-embodied-agents.md
 create mode 100644 papers/items/2026-2606-03544-sage-a-quantitative-evaluation-of-socialized-evolution-in-agent-ecosystems.md
 create mode 100644 papers/items/2026-2606-03657-diagnosing-knowledge-gaps-in-llm-tool-use-an-agentic-benchmark-for-novel-api-acq.md
 create mode 100644 papers/items/2026-2606-03895-agent-libos-a-runtime-substrate-for-capability-controlled-self-evolving-llm-agen.md
 create mode 100644 papers/items/2026-2606-04051-rubas-rubric-based-reinforcement-learning-for-agent-safety.md
 create mode 100644 papers/items/2026-2606-04120-salimory-orchestrating-cognitive-memory-for-conversational-agents.md
 create mode 100644 papers/items/2026-2606-04296-the-saturation-trap-and-the-subjectivity-of-intervention-timing-why-affect-based.md
 create mode 100644 papers/items/2026-2606-04315-exploring-cross-scenario-generality-of-agentic-memory-systems-diagnostics-and-a-.md
 create mode 100644 papers/items/2026-2606-04555-temporal-order-matters-for-agentic-memory-segment-trees-for-long-horizon-agents.md
 create mode 100644 papers/items/2026-2606-04599-plan-first-judge-later-run-better-a-dmaic-inspired-agentic-system-for-industrial.md
 create mode 100644 papers/items/2026-2606-04628-rampart-registry-based-agentic-memory-with-priority-aware-runtime-transformation.md
 create mode 100644 papers/items/2026-2606-04780-personatree-structured-lifecycle-memory-for-person-understanding-in-llm-agents.md
 create mode 100644 papers/items/2026-2606-04874-agent-planning-benchmark-a-diagnostic-framework-for-planning-capabilities-in-llm.md
 create mode 100644 papers/items/2026-2606-04990-from-agent-traces-to-trust-a-survey-of-evidence-tracing-and-execution-provenance.md
 create mode 100644 papers/items/2026-2606-05241-search-time-contamination-in-deep-research-agents-measuring-performance-inflatio.md
 create mode 100644 papers/items/2026-2606-05263-policy-conditioned-counterfactual-credit-for-verifiable-reinforcement-learning-o.md
 create mode 100644 papers/items/2026-2606-05414-when-evidence-is-sparse-weakly-supervised-early-failure-alerting-in-dialogs-and-.md
 create mode 100644 papers/items/2026-2606-05436-ten-headache-specialists-versus-artificial-intelligence-for-clinical-literature-.md
 create mode 100644 papers/items/2026-2606-05463-psebench-a-controllable-and-verifiable-benchmark-for-evaluating-llms-in-patient-.md
 create mode 100644 papers/items/2026-2606-05548-adk-arena-evaluating-agent-development-kits-via-llm-as-a-developer.md
 create mode 100644 papers/items/2026-2606-05558-autoregressive-diffusion-world-models-for-off-policy-evaluation-of-llm-agents.md
 create mode 100644 papers/items/2026-2606-05622-adaplanbench-evaluating-adaptive-planning-in-large-language-model-agents-under-w.md
 create mode 100644 papers/items/2026-2606-05658-agent-orchestrated-adaptive-rag-a-comparative-study-on-structured-and-multi-hop-.md
 create mode 100644 papers/items/2026-2606-05684-adamem-test-time-adaptive-memory-for-language-agents.md
 create mode 100644 papers/items/2026-2606-05711-beyond-tokens-a-unified-framework-for-latent-communication-in-llm-based-multi-ag.md
 create mode 100644 papers/items/2026-2606-05805-from-risk-classification-to-action-plan-remediation-a-guardrail-feedback-driven-.md
 create mode 100644 papers/items/2026-2606-06054-beyond-similarity-trustworthy-memory-search-for-personal-ai-agents.md
 create mode 100644 papers/items/2026-2606-06090-beyond-semantic-organization-memory-as-execution-state-management-for-long-horiz.md
 create mode 100644 papers/items/2026-2606-06388-humans-almanac-a-human-collaboration-dataset-of-action-level-mental-model-annota.md
 create mode 100644 papers/items/2026-2606-06399-collabsim-a-cscw-grounded-methodology-for-investigating-collaborative-competence.md
 create mode 100644 papers/items/2026-2606-06448-agent-memory-characterization-and-system-implications-of-stateful-long-horizon-w.md
 create mode 100644 papers/items/2026-2606-06462-benchmark-everything-everywhere-all-at-once.md
 create mode 100644 papers/items/2026-2606-06473-mlevolve-a-self-evolving-framework-for-automated-machine-learning-algorithm-disc.md
 create mode 100644 papers/items/2026-2606-07314-qbuglm-an-agentic-benchmarking-framework-for-llm-based-quantum-software-debuggin.md
 create mode 100644 papers/items/2026-2606-07379-do-coding-agents-deceive-us-detecting-and-preventing-cheating-via-capped-evaluat.md
 create mode 100644 papers/items/2026-2606-07402-m-3-exam-benchmarking-multimodal-memory-for-realistic-user-agent-interactions.md
 create mode 100644 papers/items/2026-2606-07591-researchclawbench-a-benchmark-for-end-to-end-autonomous-scientific-research.md
 create mode 100644 papers/items/2026-2606-07595-visualleakbench-reproducible-action-boundary-propagation-failures-in-vision-lang.md
 create mode 100644 papers/items/2026-2606-07682-swe-marathon-can-agents-autonomously-complete-ultra-long-horizon-software-work.md
 create mode 100644 papers/items/2026-2606-07711-rosetta-memory-adaptive-memory-for-cross-llm-agents.md
 create mode 100644 papers/items/2026-2606-07836-agentic-multi-fidelity-learning-of-quasiparticle-and-excitonic-properties.md
 create mode 100644 papers/items/2026-2606-07867-the-cold-start-safety-gap-in-llm-agents.md
 create mode 100644 papers/items/2026-2606-08162-silent-failure-in-llm-agent-systems-the-entropy-principle-and-the-inevitable-dis.md
 create mode 100644 papers/items/2026-2606-08172-the-governance-of-human-llm-interaction-safety-gating-civility-steering-and-affe.md
 create mode 100644 papers/items/2026-2606-08274-toward-human-centered-multi-agent-systems-integrating-cognition-culture-values-a.md
 create mode 100644 papers/items/2026-2606-08340-benchmarking-open-ended-multi-agent-coordination-in-language-agents.md
 create mode 100644 papers/items/2026-2606-08531-vesta-a-fully-automated-scenario-generation-and-safety-evaluation-framework-for-.md
 create mode 100644 papers/items/2026-2606-08625-from-holistic-evaluation-to-structured-criteria-rubrics-across-the-evolving-llm-.md
 create mode 100644 papers/items/2026-2606-08790-rails-verification-native-clearing-for-agentic-commerce.md
 create mode 100644 papers/items/2026-2606-08960-hardening-agent-benchmarks-with-adversarial-hacker-fixer-loops.md
 create mode 100644 papers/items/2026-2606-09037-a-multi-agent-system-for-ipmsm-design-optimization-via-an-fea-ai-hybrid-approach.md
 create mode 100644 papers/items/2026-2606-09071-reflect-intervention-supported-error-attribution-for-silent-failures-in-llm-agen.md
 create mode 100644 papers/items/2026-2606-09198-mass-deep-research-for-social-sciences-with-memory-augmented-social-simulation.md
 create mode 100644 papers/items/2026-2606-09316-anything2skill-compiling-external-knowledge-into-reusable-skills-for-agents.md
 create mode 100644 papers/items/2026-2606-09399-runagent-superbrowser-a-theory-of-autonomous-web-navigation-grounded-in-human-br.md
 create mode 100644 papers/items/2026-2606-09426-weavebench-a-long-horizon-real-world-benchmark-for-computer-use-agents-with-hybr.md
 create mode 100644 papers/items/2026-2606-09447-aliyunconsoleagent-training-web-agents-in-real-world-cloud-environments-via-dist.md
 create mode 100644 papers/items/2026-2606-09483-memory-beyond-recall-a-dual-process-cognitive-memory-system-for-self-evolving-ll.md
 create mode 100644 papers/items/2026-2606-09549-secureclaw-clawing-back-control-of-llm-agents.md
 create mode 100644 papers/items/2026-2606-09738-hdsl-a-hierarchical-domain-specific-language-for-structured-3d-indoor-scene-gene.md
 create mode 100644 papers/items/2026-2606-09764-iosworld-a-benchmark-for-personally-intelligent-phone-agents.md
 create mode 100644 papers/items/2026-2606-09774-auto-configuring-scientific-simulators-with-lightweight-coding-agent-adapters.md
 create mode 100644 papers/items/2026-2606-09863-from-confident-closing-to-silent-failure-characterizing-false-success-in-llm-age.md
 create mode 100644 papers/items/2026-2606-09961-3spo-state-score-supervised-policy-optimization-for-llm-agents.md
 create mode 100644 papers/items/2026-2606-10209-less-context-better-agents-efficient-context-engineering-for-long-horizon-tool-u.md
 create mode 100644 papers/items/2026-2606-10304-mirage-a-polarity-flipping-encoding-subspace-in-llm-agents.md
 create mode 100644 papers/items/2026-2606-10316-tabclaw-an-interactive-and-self-evolving-agent-for-spreadsheet-manipulation-and-.md
 create mode 100644 papers/items/2026-2606-10381-agentic-hybrid-rag-for-evidence-grounded-muon-collider-analysis.md
 create mode 100644 papers/items/2026-2606-10394-stage-claw-automated-state-based-agent-benchmarking-for-realistic-scenarios.md
 create mode 100644 papers/items/2026-2606-10423-webchallenger-a-reliable-and-efficient-generalist-web-agent.md
 create mode 100644 papers/items/2026-2606-10507-hipif-hierarchical-planning-and-information-folding-for-long-horizon-llm-agent-l.md
 create mode 100644 papers/items/2026-2606-10532-activemem-distributed-active-memory-for-long-horizon-llm-reasoning.md
 create mode 100644 papers/items/2026-2606-10577-agenticnav-zero-shot-vision-and-language-navigation-as-a-tool-calling-harness.md
 create mode 100644 papers/items/2026-2606-10616-learning-what-to-remember-observability-safe-memory-retention-via-constrained-op.md
 create mode 100644 papers/items/2026-2606-10677-infini-memory-maintainable-topic-documents-for-long-term-llm-agent-memory.md
 create mode 100644 papers/items/2026-2606-10684-divide-and-cooperate-role-decomposed-multi-agent-llm-training-with-cross-agent-l.md
 create mode 100644 papers/items/2026-2606-10742-memvenom-triggered-poisoning-of-multimodal-memories-in-web-agents.md
 create mode 100644 papers/items/2026-2606-10749-toward-secure-llm-agents-threat-surfaces-attacks-defenses-and-evaluation.md
 create mode 100644 papers/items/2026-2606-10921-trace-only-what-you-need-structure-aware-on-demand-hypergraph-memory-for-long-do.md
 create mode 100644 papers/items/2026-2606-10933-frontier-coding-agents-use-metaprogramming-to-adapt-to-unfamiliar-programming-la.md
 create mode 100644 papers/items/2026-2606-11042-workflow-gym-towards-long-horizon-evaluation-of-computer-use-agentic-tasks-in-re.md
 create mode 100644 papers/items/2026-2606-11078-a-history-aware-visually-grounded-critic-for-computer-use-agents.md
 create mode 100644 papers/items/2026-2606-11079-vista-a-versatile-interactive-user-simulation-toolkit-for-agent-evaluation.md
 create mode 100644 papers/items/2026-2606-11119-trace-a-unified-rollout-budget-allocation-framework-for-efficient-agentic-reinfo.md
 create mode 100644 papers/items/2026-2606-11176-data-journalist-agent-transforming-data-into-verifiable-multimodal-stories.md
 create mode 100644 papers/items/2026-2606-11349-knowing-when-to-ask-self-gated-clarification-for-hierarchical-language-agents.md
 create mode 100644 papers/items/2026-2606-11354-a-zero-shot-multi-agent-framework-for-human-building-interaction-via-programmati.md
 create mode 100644 papers/items/2026-2606-11680-organize-then-retrieve-hierarchical-memory-navigation-for-efficient-agents.md
 create mode 100644 papers/items/2026-2606-11688-goal-autopilot-a-verifiable-anti-fabrication-firewall-for-unattended-long-horizo.md
 create mode 100644 papers/items/2026-2606-11702-medcta-a-benchmark-for-clinical-tool-agents.md
 create mode 100644 papers/items/2026-2606-11869-agents-all-the-way-down-a-methodology-for-building-custom-ai-agents-from-substra.md
 create mode 100644 papers/items/2026-2606-12195-internvideo3-agentify-foundation-models-with-multimodal-contextual-reasoning.md
 create mode 100644 papers/items/2026-2606-12320-a-five-plane-reference-architecture-for-runtime-governance-of-production-ai-agen.md
 create mode 100644 papers/items/2026-2606-12341-ocelot-inference-leakage-budgets-for-privacy-preserving-llm-agents.md
 create mode 100644 papers/items/2026-2606-12344-claw-swe-bench-a-benchmark-for-evaluating-openclaw-style-agent-harnesses-on-codi.md
 create mode 100644 papers/items/2026-2606-12384-appo-agentic-procedural-policy-optimization.md
 create mode 100644 papers/items/2026-2606-12563-arbor-tree-search-as-a-cognition-layer-for-autonomous-agents.md
 create mode 100644 papers/items/2026-2606-12586-beyond-attack-success-rate-examining-trigger-leakage-in-vision-language-agentic-.md
 create mode 100644 papers/items/2026-2606-12634-keep-policy-gradient-in-charge-sibling-guided-credit-distillation-for-long-horiz.md
 create mode 100644 papers/items/2026-2606-12657-trajgenagent-a-hierarchical-llm-agent-for-human-mobility-trajectory-generation.md
 create mode 100644 papers/items/2026-2606-12674-evoflux-inference-time-evolution-of-executable-tool-workflows-for-compact-agents.md
 create mode 100644 papers/items/2026-2606-12703-smsr-certified-defence-against-runtime-memory-poisoning-in-persistent-llm-agent-.md
 create mode 100644 papers/items/2026-2606-12780-proplay-procedural-world-models-for-self-evolving-llm-agents.md
 create mode 100644 papers/items/2026-2606-12837-lohosearch-benchmarking-long-horizon-search-agents-beyond-the-human-difficulty-c.md
 create mode 100644 papers/items/2026-2606-12945-learning-what-to-remember-a-cognitively-grounded-multi-factor-value-model-for-ag.md
 create mode 100644 papers/items/2026-2606-13148-terrabench-can-agents-reason-over-heterogeneous-earth-system-data.md
 create mode 100644 papers/items/2026-2606-13177-memrefine-llm-guided-compression-for-long-term-agent-memory.md
 create mode 100644 papers/items/2026-2606-13192-reasoning-for-mobile-user-experience-with-multimodal-llms-task-benchmark-and-app.md
 create mode 100644 papers/items/2026-2606-13317-skillcat-contrastive-assessment-and-topology-aware-skill-self-evolution-for-llm-.md
 create mode 100644 papers/items/2026-2606-13385-who-pays-the-price-stakeholder-centric-prompt-injection-benchmarking-for-real-wo.md
 create mode 100644 papers/items/2026-2606-13602-epibench-verifiable-evaluation-of-ai-agents-on-epigenomics-analysis.md
 create mode 100644 papers/items/2026-2606-13608-agentbeats-agentifying-agent-assessment-for-openness-standardization-and-reprodu.md
 create mode 100644 papers/items/2026-2606-13643-recursive-agent-harnesses.md
 create mode 100644 papers/items/2026-2606-13663-hypertool-beyond-step-wise-tool-calls-for-tool-augmented-agents.md
 create mode 100644 papers/items/2026-2606-13686-benchmarking-web-agent-safety-under-e-commerce-deceptive-interfaces.md
 create mode 100644 papers/items/2026-2606-13904-sana-what-matters-for-qa-agents-over-massive-data-lakes.md
 create mode 100644 papers/items/2026-2606-13994-hidden-in-plain-sight-benchmarking-agent-safety-against-decomposition-attacks-wi.md
 create mode 100644 papers/items/2026-2606-14106-naive-visual-memory-is-not-enough-a-failure-mode-study-of-gui-agents.md
 create mode 100644 papers/items/2026-2606-14470-gitofthoughts-version-controlled-reasoning-and-agent-memory-you-can-replay-diff-.md
 create mode 100644 papers/items/2026-2606-14502-from-chatbot-to-digital-colleague-the-paradigm-shift-toward-persistent-autonomou.md
 create mode 100644 papers/items/2026-2606-14517-from-shield-to-target-denial-of-service-attacks-on-llm-based-agent-guardrails.md
 create mode 100644 papers/items/2026-2606-14571-streammembench-streaming-evaluation-of-agent-memory-for-future-oriented-assistan.md
 create mode 100644 papers/items/2026-2606-14574-simmer-benchmarking-latent-failures-in-llm-executable-planning-with-a-world-mode.md
 create mode 100644 papers/items/2026-2606-14790-xflow-an-executable-protocol-programming-system-for-reliable-multi-agent-workflo.md
 create mode 100644 papers/items/2026-2606-14805-knowledge-based-zero-replay-debugging-of-multi-agent-llm-traces.md
 create mode 100644 papers/items/2026-2606-15017-are-online-skill-and-memory-modules-always-worth-their-tokens-a-budget-constrain.md
 create mode 100644 papers/items/2026-2606-15034-osguard-a-benchmark-for-safety-in-computer-use-agents.md
 create mode 100644 papers/items/2026-2606-15079-ling-and-ring-2-6-technical-report-efficient-and-instant-agentic-intelligence-at.md
 create mode 100644 papers/items/2026-2606-15152-can-agents-read-the-room-benchmarking-visual-social-intelligence-in-multimodal-s.md
 create mode 100644 papers/items/2026-2606-15242-benign-in-isolation-harmful-in-composition-security-risks-in-agent-skill-ecosyst.md
 create mode 100644 papers/items/2026-2606-15376-coagent-concurrency-control-for-multi-agent-systems.md
 create mode 100644 papers/items/2026-2606-15591-agentic-retrieval-and-reinforcement-learned-equation-chains-a-controlled-generat.md
 create mode 100644 papers/items/2026-2606-15609-fragfuse-bypassing-access-control-of-large-language-model-agents-via-memory-base.md
 create mode 100644 papers/items/2026-2606-15684-multi-agent-framework-for-time-sensitive-complementary-collaboration-in-minecraf.md
 create mode 100644 papers/items/2026-2606-15709-ai-driven-framework-for-adaptive-water-network-management-with-proof-of-concept-.md
 create mode 100644 papers/items/2026-2606-15862-retailbench-benchmarking-long-horizon-reasoning-and-coherent-decision-making-of-.md
 create mode 100644 papers/items/2026-2606-15874-llm-as-code-agentic-programming-for-agent-harness.md
 create mode 100644 papers/items/2026-2606-15903-control-plane-placement-shapes-forgetting-an-architectural-study-of-agent-memory.md
 create mode 100644 papers/items/2026-2606-15906-mage-rag-multigranular-adaptive-graph-evidence-for-agentic-multimodal-rag-in-lon.md
 create mode 100644 papers/items/2026-2606-15931-deeproot-a-kg-coordinated-multi-agent-system-for-therapeutic-reasoning-over-hist.md
 create mode 100644 papers/items/2026-2606-15994-agentic-framework-for-deep-learning-workload-migration-via-in-context-learning.md
 create mode 100644 papers/items/2026-2606-16111-towards-pareto-optimal-tool-integrated-agents-with-pareto-ranking-policy-optimiz.md
 create mode 100644 papers/items/2026-2606-16295-visualclaw-a-real-time-personalized-agent-for-the-physical-world.md
 create mode 100644 papers/items/2026-2606-16420-transferable-self-evolving-playbooks-for-agentic-security-auditing.md
 create mode 100644 papers/items/2026-2606-16432-accord-action-conditioned-contextual-grounding-for-language-agents.md
 create mode 100644 papers/items/2026-2606-16481-steering-emotional-dynamics-for-art-therapy-controllable-narrative-script-genera.md
 create mode 100644 papers/items/2026-2606-16534-generated-parallel-scalable-a-study-of-agentic-ai-generated-julia-code-on-superc.md
 create mode 100644 papers/items/2026-2606-16576-can-llm-agents-infer-world-models-evidence-from-agentic-automata-learning.md
 create mode 100644 papers/items/2026-2606-16591-sing-synthetic-intention-graph-for-scalable-active-tool-discovery-in-llm-agents.md
 create mode 100644 papers/items/2026-2606-16613-coffeebench-benchmarking-long-horizon-llm-agents-in-heterogeneous-multi-agent-ec.md
 create mode 100644 papers/items/2026-2606-16659-fraudsmswalker-benchmarking-agentic-large-language-models-for-sms-to-webpage-fra.md
 create mode 100644 papers/items/2026-2606-16748-mypcbench-a-benchmark-for-personally-intelligent-computer-use-agents.md
 create mode 100644 papers/items/2026-2606-16774-openclaw-skill-collective-skill-tree-search-for-agentic-large-language-models.md
 create mode 100644 papers/items/2026-2606-16802-labosbench-benchmarking-computer-use-agents-for-scientific-instrument-control.md
 create mode 100644 papers/items/2026-2606-16813-gist-cmtf-goal-state-inference-for-causal-minimal-tool-filtering-in-llm-agents.md
 create mode 100644 papers/items/2026-2606-16839-towards-llm-accelerated-rapid-reviews-for-software-tool-discovery-case-for-log-a.md
 create mode 100644 papers/items/2026-2606-16871-human-on-the-bridge-scalable-evaluation-for-ai-agents.md
 create mode 100644 papers/items/2026-2606-17041-benchmarking-llm-agents-on-meta-analysis-articles-from-nature-portfolio.md
 create mode 100644 papers/items/2026-2606-17076-cmip-forge-an-agentic-system-that-retrieves-computes-and-self-reviews-climate-sc.md
 create mode 100644 papers/items/2026-2606-17114-an-evaluation-of-data-leakage-risks-in-tool-using-llm-agents-in-realistic-scenar.md
 create mode 100644 papers/items/2026-2606-17246-geodisaster-benchmarking-orchestrated-agents-for-operational-disaster-geo-intell.md
 create mode 100644 papers/items/2026-2606-17368-distributed-general-purpose-agent-networks-architecture-key-mechanisms-and-proto.md
 create mode 100644 papers/items/2026-2606-17383-model-validation-of-agentic-ai-systems-a-pomdp-based-framework-for-belief-state-.md
 create mode 100644 papers/items/2026-2606-17449-mode-rag-manifold-outlier-diagnosis-and-energy-based-retrieval-augmented-generat.md
 create mode 100644 papers/items/2026-2606-17453-mapsatisfybench-benchmarking-satisfaction-aware-map-agents-through-behavior-grou.md
 create mode 100644 papers/items/2026-2606-17459-can-llms-be-ceos-benchmarking-strategic-resource-reallocation-with-multi-role-ag.md
 create mode 100644 papers/items/2026-2606-17573-cordon-semantic-transactions-for-tool-using-llm-agents.md
 create mode 100644 papers/items/2026-2606-17680-envrl-learn-from-environment-dynamics-in-agentic-reinforcement-learning.md
 create mode 100644 papers/items/2026-2606-18023-loopcoder-v2-only-loop-once-for-efficient-test-time-computation-scaling.md
 create mode 100644 papers/items/2026-2606-18037-provenanceguard-source-aware-factuality-verification-for-mcp-based-llm-agents.md
 create mode 100644 papers/items/2026-2606-18051-compositional-skill-routing-for-llm-agents-decompose-retrieve-and-compose.md
 create mode 100644 papers/items/2026-2606-18068-agentic-ai-based-framework-for-mitigating-premature-diagnostic-handoff-and-silen.md
 create mode 100644 papers/items/2026-2606-18142-your-ai-travel-agent-would-book-you-a-bullfight-an-agentic-benchmark-for-implici.md
 create mode 100644 papers/items/2026-2606-18272-mitigating-anchoring-bias-in-llm-based-agents-for-energy-efficient-6g-autonomous.md
 create mode 100644 papers/items/2026-2606-18356-safeclawbench-separating-semantic-audit-evidence-and-sandbox-harm-in-tool-using-.md
 create mode 100644 papers/items/2026-2606-18363-guava-an-effective-and-universal-harness-for-embodied-manipulation.md
 create mode 100644 papers/items/2026-2606-18406-coremem-riemannian-retrieval-and-fisher-guided-distillation-for-long-term-memory.md
 create mode 100644 papers/items/2026-2606-18467-toolchain-crc-conformal-risk-control-for-agentic-ai-under-retrieval-and-tool-use.md
 create mode 100644 papers/items/2026-2606-18502-towards-scalable-customization-and-deployment-of-multi-agent-systems-for-enterpr.md
 create mode 100644 papers/items/2026-2606-18619-code-augur-agentic-vulnerability-detection-via-specification-inference.md
 create mode 100644 papers/items/2026-2606-18671-hansel-extracting-breadcrumbs-from-web-agent-trajectories-for-interactive-verifi.md
 create mode 100644 papers/items/2026-2606-18789-poweragentbench-ss-a-benchmark-for-agentic-ai-in-power-system-steady-state-studi.md
 create mode 100644 papers/items/2026-2606-18829-gatemem-benchmarking-memory-governance-in-multi-principal-shared-memory-agents.md
 create mode 100644 papers/items/2026-2606-18950-rtsgamebench-an-rts-benchmark-for-strategic-reasoning-by-vision-language-models.md
 create mode 100644 papers/items/2026-2606-19063-pypiline-malicious-pypi-package-detection-via-suspicious-api-knowledge-and-agent.md
 create mode 100644 papers/items/2026-2606-19242-runtime-compliance-verification-for-ai-agents.md
 create mode 100644 papers/items/2026-2606-19245-txbench-pp-analyzing-ai-agent-performance-on-small-molecule-preclinical-pharmaco.md
 create mode 100644 papers/items/2026-2606-19409-openrath-session-centered-runtime-state-for-agent-systems.md
 create mode 100644 papers/items/2026-2606-19464-deontic-policies-for-runtime-governance-of-agentic-ai-systems.md
 create mode 100644 papers/items/2026-2606-19613-staminabench-stress-testing-coding-agents-over-100-interaction-turns.md
 create mode 100644 papers/items/2026-2606-19704-beyond-static-leaderboards-predictive-validity-for-the-evaluation-of-llm-agents.md
 create mode 100644 papers/items/2026-2606-19787-oragentbench-can-llm-agents-solve-challenging-operations-research-tasks-end-to-e.md
 create mode 100644 papers/items/2026-2606-19812-human-on-the-loop-orchestration-for-ai-assisted-legal-discovery.md
 create mode 100644 papers/items/2026-2606-19852-prompt-plan-extract-zero-shot-agentic-llms-workflows-for-lung-pathology-extracti.md
 create mode 100644 papers/items/2026-2606-19899-measuring-biological-capabilities-and-risks-of-ai-agents.md
 create mode 100644 papers/items/2026-2606-19926-memgui-agent-an-end-to-end-long-horizon-mobile-gui-agent-with-proactive-context-.md
 create mode 100644 papers/items/2026-2606-19930-mobileforge-annotation-free-adaptation-for-mobile-gui-agents-with-hierarchical-f.md
 create mode 100644 papers/items/2026-2606-19980-enpire-agentic-robot-policy-self-improvement-in-the-real-world.md
 create mode 100644 papers/items/2026-2606-20023-when-lower-privileges-suffice-investigating-over-privileged-tool-selection-in-ll.md
 create mode 100644 papers/items/2026-2606-20041-ai-economist-agent-an-agentic-framework-for-model-grounded-economic-analysis-wit.md
 create mode 100644 papers/items/2026-2606-20047-pacms-submodular-context-selection-as-a-pluggable-engine-for-llm-agents.md
 create mode 100644 papers/items/2026-2606-20243-phoenix-safe-github-issue-resolution-via-multi-agent-llms.md
 create mode 100644 papers/items/2026-2606-20401-poweragentbench-dyn-a-benchmark-for-agentic-ai-in-power-system-dynamic-studies.md
 create mode 100644 papers/items/2026-2606-20470-analyzing-defensive-misdirection-against-model-guided-automated-attacks-on-agent.md
 create mode 100644 papers/items/2026-2606-20479-groundcontrol-anticipating-navigation-failures-in-vision-language-agents-via-tra.md
 create mode 100644 papers/items/2026-2606-20510-efficient-and-sound-probabilistic-verification-for-ai-agents.md
 create mode 100644 papers/items/2026-2606-20512-probe-and-refine-tuning-of-repository-guidance-for-coding-agents.md
 create mode 100644 papers/items/2026-2606-20515-s-agent-spatial-tool-use-elicits-reasoning-for-spatial-intelligence.md
 create mode 100644 papers/items/2026-2606-20573-aona-a-comprehensive-architecture-and-workflow-design-for-global-agentic-collabo.md
 create mode 100644 papers/items/2026-2606-20629-specialize-roles-mix-deployments-pushing-the-cost-accuracy-frontier-of-llm-agent.md
 create mode 100644 papers/items/2026-2606-20717-mirage-stealthy-visual-prompt-injection-for-vulnerability-detection-in-web-agent.md
 create mode 100644 papers/items/2026-2606-20785-fara-1-5-scalable-learning-environments-for-computer-use-agents.md
 create mode 100644 papers/items/2026-2606-20922-think-twice-before-you-act-protecting-llm-agents-against-tool-description-poison.md
 create mode 100644 papers/items/2026-2606-20950-power-systems-agent-benchmark-executable-evaluation-of-ai-agents-in-electric-pow.md
 create mode 100644 papers/items/2026-2606-20954-learning-what-not-to-forget-long-horizon-agent-memory-from-a-few-kilobytes-of-le.md
 create mode 100644 papers/items/2026-2606-21013-agentic-time-machine-as-an-infrastructure-for-future-event-forecasting.md
 create mode 100644 papers/items/2026-2606-21123-a-multi-agent-audit-framework-for-high-stakes-reasoning-evaluation-and-interpret.md
 create mode 100644 papers/items/2026-2606-21129-agenticos-an-intent-oriented-secure-operating-system-architecture-for-autonomous.md
 create mode 100644 papers/items/2026-2606-21228-sakana-fugu-technical-report.md
 create mode 100644 papers/items/2026-2606-21401-swarmx-agentic-scheduling-for-low-latency-agentic-systems.md
 create mode 100644 papers/items/2026-2606-21409-don-t-blindly-trust-it-how-unreliable-feedback-breaks-tool-using-llm-agents.md
 create mode 100644 papers/items/2026-2606-21445-autoras-learning-robust-agentic-systems-with-primitive-representations.md
 create mode 100644 papers/items/2026-2606-21553-dissecting-agentic-rag-a-component-ablation-for-multi-hop-qa-with-a-local-7b-mod.md
 create mode 100644 papers/items/2026-2606-21565-composing-verifiable-conceptual-models-via-building-blocks-towards-design-time-v.md
 create mode 100644 papers/items/2026-2606-21627-counsel-a-meta-evaluation-dataset-for-agentic-tasks.md
 create mode 100644 papers/items/2026-2606-21649-evoembedding-evolvable-representations-for-long-context-retrieval-and-agentic-me.md
 create mode 100644 papers/items/2026-2606-21710-privacyalign-contextual-privacy-alignment-for-llm-agents.md
 create mode 100644 papers/items/2026-2606-21732-safe-to-check-unsafe-to-use-relinking-at-the-compression-boundary-of-llm-agents.md
 create mode 100644 papers/items/2026-2606-21740-training-the-orchestrator-a-supervised-approach-to-end-to-end-pddl-planning-with.md
 create mode 100644 papers/items/2026-2606-21836-agentdse-reasoning-augmented-architectural-design-space-exploration.md
 create mode 100644 papers/items/2026-2606-21842-agent-assisted-side-channel-attacks-on-non-prefix-kv-cache-in-rag.md
 create mode 100644 papers/items/2026-2606-21877-agentriskbom-a-risk-scoping-security-bill-of-materials-for-agentic-ai-systems.md
 create mode 100644 papers/items/2026-2606-21963-holmes-multimodal-agentic-diagnosis-for-mixed-language-mobile-crashes-at-industr.md
 create mode 100644 papers/items/2026-2606-22030-nous-a-predictive-world-model-for-long-term-agent-memory.md
 create mode 100644 papers/items/2026-2606-22082-codeteam-an-llm-powered-multi-agent-framework-for-repository-level-code-generati.md
 create mode 100644 papers/items/2026-2606-22110-traceview-interactive-visualization-of-agentic-program-repair-trajectories.md
 create mode 100644 papers/items/2026-2606-22151-novelty-aware-agentic-retrieval-comparing-research-contributions-through-structu.md
 create mode 100644 papers/items/2026-2606-22263-revelio-cost-efficient-agentic-memory-safety-vulnerability-detection-for-reposit.md
 create mode 100644 papers/items/2026-2606-22330-hypothesis-driven-skill-optimization-for-llm-agents.md
 create mode 100644 papers/items/2026-2606-22388-planbench-xl-evaluating-long-horizon-planning-of-llm-tool-use-agents-in-large-sc.md
 create mode 100644 papers/items/2026-2606-22417-code-isn-t-memory-a-structural-codebase-index-inside-a-coding-agent.md
 create mode 100644 papers/items/2026-2606-22484-governed-ai-assisted-engineering-graduated-human-oversight-for-agentic-code-gene.md
 create mode 100644 papers/items/2026-2606-22495-grounded-scaling-why-agentic-ai-needs-deterministic-environments.md
 create mode 100644 papers/items/2026-2606-22557-macagentbench-benchmarking-ai-agents-on-real-world-macos-desktop.md
 create mode 100644 papers/items/2026-2606-22610-paperclaw-harnessing-agents-for-autonomous-research-and-human-in-the-loop-refine.md
 create mode 100644 papers/items/2026-2606-22647-raven-agentic-rag-for-automated-vulnerability-repair.md
 create mode 100644 papers/items/2026-2606-22673-agentlens-interpretable-safety-steering-via-mechanistic-subspaces-for-multi-turn.md
 create mode 100644 papers/items/2026-2606-22678-rigorbench-benchmarking-engineering-process-discipline-in-autonomous-ai-coding-a.md
 create mode 100644 papers/items/2026-2606-22737-groundeval-a-deterministic-replacement-for-llm-as-judge-in-stateful-agent-evalua.md
 create mode 100644 papers/items/2026-2606-22741-grade-graph-representation-of-llm-agent-dependency-and-execution.md
 create mode 100644 papers/items/2026-2606-22844-ramem-contextual-reinstatement-for-long-term-agentic-memory.md
 create mode 100644 papers/items/2026-2606-22864-when-auc-0-998-is-not-enough-a-candidate-evaluation-protocol-for-hidden-state-pr.md
 create mode 100644 papers/items/2026-2606-22948-envs-environment-native-verified-search-for-long-horizon-gui-agents.md
 create mode 100644 papers/items/2026-2606-22953-plans-don-t-persist-why-context-management-is-load-bearing-for-llm-agents.md
 create mode 100644 papers/items/2026-2606-23032-ipo-finance-agent-benchmark-of-llm-financial-analysts-beyond-finance-agent-v2-wi.md
 create mode 100644 papers/items/2026-2606-23130-understanding-the-in-security-of-vibe-coded-applications.md
 create mode 100644 papers/items/2026-2606-23195-memory-contagion-cross-temporal-propagation-of-evaluator-bias-via-agent-memory.md
 create mode 100644 papers/items/2026-2606-23283-towards-root-memories-benchmarking-and-enhancing-implicit-logical-memory-retriev.md
 create mode 100644 papers/items/2026-2606-23327-videoagent-all-in-one-framework-for-video-understanding-and-editing.md
 create mode 100644 papers/items/2026-2606-23343-group-selection-promotes-prosocial-prompts-in-populations-of-llm-agents.md
 create mode 100644 papers/items/2026-2606-23565-holoagent-0-a-unified-embodied-agent-framework-with-3d-spatial-memory.md
 create mode 100644 papers/items/2026-2606-23664-mas-promptbench-when-does-prompt-optimization-improve-multi-agent-llm-systems.md
 create mode 100644 papers/items/2026-2606-23752-esaa-conversational-an-event-sourced-memory-layer-for-continuity-handoff-and-cur.md
 create mode 100644 papers/items/2026-2606-23764-emergent-relational-order-in-llm-agent-societies-from-collective-affect-to-autho.md
 create mode 100644 papers/items/2026-2606-23927-rift-bench-dynamic-red-teaming-for-agentic-ai-systems.md
 create mode 100644 papers/items/2026-2606-23991-critique-of-agent-model.md
 create mode 100644 papers/items/2026-2606-24193-skychain-intelligence-a-blockchain-secured-multi-agent-drl-framework-for-low-alt.md
 create mode 100644 papers/items/2026-2606-24235-sp-mind-an-autonomous-reasoning-agent-for-spatial-proteomics-analysis.md
 create mode 100644 papers/items/2026-2606-24322-securing-llm-agent-long-term-memory-against-poisoning-non-malleable-origin-bound.md
 create mode 100644 papers/items/2026-2606-24402-poisoned-playbooks-demystifying-knowledge-poisoning-effects-on-ai-security-agent.md
 create mode 100644 papers/items/2026-2606-24416-agentic-ai-for-bilevel-long-term-optimization-of-policy-driven-physical-layer-sy.md
 create mode 100644 papers/items/2026-2606-24437-rem-moa-reasoning-memory-sustains-mixture-of-agents-scaling.md
 create mode 100644 papers/items/2026-2606-24453-bayesian-control-for-coding-agents.md
 create mode 100644 papers/items/2026-2606-24515-reinforcement-learning-for-computer-use-agents-with-autonomous-evaluation.md
 create mode 100644 papers/items/2026-2606-24525-viscritic-visual-state-comparison-as-process-reward-for-gui-agents.md
 create mode 100644 papers/items/2026-2606-24535-governed-shared-memory-for-multi-agent-llm-systems.md
 create mode 100644 papers/items/2026-2606-24551-gui-vs-cli-execution-bottlenecks-in-screen-only-and-skill-mediated-computer-use-.md
 create mode 100644 papers/items/2026-2606-24595-memprobe-probing-long-term-agent-memory-via-hidden-user-state-recovery.md
 create mode 100644 papers/items/2026-2606-24597-qwen-agentworld-language-world-models-for-general-agents.md
 create mode 100644 papers/items/2026-2606-24623-privacy-preserving-rag-via-multi-agent-semantic-rewriting-achieving-confidential.md
 create mode 100644 papers/items/2026-2606-24626-safari-scaling-long-horizon-agentic-fault-attribution-via-active-investigation.md
 create mode 100644 papers/items/2026-2606-24649-agentic-collaborative-cognition-for-zero-shot-3d-understanding.md
 create mode 100644 papers/items/2026-2606-24689-automated-summarization-of-software-documents-an-llm-based-multi-agent-approach.md
 create mode 100644 papers/items/2026-2606-24694-supplynet-supporting-visual-exploratory-learning-in-supply-chain-via-contextual-.md
 create mode 100644 papers/items/2026-2606-24775-are-we-ready-for-an-agent-native-memory-system.md
 create mode 100644 papers/items/2026-2606-24779-deepbd-a-grounded-agentic-workflow-for-variant-prioritization-and-diagnosis-of-g.md
 create mode 100644 papers/items/2026-2606-24820-sherloc-structured-diagnostic-localization-for-code-repair-agents.md
 create mode 100644 papers/items/2026-2606-24839-grading-the-grader-lessons-from-evaluating-an-agentic-data-analysis-system.md
 create mode 100644 papers/items/2026-2606-24855-openthoughts-agent-data-recipes-for-agentic-models.md
 create mode 100644 papers/items/2026-2606-24937-the-hitchhiker-s-guide-to-agentic-ai-from-foundations-to-systems.md
 create mode 100644 papers/items/2026-2606-24976-diagnosing-and-mitigating-compounding-failures-in-agentic-persuasion-via-taxonom.md
 create mode 100644 papers/items/2026-2606-25115-forget-to-improve-on-device-llm-agent-continual-learning-via-budget-curated-memo.md
 create mode 100644 papers/items/2026-2606-25139-buildrix-an-open-platform-for-sharing-and-benchmarking-agentic-ai-skills-in-buil.md
 create mode 100644 papers/items/2026-2606-25161-trustmem-learning-trustworthy-memory-consolidation-for-llm-agents-with-long-term.md
 create mode 100644 papers/items/2026-2606-25189-actplane-programmable-os-level-policy-enforcement-for-agent-harnesses.md
 create mode 100644 papers/items/2026-2606-25191-to-isolate-or-to-score-model-adaptive-assessment-for-cost-efficient-multi-agent-.md
 create mode 100644 papers/items/2026-2606-25195-sok-ai-secure-code-generation-progress-pitfalls-and-paths-forward.md
 create mode 100644 papers/items/2026-2606-25206-raven-long-horizon-reasoning-navigation-with-a-visuo-spatio-temporal-memory.md
 create mode 100644 papers/items/2026-2606-25334-bridging-the-post-discharge-gap-a-traceable-multi-agent-framework-for-safe-and-c.md
 create mode 100644 papers/items/2026-2606-25358-agentic-knowledge-tracing-a-multi-agent-llm-architecture-for-stealth-assessment-.md
 create mode 100644 papers/items/2026-2606-25361-memory-makes-the-difference-evaluating-how-different-memory-roles-shape-conversa.md
 create mode 100644 papers/items/2026-2606-25400-brainagent-a-large-language-model-driven-multi-agent-framework-for-autonomous-br.md
 create mode 100644 papers/items/2026-2606-25484-from-causal-discovery-to-implementation-an-agentic-ai-framework-for-e-scooter-mo.md
 create mode 100644 papers/items/2026-2606-25514-unlocking-model-potentials-through-adaptive-multi-agent-scaffolding-for-efficien.md
 create mode 100644 papers/items/2026-2606-25519-quantization-inflates-reasoning-token-inflation-as-a-hidden-cost-of-low-bit-reas.md
 create mode 100644 papers/items/2026-2606-25588-intenttester-intent-driven-multi-agent-framework-for-cross-library-test-migratio.md
 create mode 100644 papers/items/2026-2606-25622-probabilistic-agents-in-deterministic-audits-evaluating-multi-agent-systems-for-.md
 create mode 100644 papers/items/2026-2606-25651-medguards-multi-agent-system-for-reliable-medical-error-detection-and-correction.md
 create mode 100644 papers/items/2026-2606-25656-is-graphrag-needed-from-basic-rag-to-graph-agentic-solutions-with-context-optimi.md
 create mode 100644 papers/items/2026-2606-25705-gui-agent-guided-exploration-of-user-sensitive-screens.md
 create mode 100644 papers/items/2026-2606-25760-uncertainty-quantification-for-computer-use-agents-a-benchmark-across-vision-lan.md
 create mode 100644 papers/items/2026-2606-25819-beyond-function-calling-benchmarking-tool-using-agents-under-tool-environment-un.md
 create mode 100644 papers/items/2026-2606-25899-manipulation-is-task-dependent-a-multi-axis-multi-environment-evaluation-of-fron.md
 create mode 100644 papers/items/2026-2606-26057-the-unfireable-safety-kernel-execution-time-ai-alignment-for-ai-agents-and-other.md
 create mode 100644 papers/items/2026-2606-26203-agentic-analysis-for-agentic-infrastructure-an-llm-powered-pipeline-for-comparat.md
 create mode 100644 papers/items/2026-2606-26205-knowledge-augmented-agentic-ai-for-mental-health-medication-information-seeking.md
 create mode 100644 papers/items/2026-2606-26216-cyberchainbench-can-ai-agents-secure-smart-contracts-against-real-world-on-chain.md
 create mode 100644 papers/items/2026-2606-26300-the-verification-horizon-no-silver-bullet-for-coding-agent-rewards.md
 create mode 100644 papers/items/2026-2606-26346-how-do-tool-augmented-llm-agents-perform-on-real-world-energy-analytics-tasks.md
 create mode 100644 papers/items/2026-2606-26356-instruction-bleed-cross-module-interference-in-prompt-composed-agentic-systems.md
 create mode 100644 papers/items/2026-2606-26403-profilefoundry-a-synthetic-person-object-substrate-for-privacy-memory-and-tool-u.md
 create mode 100644 papers/items/2026-2606-26453-optimizing-cuda-like-a-human-micro-profiling-tools-as-expert-surrogates-for-llm-.md
 create mode 100644 papers/items/2026-2606-26479-adaptive-evaluation-of-out-of-band-defenses-against-prompt-injection-in-llm-agen.md
 create mode 100644 papers/items/2026-2606-26511-temporal-validity-in-retrieval-memory-eliminating-stale-fact-errors-for-ai-agent.md
 create mode 100644 papers/items/2026-2606-26524-vigil-runtime-enforcement-of-behavioral-specifications-in-ai-agent-skills.md
 create mode 100644 papers/items/2026-2606-26614-hilsva-design-and-evaluation-of-a-human-in-the-loop-agentic-system-for-scientifi.md
 create mode 100644 papers/items/2026-2606-26627-agents-that-know-too-much-a-data-centric-survey-of-privacy-in-llm-agents.md
 create mode 100644 papers/items/2026-2606-26649-autoformalization-of-agent-instructions-into-policy-as-code.md
 create mode 100644 papers/items/2026-2606-26721-knowledge-based-pull-requests-a-trusted-workflow-for-agent-mediated-knowledge-co.md
 create mode 100644 papers/items/2026-2606-26758-egg-an-expert-guided-agent-framework-for-kernel-generation.md
 create mode 100644 papers/items/2026-2606-26793-mirror-novelty-constrained-memory-guided-mcts-red-teaming-for-agentic-rag.md
 create mode 100644 papers/items/2026-2606-26806-memory-depth-not-memory-access-selective-parametric-consolidation-for-long-runni.md
 create mode 100644 papers/items/2026-2606-26883-econsimulacra-a-digital-twin-platform-of-socio-economic-systems-powered-by-llm-a.md
 create mode 100644 papers/items/2026-2606-26918-diagnosing-task-insensitivity-in-language-agents.md
 create mode 100644 papers/items/2026-2606-26924-a-deterministic-control-plane-for-llm-coding-agents.md
 create mode 100644 papers/items/2026-2606-26960-toward-agentic-sysadmin-rethinking-system-administration-with-ai-agents.md
 create mode 100644 papers/items/2026-2606-27009-semantic-early-stopping-for-iterative-llm-agent-loops.md
 create mode 100644 papers/items/2026-2606-27154-openrca-2-0-from-outcome-labels-to-causal-process-supervision.md
 create mode 100644 papers/items/2026-2606-27243-nova-a-verification-aware-agent-harness-for-architecture-evolution-in-industrial.md
 create mode 100644 papers/items/2026-2606-27330-empowering-gui-agents-via-autonomous-experience-exploration-and-hindsight-experi.md
 create mode 100644 papers/items/2026-2606-27350-chia-an-open-source-framework-for-principled-agentic-ai-driven-hardware-software.md
 create mode 100644 papers/items/2026-2606-27397-sidconarena-an-environment-evaluating-agents-in-open-ended-positive-sum-bargaini.md
 create mode 100644 papers/items/2026-2606-27406-towards-evaluation-of-implicit-software-world-models-in-coding-llms.md
 create mode 100644 papers/items/2026-2606-27416-glite-arf-verifier-driven-research-with-parallel-llm-coding-agents.md
 create mode 100644 papers/items/2026-2606-27472-supersede-diagnosing-and-training-the-memory-update-gap-in-llm-agents.md
 create mode 100644 papers/items/2026-2606-27483-internalizing-the-future-a-unified-agentic-training-paradigm-for-world-model-pla.md
 create mode 100644 papers/items/2026-2606-27492-queenbee-planner-skill-evolving-communication-topologies-for-token-efficient-llm.md
 create mode 100644 papers/items/2026-2606-27499-dmv-bench-diagnosing-long-horizon-multimodal-agents-visual-memory-with-incidenta.md
 create mode 100644 papers/items/2026-2606-27632-yuvion-llm-an-adversarially-aware-large-language-model-for-content-and-ai-safety.md
 create mode 100644 papers/items/2026-2606-27806-agent-vs-parametric-world-models-hybrid-planning-for-reliable-language-agents.md
 create mode 100644 papers/items/2026-2606-27929-when-multi-robot-systems-meet-agentic-ai-towards-embodied-collective-intelligenc.md
 create mode 100644 papers/items/2026-2606-27990-advancedshellm-a-stateful-multi-agent-llm-honeypot-for-ssh-deception.md
 create mode 100644 papers/items/2026-2606-28011-from-detection-to-action-using-llm-agents-for-fault-tolerant-control.md
 create mode 100644 papers/items/2026-2606-28061-toolprivacybench-benchmarking-purpose-bound-privacy-in-tool-using-llm-agents.md
 create mode 100644 papers/items/2026-2606-28182-llawco-learning-laws-of-cooperation-for-modeling-embodied-multi-agent-behavior.md
 create mode 100644 papers/items/2026-2606-28187-gbc-gradient-based-connections-for-optimizing-multi-agent-systems.md
 create mode 100644 papers/items/2026-2606-28270-agent-native-immune-system-architecture-taxonomy-and-engineering.md
 create mode 100644 papers/items/2026-2606-28279-agentic-hardware-design-as-repository-level-code-evolution.md
 create mode 100644 papers/items/2026-2606-28349-hmars-a-hierarchical-multi-agent-memory-system-for-long-context-reasoning.md
 create mode 100644 papers/items/2026-2606-28360-carolina-guide-a-multi-agent-rag-system-with-institutional-guardrails-for-academ.md
 create mode 100644 papers/items/2026-2606-28374-recursive-self-evolving-agents-via-held-out-selection.md
 create mode 100644 papers/items/2026-2606-28409-evidence-driven-llm-agent-for-c-to-synthesizable-c-conversion-and-verification.md
 create mode 100644 papers/items/2026-2606-28425-tool-use-enables-undetectable-steganography-in-multi-agent-llm-systems.md
 create mode 100644 papers/items/2026-2606-28430-building-to-the-test-coding-agents-deliver-what-you-check-not-what-you-requested.md
 create mode 100644 papers/items/2026-2606-28434-swe-mem-learning-adaptive-memory-management-for-long-horizon-coding-agents.md
 create mode 100644 papers/items/2026-2606-28436-dockerless-environment-free-program-verifier-for-coding-agents.md
 create mode 100644 papers/items/2026-2606-28450-llm-agents-security-duality-a-comprehensive-survey-of-self-security-and-empowere.md
 create mode 100644 papers/items/2026-2606-28456-is-lying-an-emergent-behaviour-in-llms-evidence-from-gaslighting-ai-agents-in-a-.md
 create mode 100644 papers/items/2026-2606-28467-an-agentic-ai-pipeline-for-appliance-level-energy-anomaly-detection-and-llm-driv.md
 create mode 100644 papers/items/2026-2606-28480-tua-bench-a-benchmark-for-general-purpose-terminal-use-agents.md
 create mode 100644 papers/items/2026-2606-28570-digitizing-coaching-intelligence-an-agentic-framework-for-holistic-athlete-profi.md
 create mode 100644 papers/items/2026-2606-28666-why-trust-your-agent-empirical-security-gains-from-trism-guided-agentic-workflow.md
 create mode 100644 papers/items/2026-2606-28679-capability-gates-are-not-authorization-confused-deputy-failures-in-llm-agent-fra.md
 create mode 100644 papers/items/2026-2606-28692-an-ai-agent-for-treatment-reasoning-over-a-biomedical-tool-universe.md
 create mode 100644 papers/items/2026-2606-28733-agentic-abstention-do-agents-know-when-to-stop-instead-of-act.md
 create mode 100644 papers/items/2026-2606-28739-agent-safety-is-action-alignment.md
 create mode 100644 papers/items/2026-2606-28781-hyphaedb-a-living-knowledge-topology-for-agent-first-memory.md
 create mode 100644 papers/items/2026-2606-28791-from-determinism-to-delegation-ai-native-software-engineering-and-the-evolution-.md
 create mode 100644 papers/items/2026-2606-28839-the-contagion-tensor-a-framework-for-measuring-output-distribution-coupling-in-m.md
 create mode 100644 papers/items/2026-2606-28841-lamp-lean-based-agentic-framework-with-mcp-and-proof-repair.md
 create mode 100644 papers/items/2026-2606-28896-a-task-driven-and-quality-assured-agent-framework-for-sar-data-generation.md
 create mode 100644 papers/items/2026-2606-28925-multi-agent-routing-as-set-valued-prediction-a-wildchat-benchmark-and-cost-aware.md
 create mode 100644 papers/items/2026-2606-28958-when-latent-agents-lie-kv-cache-integrity-in-multi-agent-llm-collaboration.md
 create mode 100644 papers/items/2026-2606-29014-customized-generative-ai-agent-for-transportation-engineering-practice-a-develop.md
 create mode 100644 papers/items/2026-2606-29026-preventing-error-propagation-in-multi-agent-ai-through-runtime-monitoring.md
 create mode 100644 papers/items/2026-2606-29030-memory-as-an-attack-surface-in-llm-agents-a-study-on-multiple-choice-question-an.md
 create mode 100644 papers/items/2026-2606-29116-characterizing-large-language-model-agentic-workflows-a-study-on-n8n-ecosystem.md
 create mode 100644 papers/items/2026-2606-29142-agent-security-meets-regulatory-reality-a-practitioner-systematization-of-autono.md
 create mode 100644 papers/items/2026-2606-29178-selective-memory-retention-for-long-horizon-llm-agents.md
 create mode 100644 papers/items/2026-2606-29193-a-multi-dataset-benchmark-for-evaluating-llm-agents-in-microservice-failure-diag.md
 create mode 100644 papers/items/2026-2606-29225-policyguard-a-dialogue-grounded-sub-agent-verifier-for-policy-adherence-in-llm-a.md
 create mode 100644 papers/items/2026-2606-29270-minority-sentinel-when-to-overturn-majority-voting-in-multi-agent-llm-debates.md
 create mode 100644 papers/items/2026-2606-29315-hierarchical-experimentalist-agents.md
 create mode 100644 papers/items/2026-2606-29354-when-llms-develop-languages-symbolic-communication-for-efficient-multi-agent-rea.md
 create mode 100644 papers/items/2026-2606-29445-bridging-videoqa-and-video-guided-agentic-tasks-via-generalized-keyframe-extract.md
 create mode 100644 papers/items/2026-2606-29495-cognitive-world-models-for-process-level-social-influence-evaluation.md
 create mode 100644 papers/items/2026-2606-29537-osworld2-0-benchmarking-computer-use-agents-on-long-horizon-real-world-tasks.md
 create mode 100644 papers/items/2026-2606-29654-budgeted-act-or-defer-multi-agent-llm-deliberation-with-local-reliability-bounds.md
 create mode 100644 papers/items/2026-2606-29719-a-diagnostic-framework-and-multi-evaluator-audit-of-evaluator-driven-preference-.md
 create mode 100644 papers/items/2026-2606-29742-microagent-context-augmented-multi-agent-framework-for-automatic-microservice-de.md
 create mode 100644 papers/items/2026-2606-29745-echo-learning-epistemically-adaptive-language-agents-with-turn-level-credit.md
 create mode 100644 papers/items/2026-2606-29746-deepmed-search-an-open-source-agentic-platform-for-medical-deep-research-with-in.md
 create mode 100644 papers/items/2026-2606-29762-do-recommendation-algorithms-work-when-users-are-llm-agents-a-case-study-on-molt.md
 create mode 100644 papers/items/2026-2606-29771-clqt-a-closed-loop-cost-aware-strategy-consistent-benchmark-for-diagnostic-evalu.md
 create mode 100644 papers/items/2026-2606-29774-analytic-concept-centric-memory-for-agentic-embodied-manipulation.md
 create mode 100644 papers/items/2026-2606-29778-mandol-an-agglomerative-agent-memory-system-for-long-term-conversations.md
 create mode 100644 papers/items/2026-2606-29788-memleak-diagnosing-information-leaks-in-multimodal-agent-memory.md
 create mode 100644 papers/items/2026-2606-29824-neural-procedural-memory-empowering-llm-agents-with-implicit-activation-steering.md
 create mode 100644 papers/items/2026-2606-29894-saber-math-automated-benchmark-for-information-retrieval-evaluation-in-mathemati.md
 create mode 100644 papers/items/2026-2606-29914-memdelta-controlled-baselines-and-hidden-confounds-in-agent-memory-evaluation.md
 create mode 100644 papers/items/2026-2606-29932-saga-scene-aware-goal-evolving-agents-for-long-horizon-civrealm-strategy-plannin.md
 create mode 100644 papers/items/2026-2606-29957-swe-together-evaluating-coding-agents-in-interactive-user-sessions.md
 create mode 100644 papers/items/2026-2606-29961-duomem-towards-capable-on-device-memory-agents-via-dual-space-distillation.md
 create mode 100644 papers/items/2026-2606-30005-llm-agents-are-latent-context-managers-eliciting-self-managed-context-via-a-prop.md
 create mode 100644 papers/items/2026-2606-30111-automating-the-design-of-embodied-agent-architectures.md
 create mode 100644 papers/items/2026-2606-30119-on-the-internet-nobody-knows-you-re-an-llm-bot-unmasking-web-agents-with-multi-l.md
 create mode 100644 papers/items/2026-2606-30185-dynamo-dynamic-skill-tool-evolution-for-vision-language-agents.md
 create mode 100644 papers/items/2026-2606-30251-taco-tool-augmented-credit-optimization-for-agentic-tool-use.md
 create mode 100644 papers/items/2026-2606-30259-multi-agentic-system-leveraging-open-source-llms-to-mitigate-disinformation-thre.md
 create mode 100644 papers/items/2026-2606-30266-towards-continual-motion-language-agents-lora-variants-for-incremental-motion-un.md
 create mode 100644 papers/items/2026-2606-30294-rehearsed-multi-agent-live-product-demonstrations-with-real-time-voice-question-.md
 create mode 100644 papers/items/2026-2606-30383-whose-side-is-your-agent-on-multi-party-principal-loyalty-in-llm-agents.md
 create mode 100644 papers/items/2026-2606-30454-collective-cooperation-without-individual-fidelity-in-llm-agents.md
 create mode 100644 papers/items/2026-2606-30524-the-illusion-of-agentic-complexity-in-readme-md-generation-evaluating-single-age.md
 create mode 100644 papers/items/2026-2606-30546-mas-lab-a-specification-driven-validation-framework-for-reliable-multi-agent-sys.md
 create mode 100644 papers/items/2026-2606-30555-linguistic-firewall-geometry-as-defense-in-multi-agent-systems-routing.md
 create mode 100644 papers/items/2026-2606-30560-tracelab-characterizing-coding-agent-workloads-for-llm-serving.md
 create mode 100644 papers/items/2026-2606-30566-forensic-trajectory-signatures-for-agent-memory-poisoning-detection.md
 create mode 100644 papers/items/2026-2606-30573-swe-interact-reimagining-swe-benchmarks-as-user-driven-long-horizon-coding-sessi.md
 create mode 100644 papers/items/2026-2606-30602-mesa-prioritizing-vulnerable-communication-channels-for-securing-multi-agent-sys.md
 create mode 100644 papers/items/2026-2606-30616-scaling-the-horizon-not-the-parameters-reaching-trillion-parameter-performance-w.md
 create mode 100644 papers/items/2026-2606-30639-self-evolving-world-models-for-llm-agent-planning.md
 create mode 100644 papers/items/2026-2606-30697-lumos-a-semantic-operating-system-layer-for-accessibility-grounded-ai-agents.md
 create mode 100644 papers/items/2026-2606-30755-understanding-and-evaluating-claw-like-agent-security-through-a-computer-systems.md
 create mode 100644 papers/items/2026-2606-30840-contrastive-reflection-for-iterative-prompt-optimization.md
 create mode 100644 papers/items/2026-2606-30877-a-systematic-approach-to-multi-agent-ai-from-advanced-regulatory-control-theory-.md
 create mode 100644 papers/items/2026-2606-30887-training-therapeutic-judges-and-multi-agent-systems-for-human-aligned-mental-hea.md
 create mode 100644 papers/items/2026-2606-30906-investigating-multi-agent-deliberation-in-law.md
 create mode 100644 papers/items/2026-2606-30949-agrefactor-self-evolving-agentic-workflow-for-hls-compatibility-and-performance.md
 create mode 100644 papers/items/2026-2606-30970-behavioral-governance-for-autonomous-ai-agents-the-agentbound-framework.md
 create mode 100644 papers/items/2026-2606-30986-the-organizational-behavior-of-agentic-ai-collective-intelligence-in-human-agent.md
 create mode 100644 papers/items/2026-2606-31046-openlife-toward-open-world-artificial-life-with-autonomous-llm-agents.md
 create mode 100644 papers/items/2026-2606-31073-multiuav-plat-an-llm-oriented-platform-benchmark-and-framework-for-multi-uav-col.md
 create mode 100644 papers/items/2026-2606-31085-ddiagents-mechanism-conditioned-context-flow-for-drug-drug-interaction-predictio.md
 create mode 100644 papers/items/2026-2606-31134-beyond-the-library-an-agentic-framework-for-autoformalizing-research-mathematics.md
 create mode 100644 papers/items/2026-2606-31174-clawarena-team-benchmarking-subagent-orchestration-and-dynamic-workflows-in-lang.md
 create mode 100644 papers/items/2026-2606-31179-healthagentbench-a-unified-benchmark-suite-of-realistic-agentic-healthcare-envir.md
 create mode 100644 papers/items/2026-2606-31200-agentic-rag-vlm-affordance-aware-retrieval-augmented-generation-with-self-reflec.md
 create mode 100644 papers/items/2026-2606-31209-long-term-traffic-simulation-via-structured-autoregressive-modeling.md
 create mode 100644 papers/items/2026-2606-31227-securing-the-ai-agent-a-unified-framework-for-multi-layer-agent-red-teaming.md
 create mode 100644 papers/items/2026-2606-31229-agentic-ideation-sample-efficient-agentic-trajectories-synthesis-for-scientific-.md
 create mode 100644 papers/items/2026-2606-31252-embodied-cad-solver-grounded-llm-agents-for-parametric-b-rep-assembly-modeling.md
 create mode 100644 papers/items/2026-2606-31314-a-novel-method-for-differential-algebraic-dynamic-model-discovery-in-power-syste.md
 create mode 100644 papers/items/2026-2606-31339-verification-gated-agentic-mission-state-governance-for-intelligent-industrial-m.md
 create mode 100644 papers/items/2026-2606-31410-xiaomi-gui-0-technical-report.md
 create mode 100644 papers/items/2026-2606-31471-think-while-you-map-asynchronous-vision-language-agents-for-incremental-3d-scene.md
 create mode 100644 papers/items/2026-2606-31612-what-memory-do-gui-agents-really-need-from-passive-records-to-active-task-drivin.md
 create mode 100644 papers/items/2026-2606-31635-a-tutorial-on-autonomous-fault-tolerant-control-using-knowledge-grounded-llm-age.md
 create mode 100644 papers/items/2026-2606-31639-a-lifecycle-and-application-stack-survey-of-large-language-model-vulnerabilities.md
 create mode 100644 papers/items/2026-2606-31648-think-in-english-answer-in-korean-efficient-adaptation-of-multilingual-tool-usin.md
 create mode 100644 papers/items/2026-2606-31650-echo-prune-to-act-trace-to-learn-with-selective-turn-memory-in-agentic-rl.md
 create mode 100644 papers/items/2026-2606-31665-forecastagentsearch-towards-a-multi-expert-agent-search-system-for-geopolitical-.md
 create mode 100644 papers/items/2026-2606-31693-shopx-a-foundation-model-for-intent-to-item-fulfillment-in-agentic-shopping.md
 create mode 100644 papers/items/2026-2606-31744-a-conversational-agentic-interface-to-physics-based-household-digital-twins-for-.md
 create mode 100644 papers/items/2026-2606-31767-jeto-bench-a-reproducible-benchmark-for-execution-time-improvement-patches-in-ja.md
 create mode 100644 papers/items/2026-2606-31831-an-agentic-ai-framework-to-accelerate-scientific-discovery-in-plant-phenotyping.md
 create mode 100644 papers/items/2026-2606-31916-theory-of-mind-and-persuasion-beyond-conversation-assessing-the-capacity-of-llms.md
 create mode 100644 papers/items/2026-2606-31980-digitalcoach-communication-and-grounding-gaps-in-human-and-agentic-computer-use-.md
 create mode 100644 papers/items/2026-2606-32025-generative-skill-composition-for-llm-agents.md
 create mode 100644 papers/items/2026-2606-32034-qval-cheaply-evaluating-dense-supervision-signals-for-long-horizon-llm-agents.md
 create mode 100644 papers/items/2026-2607-00038-stop-hand-holding-your-coding-agent-engineering-the-loops-that-replace-step-by-s.md
 create mode 100644 papers/items/2026-2607-00041-atm-cid-brokered-pre-write-admission-for-multi-agent-code-co-synthesis.md
 create mode 100644 papers/items/2026-2607-00233-from-signals-to-structure-how-memory-architecture-drives-language-emergence-in-l.md
 create mode 100644 papers/items/2026-2607-00255-slm-llm-or-agentic-ai-toward-intelligent-uav-enabled-wpt-systems-in-low-altitude.md
 create mode 100644 papers/items/2026-2607-00297-epc-a-standardized-protocol-for-measuring-evaluator-preference-dynamics-in-llm-a.md
 create mode 100644 papers/items/2026-2607-00334-managed-autonomy-at-runtime-gear-based-safety-and-governance-for-single-and-mult.md
 create mode 100644 papers/items/2026-2607-00345-registry-governed-agent-lifecycle-completing-eddops-with-evaluation-drivenregist.md
 create mode 100644 papers/items/2026-2607-00407-personalization-as-inverse-planning-learning-latent-design-intents-for-agentic-s.md
 create mode 100644 papers/items/2026-2607-00422-kidnaprag-a-black-box-attack-for-hijacking-reasoning-in-agentic-retrieval-augmen.md
 create mode 100644 papers/items/2026-2607-00436-phreeqc-mcq-200-a-diagnostic-benchmark-for-tool-augmented-scientific-simulator-a.md
 create mode 100644 papers/items/2026-2607-00440-minos-a-multi-agent-collaborative-framework-for-provenance-based-backward-tracki.md
 create mode 100644 papers/items/2026-2607-00454-agri-sage-simulation-grounded-multi-agent-llm-for-context-aware-agricultural-adv.md
 create mode 100644 papers/items/2026-2607-00502-a-task-state-representation-for-long-horizon-mobile-gui-agents.md
 create mode 100644 papers/items/2026-2607-00555-rise-from-the-ashes-llm-based-static-analysis-for-deep-learning-framework-bugs.md
 create mode 100644 papers/items/2026-2607-00604-vehicle-routing-problem-meets-large-language-models-an-overview-and-perspectives.md
 create mode 100644 papers/items/2026-2607-00627-agi-maze-as-a-benchmark-framework-for-world-modeling-agents.md
 create mode 100644 papers/items/2026-2607-00692-self-gc-self-governing-context-for-long-horizon-llm-agents.md
 create mode 100644 papers/items/2026-2607-00911-from-registry-to-repository-how-ai-agent-skills-are-written-adapted-and-maintain.md
 create mode 100644 papers/items/2026-2607-00918-from-personas-to-plot-character-grounded-multi-agent-story-generation-for-long-f.md
 create mode 100644 papers/items/2026-2607-00939-leveraging-llm-based-agentic-systems-to-generate-quantum-applications-for-test-o.md
 create mode 100644 papers/items/2026-2607-00972-bayesian-uncertainty-propagation-for-agentic-rag-pipelines-a-proof-of-concept-st.md
 create mode 100644 papers/items/2026-2607-00990-swe-doctor-guiding-software-engineering-agents-with-runtime-diagnosis-from-multi.md
 create mode 100644 papers/items/2026-2607-01047-conversable-complexity-agentic-llm-collectives-as-interpretable-substrates.md
 create mode 100644 papers/items/2026-2607-01061-agentic-generation-of-verifiable-rules-for-deterministic-self-expanding-reaction.md
 create mode 100644 papers/items/2026-2607-01071-memsyco-bench-benchmarking-sycophancy-in-agent-memory.md
 create mode 100644 papers/items/2026-2607-01084-can-agents-generalize-to-the-open-world-unveiling-the-fragility-of-static-traini.md
 create mode 100644 papers/items/2026-2607-01120-next-generation-agentic-reinforcement-learning-systems-enable-self-evolving-agen.md
 create mode 100644 papers/items/2026-2607-01211-are-performance-optimization-benchmarks-reliably-measuring-coding-agents.md
 create mode 100644 papers/items/2026-2607-01213-reporescue-an-empirical-study-of-llm-agents-on-whole-repository-compatibility-re.md
 create mode 100644 papers/items/2026-2607-01366-auto-fl-research-agentic-search-for-federated-learning-algorithms.md
 create mode 100644 papers/items/2026-2607-01421-risk-architecture-for-ai-native-engineering-teams-an-organizational-framework-fo.md
 create mode 100644 papers/items/2026-2607-01510-janus-a-playground-for-user-involved-agentic-permission-management.md
 create mode 100644 papers/items/2026-2607-01523-multi-head-recurrent-memory-agents.md
 create mode 100644 papers/items/2026-2607-01531-opine-world-programmatic-world-modeling-with-ontology-error-prioritized-interact.md
 create mode 100644 papers/items/2026-2607-01600-boundary-sync-measuring-communication-induced-representational-coupling-in-multi.md
 create mode 100644 papers/items/2026-2607-01640-agentflow-building-agent-dependency-graphs-for-static-analysis-of-agent-programs.md
 create mode 100644 papers/items/2026-2607-01641-when-agents-do-not-stop-uncovering-infinite-agentic-loops-in-llm-agents.md
 create mode 100644 papers/items/2026-2607-01661-diverse-evidence-better-forecasts-multi-agent-deliberation-under-information-asy.md
 create mode 100644 papers/items/2026-2607-01668-verichat-an-agentic-conversational-ai-assistant-for-hardware-security-verificati.md
 create mode 100644 papers/items/2026-2607-01709-comfyclaw-self-evolving-skill-harnesses-for-image-generation-workflows.md
 create mode 100644 papers/items/2026-2607-01766-simworlds-a-multi-agent-system-for-dynamic-3d-scene-creation.md
 create mode 100644 papers/items/2026-2607-01767-repair-the-amplifier-not-the-symptom-stable-world-model-correction-for-agent-rol.md
 create mode 100644 papers/items/2026-2607-01788-krca-an-efficient-root-cause-analysis-system-in-hyper-scale-microservice-systems.md
 create mode 100644 papers/items/2026-2607-01793-safety-testing-llm-agents-at-scale-from-risk-discovery-to-evidence-grounded-veri.md
 create mode 100644 papers/items/2026-2607-01812-to-master-an-llm-agent-framework-for-automated-topology-optimization.md
 create mode 100644 papers/items/2026-2607-01874-skillcoach-self-evolving-rubrics-for-evaluating-and-enhancing-agentic-skill-use.md
 create mode 100644 papers/items/2026-2607-01916-contextsniper-anttrail-s-token-efficient-code-memory-for-repository-level-progra.md
 create mode 100644 papers/items/2026-2607-01929-beyond-textual-repository-exploration-dual-modal-structural-reasoning-for-agenti.md
 create mode 100644 papers/items/2026-2607-01935-a-tma-decoupling-state-aware-memory-failures-in-long-term-agent-memory.md
 create mode 100644 papers/items/2026-2607-02032-pace-a-proxy-for-agentic-capability-evaluation.md
 create mode 100644 papers/items/2026-2607-02186-ua-chatdev-uncertainty-aware-multi-agent-collaboration-for-reliable-software-dev.md
 create mode 100644 papers/items/2026-2607-02210-criticality-based-guard-rail-validation-for-ai-agent-decisions-in-autonomous-tel.md
 create mode 100644 papers/items/2026-2607-02245-copewell-a-multi-agent-swarm-architecture-for-equitable-mental-wellness-support.md
 create mode 100644 papers/items/2026-2607-02255-agenticsts-a-bounded-memory-testbed-for-long-horizon-llm-agents.md
 create mode 100644 papers/items/2026-2607-02294-coding-agents-are-guessing-measuring-action-boundary-violations-in-underspecifie.md
 create mode 100644 papers/items/2026-2607-02381-hulat2-at-mer-trans-2026-governed-multi-agent-simplification-for-spanish-easy-to.md
 create mode 100644 papers/items/2026-2607-02448-agentscad-automated-design-for-manufacturing-of-fdm-parts-via-multi-agent-llm-re.md
 create mode 100644 papers/items/2026-2607-02453-adoption-and-ecosystem-health-a-longitudinal-analysis-of-open-source-multi-agent.md
 create mode 100644 papers/items/2026-2607-02507-what-llm-agents-say-when-no-one-is-watching-social-structure-and-latent-objectiv.md
 create mode 100644 papers/items/2026-2607-02577-benchmarking-the-benchmarks-a-validity-audit-of-tool-calling-evaluation.md
 create mode 100644 papers/items/2026-2607-02579-when-not-to-write-memory-governing-false-promotion-from-correlated-agent-traces.md
 create mode 100644 papers/items/2026-2607-02599-agentltl-a-trace-verification-framework-for-measuring-enforcing-and-training-pro.md
 create mode 100644 papers/items/2026-2607-02606-chainswe-benchmarking-coding-agents-on-multi-bug-software-maintenance.md
 create mode 100644 papers/items/2026-2607-02684-can-coding-agents-implement-missed-compiler-optimizations-evaluating-llm-agents-.md
 create mode 100644 papers/items/2026-2607-02689-s-ember-a-large-scale-benchmark-for-streaming-egocentric-memory-retrieval.md
 create mode 100644 papers/items/2026-2607-02703-llmoxie-exploring-agentic-ai-for-scientific-software-development.md
 create mode 100644 papers/items/2026-2607-02716-evaluating-large-language-models-for-decision-making-in-agent-based-urban-mobili.md
 create mode 100644 papers/items/2026-2607-02807-swarmresearch-orchestrating-coding-agents-for-open-ended-discovery.md
 create mode 100644 papers/items/2026-2607-02846-object-centric-environment-modeling-for-agentic-tasks.md
 create mode 100644 papers/items/2026-2607-02857-mosaic-knowledge-guided-cli-command-composition-attack-in-llm-coding-agents.md
 create mode 100644 papers/items/2026-2607-02879-medcalc-pro-solving-complex-medical-calculations-with-llm-agents.md
 create mode 100644 papers/items/2026-2607-02882-diagnosis-driven-automatic-repair-for-agentic-workflow-via-symbolic-inference.md
 create mode 100644 papers/items/2026-2607-02911-coact-action-preserving-observation-compression-for-coding-agents.md
 create mode 100644 papers/items/2026-2607-02927-videosearcher-empowering-video-deep-research-with-multi-tool-agentic-reasoning-v.md
 create mode 100644 papers/items/2026-2607-02942-a-workflow-aware-serving-layer-for-agentic-applications.md
 create mode 100644 papers/items/2026-2607-03105-orbit-q-dual-axis-benchmarking-of-autonomous-agents-in-scientific-quantum-progra.md
 create mode 100644 papers/items/2026-2607-03162-apeb-benchmarking-personalization-ability-of-large-language-model-agents.md
 create mode 100644 papers/items/2026-2607-03220-contra-red-teaming-configurations-of-personalizable-agents.md
 create mode 100644 papers/items/2026-2607-03233-agentic-and-generative-ai-for-open-source-intelligence-and-cyber-investigations-.md
 create mode 100644 papers/items/2026-2607-03269-agentic-secpbft-agentic-ai-driven-proactive-security-framework-for-wireless-pbft.md
 create mode 100644 papers/items/2026-2607-03316-is-agentic-code-review-helpful-mining-developers-feedback-to-coderabbit-reviews-.md
 create mode 100644 papers/items/2026-2607-03333-spork-self-speculative-forking-to-accelerate-agentic-llm-inference.md
 create mode 100644 papers/items/2026-2607-03423-securing-multi-tool-ai-agent-chains-with-dynamic-real-time-compositional-policie.md
 create mode 100644 papers/items/2026-2607-03441-no-time-like-the-present-agentic-test-time-training-for-llm-agents.md
 create mode 100644 papers/items/2026-2607-03510-cage-1-control-assurance-and-governance-evaluation-for-enterprise-agentic-ai.md
 create mode 100644 papers/items/2026-2607-03525-gameenginebench-evaluating-coding-agents-on-real-c-runtime-environments.md
 create mode 100644 papers/items/2026-2607-03601-archeval-measuring-ai-agents-as-computer-architects.md
 create mode 100644 papers/items/2026-2607-03628-swarm-driven-multi-agent-reasoning-for-smart-city-security.md
 create mode 100644 papers/items/2026-2607-03691-don-t-blame-the-large-language-model-how-scaffolding-evolution-shapes-coding-age.md
 create mode 100644 papers/items/2026-2607-03695-social-networks-of-llm-agents.md
 create mode 100644 papers/items/2026-2607-03702-agent-reinforcement-learning-via-pivotal-aware-self-feedback-retry.md
 create mode 100644 papers/items/2026-2607-03726-selfmem-self-optimizing-memory-for-ai-agents.md
 create mode 100644 papers/items/2026-2607-03821-dualview-preventing-indirect-prompt-injection-in-personal-ai-agents.md
 create mode 100644 papers/items/2026-2607-03853-cograd-a-cognitively-inspired-multi-agent-framework-for-radiology-report-generat.md
 create mode 100644 papers/items/2026-2607-03953-the-remarkable-effectiveness-of-providing-ai-agents-with-natural-language-tools-.md
 create mode 100644 papers/items/2026-2607-03968-refused-in-chat-written-in-code-workflow-level-jailbreak-construction-in-ide-cod.md
 create mode 100644 papers/items/2026-2607-04009-physminer-an-agentic-ai-framework-for-discovering-turbulence-physics.md
 create mode 100644 papers/items/2026-2607-04034-the-i-don-t-know-filter-enhancing-agentic-reliability-in-function-calling.md
 create mode 100644 papers/items/2026-2607-04089-placemem-toward-a-compute-aware-memory-plane-for-lifelong-agents.md
 create mode 100644 papers/items/2026-2607-04149-beyond-scene-priors-fine-grained-traffic-scene-reasoning-with-benchmarking-and-q.md
 create mode 100644 papers/items/2026-2607-04162-ace-agentic-control-for-embodied-manipulation-via-zero-shot-workflow-reasoning.md
 create mode 100644 papers/items/2026-2607-04212-an-evaluation-of-role-based-multi-agent-code-generation-on-repository-scale-prob.md
 create mode 100644 papers/items/2026-2607-04219-agentic-iot-architectures-applications-and-challenges-toward-the-internet-of-age.md
 create mode 100644 papers/items/2026-2607-04240-biological-motifs-for-agentic-control.md
 create mode 100644 papers/items/2026-2607-04293-causalgame-benchmarking-causal-thinking-of-llm-agents-in-games.md
 create mode 100644 papers/items/2026-2607-04334-do-gui-agents-believe-their-eyes-diagnosing-state-belief-reliance-on-pixels-vers.md
 create mode 100644 papers/items/2026-2607-04391-memory-orchestrated-semantic-system-moss-an-auditable-agentic-memory-architectur.md
 create mode 100644 papers/items/2026-2607-04394-mechmath-agent-team-llm-driven-agents-for-mathematical-research.md
 create mode 100644 papers/items/2026-2607-04395-nki-agent-domain-specific-fine-tuning-and-agentic-tool-use-for-neuron-kernel-gen.md
 create mode 100644 papers/items/2026-2607-04426-ace-brain-0-5-a-unified-embodied-foundational-model-for-physical-agentic-ai.md
 create mode 100644 papers/items/2026-2607-04433-autonomous-information-seeking-a-roadmap-for-agentic-recommender-systems.md
 create mode 100644 papers/items/2026-2607-04470-regime-conditional-stabilisation-of-llm-augmented-cooperative-multi-agent-reinfo.md
 create mode 100644 papers/items/2026-2607-04528-measuring-harness-induced-belief-divergence-in-multi-step-llm-agents.md
 create mode 100644 papers/items/2026-2607-04569-llms-for-agentic-home-energy-management.md
 create mode 100644 papers/items/2026-2607-04617-mrms-a-multi-resolution-memory-substrate-for-long-lived-ai-agents.md
 create mode 100644 papers/items/2026-2607-04623-can-llms-really-recover-microservice-failures-a-recovery-aware-evaluation-of-dia.md
 create mode 100644 papers/items/2026-2607-04686-toolfailbench-diagnosing-tool-use-failures-in-llm-agents.md
 create mode 100644 papers/items/2026-2607-04697-ai-agent-pull-requests-on-github-frequency-structure-and-merge-conflict-rates.md
 create mode 100644 papers/items/2026-2607-04713-rspo-reward-swap-policy-optimization-for-multi-turn-llm-agents.md
 create mode 100644 papers/items/2026-2607-04963-stapo-selective-trajectory-aware-policy-optimization-for-llm-agent-training.md
 create mode 100644 papers/items/2026-2607-05001-tactic-kg-toward-small-agent-teams-for-cyber-threat-intelligence-knowledge-graph.md
 create mode 100644 papers/items/2026-2607-05029-your-agent-s-memories-are-not-its-own-forged-reasoning-attacks-on-llm-agent-memo.md
 create mode 100644 papers/items/2026-2607-05055-toward-trustworthy-large-language-model-agents-in-healthcare.md
 create mode 100644 papers/items/2026-2607-05120-agent-data-injection-attacks-are-realistic-threats-to-ai-agents.md
 create mode 100644 papers/items/2026-2607-05132-when-agents-lie-premeditation-persistence-and-exploitation-in-repeated-games.md
 create mode 100644 papers/items/2026-2607-05174-agentgym2-benchmarking-large-language-model-agents-in-de-idealized-real-world-en.md
 create mode 100644 papers/items/2026-2607-05188-latent-programming-horizons-in-coding-agents.md
 create mode 100644 papers/items/2026-2607-05202-evoagentbench-benchmarking-agent-self-evolution-via-ability-transfer.md
 create mode 100644 papers/items/2026-2607-05297-metaskill-evolve-recursive-self-improvement-of-llm-agents-via-two-timescale-meta.md
 create mode 100644 papers/items/2026-2607-05318-pisas-benchmarking-contextual-integrity-in-multi-user-agentic-systems.md
 create mode 100644 papers/items/2026-2607-05363-sovereignpa-bench-evaluating-user-owned-personal-agents-under-evolving-intent-pl.md
 create mode 100644 papers/items/2026-2607-05378-compactionrl-reinforcement-learning-with-context-compaction-for-long-horizon-age.md
 create mode 100644 papers/items/2026-2607-05391-llm-as-a-verifier-a-general-purpose-verification-framework.md
 create mode 100644 papers/items/2026-2607-05428-charlie-an-on-premise-multi-agent-retrieval-augmented-generation-system-for-evid.md
 create mode 100644 papers/items/2026-2607-05456-prompt-to-paper-agentic-ai-system-for-bioinformatics.md
 create mode 100644 papers/items/2026-2607-05458-learning-to-control-llm-agent-harnesses-with-offline-reinforcement-learning.md
 create mode 100644 papers/items/2026-2607-05518-aiauthz-off-host-identity-bound-authorization-for-ai-agents.md
 create mode 100644 papers/items/2026-2607-05659-agents-with-feelings-personality-and-emotion-in-multi-agent-software-teams.md
 create mode 100644 papers/items/2026-2607-05666-what-do-ai-agents-actually-change-an-empirical-taxonomy-of-mutation-patterns-in-.md
 create mode 100644 papers/items/2026-2607-05677-from-conversation-to-contribution-characterizing-coding-agent-in-open-source-sof.md
 create mode 100644 papers/items/2026-2607-05690-memory-in-the-loop-in-process-retrieval-as-extendedworking-memory-for-language-a.md
 create mode 100644 papers/items/2026-2607-05743-the-balkanization-of-execution-security-research-for-ai-coding-agents-isolation-.md
 create mode 100644 papers/items/2026-2607-05772-detecting-vulnerability-inducing-commits-via-multi-stage-reasoning-with-llm-base.md
 create mode 100644 papers/items/2026-2607-05773-beyond-static-evaluation-building-simulation-environments-for-scalable-agentic-r.md
 create mode 100644 papers/items/2026-2607-05775-beyond-the-leaderboard-a-synthesis-of-tool-use-planning-and-reasoning-failures-i.md
 create mode 100644 papers/items/2026-2607-05794-from-passive-retrieval-to-active-memory-navigation-learning-to-use-memory-as-a-s.md
 create mode 100644 papers/items/2026-2607-05805-onnes-a-physics-grounded-multi-agent-llm-simulator-for-cryogenic-fault-diagnosis.md
 create mode 100644 papers/items/2026-2607-05915-pcbworld-a-benchmark-environment-for-engine-grounded-pcb-design-automation.md
 create mode 100644 papers/items/2026-2607-06000-context-to-execution-integrity-for-llm-agents.md
 create mode 100644 papers/items/2026-2607-06001-information-limits-and-attractor-dynamics-in-economies-of-frontier-llm-agents-a-.md
 create mode 100644 papers/items/2026-2607-06008-polyworkbench-benchmarking-multilingual-long-horizon-llm-agents.md
 create mode 100644 papers/items/2026-2607-06080-from-blueprint-to-reality-modeling-and-applying-putnam-s-social-capital-theory-w.md
 create mode 100644 papers/items/2026-2607-06101-agents-that-teach-towards-designing-incidental-learning-back-into-ai-assisted-so.md
 create mode 100644 papers/items/2026-2607-06118-webretriever-a-large-scale-comprehensive-benchmark-for-efficient-web-agent-evalu.md
 create mode 100644 papers/items/2026-2607-06140-curateevo-data-curation-evolving-for-agentic-post-training.md
 create mode 100644 papers/items/2026-2607-06157-llm-agents-for-deliberative-collaboration-a-study-on-joint-decision-making-under.md
 create mode 100644 papers/items/2026-2607-06195-logichunter-testing-llm-agent-frameworks-with-an-agentic-oracle.md
 create mode 100644 papers/items/2026-2607-06223-information-gain-based-rollout-policy-optimization-an-adaptive-tree-structured-r.md
 create mode 100644 papers/items/2026-2607-06273-agenttether-graph-guided-diagnosis-and-runtime-intervention-for-reliable-llm-age.md
 create mode 100644 papers/items/2026-2607-06341-harnessing-code-agents-for-automatic-software-verification.md
 create mode 100644 papers/items/2026-2607-06411-rubench-a-repository-level-agentic-coding-benchmark-with-natively-authored-russi.md
 create mode 100644 papers/items/2026-2607-06413-an-experimental-design-approach-to-evaluating-agentic-ai-s-autonomous-model-disc.md
 create mode 100644 papers/items/2026-2607-06452-from-voting-to-agent-collaboration-answer-type-aware-llm-pipelines-for-bioasq-14.md
 create mode 100644 tools/collection/collect_arxiv.py
 create mode 100644 tools/collection/promote_arxiv_manifest.py

diff --git a/collection-runs/2026-07-08-arxiv-expanded-sweep.md b/collection-runs/2026-07-08-arxiv-expanded-sweep.md
new file mode 100644
index 0000000..e21f123
--- /dev/null
+++ b/collection-runs/2026-07-08-arxiv-expanded-sweep.md
@@ -0,0 +1,55 @@
+# Collection Run: 2026-07-08 arXiv Expanded Sweep
+
+status: completed
+
+## Scope
+
+- source: arXiv API
+- date window: 2025-07-08 to 2026-07-08
+- query groups: 15
+- unique candidates seen: 1506
+- high-relevance candidates promoted to paper items: 970
+- remaining reserve candidates: 536
+- new paper files written in this run: 968
+
+## Query Groups
+
+- llm-agent
+- language-agent
+- ai-agent
+- agentic-ai
+- agent-evaluation
+- agent-memory
+- tool-use
+- function-calling
+- coding-agent
+- web-gui-agent
+- multi-agent-llm
+- agent-safety
+- rag-agent
+- planning-agent
+- autonomous-agent-llm
+
+## Outputs
+
+- `data/arxiv-agent-candidates-2025-07-08-to-2026-07-08.json`
+- `data/arxiv-agent-papers-2025-07-08-to-2026-07-08.json`
+- `papers/corpus-summary-2026-07-08.md`
+- 968 generated `papers/items/*.md` files
+
+## Method Notes
+
+- The candidate manifest stores title, arXiv URL, authors, categories, dates, matched query groups, auto topics, score, and relevance.
+- It intentionally does not store abstracts, to avoid turning the repo into a raw abstract mirror.
+- Generated paper items are `queued`; they are collection records, not reviewed summaries.
+- Auto tags are recall-oriented and need later cleanup.
+- The first API pass selected the top 400; after inspecting score distribution, all 970 `relevance=high` candidates were promoted locally from the candidate manifest.
+- arXiv returned 429 during a later refresh attempt, so the collector now includes retry/backoff controls and the promotion step can run offline.
+
+## First Trend Takeaways
+
+- Evaluation and benchmarking dominate the corpus.
+- Memory is now a central systems topic, including governance, poisoning, provenance, and procedural memory.
+- Agent safety has moved toward runtime risks: tool leakage, prompt injection, data injection, memory poisoning, and action-boundary violations.
+- Coding-agent evaluation is moving beyond single-issue SWE-Bench tasks.
+- GUI/computer-use agents and workflow/enterprise agents are now large enough to deserve separate tracks.
diff --git a/data/arxiv-agent-candidates-2025-07-08-to-2026-07-08.json b/data/arxiv-agent-candidates-2025-07-08-to-2026-07-08.json
new file mode 100644
index 0000000..dc99bd0
--- /dev/null
+++ b/data/arxiv-agent-candidates-2025-07-08-to-2026-07-08.json
@@ -0,0 +1,53788 @@
+[
+  {
+    "id": "2606.08340",
+    "title": "Benchmarking Open-Ended Multi-Agent Coordination in Language Agents",
+    "url": "https://arxiv.org/abs/2606.08340",
+    "published": "2026-06-06",
+    "updated": "2026-06-06",
+    "authors": [
+      "Kale-ab Abebe Tessera",
+      "Andras Szecsenyi",
+      "Cameron Barker",
+      "Alexander Rutherford",
+      "Davide Paglieri",
+      "Aidan Scannell",
+      "Henry Gouk",
+      "Elliot J. Crowley",
+      "Tim Rocktäschel",
+      "Amos Storkey"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 28,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm",
+      "language-agent",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.08340",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.08340",
+    "pdf_url": "https://arxiv.org/pdf/2606.08340",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.20833",
+    "title": "MemGym: a Long-Horizon Memory Environment for LLM Agents",
+    "url": "https://arxiv.org/abs/2605.20833",
+    "published": "2026-05-20",
+    "updated": "2026-05-20",
+    "authors": [
+      "Wujiang Xu",
+      "Yu Wang",
+      "Kai Mei",
+      "Kaiqu Liang",
+      "Zhenting Wang",
+      "Mingyu Jin",
+      "Han Zhang",
+      "Shi-Xiong Zhang",
+      "Wenyue Hua",
+      "Sambit Sahu",
+      "Dimitris N. Metaxas"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 26,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.20833",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.20833",
+    "pdf_url": "https://arxiv.org/pdf/2605.20833",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2607.06008",
+    "title": "PolyWorkBench: Benchmarking Multilingual Long-Horizon LLM Agents",
+    "url": "https://arxiv.org/abs/2607.06008",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Hongliang Li",
+      "Yijin Liu",
+      "Zhiwei Zhang",
+      "Zihe Liu",
+      "Xinyue Lou",
+      "Jinan Xu",
+      "Fandong Meng",
+      "Kaiyu Huang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 25,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "llm-agent",
+      "planning-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.06008",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06008",
+    "pdf_url": "https://arxiv.org/pdf/2607.06008",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.28425",
+    "title": "Tool Use Enables Undetectable Steganography in Multi-Agent LLM Systems",
+    "url": "https://arxiv.org/abs/2606.28425",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Jimmy Laurence Rippin",
+      "Simon C. Marshall",
+      "David Demitri Africa",
+      "Christian Schroeder de Witt"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 25,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "ai-agent",
+      "autonomous-agent-llm",
+      "multi-agent-llm",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.28425",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28425",
+    "pdf_url": "https://arxiv.org/pdf/2606.28425",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.24937",
+    "title": "The Hitchhiker's Guide to Agentic AI: From Foundations to Systems",
+    "url": "https://arxiv.org/abs/2606.24937",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Haggai Roitman"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.IR",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "memory",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 25,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "multi-agent-llm",
+      "rag-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.24937",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24937",
+    "pdf_url": "https://arxiv.org/pdf/2606.24937",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.10749",
+    "title": "Toward Secure LLM Agents: Threat Surfaces, Attacks, Defenses, and Evaluation",
+    "url": "https://arxiv.org/abs/2606.10749",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Yuchen Ling",
+      "Shengcheng Yu",
+      "Zhenyu Chen",
+      "Chunrong Fang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 25,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.10749",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10749",
+    "pdf_url": "https://arxiv.org/pdf/2606.10749",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.29824",
+    "title": "Neural Procedural Memory: Empowering LLM Agents with Implicit Activation Steering",
+    "url": "https://arxiv.org/abs/2606.29824",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Chengfeng Zhao",
+      "Yuqiao Tan",
+      "Shizhu He",
+      "Yequan Wang",
+      "Jun Zhao",
+      "Kang Liu"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 24,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "agent-memory",
+      "autonomous-agent-llm",
+      "llm-agent",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.29824",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29824",
+    "pdf_url": "https://arxiv.org/pdf/2606.29824",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.28011",
+    "title": "From Detection to Action: Using LLM Agents for Fault-Tolerant Control",
+    "url": "https://arxiv.org/abs/2606.28011",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Javal Vyas",
+      "Milapji Singh Gill",
+      "Artan Markaj",
+      "Felix Gehlhoff",
+      "Mehmet Mercangöz"
+    ],
+    "categories": [
+      "eess.SY",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 24,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "llm-agent",
+      "multi-agent-llm",
+      "planning-agent",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.28011",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28011",
+    "pdf_url": "https://arxiv.org/pdf/2606.28011",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.20401",
+    "title": "PowerAgentBench-Dyn: A Benchmark for Agentic AI in Power System Dynamic Studies",
+    "url": "https://arxiv.org/abs/2606.20401",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Qian Zhang",
+      "Andrea Pomarico",
+      "Costas Mylonas",
+      "Magda Foti",
+      "Alberto Berizzi",
+      "Le Xie"
+    ],
+    "categories": [
+      "eess.SY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 24,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.20401",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20401",
+    "pdf_url": "https://arxiv.org/pdf/2606.20401",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.18789",
+    "title": "PowerAgentBench-SS: A Benchmark for Agentic AI in Power System Steady-State Studies",
+    "url": "https://arxiv.org/abs/2606.18789",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Costas Mylonas",
+      "Magda Foti",
+      "Andrea Pomarico",
+      "Matheus Duarte",
+      "Qian Zhang",
+      "Emmanouel Varvarigos"
+    ],
+    "categories": [
+      "eess.SY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 24,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "planning-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.18789",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18789",
+    "pdf_url": "https://arxiv.org/pdf/2606.18789",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.08274",
+    "title": "Toward Human-Centered Multi-Agent Systems: Integrating Cognition, Culture, Values, and Cooperation in AI Agents",
+    "url": "https://arxiv.org/abs/2606.08274",
+    "published": "2026-06-06",
+    "updated": "2026-06-06",
+    "authors": [
+      "Safia Baloch",
+      "Rahemeen Khan"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 24,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.08274",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.08274",
+    "pdf_url": "https://arxiv.org/pdf/2606.08274",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2607.05775",
+    "title": "Beyond the Leaderboard: A Synthesis of Tool-Use, Planning, and Reasoning Failures in Large Language Model Agents",
+    "url": "https://arxiv.org/abs/2607.05775",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Wael Albayaydh",
+      "Rui Zhao",
+      "Ivan Flechais"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "embodied-agent",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 23,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm",
+      "planning-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.05775",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05775",
+    "pdf_url": "https://arxiv.org/pdf/2607.05775",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.02255",
+    "title": "AgenticSTS: A Bounded-Memory Testbed for Long-Horizon LLM Agents",
+    "url": "https://arxiv.org/abs/2607.02255",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Xiangchen Cheng",
+      "Yunwei Jiang",
+      "Jianwen Sun",
+      "Zizhen Li",
+      "Chuanhao Li",
+      "Xiangcheng Cao",
+      "Yihao Liu",
+      "Fanrui Zhang",
+      "Li Jin",
+      "Kaipeng Zhang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 23,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.02255",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02255",
+    "pdf_url": "https://arxiv.org/pdf/2607.02255",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.28791",
+    "title": "From Determinism to Delegation: AI-Native Software Engineering and the Evolution of the Agentic Engineer",
+    "url": "https://arxiv.org/abs/2606.28791",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Mamdouh Alenezi"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "memory",
+      "multi-agent",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 23,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "autonomous-agent-llm",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.28791",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28791",
+    "pdf_url": "https://arxiv.org/pdf/2606.28791",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.26614",
+    "title": "HiLSVA: Design and Evaluation of a Human-in-the-Loop Agentic System for Scientific Visualization",
+    "url": "https://arxiv.org/abs/2606.26614",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Kuangshi Ai",
+      "Patrick Phuoc Do",
+      "Chaoli Wang"
+    ],
+    "categories": [
+      "cs.HC",
+      "cs.AI",
+      "cs.GR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 23,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.26614",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26614",
+    "pdf_url": "https://arxiv.org/pdf/2606.26614",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2605.08442",
+    "title": "Defense effectiveness across architectural layers: a mechanistic evaluation of persistent memory attacks on stateful LLM agents",
+    "url": "https://arxiv.org/abs/2605.08442",
+    "published": "2026-05-08",
+    "updated": "2026-07-03",
+    "authors": [
+      "Jun Wen Leong"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 23,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.08442",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.08442",
+    "pdf_url": "https://arxiv.org/pdf/2605.08442",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.06869",
+    "title": "Agentick: A Unified Benchmark for General Sequential Decision-Making Agents",
+    "url": "https://arxiv.org/abs/2605.06869",
+    "published": "2026-05-07",
+    "updated": "2026-05-12",
+    "authors": [
+      "Roger Creus Castanyer",
+      "Pablo Samuel Castro",
+      "Glen Berseth"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 23,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.06869",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.06869",
+    "pdf_url": "https://arxiv.org/pdf/2605.06869",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2607.03233",
+    "title": "Agentic and Generative AI for Open-Source Intelligence and Cyber Investigations: Taxonomy, Evaluation, Challenges, and Future Directions",
+    "url": "https://arxiv.org/abs/2607.03233",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Eduardo Almeida Palmieri",
+      "Mohamed Chahine Ghanem",
+      "Dipo Dunsin",
+      "Zubair Baig",
+      "Ed de Quincey",
+      "Kim-Kwang Raymond Choo"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.IR",
+      "cs.SI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 22,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "rag-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.03233",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03233",
+    "pdf_url": "https://arxiv.org/pdf/2607.03233",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.28061",
+    "title": "ToolPrivacyBench: Benchmarking Purpose-Bound Privacy in Tool-Using LLM Agents",
+    "url": "https://arxiv.org/abs/2606.28061",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Shijing Hu",
+      "Liang Liu",
+      "Zhu Meng",
+      "Zhicheng Zhao"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 22,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "function-calling",
+      "llm-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.28061",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28061",
+    "pdf_url": "https://arxiv.org/pdf/2606.28061",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.17459",
+    "title": "Can LLMs Be CEOs? Benchmarking Strategic Resource Reallocation with Multi-Role Agent Simulation",
+    "url": "https://arxiv.org/abs/2606.17459",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Yuyang Dai",
+      "Xueqing Peng",
+      "Lingfei Qian",
+      "Zhuohan Xie"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 22,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.17459",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17459",
+    "pdf_url": "https://arxiv.org/pdf/2606.17459",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.16613",
+    "title": "CoffeeBench: Benchmarking Long-Horizon LLM Agents in Heterogeneous Multi-Agent Economies",
+    "url": "https://arxiv.org/abs/2606.16613",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Issa Sugiura",
+      "Daichi Hattori",
+      "Kazuo Araragi",
+      "Keita Ogawa",
+      "Shota Onose",
+      "Taro Makino",
+      "Teppei Usuki",
+      "Takashi Ishida"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 22,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.16613",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16613",
+    "pdf_url": "https://arxiv.org/pdf/2606.16613",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.12945",
+    "title": "Learning What to Remember: A Cognitively Grounded Multi-Factor Value Model for Agentic Memory",
+    "url": "https://arxiv.org/abs/2606.12945",
+    "published": "2026-06-11",
+    "updated": "2026-06-20",
+    "authors": [
+      "Zhibao Chen",
+      "Qian Cheng"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 22,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.12945",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12945",
+    "pdf_url": "https://arxiv.org/pdf/2606.12945",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.06399",
+    "title": "CollabSim: A CSCW-Grounded Methodology for Investigating Collaborative Competence of LLM Agents through Controlled Multi-Agent Experiments",
+    "url": "https://arxiv.org/abs/2606.06399",
+    "published": "2026-06-04",
+    "updated": "2026-06-06",
+    "authors": [
+      "Jiaju Chen",
+      "Bo Sun",
+      "Yuxuan Lu",
+      "Yun Wang",
+      "Dakuo Wang",
+      "Bingsheng Yao"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 22,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.06399",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.06399",
+    "pdf_url": "https://arxiv.org/pdf/2606.06399",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.01199",
+    "title": "Can LLM Agents Sustain Long-Horizon Organizational Dynamics?",
+    "url": "https://arxiv.org/abs/2606.01199",
+    "published": "2026-05-31",
+    "updated": "2026-05-31",
+    "authors": [
+      "Xuancheng Zhu",
+      "Yang Yue",
+      "Shuaibing Wan",
+      "Zihan Dou",
+      "Xiaohan Zhang",
+      "Yongrui Liu",
+      "Guoshun Nan"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 22,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.01199",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.01199",
+    "pdf_url": "https://arxiv.org/pdf/2606.01199",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.29861",
+    "title": "Towards Verifiable Multimodal Deep Research: A Multi-Agent Harness for Interleaved Report Generation",
+    "url": "https://arxiv.org/abs/2605.29861",
+    "published": "2026-05-28",
+    "updated": "2026-06-03",
+    "authors": [
+      "Chenghao Zhang",
+      "Guanting Dong",
+      "Yufan Liu",
+      "Tong Zhao",
+      "Xiaoxi Li",
+      "Zhicheng Dou"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 22,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.29861",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.29861",
+    "pdf_url": "https://arxiv.org/pdf/2605.29861",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.18652",
+    "title": "MementoGUI: Learning Agentic Multimodal Memory Control for Long-Horizon GUI Agents",
+    "url": "https://arxiv.org/abs/2605.18652",
+    "published": "2026-05-18",
+    "updated": "2026-05-18",
+    "authors": [
+      "Ziyun Zeng",
+      "Hang Hua",
+      "Bocheng Zou",
+      "Mu Cai",
+      "Rogerio Feris",
+      "Jiebo Luo"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 22,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.18652",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.18652",
+    "pdf_url": "https://arxiv.org/pdf/2605.18652",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.05704",
+    "title": "SafeHarbor: Hierarchical Memory-Augmented Guardrail for LLM Agent Safety",
+    "url": "https://arxiv.org/abs/2605.05704",
+    "published": "2026-05-07",
+    "updated": "2026-05-22",
+    "authors": [
+      "Zhe Liu",
+      "Zonghao Ying",
+      "Wenxin Zhang",
+      "Quanchen Zou",
+      "Deyue Zhang",
+      "Dongdong Yang",
+      "Xiangzheng Zhang",
+      "Hao Peng"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "memory",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 22,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.05704",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.05704",
+    "pdf_url": "https://arxiv.org/pdf/2605.05704",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.03242",
+    "title": "Enhancing Agent Safety Judgment: Controlled Benchmark Rewriting and Analogical Reasoning for Deceptive Out-of-Distribution Scenarios",
+    "url": "https://arxiv.org/abs/2605.03242",
+    "published": "2026-05-05",
+    "updated": "2026-05-05",
+    "authors": [
+      "Zuoyu Zhang",
+      "Yancheng Zhu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 22,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.03242",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.03242",
+    "pdf_url": "https://arxiv.org/pdf/2605.03242",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2607.06118",
+    "title": "WebRetriever: A Large-Scale Comprehensive Benchmark for Efficient Web Agent Evaluation",
+    "url": "https://arxiv.org/abs/2607.06118",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Wei Dong",
+      "Tianyu Fu",
+      "Zhe Yu",
+      "Hanning Wang",
+      "Anyang Su",
+      "Zhizhou Fang",
+      "Yuyang Chen",
+      "Shuo Wang",
+      "Minghui Wu",
+      "Ping Jiang",
+      "Zhen Lei",
+      "Chenxu Zhao"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.MM"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "embodied-agent",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2607.06118",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06118",
+    "pdf_url": "https://arxiv.org/pdf/2607.06118",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.05773",
+    "title": "Beyond Static Evaluation: Building Simulation Environments for Scalable Agentic Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2607.05773",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Akshay Arora",
+      "Ishan Nigam",
+      "Ashutosh Aggarwal",
+      "Shefali Bansal",
+      "Krishna Singh",
+      "Sweta Kumari",
+      "Nikhil Mittal",
+      "Shariq Farhan",
+      "Siddarth Malreddy"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm",
+      "tool-use",
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2607.05773",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05773",
+    "pdf_url": "https://arxiv.org/pdf/2607.05773",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2607.05456",
+    "title": "Prompt-to-Paper: Agentic AI System for Bioinformatics",
+    "url": "https://arxiv.org/abs/2607.05456",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Ramsha Kamran",
+      "Maheera Amjad",
+      "Zartasha Mustansar",
+      "Arsalan Shaukat",
+      "Salma Sherbaz",
+      "Muhammad U. S. Khan"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "q-bio.QM"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "coding-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.05456",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05456",
+    "pdf_url": "https://arxiv.org/pdf/2607.05456",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.04433",
+    "title": "Autonomous Information Seeking: A Roadmap for Agentic Recommender Systems",
+    "url": "https://arxiv.org/abs/2607.04433",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Xinyu Lin",
+      "Yashar Deldjoo",
+      "Sunhao Dai",
+      "Honghui Bao",
+      "Xiaopeng Ye",
+      "Fatemeh Nazary",
+      "Wenjie Wang",
+      "Tommaso Di Noia",
+      "Jun Xu",
+      "Tat-Seng Chua"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2607.04433",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04433",
+    "pdf_url": "https://arxiv.org/pdf/2607.04433",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2607.03601",
+    "title": "ArchEval: Measuring AI Agents as Computer Architects",
+    "url": "https://arxiv.org/abs/2607.03601",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Chenyu Wang",
+      "Zishen Wan",
+      "Jeffrey Ma",
+      "Shvetank Prakash",
+      "Zhenting Qi",
+      "Haebin Do",
+      "Andy Cheng",
+      "Arya Tschand",
+      "Jiahe Shi",
+      "Yilun Du",
+      "Vijay Janapa Reddi"
+    ],
+    "categories": [
+      "cs.AR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "llm-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.03601",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03601",
+    "pdf_url": "https://arxiv.org/pdf/2607.03601",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.02032",
+    "title": "PACE: A Proxy for Agentic Capability Evaluation",
+    "url": "https://arxiv.org/abs/2607.02032",
+    "published": "2026-07-02",
+    "updated": "2026-07-06",
+    "authors": [
+      "Yueqi Song",
+      "Lintang Sutawika",
+      "Jiarui Liu",
+      "Lindia Tjuatja",
+      "Jiayi Geng",
+      "Yunze Xiao",
+      "Daniel Lee",
+      "Aditya Bharat Soni",
+      "Vincent Lo",
+      "Xiang Yue",
+      "Graham Neubig"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "reasoning"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "coding-agent",
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.02032",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02032",
+    "pdf_url": "https://arxiv.org/pdf/2607.02032",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.01641",
+    "title": "When Agents Do Not Stop: Uncovering Infinite Agentic Loops in LLM Agents",
+    "url": "https://arxiv.org/abs/2607.01641",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Xinyi Hou",
+      "Shenao Wang",
+      "Yanjie Zhao",
+      "Haoyu Wang"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "planning-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.01641",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01641",
+    "pdf_url": "https://arxiv.org/pdf/2607.01641",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.30524",
+    "title": "The Illusion of Agentic Complexity in README.md Generation: Evaluating Single-Agent vs. Multi-Agent RAG Systems",
+    "url": "https://arxiv.org/abs/2606.30524",
+    "published": "2026-06-29",
+    "updated": "2026-07-01",
+    "authors": [
+      "Abu Saleh",
+      "Tesfay Welegebreal Tesfay",
+      "Phuong T. Nguyen",
+      "Juri Di Rocco",
+      "Muhammad Umar Zeshan",
+      "Davide Di Ruscio"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "multi-agent",
+      "planning",
+      "rag"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.30524",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30524",
+    "pdf_url": "https://arxiv.org/pdf/2606.30524",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.29537",
+    "title": "OSWorld2.0: Benchmarking Computer Use Agents on Long-Horizon Real-World Tasks",
+    "url": "https://arxiv.org/abs/2606.29537",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Mengqi Yuan",
+      "Zilong Zhou",
+      "Xinzhuang Xiong",
+      "Weiming Wu",
+      "Jiayang Sun",
+      "Jiamin Song",
+      "Kaiqian Cui",
+      "Bowen Wang",
+      "Haoyuan Wu",
+      "Yitong Li",
+      "Dunjie Lu",
+      "Haikong Lu",
+      "Qi Zhen",
+      "Xinyuan Wang",
+      "Jiaqi Deng",
+      "Yuhao Yang",
+      "Cheng Chen",
+      "Boyuan Zheng",
+      "Alex Su",
+      "Xiao Yu",
+      "Hao Zou",
+      "Saaket Agashe",
+      "Xing Han Lu",
+      "Manpreet Kaur",
+      "Zhengyang Qi",
+      "Vincent Sunn Chen",
+      "Frederic Sala",
+      "Dayiheng Liu",
+      "Junyang Lin",
+      "Zhou Yu",
+      "Yu Su",
+      "Siva Reddy",
+      "Xin Eric Wang",
+      "Peng Qi",
+      "Tianbao Xie",
+      "Tao Yu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.29537",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29537",
+    "pdf_url": "https://arxiv.org/pdf/2606.29537",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.28925",
+    "title": "Multi-Agent Routing as Set-Valued Prediction: A WildChat Benchmark and Cost-Aware Evaluation",
+    "url": "https://arxiv.org/abs/2606.28925",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Ananto Nayan Bala",
+      "Faisal Muhammad Shah"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.IR",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.28925",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28925",
+    "pdf_url": "https://arxiv.org/pdf/2606.28925",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.26511",
+    "title": "Temporal Validity in Retrieval Memory: Eliminating Stale-Fact Errors for AI Agents over Evolving Knowledge",
+    "url": "https://arxiv.org/abs/2606.26511",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Neeraj Yadav"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.ET",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.26511",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26511",
+    "pdf_url": "https://arxiv.org/pdf/2606.26511",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.22557",
+    "title": "MacAgentBench: Benchmarking AI Agents on Real-World macOS Desktop",
+    "url": "https://arxiv.org/abs/2606.22557",
+    "published": "2026-06-21",
+    "updated": "2026-06-21",
+    "authors": [
+      "Yikun Fu",
+      "Bowen Fu",
+      "Zhenyu Wu",
+      "Shuang Cheng",
+      "Xiaowei Sun",
+      "Bowen Yang",
+      "Zehao Li",
+      "Yibo Zhao",
+      "Zichen Ding",
+      "Zhoumianze Liu",
+      "Shijie Wang",
+      "Biqing Qi",
+      "Bowen Zhou"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "ai-agent",
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.22557",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22557",
+    "pdf_url": "https://arxiv.org/pdf/2606.22557",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.21877",
+    "title": "AgentRiskBOM: A Risk-Scoping Security Bill of Materials for Agentic AI Systems",
+    "url": "https://arxiv.org/abs/2606.21877",
+    "published": "2026-06-20",
+    "updated": "2026-06-20",
+    "authors": [
+      "Srimonti Dutta",
+      "Akshata Kishore Moharir"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CR",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "memory",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "ai-agent",
+      "rag-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.21877",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21877",
+    "pdf_url": "https://arxiv.org/pdf/2606.21877",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.16802",
+    "title": "LabOSBench: Benchmarking Computer Use Agents for Scientific Instrument Control",
+    "url": "https://arxiv.org/abs/2606.16802",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Anqi Zou",
+      "Han Deng",
+      "Chengyu Zhang",
+      "Junquan Hu",
+      "Yu Wang",
+      "Yuxiang Xing",
+      "Aokai Zhang",
+      "Hanling Zhang",
+      "Zhaoyang Liu",
+      "Ben Fei",
+      "Zhihui Wang",
+      "Wanli Ouyang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "planning",
+      "workflow-agent"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.16802",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16802",
+    "pdf_url": "https://arxiv.org/pdf/2606.16802",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.13994",
+    "title": "Hidden in Plain Sight: Benchmarking Agent Safety Against Decomposition Attacks with DECOMPBENCH",
+    "url": "https://arxiv.org/abs/2606.13994",
+    "published": "2026-06-12",
+    "updated": "2026-06-12",
+    "authors": [
+      "Vikhyath Kothamasu",
+      "Virginia Smith",
+      "Chhavi Yadav"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.13994",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13994",
+    "pdf_url": "https://arxiv.org/pdf/2606.13994",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.04990",
+    "title": "From Agent Traces to Trust: A Survey of Evidence Tracing and Execution Provenance in LLM Agents",
+    "url": "https://arxiv.org/abs/2606.04990",
+    "published": "2026-06-03",
+    "updated": "2026-06-28",
+    "authors": [
+      "Yiqi Wang",
+      "Jiaqi Zhang",
+      "Taotao Cai",
+      "Zirui Liu",
+      "Qingqiang Sun",
+      "Zequn Sun",
+      "Zhangkai Wu",
+      "Manqing Dong",
+      "Mingkai Zheng",
+      "Xuefei Yin",
+      "Yanming Zhu"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.04990",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.04990",
+    "pdf_url": "https://arxiv.org/pdf/2606.04990",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.02461",
+    "title": "AgentCL: Toward Rigorous Evaluation of Continual Learning in Language Agents",
+    "url": "https://arxiv.org/abs/2606.02461",
+    "published": "2026-06-01",
+    "updated": "2026-06-02",
+    "authors": [
+      "Yiheng Shu",
+      "Bernal Jiménez Gutiérrez",
+      "Saisri Padmaja Jonnalagedda",
+      "Yuguang Yao",
+      "Huan Sun",
+      "Yu Su"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.02461",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.02461",
+    "pdf_url": "https://arxiv.org/pdf/2606.02461",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.01385",
+    "title": "Bridging Requirements and Architecture: Multi-Agent Orchestration with External Knowledge and Hierarchical Memory",
+    "url": "https://arxiv.org/abs/2606.01385",
+    "published": "2026-05-31",
+    "updated": "2026-05-31",
+    "authors": [
+      "Ruiyin Li",
+      "Yiran Zhang",
+      "Xiyu Zhou",
+      "Yangxiao Cai",
+      "Peng Liang",
+      "Weisong Sun",
+      "Jifeng Xuan",
+      "Zhi Jin",
+      "Yang Liu"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "memory",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.01385",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.01385",
+    "pdf_url": "https://arxiv.org/pdf/2606.01385",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.00756",
+    "title": "CoMIC: Collaborative Memory and Insights Circulation for Long-Horizon LLM Agents in Cloud-Edge Systems",
+    "url": "https://arxiv.org/abs/2606.00756",
+    "published": "2026-05-30",
+    "updated": "2026-05-30",
+    "authors": [
+      "Yannan Wang",
+      "Longli Yang",
+      "Zhen Liu",
+      "Abhishek Kumar",
+      "Carsten Maple"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.00756",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.00756",
+    "pdf_url": "https://arxiv.org/pdf/2606.00756",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.20315",
+    "title": "Mix-Quant: Quantized Prefilling, Precise Decoding for Agentic LLMs",
+    "url": "https://arxiv.org/abs/2605.20315",
+    "published": "2026-05-19",
+    "updated": "2026-05-19",
+    "authors": [
+      "Haiquan Lu",
+      "Zigeng Chen",
+      "Gongfan Fang",
+      "Xinyin Ma",
+      "Xinchao Wang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.20315",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.20315",
+    "pdf_url": "https://arxiv.org/pdf/2605.20315",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.14498",
+    "title": "GroupMemBench: Benchmarking LLM Agent Memory in Multi-Party Conversations",
+    "url": "https://arxiv.org/abs/2605.14498",
+    "published": "2026-05-14",
+    "updated": "2026-05-16",
+    "authors": [
+      "Jingbo Yang",
+      "Kwei-Herng Lai",
+      "Xiaowen Wang",
+      "Shiyu Chang",
+      "Yaar Harari",
+      "Evgeniy Gabrilovich"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "rag",
+      "reasoning"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.14498",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.14498",
+    "pdf_url": "https://arxiv.org/pdf/2605.14498",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.11633",
+    "title": "Can LLM Agents Respond to Disasters? Benchmarking Heterogeneous Geospatial Reasoning in Emergency Operations",
+    "url": "https://arxiv.org/abs/2605.11633",
+    "published": "2026-05-12",
+    "updated": "2026-05-12",
+    "authors": [
+      "Junjue Wang",
+      "Weihao Xuan",
+      "Heli Qi",
+      "Pengyu Dai",
+      "Kunyi Liu",
+      "Hongruixuan Chen",
+      "Zhuo Zheng",
+      "Junshi Xia",
+      "Stefano Ermon",
+      "Naoto Yokoya"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.11633",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.11633",
+    "pdf_url": "https://arxiv.org/pdf/2605.11633",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.06812",
+    "title": "Towards Security-Auditable LLM Agents: A Unified Graph Representation",
+    "url": "https://arxiv.org/abs/2605.06812",
+    "published": "2026-05-07",
+    "updated": "2026-05-07",
+    "authors": [
+      "Chaofan Li",
+      "Lyuye Zhang",
+      "Jintao Zhai",
+      "Siyue Feng",
+      "Xichun Yang",
+      "Huahao Wang",
+      "Shihan Dou",
+      "Yu Ji",
+      "Yutao Hu",
+      "Yueming Wu",
+      "Yang Liu",
+      "Deqing Zou"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.06812",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.06812",
+    "pdf_url": "https://arxiv.org/pdf/2605.06812",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2602.08412",
+    "title": "From Assistant to Double Agent: Formalizing and Benchmarking Attacks on OpenClaw for Personalized Local AI Agent",
+    "url": "https://arxiv.org/abs/2602.08412",
+    "published": "2026-02-09",
+    "updated": "2026-02-11",
+    "authors": [
+      "Yuhang Wang",
+      "Feiming Xu",
+      "Zheng Lin",
+      "Guangyu He",
+      "Yuzhe Huang",
+      "Haichang Gao",
+      "Zhenxing Niu",
+      "Shiguo Lian",
+      "Zhaoxiang Liu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.08412",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.08412",
+    "pdf_url": "https://arxiv.org/pdf/2602.08412",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2508.07575",
+    "title": "MCPToolBench++: A Large Scale AI Agent Model Context Protocol MCP Tool Use Benchmark",
+    "url": "https://arxiv.org/abs/2508.07575",
+    "published": "2025-08-11",
+    "updated": "2025-08-11",
+    "authors": [
+      "Shiqing Fan",
+      "Xichen Ding",
+      "Liang Zhang",
+      "Linjian Mo"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2508.07575",
+    "source": "arxiv",
+    "source_id": "arxiv:2508.07575",
+    "pdf_url": "https://arxiv.org/pdf/2508.07575",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2607.05318",
+    "title": "PiSAs: Benchmarking Contextual Integrity in Multi-User Agentic Systems",
+    "url": "https://arxiv.org/abs/2607.05318",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Shubham Gupta",
+      "Nazanin Mohammadi Sepahvand",
+      "Abhinav Kumar",
+      "Cem Subakan",
+      "Spandana Gella",
+      "Pierre-André Noël",
+      "Perouz Taslakian",
+      "Eugene Bagdasarian",
+      "Valentina Zantedeschi"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.05318",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05318",
+    "pdf_url": "https://arxiv.org/pdf/2607.05318",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.04391",
+    "title": "Memory-Orchestrated Semantic System (MOSS): An Auditable Agentic Memory Architecture",
+    "url": "https://arxiv.org/abs/2607.04391",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Serge Lacasse",
+      "Jérémie Hatier",
+      "Alex Baker"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "ai-agent",
+      "rag-agent"
+    ],
+    "arxiv_id": "2607.04391",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04391",
+    "pdf_url": "https://arxiv.org/pdf/2607.04391",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2607.03953",
+    "title": "The Remarkable Effectiveness of Providing AI Agents with Natural Language Tools: A Replication Study Validating NLT Performance Across 14 Models",
+    "url": "https://arxiv.org/abs/2607.03953",
+    "published": "2026-07-04",
+    "updated": "2026-07-04",
+    "authors": [
+      "Alexander Somma",
+      "Isabelle Plante",
+      "Fred Premji"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "ai-agent",
+      "llm-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.03953",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03953",
+    "pdf_url": "https://arxiv.org/pdf/2607.03953",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.03726",
+    "title": "SelfMem: Self-Optimizing Memory for AI Agents",
+    "url": "https://arxiv.org/abs/2607.03726",
+    "published": "2026-07-04",
+    "updated": "2026-07-04",
+    "authors": [
+      "Shu Yang",
+      "Junchao Wu",
+      "Derek F. Wong",
+      "Di Wang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "ai-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.03726",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03726",
+    "pdf_url": "https://arxiv.org/pdf/2607.03726",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2607.01766",
+    "title": "SimWorlds: A Multi-Agent System for Dynamic 3D Scene Creation",
+    "url": "https://arxiv.org/abs/2607.01766",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Chunjiang Liu",
+      "Xiaoyuan Wang",
+      "Haoyu Chen",
+      "Yizhou Zhao",
+      "Ming-Hsuan Yang",
+      "László A. Jeni"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.01766",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01766",
+    "pdf_url": "https://arxiv.org/pdf/2607.01766",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.00454",
+    "title": "Agri-SAGE: Simulation-Grounded Multi-Agent LLM for Context-Aware Agricultural Advisory Generation",
+    "url": "https://arxiv.org/abs/2607.00454",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Vedant Balasubramaniam",
+      "Geetha Charan",
+      "Manojkumar Patil",
+      "Rohit P Suresh",
+      "V Priyanka",
+      "Kodur Sai Vinay Sathvik",
+      "Y. Narahari"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "world-model"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.00454",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00454",
+    "pdf_url": "https://arxiv.org/pdf/2607.00454",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.31073",
+    "title": "MultiUAV-Plat: An LLM-Oriented Platform, Benchmark and Framework for Multi-UAV Collaborative Task Planning",
+    "url": "https://arxiv.org/abs/2606.31073",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Sheng Zhang",
+      "Qinglin Li",
+      "Yuechao Zang",
+      "Xueqin Huang",
+      "Yijia Fu",
+      "Cheng Zhu"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA",
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "llm-agent",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.31073",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31073",
+    "pdf_url": "https://arxiv.org/pdf/2606.31073",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.31179",
+    "title": "HealthAgentBench: A Unified Benchmark Suite of Realistic Agentic Healthcare Environments for Challenging Frontier AI Agents",
+    "url": "https://arxiv.org/abs/2606.31179",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Qianchu Liu",
+      "Sheng Zhang",
+      "Guanghui Qin",
+      "Jeya Maria Jose Valanarasu",
+      "Maximilian Rokuss",
+      "Mingyu Lu",
+      "Timothy Ossowski",
+      "Juan Manuel Zambrano Chaves",
+      "Cliff Wong",
+      "Peniel Argaw",
+      "Yashna Hasija",
+      "Mu Wei",
+      "Wen-wai Yim",
+      "Qin Liu",
+      "Zilin Jing",
+      "Jason Entenmann",
+      "Naoto Usuyama",
+      "Tristan Naumann",
+      "Hoifung Poon"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.31179",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31179",
+    "pdf_url": "https://arxiv.org/pdf/2606.31179",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.31612",
+    "title": "What Memory Do GUI Agents Really Need? From Passive Records to Active Task-Driving States",
+    "url": "https://arxiv.org/abs/2606.31612",
+    "published": "2026-06-30",
+    "updated": "2026-07-02",
+    "authors": [
+      "Chen Liu",
+      "Ling Chen",
+      "Hanzhang Zhou",
+      "Xu Zhang",
+      "Quyu Kong",
+      "Panrong Tong",
+      "Wenhao Wang",
+      "Xin Yu",
+      "Steven Hoi",
+      "Yue Wang"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.31612",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31612",
+    "pdf_url": "https://arxiv.org/pdf/2606.31612",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.30906",
+    "title": "Investigating Multi-Agent Deliberation in Law",
+    "url": "https://arxiv.org/abs/2606.30906",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Cor Steging",
+      "Ludi van Leeuwen",
+      "Tadeusz Zbiegień"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "ai-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.30906",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30906",
+    "pdf_url": "https://arxiv.org/pdf/2606.30906",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.30949",
+    "title": "AgRefactor: Self-Evolving Agentic Workflow for HLS Compatibility and Performance",
+    "url": "https://arxiv.org/abs/2606.30949",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Yang Zou",
+      "Zijian Ding",
+      "Yizhou Sun",
+      "Jason Cong"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.AR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.30949",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30949",
+    "pdf_url": "https://arxiv.org/pdf/2606.30949",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.29116",
+    "title": "Characterizing Large Language Model Agentic Workflows: A Study on N8n Ecosystem",
+    "url": "https://arxiv.org/abs/2606.29116",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Yutian Tang",
+      "Yuming Zhou",
+      "Huaming Chen"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "llm-agent",
+      "planning-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.29116",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29116",
+    "pdf_url": "https://arxiv.org/pdf/2606.29116",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.24535",
+    "title": "Governed Shared Memory for Multi-Agent LLM Systems",
+    "url": "https://arxiv.org/abs/2606.24535",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Yanki Margalit",
+      "Nurit Cohen-Inger",
+      "Erni Avram",
+      "Ran Taig",
+      "Oded Margalit"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.24535",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24535",
+    "pdf_url": "https://arxiv.org/pdf/2606.24535",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.24820",
+    "title": "SHERLOC: Structured Diagnostic Localization for Code Repair Agents",
+    "url": "https://arxiv.org/abs/2606.24820",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Hovhannes Tamoyan",
+      "Sean Narenthiran",
+      "Erik Arakelyan",
+      "Mira Mezini",
+      "Boris Ginsburg"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent",
+      "multi-agent-llm",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.24820",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24820",
+    "pdf_url": "https://arxiv.org/pdf/2606.24820",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.22263",
+    "title": "Revelio: Cost-Efficient Agentic Memory Safety Vulnerability Detection For Repository-Scale Codebases",
+    "url": "https://arxiv.org/abs/2606.22263",
+    "published": "2026-06-20",
+    "updated": "2026-06-20",
+    "authors": [
+      "Yiwei Hou",
+      "Hao Wang",
+      "Muxi Lyu",
+      "Marius Momeu",
+      "Eric Nguyen",
+      "Taige Yang",
+      "Koushik Sen",
+      "Dawn Song",
+      "David Wagner"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.MA",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "memory"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.22263",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22263",
+    "pdf_url": "https://arxiv.org/pdf/2606.22263",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.21627",
+    "title": "Counsel: A Meta-Evaluation Dataset for Agentic Tasks",
+    "url": "https://arxiv.org/abs/2606.21627",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Sashank Pisupati",
+      "Henry Broomfield",
+      "Eujeong Choi",
+      "Antonia Calvi",
+      "Charlie Wang",
+      "Roman Engeler",
+      "Max Bartolo",
+      "Patrick Lewis"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "reasoning"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.21627",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21627",
+    "pdf_url": "https://arxiv.org/pdf/2606.21627",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.16871",
+    "title": "Human-on-the-Bridge: Scalable Evaluation for AI Agents",
+    "url": "https://arxiv.org/abs/2606.16871",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Fouad Bousetouane"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.16871",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16871",
+    "pdf_url": "https://arxiv.org/pdf/2606.16871",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.17246",
+    "title": "GeoDisaster: Benchmarking Orchestrated Agents for Operational Disaster Geo-Intelligence",
+    "url": "https://arxiv.org/abs/2606.17246",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Maram Hasan",
+      "Aman Verma",
+      "Savitra Roy",
+      "Hariseetharam Gunduboina",
+      "Daksh Jain",
+      "Muhammad Haris Khan",
+      "Subhasis Chaudhuri",
+      "Biplab Banerjee"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.17246",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17246",
+    "pdf_url": "https://arxiv.org/pdf/2606.17246",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.17114",
+    "title": "An Evaluation of Data Leakage Risks in Tool-Using LLM Agents in Realistic Scenarios",
+    "url": "https://arxiv.org/abs/2606.17114",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Hankyul Baek",
+      "Jaewon Noh",
+      "Sang Seo",
+      "Yongsu Kim",
+      "Gabriel Waikin Loh Matienzo",
+      "Young Il Kim",
+      "Ee Wei Seah",
+      "Akriti Vij"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.17114",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17114",
+    "pdf_url": "https://arxiv.org/pdf/2606.17114",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.16420",
+    "title": "Transferable Self-Evolving Playbooks for Agentic Security Auditing",
+    "url": "https://arxiv.org/abs/2606.16420",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Ziyue Wang",
+      "Cheuk Wang Maurice Ng",
+      "Chenchen Yu",
+      "Strick Sheng",
+      "Kaihua Qin",
+      "Liyi Zhou"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "embodied-agent",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.16420",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16420",
+    "pdf_url": "https://arxiv.org/pdf/2606.16420",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.14790",
+    "title": "XFlow: An Executable Protocol Programming System for Reliable Multi-Agent Workflows",
+    "url": "https://arxiv.org/abs/2606.14790",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Hanqi Li",
+      "Jing Peng",
+      "Zijian Wang",
+      "Lu Chen",
+      "Kai Yu"
+    ],
+    "categories": [
+      "cs.PL",
+      "cs.AI"
+    ],
+    "topics": [
+      "coding-agent",
+      "memory",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.14790",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.14790",
+    "pdf_url": "https://arxiv.org/pdf/2606.14790",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.08531",
+    "title": "VESTA: A Fully Automated Scenario Generation and Safety Evaluation Framework for LLM Agents",
+    "url": "https://arxiv.org/abs/2606.08531",
+    "published": "2026-06-07",
+    "updated": "2026-06-07",
+    "authors": [
+      "Lu Jia",
+      "Haibo Tong",
+      "Feifei Zhao",
+      "Jindong Li",
+      "Dongqi Liang",
+      "Ping Wu",
+      "Qian Zhang",
+      "Yi Zeng"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2606.08531",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.08531",
+    "pdf_url": "https://arxiv.org/pdf/2606.08531",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.07402",
+    "title": "M$^3$Exam: Benchmarking Multimodal Memory for Realistic User-Agent Interactions",
+    "url": "https://arxiv.org/abs/2606.07402",
+    "published": "2026-06-05",
+    "updated": "2026-06-05",
+    "authors": [
+      "Zhengjun Huang",
+      "Wenxuan Liu",
+      "Zhoujin Tian",
+      "Wei Chen",
+      "Junle Chen",
+      "Yuqian Wu",
+      "Fangyuan Zhang",
+      "Qintian Guo",
+      "Xiaofang Zhou"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.07402",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.07402",
+    "pdf_url": "https://arxiv.org/pdf/2606.07402",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.04780",
+    "title": "PersonaTree: Structured Lifecycle Memory for Person Understanding in LLM Agents",
+    "url": "https://arxiv.org/abs/2606.04780",
+    "published": "2026-06-03",
+    "updated": "2026-06-03",
+    "authors": [
+      "Yubo Hou",
+      "Jingwei Song",
+      "Hongbo Zhang",
+      "Zhisheng Chen",
+      "Bang Xiao",
+      "Tao Wan",
+      "Zengchang Qin"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.04780",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.04780",
+    "pdf_url": "https://arxiv.org/pdf/2606.04780",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.03657",
+    "title": "Diagnosing Knowledge Gaps in LLM Tool Use: An Agentic Benchmark for Novel API Acquisition",
+    "url": "https://arxiv.org/abs/2606.03657",
+    "published": "2026-06-02",
+    "updated": "2026-06-02",
+    "authors": [
+      "Jinnuo Liu",
+      "Yue Peng",
+      "Jinhan Niu",
+      "Hongyi Wen"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.03657",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.03657",
+    "pdf_url": "https://arxiv.org/pdf/2606.03657",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.02109",
+    "title": "BADGER: Bridging Agentic and Deterministic Evaluation for Generative Enterprise Reasoning",
+    "url": "https://arxiv.org/abs/2606.02109",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Shannon Serrao",
+      "Soumitra Chatterjee",
+      "Dorina Strori",
+      "Abhishek Sharma",
+      "Nathan Miller"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.02109",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.02109",
+    "pdf_url": "https://arxiv.org/pdf/2606.02109",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.01416",
+    "title": "Self-Healing Agentic Orchestrators for Reliable Tool-Augmented Large Language Model Systems",
+    "url": "https://arxiv.org/abs/2606.01416",
+    "published": "2026-05-31",
+    "updated": "2026-05-31",
+    "authors": [
+      "Rahul Suresh Babu",
+      "Adarsh Agrawal"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.01416",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.01416",
+    "pdf_url": "https://arxiv.org/pdf/2606.01416",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.00610",
+    "title": "MemGraphRAG: Memory-based Multi-Agent System for Graph Retrieval-Augmented Generation",
+    "url": "https://arxiv.org/abs/2606.00610",
+    "published": "2026-05-30",
+    "updated": "2026-05-30",
+    "authors": [
+      "Chuanjie Wu",
+      "Zhishang Xiang",
+      "Yunbo Tang",
+      "Zerui Chen",
+      "Qinggang Zhang",
+      "Jinsong Su"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.00610",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.00610",
+    "pdf_url": "https://arxiv.org/pdf/2606.00610",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.30883",
+    "title": "TRACE: Task-Aware Adaptive Self-Evolving Agentic Jailbreaking",
+    "url": "https://arxiv.org/abs/2605.30883",
+    "published": "2026-05-29",
+    "updated": "2026-05-29",
+    "authors": [
+      "Churui Zeng",
+      "Weiwei Qi",
+      "Kedong Xiu",
+      "Tianhang Zheng",
+      "Chaochao Lu",
+      "Liang He",
+      "Zhan Qin",
+      "Kui Ren"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.30883",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.30883",
+    "pdf_url": "https://arxiv.org/pdf/2605.30883",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.30090",
+    "title": "DirectorBench: Diagnosing Long-Form Video Generation with Personalized Multi-Agent Evaluation",
+    "url": "https://arxiv.org/abs/2605.30090",
+    "published": "2026-05-28",
+    "updated": "2026-05-28",
+    "authors": [
+      "Jiamin Chen",
+      "Qianben Chen",
+      "Jiawen Zhang",
+      "Yidi Wu",
+      "Yuchen Li",
+      "Xiaokun Zhang",
+      "Wangchunshu Zhou",
+      "Chen Ma"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2605.30090",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.30090",
+    "pdf_url": "https://arxiv.org/pdf/2605.30090",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2605.27134",
+    "title": "Scaling, Benchmarking, and Reasoning of Vision-Language Agents for Mobile GUI Navigation",
+    "url": "https://arxiv.org/abs/2605.27134",
+    "published": "2026-05-26",
+    "updated": "2026-05-26",
+    "authors": [
+      "Heng Qu",
+      "Yike Liu",
+      "Renren Jin",
+      "Wenzong Zhang",
+      "Pengzhi Gao",
+      "Wei Liu",
+      "Jian Luan"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "embodied-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.27134",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.27134",
+    "pdf_url": "https://arxiv.org/pdf/2605.27134",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.22643",
+    "title": "Boiling the Frog: A Multi-Turn Benchmark for Agentic Safety",
+    "url": "https://arxiv.org/abs/2605.22643",
+    "published": "2026-05-21",
+    "updated": "2026-05-22",
+    "authors": [
+      "Piercosma Bisconti",
+      "Matteo Prandi",
+      "Federico Pierucci",
+      "Federico Sartore",
+      "Enrico Panai",
+      "Laura Caroli",
+      "Yue Zhu",
+      "Adam Leon Smith",
+      "Luca Nannini",
+      "Marcello Galisai",
+      "Susanna Cifani",
+      "Francesco Giarrusso",
+      "Marcantonio Bracale Syrnikov",
+      "Daniele Nardi"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.22643",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.22643",
+    "pdf_url": "https://arxiv.org/pdf/2605.22643",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.15040",
+    "title": "Orchard: An Open-Source Agentic Modeling Framework",
+    "url": "https://arxiv.org/abs/2605.15040",
+    "published": "2026-05-14",
+    "updated": "2026-05-21",
+    "authors": [
+      "Baolin Peng",
+      "Wenlin Yao",
+      "Qianhui Wu",
+      "Hao Cheng",
+      "Xiao Yu",
+      "Rui Yang",
+      "Tao Ge",
+      "Alessandro Sordoni",
+      "Xingdi Yuan",
+      "Yelong Shen",
+      "Pengcheng He",
+      "Tong Zhang",
+      "Zhou Yu",
+      "Jianfeng Gao"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.15040",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.15040",
+    "pdf_url": "https://arxiv.org/pdf/2605.15040",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.13542",
+    "title": "RealICU: Do LLM Agents Understand Long-Context ICU Data? A Benchmark Beyond Behavior Imitation",
+    "url": "https://arxiv.org/abs/2605.13542",
+    "published": "2026-05-13",
+    "updated": "2026-05-13",
+    "authors": [
+      "Chengzhi Shen",
+      "Weixiang Shen",
+      "Tobias Susetzky",
+      "Chen",
+      "Chen",
+      "Jun Li",
+      "Yuyuan Liu",
+      "Xuepeng Zhang",
+      "Zhenyu Gong",
+      "Daniel Rueckert",
+      "Jiazhen Pan"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.LG",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.13542",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.13542",
+    "pdf_url": "https://arxiv.org/pdf/2605.13542",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.12015",
+    "title": "SkillSafetyBench: Evaluating Agent Safety under Skill-Facing Attack Surfaces",
+    "url": "https://arxiv.org/abs/2605.12015",
+    "published": "2026-05-12",
+    "updated": "2026-05-27",
+    "authors": [
+      "Chang Jin",
+      "An Wang",
+      "Zeming Wei",
+      "Kai Wang",
+      "Biaojie Zeng",
+      "Qiaosheng Zhang",
+      "Chao Yang",
+      "Jingjing Qu",
+      "Xia Hu",
+      "Xingcheng Xu"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.CL",
+      "cs.LG",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "memory",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.12015",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.12015",
+    "pdf_url": "https://arxiv.org/pdf/2605.12015",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.08374",
+    "title": "MemQ: Integrating Q-Learning into Self-Evolving Memory Agents over Provenance DAGs",
+    "url": "https://arxiv.org/abs/2605.08374",
+    "published": "2026-05-08",
+    "updated": "2026-05-14",
+    "authors": [
+      "Junwei Liao",
+      "Haoting Shi",
+      "Ruiwen Zhou",
+      "Jiaqian Wang",
+      "Shengtao Zhang",
+      "Wei Zhang",
+      "Ying Wen",
+      "Zhiyu Li",
+      "Feiyu Xiong",
+      "Bo Tang",
+      "Weinan Zhang",
+      "Muning Wen"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "embodied-agent",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2605.08374",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.08374",
+    "pdf_url": "https://arxiv.org/pdf/2605.08374",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2605.15206",
+    "title": "AgentStop: Terminating Local AI Agents Early to Save Energy in Consumer Devices",
+    "url": "https://arxiv.org/abs/2605.15206",
+    "published": "2026-05-01",
+    "updated": "2026-05-01",
+    "authors": [
+      "Dzung Pham",
+      "Kleomenis Katevas",
+      "Ali Shahin Shamsabadi",
+      "Hamed Haddadi"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.DC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.15206",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.15206",
+    "pdf_url": "https://arxiv.org/pdf/2605.15206",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.16282",
+    "title": "Taxonomy and Consistency Analysis of Safety Benchmarks for AI Agents",
+    "url": "https://arxiv.org/abs/2605.16282",
+    "published": "2026-04-11",
+    "updated": "2026-04-11",
+    "authors": [
+      "Miles Q. Li",
+      "Benjamin C. M. Fung",
+      "Boyang Li",
+      "Heba Ismail",
+      "Farkhund Iqbal"
+    ],
+    "categories": [
+      "cs.CY",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.16282",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.16282",
+    "pdf_url": "https://arxiv.org/pdf/2605.16282",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.02022",
+    "title": "ATBench: A Diverse and Realistic Agent Trajectory Benchmark for Safety Evaluation and Diagnosis",
+    "url": "https://arxiv.org/abs/2604.02022",
+    "published": "2026-04-02",
+    "updated": "2026-05-13",
+    "authors": [
+      "Yu Li",
+      "Haoyu Luo",
+      "Yuejin Xie",
+      "Yuqian Fu",
+      "Zhonghao Yang",
+      "Shuai Shao",
+      "Qihan Ren",
+      "Wanying Qu",
+      "Yanwei Fu",
+      "Yujiu Yang",
+      "Jing Shao",
+      "Xia Hu",
+      "Dongrui Liu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.02022",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.02022",
+    "pdf_url": "https://arxiv.org/pdf/2604.02022",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.16734",
+    "title": "Differential Harm Propensity in Personalized LLM Agents: The Curious Case of Mental Health Disclosure",
+    "url": "https://arxiv.org/abs/2603.16734",
+    "published": "2026-03-17",
+    "updated": "2026-03-17",
+    "authors": [
+      "Caglar Yildirim"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.16734",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.16734",
+    "pdf_url": "https://arxiv.org/pdf/2603.16734",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2509.20998",
+    "title": "CORE: Full-Path Evaluation of LLM Agents Beyond Final State",
+    "url": "https://arxiv.org/abs/2509.20998",
+    "published": "2025-09-25",
+    "updated": "2025-09-25",
+    "authors": [
+      "Panagiotis Michelakis",
+      "Yiannis Hadjiyiannis",
+      "Dimitrios Stamoulis"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2509.20998",
+    "source": "arxiv",
+    "source_id": "arxiv:2509.20998",
+    "pdf_url": "https://arxiv.org/pdf/2509.20998",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2607.05174",
+    "title": "AgentGym2: Benchmarking Large Language Model Agents in De-Idealized Real-World Environments",
+    "url": "https://arxiv.org/abs/2607.05174",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Zhiheng Xi",
+      "Dingwen Yang",
+      "Jiaqi Liu",
+      "Jixuan Huang",
+      "Honglin Guo",
+      "Baodai Huang",
+      "Tinggang Chen",
+      "Qi Zhang",
+      "Zhonghang Lu",
+      "Chenyu Liu",
+      "Jiajun Sun",
+      "Jiazheng Zhang",
+      "Dingwei Zhu",
+      "Xin Guo",
+      "Junzhe Wang",
+      "Zhihao Zhang",
+      "Yuming Yang",
+      "Junjie Ye",
+      "Minghe Gao",
+      "Dongrui Liu",
+      "Jiaming Ji",
+      "Guohao Li",
+      "Tao Gui",
+      "Qi Zhang",
+      "Xuanjing Huang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent",
+      "llm-agent",
+      "planning-agent"
+    ],
+    "arxiv_id": "2607.05174",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05174",
+    "pdf_url": "https://arxiv.org/pdf/2607.05174",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2607.05029",
+    "title": "Your Agent's Memories Are Not Its Own: Forged Reasoning Attacks on LLM Agent Memory and Defenses",
+    "url": "https://arxiv.org/abs/2607.05029",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Neeraj Karamchandani",
+      "Piyush Nagasubramaniam",
+      "Sencun Zhu",
+      "Dinghao Wu"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.05029",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05029",
+    "pdf_url": "https://arxiv.org/pdf/2607.05029",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2607.05120",
+    "title": "Agent Data Injection Attacks are Realistic Threats to AI Agents",
+    "url": "https://arxiv.org/abs/2607.05120",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Woohyuk Choi",
+      "Juhee Kim",
+      "Taehyun Kang",
+      "Jihyeon Jeong",
+      "Luyi Xing",
+      "Byoungyoung Lee"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "ai-agent",
+      "coding-agent",
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2607.05120",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05120",
+    "pdf_url": "https://arxiv.org/pdf/2607.05120",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2607.05202",
+    "title": "EvoAgentBench: Benchmarking Agent Self-Evolution via Ability Transfer",
+    "url": "https://arxiv.org/abs/2607.05202",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Xingze Gao",
+      "Chuanrui Hu",
+      "Hongda Chen",
+      "Pengfei Yao",
+      "Zhao Wang",
+      "Yi Bai",
+      "Zhengwei Wu",
+      "Yunyun Han",
+      "Xiaofeng Cong",
+      "Jie Gui",
+      "Yafeng Deng",
+      "Teng Li"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "memory",
+      "planning",
+      "reasoning"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2607.05202",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05202",
+    "pdf_url": "https://arxiv.org/pdf/2607.05202",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.04395",
+    "title": "NKI-Agent: Domain-Specific Fine-Tuning and Agentic Tool Use for Neuron Kernel Generation",
+    "url": "https://arxiv.org/abs/2607.04395",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Junjie Tang",
+      "Jun Huan",
+      "Hao Zhou",
+      "Yuhao Zhang",
+      "Lin Wang"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2607.04395",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04395",
+    "pdf_url": "https://arxiv.org/pdf/2607.04395",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2607.03510",
+    "title": "CAGE-1: Control, Assurance, and Governance Evaluation for Enterprise Agentic AI",
+    "url": "https://arxiv.org/abs/2607.03510",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Roopam W. Sure"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.CY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.03510",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03510",
+    "pdf_url": "https://arxiv.org/pdf/2607.03510",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.02684",
+    "title": "Can Coding Agents Implement Missed Compiler Optimizations? Evaluating LLM Agents on LLVM Peephole Optimizations",
+    "url": "https://arxiv.org/abs/2607.02684",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Hongxu Xu",
+      "Chunhao Liao",
+      "Xintong Zhou",
+      "Chengnian Sun"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "reasoning"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent",
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.02684",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02684",
+    "pdf_url": "https://arxiv.org/pdf/2607.02684",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.02507",
+    "title": "What LLM Agents Say When No One Is Watching: Social Structure and Latent Objective Emergence in Multi-Agent Debates",
+    "url": "https://arxiv.org/abs/2607.02507",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Arman Ghaffarizadeh",
+      "Danyal Mohaddes",
+      "Aliakbar Izadkhah",
+      "Shahriar Noroozizadeh"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.LG",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.02507",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02507",
+    "pdf_url": "https://arxiv.org/pdf/2607.02507",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.30986",
+    "title": "The Organizational Behavior of Agentic AI: Collective Intelligence in Human-Agent Workflows",
+    "url": "https://arxiv.org/abs/2606.30986",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Canhui Liu"
+    ],
+    "categories": [
+      "cs.CY",
+      "cs.HC",
+      "cs.MA",
+      "econ.GN"
+    ],
+    "topics": [
+      "memory",
+      "planning",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.30986",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30986",
+    "pdf_url": "https://arxiv.org/pdf/2606.30986",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.30639",
+    "title": "Self-Evolving World Models for LLM Agent Planning",
+    "url": "https://arxiv.org/abs/2606.30639",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Xuan Zhang",
+      "Wenxuan Zhang",
+      "See-Kiong Ng",
+      "Yang Deng"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.30639",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30639",
+    "pdf_url": "https://arxiv.org/pdf/2606.30639",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.29771",
+    "title": "CLQT: A Closed-Loop, Cost-Aware, Strategy-Consistent Benchmark for Diagnostic Evaluation of LLM Portfolio-Management Agents",
+    "url": "https://arxiv.org/abs/2606.29771",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Bo Qu",
+      "Mingguang Chen"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG",
+      "q-fin.CP",
+      "q-fin.PM"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.29771",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29771",
+    "pdf_url": "https://arxiv.org/pdf/2606.29771",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.00041",
+    "title": "ATM: CID-Brokered Pre-Write Admission for Multi-Agent Code Co-Synthesis",
+    "url": "https://arxiv.org/abs/2607.00041",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Eagl Huang"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "multi-agent",
+      "planning",
+      "rag"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.00041",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00041",
+    "pdf_url": "https://arxiv.org/pdf/2607.00041",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.29774",
+    "title": "Analytic Concept-Centric Memory for Agentic Embodied Manipulation",
+    "url": "https://arxiv.org/abs/2606.29774",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Mingyang Sun",
+      "Xiujian Liang",
+      "Jiude Wei",
+      "Qichen He",
+      "Donglin Wang",
+      "Cewu Lu",
+      "Jianhua Sun"
+    ],
+    "categories": [
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.29774",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29774",
+    "pdf_url": "https://arxiv.org/pdf/2606.29774",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.29193",
+    "title": "A Multi-Dataset Benchmark for Evaluating LLM Agents in Microservice Failure Diagnosis",
+    "url": "https://arxiv.org/abs/2606.29193",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Yuanhong Cai",
+      "Xiaohui Nie",
+      "Kanglin Yin",
+      "Changhua Pei",
+      "Yongqian Sun",
+      "Shenglin Zhang",
+      "Haibin Liu",
+      "Guiyang Liu",
+      "Xidao Wen",
+      "Fang Situ",
+      "Dan Pei"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.29193",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29193",
+    "pdf_url": "https://arxiv.org/pdf/2606.29193",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.29030",
+    "title": "Memory as an Attack Surface in LLM Agents: A Study on Multiple-Choice Question Answering",
+    "url": "https://arxiv.org/abs/2606.29030",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Shahnewaz Karim Sakib",
+      "Anindya Bijoy Das"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.ET"
+    ],
+    "topics": [
+      "memory",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "llm-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.29030",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29030",
+    "pdf_url": "https://arxiv.org/pdf/2606.29030",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.28692",
+    "title": "An AI agent for treatment reasoning over a biomedical tool universe",
+    "url": "https://arxiv.org/abs/2606.28692",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Shanghua Gao",
+      "Ayush Noori",
+      "Richard Zhu",
+      "Curtis Ginder",
+      "Zhenglun Kong",
+      "Xiaorui Su",
+      "Justin Kauffman",
+      "Benjamin S. Glicksberg",
+      "Joshua Lampert",
+      "Ankit Sakhuja",
+      "Ashwin Sawant",
+      "ATHENA-R1 Evaluation Consortium",
+      "David A. Clifton",
+      "Noa Dagan",
+      "Ran Balicer",
+      "Marinka Zitnik"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.28692",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28692",
+    "pdf_url": "https://arxiv.org/pdf/2606.28692",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.27806",
+    "title": "Agent vs. Parametric World Models: Hybrid Planning for Reliable Language Agents",
+    "url": "https://arxiv.org/abs/2606.27806",
+    "published": "2026-06-26",
+    "updated": "2026-07-05",
+    "authors": [
+      "Xinyuan Song",
+      "Zekun Cai"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.27806",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27806",
+    "pdf_url": "https://arxiv.org/pdf/2606.27806",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.28467",
+    "title": "An Agentic AI Pipeline for Appliance-Level Energy Anomaly Detection and LLM-Driven Recommendations",
+    "url": "https://arxiv.org/abs/2606.28467",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Dihia Falouz",
+      "Aida Douaibia",
+      "Amine Bechar",
+      "Youssef Elmir",
+      "Abbes Amira",
+      "Adel Oulefki"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.28467",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28467",
+    "pdf_url": "https://arxiv.org/pdf/2606.28467",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.26960",
+    "title": "Toward Agentic SysAdmin: Rethinking System Administration with AI Agents",
+    "url": "https://arxiv.org/abs/2606.26960",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Gianmaria Frigo",
+      "Davide Saladino",
+      "Alberto Castagnaro",
+      "Francesco Marchiori",
+      "Denis Donadel",
+      "Luca Pajola",
+      "Mauro Conti"
+    ],
+    "categories": [
+      "cs.NI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.26960",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26960",
+    "pdf_url": "https://arxiv.org/pdf/2606.26960",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.26346",
+    "title": "How Do Tool-Augmented LLM Agents Perform on Real-World Energy Analytics Tasks?",
+    "url": "https://arxiv.org/abs/2606.26346",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "David Akinpelu",
+      "Akintonde Abbas",
+      "Rereloluwa Alimi",
+      "Ayodeji Lana"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.26346",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26346",
+    "pdf_url": "https://arxiv.org/pdf/2606.26346",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.26403",
+    "title": "ProfileFoundry: A Synthetic Person-Object Substrate for Privacy, Memory, and Tool-Use Evaluation in LLM Agent",
+    "url": "https://arxiv.org/abs/2606.26403",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Sriram Selvam",
+      "Anneswa Ghosh"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.26403",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26403",
+    "pdf_url": "https://arxiv.org/pdf/2606.26403",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.25161",
+    "title": "TRUSTMEM: Learning Trustworthy Memory Consolidation for LLM Agents with Long-Term Memory",
+    "url": "https://arxiv.org/abs/2606.25161",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Tianyu Yang",
+      "Sudipta Paul",
+      "Vijay Srinivasan",
+      "Vivek Kulkarni",
+      "Srinivas Chappidi"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.25161",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25161",
+    "pdf_url": "https://arxiv.org/pdf/2606.25161",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.24626",
+    "title": "SAFARI: Scaling Long Horizon Agentic Fault Attribution via Active Investigation",
+    "url": "https://arxiv.org/abs/2606.24626",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Chenyang Zhu",
+      "Jiayu Yao",
+      "Kushal Chawla",
+      "Youbing Yin",
+      "Nathan Wolfe",
+      "Pengshan Cai",
+      "Jingyu Wu",
+      "Spencer Hong",
+      "Sangwoo Cho",
+      "Shi-Xiong Zhang",
+      "Daben Liu",
+      "Sambit Sahu",
+      "Erin Babinsky"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.24626",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24626",
+    "pdf_url": "https://arxiv.org/pdf/2606.24626",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.23991",
+    "title": "Critique of Agent Model",
+    "url": "https://arxiv.org/abs/2606.23991",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Eric Xing",
+      "Mingkai Deng",
+      "Jinyu Hou"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG",
+      "cs.MA",
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "ai-agent",
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.23991",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23991",
+    "pdf_url": "https://arxiv.org/pdf/2606.23991",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.22844",
+    "title": "RaMem: Contextual Reinstatement for Long-term Agentic Memory",
+    "url": "https://arxiv.org/abs/2606.22844",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Wei Yang",
+      "Bryce Kan",
+      "Shixuan Li",
+      "Li Li",
+      "Yuehan Qin",
+      "Jiate Li",
+      "Paul Bogdan",
+      "Jesse Thomason"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.22844",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22844",
+    "pdf_url": "https://arxiv.org/pdf/2606.22844",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.23565",
+    "title": "HoloAgent-0: A Unified Embodied Agent Framework with 3D Spatial Memory",
+    "url": "https://arxiv.org/abs/2606.23565",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Xiaolin Zhou",
+      "Liu Liu",
+      "Tingyang Xiao",
+      "Wei Feng",
+      "Fa Fu",
+      "Xinrui Meng",
+      "Xinjie Wang",
+      "Jialiang Han",
+      "Boyang Yu",
+      "Yun Du",
+      "Wei Sui",
+      "Zhizhong Su"
+    ],
+    "categories": [
+      "cs.RO",
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.23565",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23565",
+    "pdf_url": "https://arxiv.org/pdf/2606.23565",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.22678",
+    "title": "RigorBench: Benchmarking Engineering Process Discipline in Autonomous AI Coding Agents",
+    "url": "https://arxiv.org/abs/2606.22678",
+    "published": "2026-06-21",
+    "updated": "2026-06-29",
+    "authors": [
+      "Meher Bhaskar Madiraju",
+      "Meher Sai Preetam Madiraju"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.22678",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22678",
+    "pdf_url": "https://arxiv.org/pdf/2606.22678",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.22417",
+    "title": "Code Isn't Memory: A Structural Codebase Index Inside a Coding Agent",
+    "url": "https://arxiv.org/abs/2606.22417",
+    "published": "2026-06-21",
+    "updated": "2026-06-21",
+    "authors": [
+      "Ishaan Bhola",
+      "Adithyan Krishnan",
+      "Sravanth Kurmala",
+      "Mukunda NS"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "memory",
+      "rag"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.22417",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22417",
+    "pdf_url": "https://arxiv.org/pdf/2606.22417",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.21129",
+    "title": "AgenticOS: An Intent-Oriented Secure Operating System Architecture for Autonomous AI Agents",
+    "url": "https://arxiv.org/abs/2606.21129",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Zhen Zhao",
+      "Yu Zhang",
+      "Yanpeng Zhu",
+      "Jia Wang",
+      "Songqiao Tao",
+      "Xin Cheng",
+      "Jiexin Gao"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.OS"
+    ],
+    "topics": [
+      "agent-safety",
+      "planning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "autonomous-agent-llm",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.21129",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21129",
+    "pdf_url": "https://arxiv.org/pdf/2606.21129",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.21649",
+    "title": "EvoEmbedding: Evolvable Representations for Long-Context Retrieval and Agentic Memory",
+    "url": "https://arxiv.org/abs/2606.21649",
+    "published": "2026-06-19",
+    "updated": "2026-06-25",
+    "authors": [
+      "Chang Nie",
+      "Chaoyou Fu",
+      "Junlan Feng",
+      "Caifeng Shan"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "memory",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "agentic-ai",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.21649",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21649",
+    "pdf_url": "https://arxiv.org/pdf/2606.21649",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.20950",
+    "title": "Power Systems Agent Benchmark: Executable Evaluation of AI Agents in Electric Power Engineering",
+    "url": "https://arxiv.org/abs/2606.20950",
+    "published": "2026-06-18",
+    "updated": "2026-07-02",
+    "authors": [
+      "Sergei Trashchenkov"
+    ],
+    "categories": [
+      "cs.AI",
+      "eess.SY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "ai-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.20950",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20950",
+    "pdf_url": "https://arxiv.org/pdf/2606.20950",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.19704",
+    "title": "Beyond Static Leaderboards: Predictive Validity for the Evaluation of LLM Agents",
+    "url": "https://arxiv.org/abs/2606.19704",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Dhaval C. Patel",
+      "Kaoutar El Maghraoui",
+      "Shuxin Lin",
+      "Yusheng Li",
+      "Tianjun Feng",
+      "Chun-Yi Tsai",
+      "Yihan Sun",
+      "Wei Alexander Xin",
+      "Akshat Bhandari",
+      "Tanisha Rathod",
+      "Aaron Fan",
+      "Sanskruti Vijay Shejwal",
+      "Tomas Pasiecznik",
+      "Sagar Chethan Kumar",
+      "Tanmay Agarwal",
+      "Rohith Kanathur",
+      "Sam Colman",
+      "Amaan Sheikh",
+      "Dev Bahl",
+      "Ann Li",
+      "Krish Veera",
+      "Alimurtaza Mustafa Merchant",
+      "Shambhawi Baswaraj Bhure",
+      "Sajal Kumar Goyla",
+      "Chengrui Li",
+      "Kirthana Natarajan",
+      "Rui Li",
+      "Thomas Ajai",
+      "Rujing Li",
+      "Vivek G. Iyer",
+      "Sanjaii Vijayakumar",
+      "Yitong Bai",
+      "Ayal Yakobe",
+      "Darief Maes",
+      "Yassine Jebbouri",
+      "Tianyang Xu",
+      "Thai Quoc On",
+      "Vera Mazeeva",
+      "Winston Li",
+      "Yuval Shemla",
+      "Yeshitha Bhuvanesh",
+      "Rushin Bhatt",
+      "Siddharth Chethan Gowda",
+      "Alisha Vinod",
+      "Caroline Cahill",
+      "Shriya Aishani Rachakonda",
+      "Yunfeng Chen",
+      "Aryaman Agrawal",
+      "Aman Upganlawar",
+      "Mao Le Jonathan Ang",
+      "Yubin Sally Go",
+      "Madhav Rajkondawar",
+      "Yang-Jung Chen",
+      "Trisha Maturi",
+      "Ananya Kapoor",
+      "Andrew Li",
+      "Shrey Arora",
+      "Mana Abbaszadeh",
+      "Shen Li",
+      "Charles Xu",
+      "Byeolah Kwon"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.19704",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19704",
+    "pdf_url": "https://arxiv.org/pdf/2606.19704",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.20512",
+    "title": "Probe-and-Refine Tuning of Repository Guidance for Coding Agents",
+    "url": "https://arxiv.org/abs/2606.20512",
+    "published": "2026-06-18",
+    "updated": "2026-06-19",
+    "authors": [
+      "Asa Shepard",
+      "Jeannie Albrecht"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.20512",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20512",
+    "pdf_url": "https://arxiv.org/pdf/2606.20512",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.18829",
+    "title": "GateMem: Benchmarking Memory Governance in Multi-Principal Shared-Memory Agents",
+    "url": "https://arxiv.org/abs/2606.18829",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Zhe Ren",
+      "Yibo Yang",
+      "Yimeng Chen",
+      "Zijun Zhao",
+      "Benshuo Fu",
+      "Zhihao Shu",
+      "Bingjie Zhang",
+      "Yangyang Xu",
+      "Dandan Guo",
+      "Shuicheng Yan"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.18829",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18829",
+    "pdf_url": "https://arxiv.org/pdf/2606.18829",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.18356",
+    "title": "SafeClawBench: Separating Semantic, Audit-Evidence, and Sandbox Harm in Tool-Using LLM Agents",
+    "url": "https://arxiv.org/abs/2606.18356",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Yuchuan Tian",
+      "Mengyu Zheng",
+      "Haocheng Mei",
+      "Ye Yuan",
+      "Chao Xu",
+      "Xinghao Chen",
+      "Hanting Chen",
+      "Yu Wang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "memory",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.18356",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18356",
+    "pdf_url": "https://arxiv.org/pdf/2606.18356",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.16774",
+    "title": "OpenClaw-Skill: Collective Skill Tree Search for Agentic Large Language Models",
+    "url": "https://arxiv.org/abs/2606.16774",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Tianyi Lin",
+      "Chuanyu Sun",
+      "Jingyi Zhang",
+      "Changxu Wei",
+      "Huanjin Yao",
+      "Shunyu Liu",
+      "Xikun Zhang",
+      "Liu Liu",
+      "Jiaxing Huang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.16774",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16774",
+    "pdf_url": "https://arxiv.org/pdf/2606.16774",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.15862",
+    "title": "RetailBench: Benchmarking long horizon reasoning and coherent decision making of LLM agents in realistic retail environments",
+    "url": "https://arxiv.org/abs/2606.15862",
+    "published": "2026-06-14",
+    "updated": "2026-06-19",
+    "authors": [
+      "Linghua Zhang",
+      "Jun Wang",
+      "Jingtong Wu",
+      "Zhisong Zhang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.15862",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15862",
+    "pdf_url": "https://arxiv.org/pdf/2606.15862",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.12586",
+    "title": "Beyond Attack Success Rate: Examining Trigger Leakage in Vision-Language Agentic Systems",
+    "url": "https://arxiv.org/abs/2606.12586",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Jiamin Chang",
+      "Salil Kanhere",
+      "Piotr Koniusz",
+      "Jason",
+      "Xue",
+      "Hammond Pearce"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.12586",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12586",
+    "pdf_url": "https://arxiv.org/pdf/2606.12586",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.10507",
+    "title": "HIPIF: Hierarchical Planning and Information Folding for Long-Horizon LLM Agent Learning",
+    "url": "https://arxiv.org/abs/2606.10507",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Juncheng Diao",
+      "Zhicong Lu",
+      "Peiguang Li",
+      "Yongwei Zhou",
+      "Changyuan Tian",
+      "Qingbin Li",
+      "Rongxiang Weng",
+      "Jingang Wang",
+      "Xunliang Cai"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "reasoning"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "autonomous-agent-llm",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.10507",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10507",
+    "pdf_url": "https://arxiv.org/pdf/2606.10507",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.11042",
+    "title": "Workflow-GYM: Towards Long-Horizon Evaluation of Computer-use Agentic tasks in Real-World Professional Fields",
+    "url": "https://arxiv.org/abs/2606.11042",
+    "published": "2026-06-09",
+    "updated": "2026-06-11",
+    "authors": [
+      "Liya Zhu",
+      "Jingzhe Ding",
+      "Jian Zhang",
+      "Jianbo Xue",
+      "Shihao Liang",
+      "Ge Zhang",
+      "Yi Zhu",
+      "Duju Zeng",
+      "Xiang Gao",
+      "Qingshui Gu",
+      "Mailun Gao",
+      "Huimin Che",
+      "Yan Zhao",
+      "Peiheng Zhou",
+      "Haojun Wang",
+      "Chaobo Xian",
+      "Lili Le",
+      "Chi Wu",
+      "Yiwei Liu",
+      "Shengda Long",
+      "Jiale Yang",
+      "Fangzhi Xu",
+      "Sijin Wu",
+      "Haodong Duan",
+      "Chao He",
+      "Zhaojian Li",
+      "Minchao Wang",
+      "Huan Zhou",
+      "Jiani Hou",
+      "Chuqian Yu",
+      "Weiran Shi",
+      "Hongwan Gao",
+      "Jiamin Chen",
+      "Guanhong Chen",
+      "Tingqin Luo",
+      "Kaiyuan Zhang",
+      "Zhixin Yao",
+      "Qing Hua",
+      "Yuhao Jiang",
+      "Jin Chen",
+      "Pu Chen",
+      "Zhenyu Hu",
+      "Xingyu Li",
+      "Zhengxuan Jiang",
+      "Meng Cao",
+      "Tianfeng Long",
+      "Haozhe Wang",
+      "Mingzhang Wang",
+      "Yichen Zhang",
+      "Yiming Dai",
+      "Chenchen Zhang",
+      "Jiaying Wang",
+      "Xinying Liu",
+      "Xingzu Liu",
+      "Lingling Zhang",
+      "Xinjie Chen",
+      "Yujia Qin",
+      "Wangchunshu Zhou",
+      "Zhiyong Wu",
+      "Yang Liu",
+      "Jiaheng Liu",
+      "Lei Zhang",
+      "Shen Yan",
+      "Wenhao Huang",
+      "Zaiyuan Wang",
+      "Xiaolong Chang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.11042",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.11042",
+    "pdf_url": "https://arxiv.org/pdf/2606.11042",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.09483",
+    "title": "Memory Beyond Recall: A Dual-Process Cognitive Memory System for Self-Evolving LLM Agents",
+    "url": "https://arxiv.org/abs/2606.09483",
+    "published": "2026-06-08",
+    "updated": "2026-06-08",
+    "authors": [
+      "Tianxiang Fei",
+      "Mingyang Song",
+      "Mao Zheng",
+      "Xiang Yu"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.09483",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09483",
+    "pdf_url": "https://arxiv.org/pdf/2606.09483",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.07314",
+    "title": "QBugLM: An Agentic Benchmarking Framework for LLM-based Quantum Software Debugging",
+    "url": "https://arxiv.org/abs/2606.07314",
+    "published": "2026-06-05",
+    "updated": "2026-06-05",
+    "authors": [
+      "An B. B. Pham",
+      "Hoa T. Nguyen",
+      "Muhammad Usman"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.ET",
+      "quant-ph"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "reasoning",
+      "world-model"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.07314",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.07314",
+    "pdf_url": "https://arxiv.org/pdf/2606.07314",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.05684",
+    "title": "AdaMEM: Test-Time Adaptive Memory for Language Agents",
+    "url": "https://arxiv.org/abs/2606.05684",
+    "published": "2026-06-04",
+    "updated": "2026-06-04",
+    "authors": [
+      "Yunxiang Zhang",
+      "Yiheng Li",
+      "Ali Payani",
+      "Lu Wang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "language-agent"
+    ],
+    "arxiv_id": "2606.05684",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.05684",
+    "pdf_url": "https://arxiv.org/pdf/2606.05684",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.06448",
+    "title": "Agent Memory: Characterization and System Implications of Stateful Long-Horizon Workloads",
+    "url": "https://arxiv.org/abs/2606.06448",
+    "published": "2026-06-04",
+    "updated": "2026-06-04",
+    "authors": [
+      "Yasmine Omri",
+      "Ziyu Gan",
+      "Zachary Broveak",
+      "Robin Geens",
+      "Zexue He",
+      "Alex Pentland",
+      "Marian Verhelst",
+      "Tsachy Weissman",
+      "Thierry Tambe"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.06448",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.06448",
+    "pdf_url": "https://arxiv.org/pdf/2606.06448",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.04874",
+    "title": "Agent Planning Benchmark: A Diagnostic Framework for Planning Capabilities in LLM Agents",
+    "url": "https://arxiv.org/abs/2606.04874",
+    "published": "2026-06-03",
+    "updated": "2026-06-05",
+    "authors": [
+      "Haoyu Sun",
+      "Wenxuan Wang",
+      "Mingyang Song",
+      "Jujie He",
+      "Weinan Zhang",
+      "Yang Liu",
+      "Yang Yang",
+      "Yu Cheng"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.04874",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.04874",
+    "pdf_url": "https://arxiv.org/pdf/2606.04874",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.28349",
+    "title": "HMARS: A Hierarchical Multi-Agent Memory System for Long-Context Reasoning",
+    "url": "https://arxiv.org/abs/2606.28349",
+    "published": "2026-06-03",
+    "updated": "2026-06-03",
+    "authors": [
+      "Zeju Li",
+      "Ziyang Zheng",
+      "Yizhou Zhou",
+      "Qiang Xu"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.28349",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28349",
+    "pdf_url": "https://arxiv.org/pdf/2606.28349",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.04315",
+    "title": "Exploring Cross-Scenario Generality of Agentic Memory Systems: Diagnostics and a Strong Baseline",
+    "url": "https://arxiv.org/abs/2606.04315",
+    "published": "2026-06-03",
+    "updated": "2026-06-03",
+    "authors": [
+      "Zhikai Chen",
+      "Jialiang Gu",
+      "Junyu Yin",
+      "Xianxuan Long",
+      "Shenglai Zeng",
+      "Xiaoze Liu",
+      "Kai Guo",
+      "Keren Zhou",
+      "Jiliang Tang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.04315",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.04315",
+    "pdf_url": "https://arxiv.org/pdf/2606.04315",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.03374",
+    "title": "eMEM: A Hybrid Spatio-Temporal Memory System For Embodied Agents",
+    "url": "https://arxiv.org/abs/2606.03374",
+    "published": "2026-06-02",
+    "updated": "2026-06-22",
+    "authors": [
+      "A. Haroon Rasheed",
+      "Maria Kabtoul"
+    ],
+    "categories": [
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.03374",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.03374",
+    "pdf_url": "https://arxiv.org/pdf/2606.03374",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.02372",
+    "title": "COMAP: Co-Evolving World Models and Agent Policies for LLM Agents",
+    "url": "https://arxiv.org/abs/2606.02372",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Youwei Liu",
+      "Jian Wang",
+      "Hanlin Wang",
+      "Wenjie Li"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.02372",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.02372",
+    "pdf_url": "https://arxiv.org/pdf/2606.02372",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.01613",
+    "title": "TechRAG: Evidence-Gated Multimodal Agentic RAG for Technical Literature Reasoning",
+    "url": "https://arxiv.org/abs/2606.01613",
+    "published": "2026-06-01",
+    "updated": "2026-06-13",
+    "authors": [
+      "Kanwar Bharat Singh"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.01613",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.01613",
+    "pdf_url": "https://arxiv.org/pdf/2606.01613",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.00939",
+    "title": "FinCom: A Financial Multi-Agent Demo with Disagree-or-Commit Deliberation",
+    "url": "https://arxiv.org/abs/2606.00939",
+    "published": "2026-05-31",
+    "updated": "2026-05-31",
+    "authors": [
+      "Chao Peter Yang",
+      "Zixiao Tan",
+      "Kaisen Yao",
+      "Ziyu Zhou",
+      "Eleanor Jiang",
+      "Michael Wu"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.00939",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.00939",
+    "pdf_url": "https://arxiv.org/pdf/2606.00939",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2605.30690",
+    "title": "ElasticMem: Latent Memory as a Learnable Resource for LLM Agents",
+    "url": "https://arxiv.org/abs/2605.30690",
+    "published": "2026-05-29",
+    "updated": "2026-05-29",
+    "authors": [
+      "Tao Feng",
+      "Chongrui Ye",
+      "Tianyang Luo",
+      "Jingjun Xu",
+      "Xueqiang Xu",
+      "Haozhen Zhang",
+      "Ge Liu",
+      "Jiaxuan You"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.30690",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.30690",
+    "pdf_url": "https://arxiv.org/pdf/2605.30690",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.20629",
+    "title": "Specialize Roles, Mix Deployments: Pushing the Cost-Accuracy Frontier of LLM Agent Teams",
+    "url": "https://arxiv.org/abs/2606.20629",
+    "published": "2026-05-28",
+    "updated": "2026-05-28",
+    "authors": [
+      "Yinsicheng Jiang",
+      "Liang Cheng",
+      "Yeqi Huang",
+      "Yufan Zhao",
+      "Zhan Lu",
+      "Li Dong",
+      "Wenda Li",
+      "Edoardo Ponti",
+      "Luo Mai"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.20629",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20629",
+    "pdf_url": "https://arxiv.org/pdf/2606.20629",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2605.30604",
+    "title": "An Organization-Scoped LLM Agent Runtime Architecture for Regulated Cybersecurity Operations",
+    "url": "https://arxiv.org/abs/2605.30604",
+    "published": "2026-05-28",
+    "updated": "2026-05-28",
+    "authors": [
+      "George Fatouros",
+      "Georgios Makridis",
+      "George Kousiouris",
+      "John Soldatos",
+      "Dimosthenis Kyriazis"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.CL",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.30604",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.30604",
+    "pdf_url": "https://arxiv.org/pdf/2605.30604",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.27240",
+    "title": "ENPMR-Bench: Benchmarking Proactive Memory Retrieval for Emotional Support Agents",
+    "url": "https://arxiv.org/abs/2605.27240",
+    "published": "2026-05-26",
+    "updated": "2026-05-26",
+    "authors": [
+      "Xing Fu",
+      "Yulin Hu",
+      "Mengtong Ji",
+      "Haozhen Li",
+      "Yixin Sun",
+      "Weixiang Zhao",
+      "Yanyan Zhao",
+      "Bing Qin"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.27240",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.27240",
+    "pdf_url": "https://arxiv.org/pdf/2605.27240",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.25200",
+    "title": "GroupTravelBench: Benchmarking LLM Agents on Multi-Person Travel Planning",
+    "url": "https://arxiv.org/abs/2605.25200",
+    "published": "2026-05-24",
+    "updated": "2026-06-03",
+    "authors": [
+      "Xiang Cheng",
+      "Yulan Hu",
+      "Lulu Zheng",
+      "Zheng Pan",
+      "Xin Li",
+      "Yong Liu"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.25200",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.25200",
+    "pdf_url": "https://arxiv.org/pdf/2605.25200",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.24220",
+    "title": "Polar: Agentic RL on Any Harness at Scale",
+    "url": "https://arxiv.org/abs/2605.24220",
+    "published": "2026-05-22",
+    "updated": "2026-05-22",
+    "authors": [
+      "Binfeng Xu",
+      "Hao Zhang",
+      "Shaokun Zhang",
+      "Songyang Han",
+      "Mingjie Liu",
+      "Jian Hu",
+      "Shizhe Diao",
+      "Zhenghui Jin",
+      "Yunheng Zou",
+      "Michael Demoret",
+      "Jan Kautz",
+      "Yi Dong"
+    ],
+    "categories": [
+      "cs.DC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.24220",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.24220",
+    "pdf_url": "https://arxiv.org/pdf/2605.24220",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.23574",
+    "title": "Push Your Agent: Measuring and Enforcing Quantitative Goal Persistence in Long-Horizon LLM Agents",
+    "url": "https://arxiv.org/abs/2605.23574",
+    "published": "2026-05-22",
+    "updated": "2026-05-22",
+    "authors": [
+      "Yuandao Cai",
+      "Yuzhang Zhu",
+      "Liyou Gao",
+      "Wensheng Tang",
+      "Shengchao Qin"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.23574",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.23574",
+    "pdf_url": "https://arxiv.org/pdf/2605.23574",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.24069",
+    "title": "When the Manual Lies: A Realistic Benchmark to Evaluate MCP Poisoning Attacks for LLM Agents",
+    "url": "https://arxiv.org/abs/2605.24069",
+    "published": "2026-05-22",
+    "updated": "2026-05-22",
+    "authors": [
+      "Shi Liu",
+      "Xuehai Tang",
+      "Xikang Yang",
+      "Liang Lin",
+      "Biyu Zhou",
+      "Wenjie Xiao",
+      "Wantao Liu"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.24069",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.24069",
+    "pdf_url": "https://arxiv.org/pdf/2605.24069",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.24216",
+    "title": "Agent-ToM: Learning to Monitor Autonomous LLM Agents via Theory-of-Mind Reasoning",
+    "url": "https://arxiv.org/abs/2605.24216",
+    "published": "2026-05-22",
+    "updated": "2026-05-22",
+    "authors": [
+      "Nesreen K. Ahmed",
+      "Nima Nafisi"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.24216",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.24216",
+    "pdf_url": "https://arxiv.org/pdf/2605.24216",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.18421",
+    "title": "EvoMemBench: Benchmarking Agent Memory from a Self-Evolving Perspective",
+    "url": "https://arxiv.org/abs/2605.18421",
+    "published": "2026-05-18",
+    "updated": "2026-06-15",
+    "authors": [
+      "Yuyao Wang",
+      "Zhongjian Zhang",
+      "Mo Chi",
+      "Kaichi Yu",
+      "Yuhan Li",
+      "Miao Peng",
+      "Bing Tong",
+      "Chen Zhang",
+      "Yan Zhou",
+      "Jia Li"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.18421",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.18421",
+    "pdf_url": "https://arxiv.org/pdf/2605.18421",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.10779",
+    "title": "LITMUS: Benchmarking Behavioral Jailbreaks of LLM Agents in Real OS Environments",
+    "url": "https://arxiv.org/abs/2605.10779",
+    "published": "2026-05-11",
+    "updated": "2026-05-11",
+    "authors": [
+      "Chiyu Zhang",
+      "Huiqin Yang",
+      "Bendong Jiang",
+      "Xiaolei Zhang",
+      "Yiran Zhao",
+      "Ruyi Chen",
+      "Lu Zhou",
+      "Xiaogang Xu",
+      "Jiafei Wu",
+      "Liming Fang",
+      "Zhe Liu"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.10779",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.10779",
+    "pdf_url": "https://arxiv.org/pdf/2605.10779",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.03312",
+    "title": "MemFlow: Intent-Driven Memory Orchestration for Small Language Model Agents",
+    "url": "https://arxiv.org/abs/2605.03312",
+    "published": "2026-05-05",
+    "updated": "2026-05-05",
+    "authors": [
+      "Jiayi Chen",
+      "Yingcong Li",
+      "Guiling Wang"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.03312",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.03312",
+    "pdf_url": "https://arxiv.org/pdf/2605.03312",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.01101",
+    "title": "Virtual Speech Therapist: A Clinician-in-the-Loop AI Speech Therapy Agent for Personalized and Supervised Therapy",
+    "url": "https://arxiv.org/abs/2605.01101",
+    "published": "2026-05-01",
+    "updated": "2026-06-15",
+    "authors": [
+      "Shakeel Sheikh",
+      "Patrick Marmaroli",
+      "MD Sahidullah",
+      "Slim Ouni",
+      "Fabrice Hirsch",
+      "Goncalo Leal",
+      "Bjorn W Schuller"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.SD",
+      "eess.AS"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.01101",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.01101",
+    "pdf_url": "https://arxiv.org/pdf/2605.01101",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2604.19844",
+    "title": "If you're waiting for a sign... that might not be it! Mitigating Trust Boundary Confusion from Visual Injections on Vision-Language Agentic Systems",
+    "url": "https://arxiv.org/abs/2604.19844",
+    "published": "2026-04-21",
+    "updated": "2026-04-21",
+    "authors": [
+      "Jiamin Chang",
+      "Minhui Xue",
+      "Ruoxi Sun",
+      "Shuchao Pang",
+      "Salil S. Kanhere",
+      "Hammond Pearce"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "embodied-agent",
+      "multi-agent"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2604.19844",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.19844",
+    "pdf_url": "https://arxiv.org/pdf/2604.19844",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2604.18658",
+    "title": "Owner-Harm: A Missing Threat Model for AI Agent Safety",
+    "url": "https://arxiv.org/abs/2604.18658",
+    "published": "2026-04-20",
+    "updated": "2026-04-20",
+    "authors": [
+      "Dongcheng Zhang",
+      "Yiqing Jiang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.18658",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.18658",
+    "pdf_url": "https://arxiv.org/pdf/2604.18658",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.17562",
+    "title": "SafeAgent: A Runtime Protection Architecture for Agentic Systems",
+    "url": "https://arxiv.org/abs/2604.17562",
+    "published": "2026-04-19",
+    "updated": "2026-04-19",
+    "authors": [
+      "Hailin Liu",
+      "Eugene Ilyushin",
+      "Jie Ni",
+      "Min Zhu"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "memory",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.17562",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.17562",
+    "pdf_url": "https://arxiv.org/pdf/2604.17562",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.00623",
+    "title": "TraceSIR: A Multi-Agent Framework for Structured Analysis and Reporting of Agentic Execution Traces",
+    "url": "https://arxiv.org/abs/2603.00623",
+    "published": "2026-02-28",
+    "updated": "2026-02-28",
+    "authors": [
+      "Shu-Xun Yang",
+      "Cunxiang Wang",
+      "Haoke Zhang",
+      "Wenbo Yu",
+      "Lindong Wu",
+      "Jiayi Gui",
+      "Dayong Yang",
+      "Yukuo Cen",
+      "Zhuoer Feng",
+      "Bosi Wen",
+      "Yidong Wang",
+      "Lucen Zhong",
+      "Jiamin Ren",
+      "Linfeng Zhang",
+      "Jie Tang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2603.00623",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.00623",
+    "pdf_url": "https://arxiv.org/pdf/2603.00623",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2602.13530",
+    "title": "REMem: Reasoning with Episodic Memory in Language Agent",
+    "url": "https://arxiv.org/abs/2602.13530",
+    "published": "2026-02-13",
+    "updated": "2026-02-28",
+    "authors": [
+      "Yiheng Shu",
+      "Saisri Padmaja Jonnalagedda",
+      "Xiang Gao",
+      "Bernal Jiménez Gutiérrez",
+      "Weijian Qi",
+      "Kamalika Das",
+      "Huan Sun",
+      "Yu Su"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2602.13530",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.13530",
+    "pdf_url": "https://arxiv.org/pdf/2602.13530",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2510.03847",
+    "title": "Small Language Models for Agentic Systems: A Survey of Architectures, Capabilities, and Deployment Trade offs",
+    "url": "https://arxiv.org/abs/2510.03847",
+    "published": "2025-10-04",
+    "updated": "2025-10-04",
+    "authors": [
+      "Raghav Sharma",
+      "Manan Mehta"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2510.03847",
+    "source": "arxiv",
+    "source_id": "arxiv:2510.03847",
+    "pdf_url": "https://arxiv.org/pdf/2510.03847",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2607.06157",
+    "title": "LLM Agents for Deliberative Collaboration: A Study on Joint Decision Making Under Partial Observability",
+    "url": "https://arxiv.org/abs/2607.06157",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Chenxu Wang",
+      "Yongkun Yang",
+      "Boyuan Du",
+      "Shiwei Lin",
+      "Huaping Liu"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.06157",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06157",
+    "pdf_url": "https://arxiv.org/pdf/2607.06157",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.04686",
+    "title": "ToolFailBench: Diagnosing Tool-Use Failures in LLM Agents",
+    "url": "https://arxiv.org/abs/2607.04686",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Harsh Soni"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "llm-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.04686",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04686",
+    "pdf_url": "https://arxiv.org/pdf/2607.04686",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.04240",
+    "title": "Biological Motifs for Agentic Control",
+    "url": "https://arxiv.org/abs/2607.04240",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Bogdan Banu"
+    ],
+    "categories": [
+      "cs.AI",
+      "q-bio.CB"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "autonomous-agent-llm",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.04240",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04240",
+    "pdf_url": "https://arxiv.org/pdf/2607.04240",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.03441",
+    "title": "No Time Like the Present: Agentic Test-Time Training for LLM Agents",
+    "url": "https://arxiv.org/abs/2607.03441",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Yanbo Wang",
+      "Jinhua Hao",
+      "Yuze Shi",
+      "Kun Yuan",
+      "Ming Sun"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "coding-agent",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent",
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.03441",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03441",
+    "pdf_url": "https://arxiv.org/pdf/2607.03441",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.01874",
+    "title": "SkillCoach: Self-Evolving Rubrics for Evaluating and Enhancing Agentic Skill-Use",
+    "url": "https://arxiv.org/abs/2607.01874",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Jiayin Zhu",
+      "Kelong Mao",
+      "Yudong Guo",
+      "Dengbo He",
+      "Sulong Xu",
+      "Simiu Gu",
+      "Yutao Yue"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.01874",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01874",
+    "pdf_url": "https://arxiv.org/pdf/2607.01874",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.01793",
+    "title": "Safety Testing LLM Agents at Scale: From Risk Discovery to Evidence-Grounded Verification",
+    "url": "https://arxiv.org/abs/2607.01793",
+    "published": "2026-07-02",
+    "updated": "2026-07-04",
+    "authors": [
+      "Yunhao Feng",
+      "Ruixiao Lin",
+      "Ming Wen",
+      "Qinqin He",
+      "Yanming Guo",
+      "Yifan Ding",
+      "Yutao Wu",
+      "Jialuo Chen",
+      "Zhuoer Xu",
+      "Xiaohu Du",
+      "Jianan Ma",
+      "Zixing Chen",
+      "Xingjun Ma",
+      "Yunhao Chen",
+      "Xinhao Deng"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.01793",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01793",
+    "pdf_url": "https://arxiv.org/pdf/2607.01793",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.02703",
+    "title": "LLMoxie: Exploring Agentic AI for Scientific Software Development",
+    "url": "https://arxiv.org/abs/2607.02703",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Landung Setiawan",
+      "Anant Mittal",
+      "Cordero Core",
+      "Anshul Tambay",
+      "Carlos Garcia Jurado Suarez",
+      "David A. C. Beck",
+      "Andrew J. Connolly",
+      "Vani Mandava"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.DC",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.02703",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02703",
+    "pdf_url": "https://arxiv.org/pdf/2607.02703",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.00627",
+    "title": "AGI Maze as a Benchmark Framework for World-Modeling Agents",
+    "url": "https://arxiv.org/abs/2607.00627",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Alexey Potapov"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.00627",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00627",
+    "pdf_url": "https://arxiv.org/pdf/2607.00627",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.02606",
+    "title": "ChainSWE: Benchmarking Coding Agents on Multi-Bug Software Maintenance",
+    "url": "https://arxiv.org/abs/2607.02606",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Qirui Jin",
+      "Lingching Tung",
+      "Kenan Li",
+      "Qiyang Shi",
+      "Yushi She",
+      "Huanzhong Jia",
+      "Harrison Zhao",
+      "Kejing Xia",
+      "Zhenbang Du",
+      "Yikai Zhang",
+      "Jiaxin Pei",
+      "Zhenyu Zhang",
+      "Zhen Qi",
+      "Yuyan Duan",
+      "Wenke Lee",
+      "Zijian Jin"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.02606",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02606",
+    "pdf_url": "https://arxiv.org/pdf/2607.02606",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.32025",
+    "title": "Generative Skill Composition for LLM Agents",
+    "url": "https://arxiv.org/abs/2606.32025",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Xinyu Zhao",
+      "Zhen Tan",
+      "Vaishnav Tadiparthi",
+      "Nakul Agarwal",
+      "Kwonjoon Lee",
+      "Ehsan Moradi Pari",
+      "Hossein Nourkhiz Mahjoub",
+      "Tianlong Chen"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent",
+      "llm-agent",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.32025",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.32025",
+    "pdf_url": "https://arxiv.org/pdf/2606.32025",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.31229",
+    "title": "Agentic-Ideation: Sample Efficient Agentic Trajectories Synthesis for Scientific Ideation Agents",
+    "url": "https://arxiv.org/abs/2606.31229",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Keyu Zhao",
+      "Lingyan Kong",
+      "Fengli Xu",
+      "Yong Li"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "computer-use",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.31229",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31229",
+    "pdf_url": "https://arxiv.org/pdf/2606.31229",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.31410",
+    "title": "Xiaomi-GUI-0 Technical Report",
+    "url": "https://arxiv.org/abs/2606.31410",
+    "published": "2026-06-30",
+    "updated": "2026-07-01",
+    "authors": [
+      "Wanxia Cao",
+      "Chengzhen Duan",
+      "Pei Fu",
+      "Pengzhi Gao",
+      "Niu Lian",
+      "Fazhan Liu",
+      "Hui Liu",
+      "Heng Qu",
+      "Qinzhuo Wu",
+      "Zhehao Yu",
+      "Tongbo Chen",
+      "Shiqi Cui",
+      "Anan Du",
+      "Shukai Jia",
+      "Yuanfa Li",
+      "Wei Liu",
+      "Yike Liu",
+      "Wenchao Lu",
+      "Zhenbo Luo",
+      "Haoyuan Sun",
+      "Jiatong Sun",
+      "Cheng Tan",
+      "Yajie Wang",
+      "Changqiao Wu",
+      "Tao Xiong",
+      "Jiahui Yang",
+      "Yuxuan Yuan",
+      "Ruoceng Zhang",
+      "Shaojie Zhang",
+      "Jian Zhu",
+      "Jian Luan",
+      "Cong Zou"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.31410",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31410",
+    "pdf_url": "https://arxiv.org/pdf/2606.31410",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.29178",
+    "title": "Selective Memory Retention for Long-Horizon LLM Agents",
+    "url": "https://arxiv.org/abs/2606.29178",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Pranath Reddy"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.29178",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29178",
+    "pdf_url": "https://arxiv.org/pdf/2606.29178",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.28456",
+    "title": "Is Lying an Emergent Behaviour in LLMs? Evidence from Gaslighting AI agents in a Sustainability Game",
+    "url": "https://arxiv.org/abs/2606.28456",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Subhendu Bhandary",
+      "Federico Carucci",
+      "Christos Charalambous",
+      "Francesca Dilisante",
+      "Ksenia Dvorkina",
+      "Anna Garbo",
+      "Jiaqi Liang",
+      "Riccardo Vasellini",
+      "Francesco Bertolotti"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "memory",
+      "multi-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.28456",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28456",
+    "pdf_url": "https://arxiv.org/pdf/2606.28456",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.27406",
+    "title": "Towards Evaluation of Implicit Software World Models in Coding LLMs",
+    "url": "https://arxiv.org/abs/2606.27406",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Egor Bogomolov",
+      "Yaroslav Zharov"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "memory",
+      "reasoning",
+      "world-model"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.27406",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27406",
+    "pdf_url": "https://arxiv.org/pdf/2606.27406",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.26627",
+    "title": "Agents That Know Too Much: A Data-Centric Survey of Privacy in LLM Agents",
+    "url": "https://arxiv.org/abs/2606.26627",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Nada Lahjouji",
+      "Ashwin Gerard Colaco"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.26627",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26627",
+    "pdf_url": "https://arxiv.org/pdf/2606.26627",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.26479",
+    "title": "Adaptive Evaluation of Out-of-Band Defenses Against Prompt Injection in LLM Agents",
+    "url": "https://arxiv.org/abs/2606.26479",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Praneeth Narisetty",
+      "Shiva Nagendra Babu Kore",
+      "Uday Kumar Reddy Kattamanchi",
+      "Jayaram Kumarapu"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.26479",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26479",
+    "pdf_url": "https://arxiv.org/pdf/2606.26479",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.26793",
+    "title": "MIRROR: Novelty-Constrained Memory-Guided MCTS Red-Teaming for Agentic RAG",
+    "url": "https://arxiv.org/abs/2606.26793",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Inderjeet Singh",
+      "Andrés Murillo",
+      "Motoyoshi Sekiya",
+      "Yuki Unno",
+      "Junichi Suga"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.26793",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26793",
+    "pdf_url": "https://arxiv.org/pdf/2606.26793",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.27472",
+    "title": "Supersede: Diagnosing and Training the Memory-Update Gap in LLM Agents",
+    "url": "https://arxiv.org/abs/2606.27472",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Vedant Patel"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.27472",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27472",
+    "pdf_url": "https://arxiv.org/pdf/2606.27472",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.25622",
+    "title": "Probabilistic Agents in Deterministic Audits: Evaluating Multi-Agent Systems for Automated Audits Based on the German IT-Grundschutz",
+    "url": "https://arxiv.org/abs/2606.25622",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Lea Roxanne Muth",
+      "Marian Margraf"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.25622",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25622",
+    "pdf_url": "https://arxiv.org/pdf/2606.25622",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.25358",
+    "title": "Agentic Knowledge Tracing: A Multi-Agent LLM Architecture for Stealth Assessment of Financial Literacy in Serious Games",
+    "url": "https://arxiv.org/abs/2606.25358",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Gabriel Santos",
+      "Rita Julia",
+      "Marcelo Nascimento"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.25358",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25358",
+    "pdf_url": "https://arxiv.org/pdf/2606.25358",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.25334",
+    "title": "Bridging the Post-discharge Gap: A Traceable Multi-agent Framework for Safe and Continuous Care",
+    "url": "https://arxiv.org/abs/2606.25334",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Runwei Guan",
+      "Yi Zhou",
+      "Heyi Lin",
+      "Jinjing Zhu",
+      "Mingyuan Hou",
+      "Yang Yang",
+      "Fang Yuan",
+      "Xiaohong Lin",
+      "Shaofeng Liang",
+      "Xuming Hu",
+      "Tao Li",
+      "Tianbin Zhao",
+      "Yutao Yue",
+      "Zhiyuan Wang",
+      "Hui Xiong"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "multi-agent",
+      "rag"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.25334",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25334",
+    "pdf_url": "https://arxiv.org/pdf/2606.25334",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.22673",
+    "title": "AgentLens: Interpretable Safety Steering via Mechanistic Subspaces for Multi-Turn Coding Agent",
+    "url": "https://arxiv.org/abs/2606.22673",
+    "published": "2026-06-21",
+    "updated": "2026-06-21",
+    "authors": [
+      "Weidi Luo",
+      "Qiming Zhang",
+      "Yihao Quan",
+      "Mingyu Jin",
+      "Jie Cai",
+      "Chaowei Xiao",
+      "Jingcheng Niu",
+      "Zhen Xiang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.22673",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22673",
+    "pdf_url": "https://arxiv.org/pdf/2606.22673",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.21710",
+    "title": "PrivacyAlign: Contextual Privacy Alignment for LLM Agents",
+    "url": "https://arxiv.org/abs/2606.21710",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Manveer Singh Tamber",
+      "Abhay Puri",
+      "Marc-Etienne Brunet",
+      "Perouz Taslakian",
+      "Jimmy Lin",
+      "Spandana Gella"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.21710",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21710",
+    "pdf_url": "https://arxiv.org/pdf/2606.21710",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.21013",
+    "title": "Agentic Time Machine as an Infrastructure for Future-Event Forecasting",
+    "url": "https://arxiv.org/abs/2606.21013",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Jingyi Chai",
+      "Bingyang Zheng",
+      "Xiangrui Liu",
+      "Hao Lu",
+      "Zihang Zhou",
+      "Tianchen Wang",
+      "Kemeng Zhang",
+      "Siheng Chen"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.21013",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21013",
+    "pdf_url": "https://arxiv.org/pdf/2606.21013",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.21123",
+    "title": "A Multi-Agent Audit Framework for High-Stakes Reasoning: Evaluation and Interpretability in Clinical Mental Health Screening",
+    "url": "https://arxiv.org/abs/2606.21123",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Jingchen Ye",
+      "Yanpei Yu",
+      "Luyao Zhang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.21123",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21123",
+    "pdf_url": "https://arxiv.org/pdf/2606.21123",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.19899",
+    "title": "Measuring Biological Capabilities and Risks of AI Agents",
+    "url": "https://arxiv.org/abs/2606.19899",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Patricia Paskov",
+      "Jeffrey Lee",
+      "Kyle Brady",
+      "Alyssa Worland"
+    ],
+    "categories": [
+      "cs.CY",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "agentic-ai",
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.19899",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19899",
+    "pdf_url": "https://arxiv.org/pdf/2606.19899",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.20243",
+    "title": "Phoenix: Safe GitHub Issue Resolution via Multi-Agent LLMs",
+    "url": "https://arxiv.org/abs/2606.20243",
+    "published": "2026-06-18",
+    "updated": "2026-06-22",
+    "authors": [
+      "Kipngeno Koech",
+      "Muhammad Adam",
+      "Baimam Boukar Jean Jacques",
+      "Joao Barros"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "multi-agent",
+      "planning",
+      "rag"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.20243",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20243",
+    "pdf_url": "https://arxiv.org/pdf/2606.20243",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.18950",
+    "title": "RTSGameBench: An RTS Benchmark for Strategic Reasoning by Vision-Language Models",
+    "url": "https://arxiv.org/abs/2606.18950",
+    "published": "2026-06-17",
+    "updated": "2026-06-18",
+    "authors": [
+      "San Kim",
+      "Daechul Ahn",
+      "Reokyoung Kim",
+      "Hyeonbeom Choi",
+      "Seungyeon Jwa",
+      "Jonghyun Choi"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.18950",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18950",
+    "pdf_url": "https://arxiv.org/pdf/2606.18950",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.16659",
+    "title": "FraudSMSWalker: Benchmarking Agentic Large Language Models for SMS-to-Webpage Fraud Detection",
+    "url": "https://arxiv.org/abs/2606.16659",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Y. H. Zhou",
+      "Z. M. Ma",
+      "Y. J. Zhou",
+      "Y. T. Li",
+      "H. X. Xiang",
+      "Y. M. Cheng",
+      "T. L. Chen",
+      "K. J. Zhang",
+      "Z. H. Nan",
+      "J. H. Ni",
+      "Z. Wu",
+      "Q. Y. Pan",
+      "S. Zhang",
+      "S. Cheng",
+      "M. Y. Luo"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.16659",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16659",
+    "pdf_url": "https://arxiv.org/pdf/2606.16659",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.17041",
+    "title": "Benchmarking LLM Agents on Meta-Analysis Articles from Nature Portfolio",
+    "url": "https://arxiv.org/abs/2606.17041",
+    "published": "2026-06-15",
+    "updated": "2026-07-01",
+    "authors": [
+      "Anzhe Xie",
+      "Weihang Su",
+      "Yujia Zhou",
+      "Yiqun Liu",
+      "Qingyao Ai"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.17041",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17041",
+    "pdf_url": "https://arxiv.org/pdf/2606.17041",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.16576",
+    "title": "Can LLM Agents Infer World Models? Evidence from Agentic Automata Learning",
+    "url": "https://arxiv.org/abs/2606.16576",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Reef Menaged",
+      "Gili Lior",
+      "Shauli Ravfogel",
+      "Roee Aharoni",
+      "Gabriel Stanovsky"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.16576",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16576",
+    "pdf_url": "https://arxiv.org/pdf/2606.16576",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.12780",
+    "title": "ProPlay: Procedural World Models for Self-Evolving LLM Agents",
+    "url": "https://arxiv.org/abs/2606.12780",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Yijun Ma",
+      "Zehong Wang",
+      "Yiyang Li",
+      "Ziming Li",
+      "Xiaoguang Guo",
+      "Weixiang Sun",
+      "Chuxu Zhang",
+      "Yanfang Ye"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.12780",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12780",
+    "pdf_url": "https://arxiv.org/pdf/2606.12780",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.12320",
+    "title": "A Five-Plane Reference Architecture for Runtime Governance of Production AI Agents",
+    "url": "https://arxiv.org/abs/2606.12320",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Krti Tallam"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CC",
+      "cs.CR",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.12320",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12320",
+    "pdf_url": "https://arxiv.org/pdf/2606.12320",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.11680",
+    "title": "Organize then Retrieve: Hierarchical Memory Navigation for Efficient Agents",
+    "url": "https://arxiv.org/abs/2606.11680",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Hao-Lun Hsu",
+      "Nikki Lijing Kuang",
+      "Boyi Liu",
+      "Zhewei Yao",
+      "Yuxiong He"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.11680",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.11680",
+    "pdf_url": "https://arxiv.org/pdf/2606.11680",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.12657",
+    "title": "TrajGenAgent: A Hierarchical LLM Agent for Human Mobility Trajectory Generation",
+    "url": "https://arxiv.org/abs/2606.12657",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Siyu Li",
+      "Toan Tran",
+      "Lingyi Zhao",
+      "Khurram Shafique",
+      "Li Xiong"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.DB",
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.12657",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12657",
+    "pdf_url": "https://arxiv.org/pdf/2606.12657",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.10677",
+    "title": "Infini Memory: Maintainable Topic Documents for Long-Term LLM Agent Memory",
+    "url": "https://arxiv.org/abs/2606.10677",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Suozhao Ji",
+      "Baodong Wu",
+      "Zehao Wang",
+      "Lei Xia",
+      "Qingping Li",
+      "Ruisong Wang",
+      "Wenbo Ding",
+      "Zhenhua Zhu",
+      "Boxun Li",
+      "Guohao Dai",
+      "Yu Wang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.10677",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10677",
+    "pdf_url": "https://arxiv.org/pdf/2606.10677",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.11354",
+    "title": "A Zero-Shot Multi-Agent Framework for Human-Building Interaction via Programmatic Reasoning",
+    "url": "https://arxiv.org/abs/2606.11354",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Yuqi Wang",
+      "Gulai Shen",
+      "Ali Mehmani"
+    ],
+    "categories": [
+      "cs.ET"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.11354",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.11354",
+    "pdf_url": "https://arxiv.org/pdf/2606.11354",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.05558",
+    "title": "Autoregressive Diffusion World Models for Off-Policy Evaluation of LLM Agents",
+    "url": "https://arxiv.org/abs/2606.05558",
+    "published": "2026-06-04",
+    "updated": "2026-06-04",
+    "authors": [
+      "Kaixuan Liu",
+      "Guojun Xiong",
+      "Weinan Zhang",
+      "Shengpu Tang"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.05558",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.05558",
+    "pdf_url": "https://arxiv.org/pdf/2606.05558",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.04555",
+    "title": "Temporal Order Matters for Agentic Memory: Segment Trees for Long-Horizon Agents",
+    "url": "https://arxiv.org/abs/2606.04555",
+    "published": "2026-06-03",
+    "updated": "2026-06-03",
+    "authors": [
+      "Yifan Simon Liu",
+      "Liam Gallagher",
+      "Faeze Moradi Kalarde",
+      "Jiazhou Liang",
+      "Armin Toroghi",
+      "Scott Sanner"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.04555",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.04555",
+    "pdf_url": "https://arxiv.org/pdf/2606.04555",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.03895",
+    "title": "Agent libOS: A Runtime Substrate for Capability-Controlled Self-Evolving LLM Agents",
+    "url": "https://arxiv.org/abs/2606.03895",
+    "published": "2026-06-02",
+    "updated": "2026-06-29",
+    "authors": [
+      "Yingqi Zhang"
+    ],
+    "categories": [
+      "cs.OS",
+      "cs.AI",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "planning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.03895",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.03895",
+    "pdf_url": "https://arxiv.org/pdf/2606.03895",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.02302",
+    "title": "SeClaw: Spec-Driven Security Task Synthesis for Evaluating Autonomous Agents",
+    "url": "https://arxiv.org/abs/2606.02302",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Hao Cheng",
+      "Changtao Miao",
+      "Tianle Song",
+      "Yin Wu",
+      "He Liu",
+      "Erjia Xiao",
+      "Junchi Chen",
+      "Xiaoyu Shi",
+      "Yichi Wang",
+      "Jing Yang",
+      "Taowen Wang",
+      "Jinhao Duan",
+      "Mengshu Sun",
+      "Peiyan Dong",
+      "Xuan Shen",
+      "Yang Cao",
+      "Renjing Xu",
+      "Kaidi Xu",
+      "Jindong Gu",
+      "Bo Zhang",
+      "Jize Zhang",
+      "Chenhao Lin",
+      "Philip Torr",
+      "Chao Shen"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.02302",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.02302",
+    "pdf_url": "https://arxiv.org/pdf/2606.02302",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.30711",
+    "title": "SAGE: A Novelty Gate for Efficient Memory Evolution in Agentic LLMs",
+    "url": "https://arxiv.org/abs/2605.30711",
+    "published": "2026-05-29",
+    "updated": "2026-06-18",
+    "authors": [
+      "Sijia Wang",
+      "Dhanajit Brahma",
+      "Ricardo Henao"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.LG",
+      "stat.ML"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.30711",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.30711",
+    "pdf_url": "https://arxiv.org/pdf/2605.30711",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.29341",
+    "title": "WorldMemArena: Evaluating Multimodal Agent Memory Through Action-World Interaction",
+    "url": "https://arxiv.org/abs/2605.29341",
+    "published": "2026-05-28",
+    "updated": "2026-06-01",
+    "authors": [
+      "Chengzhi Liu",
+      "Yuzhe Yang",
+      "Sophia Xiao Pu",
+      "Yepeng Liu",
+      "Lin Long",
+      "Yichen Guo",
+      "Nuo Chen",
+      "Zhaotian Weng",
+      "Elena Kochkina",
+      "Simerjot Kaur",
+      "Charese Smiley",
+      "Xiaomo Liu",
+      "James Zou",
+      "Sheng Liu",
+      "Yuheng Bu",
+      "Songyou Peng",
+      "Xin Eric Wang"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.29341",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.29341",
+    "pdf_url": "https://arxiv.org/pdf/2605.29341",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.27690",
+    "title": "TRACES: Proactive Safety Auditing for Multi-Turn LLM Agents via Trajectory-State Modeling",
+    "url": "https://arxiv.org/abs/2605.27690",
+    "published": "2026-05-26",
+    "updated": "2026-05-26",
+    "authors": [
+      "Jiaqian Li",
+      "Yanshu Li",
+      "Boxuan Zhang",
+      "Ruixiang Tang",
+      "Kuan-Hao Huang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.27690",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.27690",
+    "pdf_url": "https://arxiv.org/pdf/2605.27690",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.25141",
+    "title": "LLM Agent Based Renewable Energy Forecasting Using Edge and IoT Data A Review of Solar Wind Weather and Grid Aware Decision Support",
+    "url": "https://arxiv.org/abs/2605.25141",
+    "published": "2026-05-24",
+    "updated": "2026-05-24",
+    "authors": [
+      "Pavan Manjunath",
+      "Thomas Pruefer"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.25141",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.25141",
+    "pdf_url": "https://arxiv.org/pdf/2605.25141",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.19952",
+    "title": "Rethinking How to Remember: Beyond Atomic Facts in Lifelong LLM Agent Memory",
+    "url": "https://arxiv.org/abs/2605.19952",
+    "published": "2026-05-19",
+    "updated": "2026-05-19",
+    "authors": [
+      "Jingwei Sun",
+      "Jianing Zhu",
+      "Jiangchao Yao",
+      "Tongliang Liu",
+      "Bo Han"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.19952",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.19952",
+    "pdf_url": "https://arxiv.org/pdf/2605.19952",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.17625",
+    "title": "Episodic-Semantic Memory Architecture for Long-Horizon Scientific Agents",
+    "url": "https://arxiv.org/abs/2605.17625",
+    "published": "2026-05-17",
+    "updated": "2026-05-17",
+    "authors": [
+      "Nikola Milosevic"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.17625",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.17625",
+    "pdf_url": "https://arxiv.org/pdf/2605.17625",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.16821",
+    "title": "Multi-Paradigm Agent Interaction in Practice:A Systematic Analysis of Generator-Evaluator, ReAct Loop,and Adversarial Evaluation in the buddyMe Framework",
+    "url": "https://arxiv.org/abs/2605.16821",
+    "published": "2026-05-16",
+    "updated": "2026-05-16",
+    "authors": [
+      "Xiaohua Wang",
+      "Chao Han",
+      "Kai Yu",
+      "XiaoLiang Xu",
+      "Liang Wang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "multi-agent",
+      "planning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.16821",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.16821",
+    "pdf_url": "https://arxiv.org/pdf/2605.16821",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.16481",
+    "title": "Visual Agentic Memory: Enabling Online Long Video Understanding via Online Indexing, Hierarchical Memory, and Agentic Retrieval",
+    "url": "https://arxiv.org/abs/2605.16481",
+    "published": "2026-05-15",
+    "updated": "2026-05-15",
+    "authors": [
+      "Aiden Yiliu Li",
+      "Nels Numan",
+      "Anthony Steed"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.16481",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.16481",
+    "pdf_url": "https://arxiv.org/pdf/2605.16481",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.12061",
+    "title": "SAGE: A Self-Evolving Agentic Graph-Memory Engine for Structure-Aware Associative Memory",
+    "url": "https://arxiv.org/abs/2605.12061",
+    "published": "2026-05-12",
+    "updated": "2026-05-12",
+    "authors": [
+      "Juntong Wang",
+      "Haoyue Zhao",
+      "guanghui Pan",
+      "Xiyuan Wang",
+      "Yanbo Wang",
+      "Qiyan Deng",
+      "Muhan Zhang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.12061",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.12061",
+    "pdf_url": "https://arxiv.org/pdf/2605.12061",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.06890",
+    "title": "Beyond the Black Box: Interpretability of Agentic AI Tool Use",
+    "url": "https://arxiv.org/abs/2605.06890",
+    "published": "2026-05-07",
+    "updated": "2026-07-05",
+    "authors": [
+      "Hariom Tatsat",
+      "Ariye Shater"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2605.06890",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.06890",
+    "pdf_url": "https://arxiv.org/pdf/2605.06890",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2605.05716",
+    "title": "More Is Not Always Better: Cross-Component Interference in LLM Agent Scaffolding",
+    "url": "https://arxiv.org/abs/2605.05716",
+    "published": "2026-05-07",
+    "updated": "2026-05-07",
+    "authors": [
+      "Ming Liu"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.05716",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.05716",
+    "pdf_url": "https://arxiv.org/pdf/2605.05716",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.06716",
+    "title": "From Storage to Experience: A Survey on the Evolution of LLM Agent Memory Mechanisms",
+    "url": "https://arxiv.org/abs/2605.06716",
+    "published": "2026-05-07",
+    "updated": "2026-05-07",
+    "authors": [
+      "Jinghao Luo",
+      "Yuchen Tian",
+      "Chuxue Cao",
+      "Ziyang Luo",
+      "Hongzhan Lin",
+      "Kaixin Li",
+      "Chuyi Kong",
+      "Ruichao Yang",
+      "Jing Ma"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.06716",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.06716",
+    "pdf_url": "https://arxiv.org/pdf/2605.06716",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2604.25318",
+    "title": "Cutscene Agent: An LLM Agent Framework for Automated 3D Cutscene Generation",
+    "url": "https://arxiv.org/abs/2604.25318",
+    "published": "2026-04-28",
+    "updated": "2026-04-28",
+    "authors": [
+      "Lanshan He",
+      "Haozhou Pang",
+      "Qi Gan",
+      "Xin Shen",
+      "Ziwei Zhang",
+      "Yibo Liu",
+      "Gang Fang",
+      "Bo Liu",
+      "Kai Sheng",
+      "Shengfeng Zeng",
+      "Chaofan Li",
+      "Zhen Hui",
+      "Keer Zhou",
+      "Lan Zhou",
+      "Shujun Dai"
+    ],
+    "categories": [
+      "cs.GR",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2604.25318",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.25318",
+    "pdf_url": "https://arxiv.org/pdf/2604.25318",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2604.25135",
+    "title": "FAMA: Failure-Aware Meta-Agentic Framework for Open-Source LLMs in Interactive Tool Use Environments",
+    "url": "https://arxiv.org/abs/2604.25135",
+    "published": "2026-04-28",
+    "updated": "2026-04-28",
+    "authors": [
+      "Amir Saeidi",
+      "Venkatesh Mishra",
+      "Souradeep Mukhopadhyay",
+      "Gaowen Liu",
+      "Ali Payani",
+      "Jayanth Srinivasa",
+      "Chitta Baral"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2604.25135",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.25135",
+    "pdf_url": "https://arxiv.org/pdf/2604.25135",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2604.23459",
+    "title": "Architecture Matters for Multi-Agent Security",
+    "url": "https://arxiv.org/abs/2604.23459",
+    "published": "2026-04-25",
+    "updated": "2026-04-25",
+    "authors": [
+      "Ben Hagag",
+      "William L. Anderson",
+      "Christian Schroeder de Witt",
+      "Sarah Scheffler"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.CR",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "memory",
+      "multi-agent",
+      "planning"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.23459",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.23459",
+    "pdf_url": "https://arxiv.org/pdf/2604.23459",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.23374",
+    "title": "Ghost in the Agent: Redefining Information Flow Tracking for LLM Agents",
+    "url": "https://arxiv.org/abs/2604.23374",
+    "published": "2026-04-25",
+    "updated": "2026-04-25",
+    "authors": [
+      "Yuandao Cai",
+      "Wensheng Tang",
+      "Cheng Wen",
+      "Shengchao Qin"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.23374",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.23374",
+    "pdf_url": "https://arxiv.org/pdf/2604.23374",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.22879",
+    "title": "Beyond Single-Agent Alignment: Preventing Context-Fragmented Violations in Multi-Agent Systems",
+    "url": "https://arxiv.org/abs/2604.22879",
+    "published": "2026-04-24",
+    "updated": "2026-04-24",
+    "authors": [
+      "Jie Wu",
+      "Ming Gong"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.AI",
+      "cs.CR",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "rag",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.22879",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.22879",
+    "pdf_url": "https://arxiv.org/pdf/2604.22879",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.18847",
+    "title": "Human-Guided Harm Recovery for Computer Use Agents",
+    "url": "https://arxiv.org/abs/2604.18847",
+    "published": "2026-04-20",
+    "updated": "2026-05-28",
+    "authors": [
+      "Christy Li",
+      "Sky CH-Wang",
+      "Andi Peng",
+      "Andreea Bobu"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.18847",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.18847",
+    "pdf_url": "https://arxiv.org/pdf/2604.18847",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.18245",
+    "title": "Who Tests the Testers? Systematic Enumeration and Coverage Audit of LLM Agent Tool Call Safety",
+    "url": "https://arxiv.org/abs/2603.18245",
+    "published": "2026-03-18",
+    "updated": "2026-03-18",
+    "authors": [
+      "Xuan Chen",
+      "Lu Yan",
+      "Ruqi Zhang",
+      "Xiangyu Zhang"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.18245",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.18245",
+    "pdf_url": "https://arxiv.org/pdf/2603.18245",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.07980",
+    "title": "\\$OneMillion-Bench: How Far are Language Agents from Human Experts?",
+    "url": "https://arxiv.org/abs/2603.07980",
+    "published": "2026-03-09",
+    "updated": "2026-03-09",
+    "authors": [
+      "Qianyu Yang",
+      "Yang Liu",
+      "Jiaqi Li",
+      "Jun Bai",
+      "Hao Chen",
+      "Kaiyuan Chen",
+      "Tiliang Duan",
+      "Jiayun Dong",
+      "Xiaobo Hu",
+      "Zixia Jia",
+      "Yang Liu",
+      "Tao Peng",
+      "Yixin Ren",
+      "Ran Tian",
+      "Zaiyuan Wang",
+      "Yanglihong Xiao",
+      "Gang Yao",
+      "Lingyue Yin",
+      "Ge Zhang",
+      "Chun Zhang",
+      "Jianpeng Jiao",
+      "Zilong Zheng",
+      "Yuan Gong"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.07980",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.07980",
+    "pdf_url": "https://arxiv.org/pdf/2603.07980",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.09002",
+    "title": "Security Considerations for Multi-agent Systems",
+    "url": "https://arxiv.org/abs/2603.09002",
+    "published": "2026-03-09",
+    "updated": "2026-04-26",
+    "authors": [
+      "Tam Nguyen",
+      "Moses Ndebugre",
+      "Dheeraj Arremsetty"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.09002",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.09002",
+    "pdf_url": "https://arxiv.org/pdf/2603.09002",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2602.07962",
+    "title": "LOCA-bench: Benchmarking Language Agents Under Controllable and Extreme Context Growth",
+    "url": "https://arxiv.org/abs/2602.07962",
+    "published": "2026-02-08",
+    "updated": "2026-02-08",
+    "authors": [
+      "Weihao Zeng",
+      "Yuzhen Huang",
+      "Junxian He"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2602.07962",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.07962",
+    "pdf_url": "https://arxiv.org/pdf/2602.07962",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2602.05302",
+    "title": "PieArena: Ranking and Profiling Language Agents in Realistic Negotiation Scenarios",
+    "url": "https://arxiv.org/abs/2602.05302",
+    "published": "2026-02-05",
+    "updated": "2026-06-01",
+    "authors": [
+      "Chris Zhu",
+      "Sasha Cui",
+      "Will Sanok Dufallo",
+      "Runzhi Jin",
+      "Zhen Xu",
+      "Linjun Zhang",
+      "Daylian Cain"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2602.05302",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.05302",
+    "pdf_url": "https://arxiv.org/pdf/2602.05302",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2601.06007",
+    "title": "Don't Break the Cache: An Evaluation of Prompt Caching for Long-Horizon Agentic Tasks",
+    "url": "https://arxiv.org/abs/2601.06007",
+    "published": "2026-01-09",
+    "updated": "2026-01-31",
+    "authors": [
+      "Elias Lumer",
+      "Faheem Nizar",
+      "Akshaya Jangiti",
+      "Kevin Frank",
+      "Anmol Gulati",
+      "Mandar Phadate",
+      "Vamse Kumar Subbiah"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2601.06007",
+    "source": "arxiv",
+    "source_id": "arxiv:2601.06007",
+    "pdf_url": "https://arxiv.org/pdf/2601.06007",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2511.04847",
+    "title": "Test-Time Adaptation for LLM Agents via Environment Interaction",
+    "url": "https://arxiv.org/abs/2511.04847",
+    "published": "2025-11-06",
+    "updated": "2026-02-22",
+    "authors": [
+      "Arthur Chen",
+      "Zuxin Liu",
+      "Jianguo Zhang",
+      "Akshara Prabhakar",
+      "Zhiwei Liu",
+      "Shelby Heinecke",
+      "Silvio Savarese",
+      "Victor Zhong",
+      "Caiming Xiong"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "embodied-agent",
+      "rag",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2511.04847",
+    "source": "arxiv",
+    "source_id": "arxiv:2511.04847",
+    "pdf_url": "https://arxiv.org/pdf/2511.04847",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2509.10769",
+    "title": "AgentArch: A Comprehensive Benchmark to Evaluate Agent Architectures in Enterprise",
+    "url": "https://arxiv.org/abs/2509.10769",
+    "published": "2025-09-13",
+    "updated": "2026-01-06",
+    "authors": [
+      "Tara Bogavelli",
+      "Roshnee Sharma",
+      "Hari Subramani"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2509.10769",
+    "source": "arxiv",
+    "source_id": "arxiv:2509.10769",
+    "pdf_url": "https://arxiv.org/pdf/2509.10769",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2607.06273",
+    "title": "AgentTether: Graph-Guided Diagnosis and Runtime Intervention for Reliable LLM Agent Operation",
+    "url": "https://arxiv.org/abs/2607.06273",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Chenyu Zhao",
+      "Shenglin Zhang",
+      "Wenwei Gu",
+      "Yongqian Sun",
+      "Dan Pei",
+      "Chetan Bansal",
+      "Saravan Rajmohan",
+      "Minghua Ma"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.06273",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06273",
+    "pdf_url": "https://arxiv.org/pdf/2607.06273",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.06195",
+    "title": "LogicHunter: Testing LLM Agent Frameworks with an Agentic Oracle",
+    "url": "https://arxiv.org/abs/2607.06195",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Minghui Long",
+      "Yanjie Zhao",
+      "Haoyu Wang"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.06195",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06195",
+    "pdf_url": "https://arxiv.org/pdf/2607.06195",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.06080",
+    "title": "From Blueprint to Reality: Modeling and Applying Putnam's Social Capital Theory with LLM-based Multi-agent Simulations",
+    "url": "https://arxiv.org/abs/2607.06080",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Shiyi Ling",
+      "Zhi Zheng",
+      "Hui Zheng",
+      "Wenjun Xue",
+      "Feng Ye",
+      "Tong Xu"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.SI"
+    ],
+    "topics": [
+      "agent-safety",
+      "multi-agent",
+      "rag",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.06080",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06080",
+    "pdf_url": "https://arxiv.org/pdf/2607.06080",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.05297",
+    "title": "MetaSkill-Evolve: Recursive Self-Improvement of LLM Agents via Two-Timescale Meta-Skill Evolution",
+    "url": "https://arxiv.org/abs/2607.05297",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Zefeng Wang",
+      "Minxi Yan",
+      "Jinhe Bi",
+      "Sikuan Yan",
+      "Volker Tresp",
+      "Yunpu Ma"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.05297",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05297",
+    "pdf_url": "https://arxiv.org/pdf/2607.05297",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.04528",
+    "title": "Measuring Harness-Induced Belief Divergence in Multi-Step LLM Agents",
+    "url": "https://arxiv.org/abs/2607.04528",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Haiwen Yi",
+      "Xinyuan Song"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.04528",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04528",
+    "pdf_url": "https://arxiv.org/pdf/2607.04528",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.04293",
+    "title": "CausalGame: Benchmarking Causal Thinking of LLM Agents in Games",
+    "url": "https://arxiv.org/abs/2607.04293",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Zhenhao Chen",
+      "Yongqiang Chen",
+      "Chenxi Liu",
+      "Junchi Yu",
+      "Xiangchen Song",
+      "Zijian Li",
+      "Jialin Li",
+      "Philip Torr",
+      "Bo Han",
+      "Kun Zhang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.LG",
+      "stat.ML"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "reasoning"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.04293",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04293",
+    "pdf_url": "https://arxiv.org/pdf/2607.04293",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.04426",
+    "title": "ACE-Brain-0.5: A Unified Embodied Foundational Model for Physical Agentic AI",
+    "url": "https://arxiv.org/abs/2607.04426",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "ACE-Brain Team",
+      ":",
+      "Ziyang Gong",
+      "Haoming Gu",
+      "Zehang Luo",
+      "Tianyi Zhang",
+      "Tao Tao",
+      "Yixiao Chi",
+      "Zhe Liu",
+      "Lingsi Zhu",
+      "Jingyuan Liu",
+      "Anke Tang",
+      "Songze Li",
+      "Yilun Kong",
+      "Ningjing Liu",
+      "Tianyu Zhu",
+      "Yunpeng Qing",
+      "Shuang Luo",
+      "Xiang Liu",
+      "Shi Fu",
+      "Dawei Nie",
+      "Sixiang Liu",
+      "Zhexi Wen",
+      "Feng Pan",
+      "Xiaofeng Wang",
+      "Zhi Hou",
+      "Chunxiao Liu",
+      "Xue Yang",
+      "Junchi Yan",
+      "Hengshuang Zhao",
+      "Dacheng Tao",
+      "Xiaogang Wang"
+    ],
+    "categories": [
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.04426",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04426",
+    "pdf_url": "https://arxiv.org/pdf/2607.04426",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.04162",
+    "title": "ACE: Agentic Control for Embodied Manipulation via Zero-shot Workflow Reasoning",
+    "url": "https://arxiv.org/abs/2607.04162",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Iok Tong Lei",
+      "QianZhi Li",
+      "Ying Jie Yap",
+      "Yujie Zhang",
+      "Rui Zhong",
+      "Haichao Gui",
+      "Xiaolong Liu",
+      "Zhidong Deng"
+    ],
+    "categories": [
+      "cs.RO",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.04162",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04162",
+    "pdf_url": "https://arxiv.org/pdf/2607.04162",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.03333",
+    "title": "SPORK: Self-Speculative Forking to Accelerate Agentic LLM Inference",
+    "url": "https://arxiv.org/abs/2607.03333",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Huajun Bai",
+      "Weiwei Lv",
+      "Huichuan Zheng",
+      "Youyou Lu",
+      "Jiwu Shu"
+    ],
+    "categories": [
+      "cs.DC",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.03333",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03333",
+    "pdf_url": "https://arxiv.org/pdf/2607.03333",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.02857",
+    "title": "MOSAIC: Knowledge-Guided CLI Command Composition Attack in LLM Coding Agents",
+    "url": "https://arxiv.org/abs/2607.02857",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Jiangrong Wu",
+      "Huaijin Wang",
+      "Yihao Zhang",
+      "Yuhong Nan",
+      "Shuai Wang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.02857",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02857",
+    "pdf_url": "https://arxiv.org/pdf/2607.02857",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.01935",
+    "title": "A-TMA: Decoupling State-Aware Memory Failures in Long-Term Agent Memory",
+    "url": "https://arxiv.org/abs/2607.01935",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Zitong Shi",
+      "Yixuan Tang",
+      "Anthony Kum Hoe Tung"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.01935",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01935",
+    "pdf_url": "https://arxiv.org/pdf/2607.01935",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2607.01640",
+    "title": "AgentFlow: Building Agent Dependency Graphs for Static Analysis of Agent Programs",
+    "url": "https://arxiv.org/abs/2607.01640",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Shenao Wang",
+      "Xinyi Hou",
+      "Yanjie Zhao",
+      "Xiao Cheng",
+      "Haoyu Wang"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.01640",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01640",
+    "pdf_url": "https://arxiv.org/pdf/2607.01640",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.01668",
+    "title": "VeriChat: An Agentic Conversational AI Assistant for Hardware Security Verification",
+    "url": "https://arxiv.org/abs/2607.01668",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Dipayan Saha",
+      "Khan Thamid Hasan",
+      "Shams Tarek",
+      "Sujan Kumar Saha",
+      "Mark Tehranipoor",
+      "Farimah Farahmandi"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "rag",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.01668",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01668",
+    "pdf_url": "https://arxiv.org/pdf/2607.01668",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.02294",
+    "title": "Coding Agents Are Guessing: Measuring Action-Boundary Violations in Underspecified DevOps Instructions",
+    "url": "https://arxiv.org/abs/2607.02294",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Zimo Ji",
+      "Zekai Zhang",
+      "Congying Xu",
+      "Zongjie Li",
+      "Yudong Gao",
+      "Shuai Wang",
+      "Shing-Chi Cheung"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.02294",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02294",
+    "pdf_url": "https://arxiv.org/pdf/2607.02294",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.01916",
+    "title": "ContextSniper: AntTrail's Token-Efficient Code Memory for Repository-Level Program Repair",
+    "url": "https://arxiv.org/abs/2607.01916",
+    "published": "2026-07-02",
+    "updated": "2026-07-06",
+    "authors": [
+      "Chiwang Luk",
+      "Matin Mohammad Najafi",
+      "Zhifeng Jia",
+      "Wei Yang",
+      "Xiuchang Li",
+      "Jinwei Zhu",
+      "Yang Ren",
+      "Lei Chen",
+      "Gao Cong"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "coding-agent",
+      "rag-agent"
+    ],
+    "arxiv_id": "2607.01916",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01916",
+    "pdf_url": "https://arxiv.org/pdf/2607.01916",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2607.01929",
+    "title": "Beyond Textual Repository Exploration: Dual-Modal Structural Reasoning for Agentic Issue Resolution",
+    "url": "https://arxiv.org/abs/2607.01929",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Jiayi Zhang",
+      "Kai Huang",
+      "Yang Liu",
+      "Chunyang Chen"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "embodied-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent",
+      "function-calling"
+    ],
+    "arxiv_id": "2607.01929",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01929",
+    "pdf_url": "https://arxiv.org/pdf/2607.01929",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.01071",
+    "title": "MemSyco-Bench: Benchmarking Sycophancy in Agent Memory",
+    "url": "https://arxiv.org/abs/2607.01071",
+    "published": "2026-07-01",
+    "updated": "2026-07-02",
+    "authors": [
+      "Zhishang Xiang",
+      "Zerui Chen",
+      "Yunbo Tang",
+      "Zhimin Wei",
+      "Ruqin Ning",
+      "Yujie Lin",
+      "Qinggang Zhang",
+      "Jinsong Su"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "reasoning"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2607.01071",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01071",
+    "pdf_url": "https://arxiv.org/pdf/2607.01071",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2607.00939",
+    "title": "Leveraging LLM-Based Agentic Systems to Generate Quantum Applications for Test Optimization",
+    "url": "https://arxiv.org/abs/2607.00939",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Ming Tao",
+      "Yuechen Li",
+      "Tao Yue",
+      "Man Zhang",
+      "Aitor Arrieta Marcos"
+    ],
+    "categories": [
+      "cs.SE",
+      "quant-ph"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.00939",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00939",
+    "pdf_url": "https://arxiv.org/pdf/2607.00939",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.00334",
+    "title": "Managed Autonomy at Runtime: Gear-Based Safety and Governance for Single- and Multi-Agent Cyber-Physical Systems",
+    "url": "https://arxiv.org/abs/2607.00334",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Srini Ramaswamy",
+      "Wang Miaosheng"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "embodied-agent",
+      "multi-agent",
+      "planning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.00334",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00334",
+    "pdf_url": "https://arxiv.org/pdf/2607.00334",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2607.05428",
+    "title": "CHARLIE: An On-Premise Multi-Agent Retrieval-Augmented Generation System for Evidential Reasoning in Forensic Science",
+    "url": "https://arxiv.org/abs/2607.05428",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Leandro D. Carneiro",
+      "Andre L. S. Meirelles",
+      "Juliano de A. Gomes",
+      "Rafael C. A. Cabral"
+    ],
+    "categories": [
+      "cs.DL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2607.05428",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05428",
+    "pdf_url": "https://arxiv.org/pdf/2607.05428",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.31693",
+    "title": "ShopX: A Foundation Model for Intent-to-Item Fulfillment in Agentic Shopping",
+    "url": "https://arxiv.org/abs/2606.31693",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Jiacheng Chen",
+      "Tao Zhang",
+      "Manxi Lin",
+      "Dunxian Huang",
+      "Teng Shi",
+      "Honghao Fu",
+      "Mengyan Li",
+      "Xinming Zhang",
+      "Chenchi Zhang",
+      "Xuan Lu",
+      "Xiaoxiong Du",
+      "Haibin Chen",
+      "Shaolin Ye",
+      "Hao Chang",
+      "Xiaoqi Li",
+      "Shuwen Xiao",
+      "Yujin Yuan",
+      "Jingxuan Feng",
+      "Shaopan Xiong",
+      "Huimin Yi",
+      "Ju Huang",
+      "Qiu Shen",
+      "Ying Chen",
+      "Junjun Zheng",
+      "Xiangheng Kong",
+      "Yuning Jiang"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.31693",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31693",
+    "pdf_url": "https://arxiv.org/pdf/2606.31693",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.31252",
+    "title": "Embodied CAD: Solver-Grounded LLM Agents for Parametric B-Rep Assembly Modeling",
+    "url": "https://arxiv.org/abs/2606.31252",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Fumin Liu",
+      "Haoyu Zhou",
+      "Fei Hao",
+      "Lin Yang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.31252",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31252",
+    "pdf_url": "https://arxiv.org/pdf/2606.31252",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.31174",
+    "title": "ClawArena-Team: Benchmarking Subagent Orchestration and Dynamic Workflows in Language-Model Agents",
+    "url": "https://arxiv.org/abs/2606.31174",
+    "published": "2026-06-30",
+    "updated": "2026-07-02",
+    "authors": [
+      "Kaiwen Xiong",
+      "Haonian Ji",
+      "Shi Qiu",
+      "Zeyu Zheng",
+      "Cihang Xie",
+      "Xinyu Ye",
+      "Huaxiu Yao"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.31174",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31174",
+    "pdf_url": "https://arxiv.org/pdf/2606.31174",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.31046",
+    "title": "OpenLife: Toward Open-World Artificial Life with Autonomous LLM Agents",
+    "url": "https://arxiv.org/abs/2606.31046",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Atsushi Masumori",
+      "Itsuki Doi",
+      "Norihiro Maruyama",
+      "Ryosuke Takata",
+      "Takashi Ikegami"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.31046",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31046",
+    "pdf_url": "https://arxiv.org/pdf/2606.31046",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.31650",
+    "title": "ECHO: Prune to act, trace to learn with selective turn memory in agentic RL",
+    "url": "https://arxiv.org/abs/2606.31650",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Zijun Xie",
+      "Binbin Zheng",
+      "Enlei Gong",
+      "Jihua Liu",
+      "Yuyang You",
+      "Lingfeng Liu",
+      "Jiayao Tang",
+      "Guanqun Zhao",
+      "Aoqi Hu",
+      "Zeyu Chen"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.31650",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31650",
+    "pdf_url": "https://arxiv.org/pdf/2606.31650",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.31639",
+    "title": "A Lifecycle and Application-Stack Survey of Large Language Model Vulnerabilities: Attacks, Risks, Defenses, and Open Problems",
+    "url": "https://arxiv.org/abs/2606.31639",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Seyed Bagher Hashemi Natanzi",
+      "Bo Tang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.GT",
+      "cs.LO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.31639",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31639",
+    "pdf_url": "https://arxiv.org/pdf/2606.31639",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.30566",
+    "title": "Forensic Trajectory Signatures for Agent Memory Poisoning Detection",
+    "url": "https://arxiv.org/abs/2606.30566",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Jun Wen Leong"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.30566",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30566",
+    "pdf_url": "https://arxiv.org/pdf/2606.30566",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.29914",
+    "title": "MemDelta: Controlled Baselines and Hidden Confounds in Agent Memory Evaluation",
+    "url": "https://arxiv.org/abs/2606.29914",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Kuan Wang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.29914",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29914",
+    "pdf_url": "https://arxiv.org/pdf/2606.29914",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.30555",
+    "title": "Linguistic Firewall: Geometry as Defense in Multi-Agent Systems Routing",
+    "url": "https://arxiv.org/abs/2606.30555",
+    "published": "2026-06-29",
+    "updated": "2026-07-05",
+    "authors": [
+      "Dvir Alsheich",
+      "Adar Peleg",
+      "Ben Hagag",
+      "Rom Himelstein",
+      "Amit Levi",
+      "Avi Mendelson"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.30555",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30555",
+    "pdf_url": "https://arxiv.org/pdf/2606.30555",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.30546",
+    "title": "MAS-Lab: A Specification-Driven Validation Framework for Reliable Multi-Agent Systems",
+    "url": "https://arxiv.org/abs/2606.30546",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Jordan Augé",
+      "Giovanna Carofiglio",
+      "Giulio Grassi",
+      "Jacques Samain"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.30546",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30546",
+    "pdf_url": "https://arxiv.org/pdf/2606.30546",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.30259",
+    "title": "Multi-Agentic System Leveraging Open-Source LLMs to Mitigate Disinformation Threats",
+    "url": "https://arxiv.org/abs/2606.30259",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Sebastian Kula",
+      "Martin Tamajka"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.30259",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30259",
+    "pdf_url": "https://arxiv.org/pdf/2606.30259",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.29742",
+    "title": "MicroAgent: Context-Augmented Multi-Agent Framework for Automatic Microservice Decomposition",
+    "url": "https://arxiv.org/abs/2606.29742",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Zishan Su",
+      "Junjie Huang",
+      "Shiwen Shan",
+      "Xingyan Chen",
+      "Hui Zeng",
+      "Yuxin Su",
+      "Yanlin Wang",
+      "Michael R. Lyu"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.29742",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29742",
+    "pdf_url": "https://arxiv.org/pdf/2606.29742",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.29270",
+    "title": "Minority Sentinel: When to Overturn Majority Voting in Multi-Agent LLM Debates",
+    "url": "https://arxiv.org/abs/2606.29270",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Chuan He",
+      "Zebin Chen",
+      "Zhengyi Yang",
+      "Shaobo Qiao",
+      "Mingchen Ju",
+      "Jiate Liu",
+      "Dong Wen",
+      "Guanfeng Liu"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "reasoning"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.29270",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29270",
+    "pdf_url": "https://arxiv.org/pdf/2606.29270",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.29654",
+    "title": "Budgeted Act-or-Defer Multi-Agent LLM Deliberation with Local Reliability Bounds",
+    "url": "https://arxiv.org/abs/2606.29654",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Mengdie Flora Wang",
+      "Haochen Xie",
+      "Guanghui Wang",
+      "Devin Zhang",
+      "Jae Oh Woo"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.29654",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29654",
+    "pdf_url": "https://arxiv.org/pdf/2606.29654",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.28958",
+    "title": "When Latent Agents Lie: KV-Cache Integrity in Multi-Agent LLM Collaboration",
+    "url": "https://arxiv.org/abs/2606.28958",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Luís Brito",
+      "Carlos Baquero"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-safety",
+      "memory",
+      "multi-agent",
+      "reasoning"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.28958",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28958",
+    "pdf_url": "https://arxiv.org/pdf/2606.28958",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.28666",
+    "title": "Why Trust Your Agent? Empirical Security Gains from TRiSM-Guided Agentic Workflows in Healthcare",
+    "url": "https://arxiv.org/abs/2606.28666",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Liam Kearns"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.28666",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28666",
+    "pdf_url": "https://arxiv.org/pdf/2606.28666",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.27990",
+    "title": "AdvancedShelLM: A Stateful Multi-Agent LLM Honeypot for SSH Deception",
+    "url": "https://arxiv.org/abs/2606.27990",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Muris Sladić",
+      "Eman Alibalić",
+      "Veronica Valeros",
+      "Carlos Catania",
+      "Sebastian Garcia"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.27990",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27990",
+    "pdf_url": "https://arxiv.org/pdf/2606.27990",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.27929",
+    "title": "When Multi-Robot Systems Meet Agentic AI:Towards Embodied Collective Intelligence",
+    "url": "https://arxiv.org/abs/2606.27929",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Yuxuan Yan",
+      "Yuanyuan Jia",
+      "Qianqian Yang"
+    ],
+    "categories": [
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.27929",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27929",
+    "pdf_url": "https://arxiv.org/pdf/2606.27929",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.28434",
+    "title": "SWE-MeM: Learning Adaptive Memory Management for Long-Horizon Coding Agents",
+    "url": "https://arxiv.org/abs/2606.28434",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Shuzheng Gao",
+      "Wenhao Zeng",
+      "Zhaojian Yu",
+      "Jianqiao Wangni",
+      "Chaozheng Wang",
+      "Kai Cai",
+      "Shilin He",
+      "Michael R. Lyu"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "coding-agent",
+      "memory",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.28434",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28434",
+    "pdf_url": "https://arxiv.org/pdf/2606.28434",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.28570",
+    "title": "Digitizing Coaching Intelligence: An Agentic Framework for Holistic Athlete Profiling using VLM and RAG",
+    "url": "https://arxiv.org/abs/2606.28570",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Deep Ghosal",
+      "Ishani Sen",
+      "Wazib Ansar",
+      "Amlan Chakrabarti"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.28570",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28570",
+    "pdf_url": "https://arxiv.org/pdf/2606.28570",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.28182",
+    "title": "LLawCo: Learning Laws of Cooperation for Modeling Embodied Multi-Agent Behavior",
+    "url": "https://arxiv.org/abs/2606.28182",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Qinhong Zhou",
+      "Chuang Gan",
+      "Anoop Cherian"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CV",
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.28182",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28182",
+    "pdf_url": "https://arxiv.org/pdf/2606.28182",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.25514",
+    "title": "Unlocking Model Potentials Through Adaptive Multi-Agent Scaffolding for Efficient Issue Resolution",
+    "url": "https://arxiv.org/abs/2606.25514",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Yang Chen",
+      "Aliya Ahmad",
+      "Yiheng Zhou",
+      "Reyhaneh Jabbarvand"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.25514",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25514",
+    "pdf_url": "https://arxiv.org/pdf/2606.25514",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.25588",
+    "title": "IntentTester: Intent-Driven Multi-agent Framework for Cross-Library Test Migration",
+    "url": "https://arxiv.org/abs/2606.25588",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Yi Gao",
+      "Ziyuan Zhang",
+      "Xing Hu",
+      "Xiaohu Yang",
+      "Xin Xia"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.25588",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25588",
+    "pdf_url": "https://arxiv.org/pdf/2606.25588",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.25400",
+    "title": "BrainAgent: A Large Language Model-Driven Multi-Agent Framework for Autonomous Brain Signal Understanding",
+    "url": "https://arxiv.org/abs/2606.25400",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Yangxuan Zhou",
+      "Sha Zhao",
+      "Jiquan Wang",
+      "Shijian Li",
+      "Gang Pan"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.25400",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25400",
+    "pdf_url": "https://arxiv.org/pdf/2606.25400",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.24775",
+    "title": "Are We Ready For An Agent-Native Memory System?",
+    "url": "https://arxiv.org/abs/2606.24775",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Wei Zhou",
+      "Xuanhe Zhou",
+      "Shaokun Han",
+      "Hongming Xu",
+      "Guoliang Li",
+      "Zhiyu Li",
+      "Feiyu Xiong",
+      "Fan Wu"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.DB",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.24775",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24775",
+    "pdf_url": "https://arxiv.org/pdf/2606.24775",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.24649",
+    "title": "Agentic Collaborative Cognition for Zero-Shot 3D Understanding",
+    "url": "https://arxiv.org/abs/2606.24649",
+    "published": "2026-06-23",
+    "updated": "2026-06-25",
+    "authors": [
+      "Wenxin Wang",
+      "Bo Zhang",
+      "Feng Chen",
+      "Zixuan Wang",
+      "Wen Li",
+      "Changsheng Li",
+      "Yinjie Lei"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent",
+      "planning",
+      "rag"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.24649",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24649",
+    "pdf_url": "https://arxiv.org/pdf/2606.24649",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.24437",
+    "title": "ReM-MoA: Reasoning Memory Sustains Mixture-of-Agents Scaling",
+    "url": "https://arxiv.org/abs/2606.24437",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Heng Ping",
+      "Arijit Bhattacharjee",
+      "Peiyu Zhang",
+      "Shixuan Li",
+      "Wei Yang",
+      "Ali Jannesari",
+      "Nesreen Ahmed",
+      "Paul Bogdan"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "reasoning"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.24437",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24437",
+    "pdf_url": "https://arxiv.org/pdf/2606.24437",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.22647",
+    "title": "RAVEN: Agentic RAG for Automated Vulnerability Repair",
+    "url": "https://arxiv.org/abs/2606.22647",
+    "published": "2026-06-21",
+    "updated": "2026-06-21",
+    "authors": [
+      "Varun Gadey",
+      "Zijie Liu",
+      "Alexandra Dmitrienko"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.LG",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "memory",
+      "rag"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.22647",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22647",
+    "pdf_url": "https://arxiv.org/pdf/2606.22647",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.22330",
+    "title": "Hypothesis-Driven Skill Optimization for LLM Agents",
+    "url": "https://arxiv.org/abs/2606.22330",
+    "published": "2026-06-21",
+    "updated": "2026-06-21",
+    "authors": [
+      "Fangxin Shang",
+      "Yehui Yang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.22330",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22330",
+    "pdf_url": "https://arxiv.org/pdf/2606.22330",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.21740",
+    "title": "Training the Orchestrator: A Supervised Approach to End-to-End PDDL Planning with LLM Agents",
+    "url": "https://arxiv.org/abs/2606.21740",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Rajesh Mangannavar",
+      "Zachary Coalson",
+      "Pranay Dugar",
+      "Prasad Tadepalli"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.21740",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21740",
+    "pdf_url": "https://arxiv.org/pdf/2606.21740",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.18467",
+    "title": "ToolChain-CRC: Conformal Risk Control for Agentic AI Under Retrieval and Tool-Use Drift",
+    "url": "https://arxiv.org/abs/2606.18467",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Jeffery Opoku",
+      "David Banahene"
+    ],
+    "categories": [
+      "stat.ML",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "agentic-ai",
+      "ai-agent",
+      "rag-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.18467",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18467",
+    "pdf_url": "https://arxiv.org/pdf/2606.18467",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.18142",
+    "title": "Your AI Travel Agent Would Book You a Bullfight: An Agentic Benchmark for Implicit Animal Welfare in Frontier AI Models",
+    "url": "https://arxiv.org/abs/2606.18142",
+    "published": "2026-06-16",
+    "updated": "2026-07-06",
+    "authors": [
+      "Jasmine Brazilek",
+      "Joel Christoph",
+      "Maheep Chaudhary",
+      "Oliver Tullio",
+      "Carol Kline",
+      "Miles Tidmarsh",
+      "Arturs Kanepajs"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.CY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.18142",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18142",
+    "pdf_url": "https://arxiv.org/pdf/2606.18142",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.16591",
+    "title": "SING: Synthetic Intention Graph for Scalable Active Tool Discovery in LLM Agents",
+    "url": "https://arxiv.org/abs/2606.16591",
+    "published": "2026-06-15",
+    "updated": "2026-06-16",
+    "authors": [
+      "Qiao Xiao",
+      "Haochen Shi",
+      "Yisen Gao",
+      "Wenbin Hu",
+      "Huihao Jing",
+      "Tianshi Zheng",
+      "Baixuan Xu",
+      "Ziheng Zhang",
+      "Weiqi Wang",
+      "Haoran Li",
+      "Jiaxin Bai",
+      "Yangqiu Song"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.16591",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16591",
+    "pdf_url": "https://arxiv.org/pdf/2606.16591",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.15684",
+    "title": "Multi-agent Framework for Time-Sensitive Complementary Collaboration in Minecraft",
+    "url": "https://arxiv.org/abs/2606.15684",
+    "published": "2026-06-14",
+    "updated": "2026-06-14",
+    "authors": [
+      "Juheon Yi",
+      "Jinglu Wang",
+      "Xiaoyi Zhang",
+      "Yan Lu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.15684",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15684",
+    "pdf_url": "https://arxiv.org/pdf/2606.15684",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.15931",
+    "title": "DeepRoot: A KG-Coordinated Multi-Agent System for Therapeutic Reasoning over Historical Medical Texts",
+    "url": "https://arxiv.org/abs/2606.15931",
+    "published": "2026-06-14",
+    "updated": "2026-06-14",
+    "authors": [
+      "Zijian Carl Ma",
+      "Sean J. Wang",
+      "Sijbren Kramer",
+      "Li Erran Li"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.15931",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15931",
+    "pdf_url": "https://arxiv.org/pdf/2606.15931",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.12703",
+    "title": "SMSR: Certified Defence Against Runtime Memory Poisoning in Persistent LLM Agent Systems",
+    "url": "https://arxiv.org/abs/2606.12703",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Tarun Sharma"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.12703",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12703",
+    "pdf_url": "https://arxiv.org/pdf/2606.12703",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.10684",
+    "title": "Divide and Cooperate: Role-Decomposed Multi-Agent LLM Training with Cross-Agent Learning Signals",
+    "url": "https://arxiv.org/abs/2606.10684",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Jaewan Park",
+      "Solbee Cho",
+      "Jay-Yoon Lee"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.10684",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10684",
+    "pdf_url": "https://arxiv.org/pdf/2606.10684",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.10616",
+    "title": "Learning What to Remember: Observability-Safe Memory Retention via Constrained Optimization for Long-Horizon Language Agents",
+    "url": "https://arxiv.org/abs/2606.10616",
+    "published": "2026-06-09",
+    "updated": "2026-06-29",
+    "authors": [
+      "Qingcan Kang",
+      "Liu Mingyang",
+      "Shixiong Kai",
+      "Kaichao Liang",
+      "Tao Zhong",
+      "Mingxuan Yuan"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.10616",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10616",
+    "pdf_url": "https://arxiv.org/pdf/2606.10616",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.10933",
+    "title": "Frontier Coding Agents Use Metaprogramming to Adapt to Unfamiliar Programming Languages",
+    "url": "https://arxiv.org/abs/2606.10933",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Aman Sharma",
+      "Sushrut Thorat",
+      "Paras Chopra"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.10933",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10933",
+    "pdf_url": "https://arxiv.org/pdf/2606.10933",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.10577",
+    "title": "AgenticNav: Zero-Shot Vision-and-Language Navigation as a Tool-Calling Harness",
+    "url": "https://arxiv.org/abs/2606.10577",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Yijian Li",
+      "Changze Li",
+      "Hantian Shi",
+      "Jiaying Luo",
+      "Jiyuan Cai",
+      "Ming Yang",
+      "Tong Qin"
+    ],
+    "categories": [
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.10577",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10577",
+    "pdf_url": "https://arxiv.org/pdf/2606.10577",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.10304",
+    "title": "MIRAGE: A Polarity-Flipping Encoding Subspace in LLM Agents",
+    "url": "https://arxiv.org/abs/2606.10304",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Pratibha Revankar",
+      "Kargi Chauhan",
+      "Jihye Kim",
+      "Sadiba Nusrat Nur",
+      "Vincent Siu",
+      "Chenguang Wang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.10304",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10304",
+    "pdf_url": "https://arxiv.org/pdf/2606.10304",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.07682",
+    "title": "SWE-Marathon: Can Agents Autonomously Complete Ultra-Long-Horizon Software Work?",
+    "url": "https://arxiv.org/abs/2606.07682",
+    "published": "2026-06-05",
+    "updated": "2026-06-05",
+    "authors": [
+      "Rishi Desai",
+      "Jesse Hu",
+      "Joan Cabezas",
+      "Neel Harsola",
+      "Pratyush Shukla",
+      "Roey Ben Chaim",
+      "Adnan El Assadi",
+      "Omkaar Mukund Kamath",
+      "Fenil Faldu",
+      "Prannay Hebbar",
+      "Jiankai Sun",
+      "Yiyuan Li",
+      "Pramod Srinivasan",
+      "Ishan Gupta",
+      "Christopher Settles",
+      "Daniel Wang",
+      "Derek Chen",
+      "Pranav Raja",
+      "Albert Liu",
+      "Marek Šuppa",
+      "Nevasini Sasikumar",
+      "Luyang Kong",
+      "Erik Quintanilla",
+      "Xiangyi Li",
+      "Ivan Bercovich",
+      "Steven Dillmann"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.07682",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.07682",
+    "pdf_url": "https://arxiv.org/pdf/2606.07682",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.07867",
+    "title": "The Cold-Start Safety Gap in LLM Agents",
+    "url": "https://arxiv.org/abs/2606.07867",
+    "published": "2026-06-05",
+    "updated": "2026-06-05",
+    "authors": [
+      "Chung-En Sun",
+      "Linbo Liu",
+      "Tsui-Wei Weng"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2606.07867",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.07867",
+    "pdf_url": "https://arxiv.org/pdf/2606.07867",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.07711",
+    "title": "Rosetta Memory: Adaptive Memory for Cross-LLM Agents",
+    "url": "https://arxiv.org/abs/2606.07711",
+    "published": "2026-06-05",
+    "updated": "2026-06-05",
+    "authors": [
+      "Hao Yang",
+      "Shiqi Shen",
+      "Haoxuan Li",
+      "Zhipeng Wang",
+      "Zhi Gong",
+      "Xu Chen"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "coding-agent",
+      "memory",
+      "planning",
+      "reasoning"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.07711",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.07711",
+    "pdf_url": "https://arxiv.org/pdf/2606.07711",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.06054",
+    "title": "Beyond Similarity: Trustworthy Memory Search for Personal AI Agents",
+    "url": "https://arxiv.org/abs/2606.06054",
+    "published": "2026-06-04",
+    "updated": "2026-06-04",
+    "authors": [
+      "Jiawen Zhang",
+      "Kejia Chen",
+      "Jiachen Ma",
+      "Yangfan Hu",
+      "Lipeng He",
+      "Yechao Zhang",
+      "Jian Liu",
+      "Xiaohu Yang",
+      "Tianwei Zhang",
+      "Ruoxi Jia"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.06054",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.06054",
+    "pdf_url": "https://arxiv.org/pdf/2606.06054",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.05805",
+    "title": "From Risk Classification to Action Plan Remediation: A Guardrail Feedback Driven Framework for LLM Agents",
+    "url": "https://arxiv.org/abs/2606.05805",
+    "published": "2026-06-04",
+    "updated": "2026-06-04",
+    "authors": [
+      "Yuhao Sun",
+      "Jiacheng Zhang",
+      "Shaanan Cohney",
+      "Zhexin Zhang",
+      "Feng Liu",
+      "Xingliang Yuan"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.05805",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.05805",
+    "pdf_url": "https://arxiv.org/pdf/2606.05805",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.04599",
+    "title": "Plan First, Judge Later, Run Better: A DMAIC-Inspired Agentic System for Industrial Anomaly Detection",
+    "url": "https://arxiv.org/abs/2606.04599",
+    "published": "2026-06-03",
+    "updated": "2026-06-03",
+    "authors": [
+      "Yongzi Yu",
+      "Ao Li",
+      "Le Wang",
+      "Ziyue Li",
+      "Fugee Tsung",
+      "Yuxuan Liang",
+      "Man Li"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CE"
+    ],
+    "topics": [
+      "agent-safety",
+      "multi-agent",
+      "planning",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.04599",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.04599",
+    "pdf_url": "https://arxiv.org/pdf/2606.04599",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.04051",
+    "title": "RUBAS: Rubric-Based Reinforcement Learning for Agent Safety",
+    "url": "https://arxiv.org/abs/2606.04051",
+    "published": "2026-06-02",
+    "updated": "2026-06-02",
+    "authors": [
+      "Xian Qi Loye",
+      "Qinglin Su",
+      "Zhexin Zhang",
+      "Shiyao Cui",
+      "Qi Zhu",
+      "Fei Mi",
+      "Hongning Wang",
+      "Minlie Huang"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2606.04051",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.04051",
+    "pdf_url": "https://arxiv.org/pdf/2606.04051",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.02812",
+    "title": "Traj-Evolve: A Self-Evolving Multi-Agent System for Patient Trajectory Modeling in Lung Cancer Early Detection",
+    "url": "https://arxiv.org/abs/2606.02812",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Sihang Zeng",
+      "Matthew Thompson",
+      "Ruth Etzioni",
+      "Meliha Yetisgen"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "multi-agent",
+      "rag",
+      "reasoning"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.02812",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.02812",
+    "pdf_url": "https://arxiv.org/pdf/2606.02812",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.01528",
+    "title": "Joint Agent Memory and Exploration Learning via Novelty Signals",
+    "url": "https://arxiv.org/abs/2606.01528",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Shizuo Tian",
+      "Xiaohong Weng",
+      "Rui Kong",
+      "Yuxuan Chen",
+      "Guohong Liu",
+      "Yuebing Song",
+      "Jiacheng Liu",
+      "Yuchen Li",
+      "Dawei Yin",
+      "Ting Cao",
+      "Yunxin Liu",
+      "Yuanchun Li"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.01528",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.01528",
+    "pdf_url": "https://arxiv.org/pdf/2606.01528",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.01041",
+    "title": "ExpWeaver: LLM Agents Learn from Experience via Latent RAG",
+    "url": "https://arxiv.org/abs/2606.01041",
+    "published": "2026-05-31",
+    "updated": "2026-05-31",
+    "authors": [
+      "Tao Feng",
+      "Tianyang Luo",
+      "Jingjun Xu",
+      "Zhigang Hua",
+      "Yan Xie",
+      "Shuang Yang",
+      "Ge Liu",
+      "Jiaxuan You"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.01041",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.01041",
+    "pdf_url": "https://arxiv.org/pdf/2606.01041",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.30907",
+    "title": "BlueFin: Benchmarking LLM Agents on Financial Spreadsheets",
+    "url": "https://arxiv.org/abs/2605.30907",
+    "published": "2026-05-29",
+    "updated": "2026-05-29",
+    "authors": [
+      "Srivatsa Kundurthy",
+      "Clara Na",
+      "Colton Moraine",
+      "Anoushka Mohta",
+      "Case Winter",
+      "George Fang",
+      "John Ling",
+      "Emma Strubell",
+      "Zach Kirshner"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2605.30907",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.30907",
+    "pdf_url": "https://arxiv.org/pdf/2605.30907",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2605.30858",
+    "title": "ForecastCompass: Guiding Agentic Forecasting with Adaptive Factor Memory",
+    "url": "https://arxiv.org/abs/2605.30858",
+    "published": "2026-05-29",
+    "updated": "2026-05-29",
+    "authors": [
+      "Yurui Chang",
+      "Yongkang Du",
+      "Yuanpu Cao",
+      "Jinghui Chen",
+      "Lu Lin"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.30858",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.30858",
+    "pdf_url": "https://arxiv.org/pdf/2605.30858",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.30058",
+    "title": "HEART-Bench: Do LLM Agents Exhibit Human-like Psychology?",
+    "url": "https://arxiv.org/abs/2605.30058",
+    "published": "2026-05-28",
+    "updated": "2026-05-28",
+    "authors": [
+      "Weihan Peng",
+      "Chenxu Zhang",
+      "Qianao Wang",
+      "Yuling Shi",
+      "Heng Lian",
+      "Qihong Mao",
+      "Jiahao Pang",
+      "Chunliang Feng",
+      "Bowen Li",
+      "Xiaodong Gu"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.30058",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.30058",
+    "pdf_url": "https://arxiv.org/pdf/2605.30058",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.27762",
+    "title": "PEAM: Parametric Embodied Agent Memory through Contrastive Internalization of Experience in Minecraft",
+    "url": "https://arxiv.org/abs/2605.27762",
+    "published": "2026-05-26",
+    "updated": "2026-06-01",
+    "authors": [
+      "Yuchen Guo",
+      "Junli Gong",
+      "Weicheng Wang",
+      "Hongmin Cai",
+      "Yiu-ming Cheung",
+      "Weifeng Su"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.27762",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.27762",
+    "pdf_url": "https://arxiv.org/pdf/2605.27762",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.24659",
+    "title": "IterInject: Indirect Prompt Injection Against LLM Agents via Feedback-Guided Iterative Optimization",
+    "url": "https://arxiv.org/abs/2605.24659",
+    "published": "2026-05-23",
+    "updated": "2026-05-23",
+    "authors": [
+      "Zixuan Chen",
+      "Jiaxiang Chen",
+      "Li Luo",
+      "Ke Xu",
+      "Xiaoxiang Huang",
+      "Tanfeng Sun",
+      "Xinghao Jiang"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.24659",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.24659",
+    "pdf_url": "https://arxiv.org/pdf/2605.24659",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.23723",
+    "title": "MemAudit: Post-hoc Auditing of Poisoned Agent Memory via Causal Attribution and Structural Anomaly Detection",
+    "url": "https://arxiv.org/abs/2605.23723",
+    "published": "2026-05-22",
+    "updated": "2026-05-22",
+    "authors": [
+      "Zhewen Tan",
+      "Yilun Yao",
+      "Huiyan Jin",
+      "Wenhan Yu",
+      "Guoan Wang",
+      "Mengyuan Fan",
+      "liang lu",
+      "Feng Liu",
+      "Xiangzheng Zhang",
+      "Duohe Ma",
+      "Tong Yang",
+      "Lin Sun"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.23723",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.23723",
+    "pdf_url": "https://arxiv.org/pdf/2605.23723",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.24219",
+    "title": "Beyond Final Answers: Auditing Trajectory-Level Hallucinations in Multi-Agent Industrial Workflows",
+    "url": "https://arxiv.org/abs/2605.24219",
+    "published": "2026-05-22",
+    "updated": "2026-05-26",
+    "authors": [
+      "Harshada Badave",
+      "Santosh Borse",
+      "Andrea Gomez",
+      "Harshitha Narahari",
+      "Sara Carter",
+      "Vishwa Bhatt",
+      "Aishani Rachakonda",
+      "Shuxin Lin",
+      "Dhaval Patel"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.24219",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.24219",
+    "pdf_url": "https://arxiv.org/pdf/2605.24219",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.22154",
+    "title": "IdleSpec: Exploiting Idle Time via Speculative Planning for LLM Agents",
+    "url": "https://arxiv.org/abs/2605.22154",
+    "published": "2026-05-21",
+    "updated": "2026-05-21",
+    "authors": [
+      "Daewon Choi",
+      "Kyunghyun Park",
+      "Woomin Song",
+      "Saket Dingliwal",
+      "Sai Muralidhar Jayanthi",
+      "Jinwoo Shin",
+      "Aram Galstyan"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.22154",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.22154",
+    "pdf_url": "https://arxiv.org/pdf/2605.22154",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.21240",
+    "title": "APEX: Autonomous Policy Exploration for Self-Evolving LLM Agents",
+    "url": "https://arxiv.org/abs/2605.21240",
+    "published": "2026-05-20",
+    "updated": "2026-05-20",
+    "authors": [
+      "Yibo Li",
+      "Jiashuo Yang",
+      "Zhi Zheng",
+      "Zhiyuan Hu",
+      "Yuan Sui",
+      "Shizun Wang",
+      "Yufei He",
+      "Bryan Hooi"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.21240",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.21240",
+    "pdf_url": "https://arxiv.org/pdf/2605.21240",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.18930",
+    "title": "OEP: Poisoning Self-Evolving LLM Agents via Locally Correct but Non-Transferable Experiences",
+    "url": "https://arxiv.org/abs/2605.18930",
+    "published": "2026-05-18",
+    "updated": "2026-05-18",
+    "authors": [
+      "Kaixiang Wang",
+      "Jiong Lou",
+      "Zhaojiacheng Zhou",
+      "Jie Li"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "reasoning"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.18930",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.18930",
+    "pdf_url": "https://arxiv.org/pdf/2605.18930",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.18284",
+    "title": "CommitDistill: A Lightweight Knowledge-Centric Memory Layer for Software Repositories",
+    "url": "https://arxiv.org/abs/2605.18284",
+    "published": "2026-05-18",
+    "updated": "2026-05-18",
+    "authors": [
+      "Divya Chukkapalli",
+      "Thejesh Avula",
+      "Aditya Aggarwal",
+      "Harsimran Singh",
+      "Amith Tallanki"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.18284",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.18284",
+    "pdf_url": "https://arxiv.org/pdf/2605.18284",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.14421",
+    "title": "MemLineage: Lineage-Guided Enforcement for LLM Agent Memory",
+    "url": "https://arxiv.org/abs/2605.14421",
+    "published": "2026-05-14",
+    "updated": "2026-05-14",
+    "authors": [
+      "Ciyan Ouyang",
+      "Rui Hou"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.14421",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.14421",
+    "pdf_url": "https://arxiv.org/pdf/2605.14421",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.14892",
+    "title": "Beyond Individual Intelligence: Surveying Collaboration, Failure Attribution, and Self-Evolution in LLM-based Multi-Agent Systems",
+    "url": "https://arxiv.org/abs/2605.14892",
+    "published": "2026-05-14",
+    "updated": "2026-05-15",
+    "authors": [
+      "Shihao Qi",
+      "Jie Ma",
+      "Rui Xing",
+      "Wei Guo",
+      "Xiao Huang",
+      "Zhitao Gao",
+      "Jianhao Deng",
+      "Jun Liu",
+      "Lingling Zhang",
+      "Bifan Wei",
+      "Boqian Yang",
+      "Pinghui Wang",
+      "Jianwen Sun",
+      "Jing Tao",
+      "Yaqiang Wu",
+      "Hui Liu",
+      "Yu Yao",
+      "Tongliang Liu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.14892",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.14892",
+    "pdf_url": "https://arxiv.org/pdf/2605.14892",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.14527",
+    "title": "Lang2MLIP: End-to-End Language-to-Machine Learning Interatomic Potential Development with Autonomous Agentic Workflows",
+    "url": "https://arxiv.org/abs/2605.14527",
+    "published": "2026-05-14",
+    "updated": "2026-05-14",
+    "authors": [
+      "Wenwen Li",
+      "Yuki Orimo",
+      "Nontawat Charoenphakdee"
+    ],
+    "categories": [
+      "cs.LG",
+      "cond-mat.mtrl-sci",
+      "physics.comp-ph"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.14527",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.14527",
+    "pdf_url": "https://arxiv.org/pdf/2605.14527",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.13481",
+    "title": "PersonalAI 2.0: Enhancing knowledge graph traversal/retrieval with planning mechanism for Personalized LLM Agents",
+    "url": "https://arxiv.org/abs/2605.13481",
+    "published": "2026-05-13",
+    "updated": "2026-05-13",
+    "authors": [
+      "Mikhail Menschikov",
+      "Matvey Iskornev",
+      "Alexander Kharitonov",
+      "Alina Bogdanova",
+      "Mikhail Belkin",
+      "Ekaterina Lisitsyna",
+      "Artyom Sosedka",
+      "Victoria Dochkina",
+      "Ruslan Kostoev",
+      "Ilia Perepechkin",
+      "Evgeny Burnaev"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.13481",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.13481",
+    "pdf_url": "https://arxiv.org/pdf/2605.13481",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.12260",
+    "title": "PRISM: Pareto-Efficient Retrieval over Intent-Aware Structured Memory for Long-Horizon Agents",
+    "url": "https://arxiv.org/abs/2605.12260",
+    "published": "2026-05-12",
+    "updated": "2026-05-22",
+    "authors": [
+      "Jingyi Peng",
+      "Zhongwei Wan",
+      "Weiting Liu",
+      "Qiuzhuang Sun"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.12260",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.12260",
+    "pdf_url": "https://arxiv.org/pdf/2605.12260",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.11225",
+    "title": "PIVOT: Bridging Planning and Execution in LLM Agents via Trajectory Refinement",
+    "url": "https://arxiv.org/abs/2605.11225",
+    "published": "2026-05-11",
+    "updated": "2026-05-11",
+    "authors": [
+      "Tuo Zhang",
+      "Alin-Ionut Popa",
+      "Yan Xu",
+      "Rui Song",
+      "Dimitrios Dimitriadis"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "planning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm",
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.11225",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.11225",
+    "pdf_url": "https://arxiv.org/pdf/2605.11225",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.09692",
+    "title": "Causal state binding predicts action control in language agents",
+    "url": "https://arxiv.org/abs/2605.09692",
+    "published": "2026-05-10",
+    "updated": "2026-06-01",
+    "authors": [
+      "Xiao Jia"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "memory",
+      "planning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.09692",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.09692",
+    "pdf_url": "https://arxiv.org/pdf/2605.09692",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.07251",
+    "title": "Can Agents Price a Reaction? Evaluating LLMs on Chemical Cost Reasoning",
+    "url": "https://arxiv.org/abs/2605.07251",
+    "published": "2026-05-08",
+    "updated": "2026-05-08",
+    "authors": [
+      "Yuyang Wu",
+      "Yue Huang",
+      "Shuaike Shen",
+      "Xujian Wang",
+      "Shuhao Zhang",
+      "Qiyao Xue",
+      "Weichen Liu",
+      "Runtian Gao",
+      "Jian Ma",
+      "Xiangliang Zhang",
+      "Olexandr Isayev"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.07251",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.07251",
+    "pdf_url": "https://arxiv.org/pdf/2605.07251",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.06713",
+    "title": "Agentic AI and the Industrialization of Cyber Offense: Forecast, Consequences, and Defensive Priorities for Enterprises and the Mittelstand",
+    "url": "https://arxiv.org/abs/2605.06713",
+    "published": "2026-05-06",
+    "updated": "2026-05-06",
+    "authors": [
+      "Christopher Koch"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.06713",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.06713",
+    "pdf_url": "https://arxiv.org/pdf/2605.06713",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.02240",
+    "title": "PhysicianBench: Evaluating LLM Agents in Real-World EHR Environments",
+    "url": "https://arxiv.org/abs/2605.02240",
+    "published": "2026-05-04",
+    "updated": "2026-05-04",
+    "authors": [
+      "Ruoqi Liu",
+      "Imran Q. Mohiuddin",
+      "Austin J. Schoeffler",
+      "Kavita Renduchintala",
+      "Ashwin Nayak",
+      "Prasantha L. Vemu",
+      "Shivam C. Vedak",
+      "Kameron C. Black",
+      "John L. Havlik",
+      "Isaac Ogunmola",
+      "Stephen P. Ma",
+      "Roopa Dhatt",
+      "Jonathan H. Chen"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.02240",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.02240",
+    "pdf_url": "https://arxiv.org/pdf/2605.02240",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2604.11557",
+    "title": "UniToolCall: Unifying Tool-Use Representation, Data, and Evaluation for LLM Agents",
+    "url": "https://arxiv.org/abs/2604.11557",
+    "published": "2026-04-13",
+    "updated": "2026-05-25",
+    "authors": [
+      "Yijuan Liang",
+      "Xinghao Chen",
+      "Yifan Ge",
+      "Ziyi Wu",
+      "Hao Wu",
+      "Changyu Zeng",
+      "Wei Xing",
+      "Xiaoyu Shen"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2604.11557",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.11557",
+    "pdf_url": "https://arxiv.org/pdf/2604.11557",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2604.10577",
+    "title": "The Blind Spot of Agent Safety: How Benign User Instructions Expose Critical Vulnerabilities in Computer-Use Agents",
+    "url": "https://arxiv.org/abs/2604.10577",
+    "published": "2026-04-12",
+    "updated": "2026-04-17",
+    "authors": [
+      "Xuwei Ding",
+      "Skylar Zhai",
+      "Linxin Song",
+      "Jiate Li",
+      "Taiwei Shi",
+      "Nicholas Meade",
+      "Siva Reddy",
+      "Jian Kang",
+      "Jieyu Zhao"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.10577",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.10577",
+    "pdf_url": "https://arxiv.org/pdf/2604.10577",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.24257",
+    "title": "Memory-Augmented Vision-Language Agents for Persistent and Semantically Consistent Object Captioning",
+    "url": "https://arxiv.org/abs/2603.24257",
+    "published": "2026-03-25",
+    "updated": "2026-03-30",
+    "authors": [
+      "Tommaso Galliena",
+      "Stefano Rosa",
+      "Tommaso Apicella",
+      "Pietro Morerio",
+      "Alessio Del Bue",
+      "Lorenzo Natale"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.24257",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.24257",
+    "pdf_url": "https://arxiv.org/pdf/2603.24257",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.10492",
+    "title": "Human-AI Co-reasoning for Clinical Diagnosis with Evidence-Integrated Language Agent",
+    "url": "https://arxiv.org/abs/2603.10492",
+    "published": "2026-03-11",
+    "updated": "2026-03-18",
+    "authors": [
+      "Zhongzhen Huang",
+      "Yan Ling",
+      "Hong Chen",
+      "Ye Feng",
+      "Li Wu",
+      "Linjie Mu",
+      "Shaoting Zhang",
+      "Xiaofan Zhang",
+      "Kun Qian",
+      "Xiaomu Li"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.10492",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.10492",
+    "pdf_url": "https://arxiv.org/pdf/2603.10492",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.07496",
+    "title": "From Thinker to Society: Security in Hierarchical Autonomy Evolution of AI Agents",
+    "url": "https://arxiv.org/abs/2603.07496",
+    "published": "2026-03-08",
+    "updated": "2026-03-21",
+    "authors": [
+      "Xiaolei Zhang",
+      "Lu Zhou",
+      "Xiaogang Xu",
+      "Jiafei Wu",
+      "Tianyu Du",
+      "Heqing Huang",
+      "Hao Peng",
+      "Zhe Liu"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.07496",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.07496",
+    "pdf_url": "https://arxiv.org/pdf/2603.07496",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.03680",
+    "title": "MAGE: Meta-Reinforcement Learning for Language Agents toward Strategic Exploration and Exploitation",
+    "url": "https://arxiv.org/abs/2603.03680",
+    "published": "2026-03-04",
+    "updated": "2026-03-04",
+    "authors": [
+      "Lu Yang",
+      "Zelai Xu",
+      "Minyang Xie",
+      "Jiaxuan Gao",
+      "Zhao Shok",
+      "Yu Wang",
+      "Yi Wu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "memory",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.03680",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.03680",
+    "pdf_url": "https://arxiv.org/pdf/2603.03680",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.02711",
+    "title": "A Natural Language Agentic Approach to Study Affective Polarization",
+    "url": "https://arxiv.org/abs/2603.02711",
+    "published": "2026-03-03",
+    "updated": "2026-03-03",
+    "authors": [
+      "Stephanie Anneris Malvicini",
+      "Ewelina Gajewska",
+      "Arda Derbent",
+      "Katarzyna Budzynska",
+      "Jarosław A. Chudziak",
+      "Maria Vanina Martinez"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.02711",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.02711",
+    "pdf_url": "https://arxiv.org/pdf/2603.02711",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.03515",
+    "title": "The Controllability Trap: A Governance Framework for Military AI Agents",
+    "url": "https://arxiv.org/abs/2603.03515",
+    "published": "2026-03-03",
+    "updated": "2026-03-03",
+    "authors": [
+      "Subramanyam Sahoo"
+    ],
+    "categories": [
+      "cs.CY",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.03515",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.03515",
+    "pdf_url": "https://arxiv.org/pdf/2603.03515",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.03242",
+    "title": "DRAFT: Task Decoupled Latent Reasoning for Agent Safety",
+    "url": "https://arxiv.org/abs/2604.03242",
+    "published": "2026-02-11",
+    "updated": "2026-02-11",
+    "authors": [
+      "Lin Wang",
+      "Junfeng Fang",
+      "Dan Zhang",
+      "Fei Shen",
+      "Xiang Wang",
+      "Tat-Seng Chua"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.03242",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.03242",
+    "pdf_url": "https://arxiv.org/pdf/2604.03242",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.08721",
+    "title": "KernelCraft: Benchmarking for Agentic Close-to-Metal Kernel Generation on Emerging Hardware",
+    "url": "https://arxiv.org/abs/2603.08721",
+    "published": "2026-02-10",
+    "updated": "2026-05-29",
+    "authors": [
+      "Jiayi Nie",
+      "Haoran Wu",
+      "Yao Lai",
+      "Zeyu Cao",
+      "Cheng Zhang",
+      "Binglei Lou",
+      "Erwei Wang",
+      "Jianyi Cheng",
+      "Timothy M. Jones",
+      "Robert Mullins",
+      "Rika Antonova",
+      "Yiren Zhao"
+    ],
+    "categories": [
+      "cs.AR",
+      "cs.LG",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2603.08721",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.08721",
+    "pdf_url": "https://arxiv.org/pdf/2603.08721",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2602.03224",
+    "title": "TAME: A Trustworthy Test-Time Evolution of Agent Memory with Systematic Benchmarking",
+    "url": "https://arxiv.org/abs/2602.03224",
+    "published": "2026-02-03",
+    "updated": "2026-06-06",
+    "authors": [
+      "Yu Cheng",
+      "Yongkang Hu",
+      "Jiuan Zhou",
+      "Yushuo Zhang",
+      "Yihang Chen",
+      "Huichi Zhou",
+      "Mingang Chen",
+      "Zhizhong Zhang",
+      "Kun Shao",
+      "Yuan Xie",
+      "Zhaoxia Yin"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "memory",
+      "reasoning"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.03224",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.03224",
+    "pdf_url": "https://arxiv.org/pdf/2602.03224",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2512.11682",
+    "title": "MedAI: Evaluating TxAgent's Therapeutic Agentic Reasoning in the NeurIPS CURE-Bench Competition",
+    "url": "https://arxiv.org/abs/2512.11682",
+    "published": "2025-12-12",
+    "updated": "2026-06-15",
+    "authors": [
+      "Tim Cofala",
+      "Christian Kalfar",
+      "Jingge Xiao",
+      "Johanna Schrader",
+      "Michelle Tang",
+      "Wolfgang Nejdl"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2512.11682",
+    "source": "arxiv",
+    "source_id": "arxiv:2512.11682",
+    "pdf_url": "https://arxiv.org/pdf/2512.11682",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2511.15203",
+    "title": "Taxonomy, Evaluation and Exploitation of IPI-Centric LLM Agent Defense Frameworks",
+    "url": "https://arxiv.org/abs/2511.15203",
+    "published": "2025-11-19",
+    "updated": "2025-11-19",
+    "authors": [
+      "Zimo Ji",
+      "Xunguang Wang",
+      "Zongjie Li",
+      "Pingchuan Ma",
+      "Yudong Gao",
+      "Daoyuan Wu",
+      "Xincheng Yan",
+      "Tian Tian",
+      "Shuai Wang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2511.15203",
+    "source": "arxiv",
+    "source_id": "arxiv:2511.15203",
+    "pdf_url": "https://arxiv.org/pdf/2511.15203",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2510.18586",
+    "title": "TokenCake: A KV-Cache-centric Serving Framework for LLM-based Multi-Agent Applications",
+    "url": "https://arxiv.org/abs/2510.18586",
+    "published": "2025-10-21",
+    "updated": "2026-05-20",
+    "authors": [
+      "Zhuohang Bian",
+      "Feiyang Wu",
+      "Zhuoran Li",
+      "Teng Ma",
+      "Youwei Zhuo"
+    ],
+    "categories": [
+      "cs.DC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "multi-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2510.18586",
+    "source": "arxiv",
+    "source_id": "arxiv:2510.18586",
+    "pdf_url": "https://arxiv.org/pdf/2510.18586",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2607.06001",
+    "title": "Information Limits and Attractor Dynamics in Economies of Frontier LLM Agents: A Pre-Registered Test",
+    "url": "https://arxiv.org/abs/2607.06001",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Cheng Qian"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-safety",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.06001",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06001",
+    "pdf_url": "https://arxiv.org/pdf/2607.06001",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.06000",
+    "title": "Context-to-Execution Integrity for LLM Agents",
+    "url": "https://arxiv.org/abs/2607.06000",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Igor Santos-Grueiro"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "coding-agent",
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.06000",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06000",
+    "pdf_url": "https://arxiv.org/pdf/2607.06000",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.06413",
+    "title": "An Experimental Design Approach to Evaluating Agentic AI's Autonomous Model Discovery",
+    "url": "https://arxiv.org/abs/2607.06413",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Hao He",
+      "Xueying Liu",
+      "Chris J. Kuhlman",
+      "Xinwei Deng"
+    ],
+    "categories": [
+      "stat.ME",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.06413",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06413",
+    "pdf_url": "https://arxiv.org/pdf/2607.06413",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.06411",
+    "title": "RuBench: A Repository-Level Agentic Coding Benchmark with Natively Authored Russian Task Specifications",
+    "url": "https://arxiv.org/abs/2607.06411",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Evgeny Shilov"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.06411",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06411",
+    "pdf_url": "https://arxiv.org/pdf/2607.06411",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.06101",
+    "title": "Agents That Teach: Towards Designing Incidental Learning Back into AI-Assisted Software Development",
+    "url": "https://arxiv.org/abs/2607.06101",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Rohit Mehra",
+      "Samdyuti Suri",
+      "Prithviraj K Tagadinamani",
+      "Kapil Singi",
+      "Vikrant Kaulgud",
+      "Adam P. Burden"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.CY",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.06101",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06101",
+    "pdf_url": "https://arxiv.org/pdf/2607.06101",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.05659",
+    "title": "Agents with Feelings? Personality and Emotion in Multi-Agent Software Teams",
+    "url": "https://arxiv.org/abs/2607.05659",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Yunyan Ding",
+      "Thomas Zimmermann",
+      "Iftekhar Ahmed"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.05659",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05659",
+    "pdf_url": "https://arxiv.org/pdf/2607.05659",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.04713",
+    "title": "RSPO: Reward-Swap Policy Optimization for Multi-Turn LLM Agents",
+    "url": "https://arxiv.org/abs/2607.04713",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Qiang Liu",
+      "Taian Guo",
+      "Ruizhi Qiao",
+      "Xing Sun"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.04713",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04713",
+    "pdf_url": "https://arxiv.org/pdf/2607.04713",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.04569",
+    "title": "LLMs for Agentic Home Energy Management",
+    "url": "https://arxiv.org/abs/2607.04569",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Sokipriala Jonah"
+    ],
+    "categories": [
+      "eess.SY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling",
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.04569",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04569",
+    "pdf_url": "https://arxiv.org/pdf/2607.04569",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2607.04617",
+    "title": "MRMS: A Multi-Resolution Memory Substrate for Long-Lived AI Agents",
+    "url": "https://arxiv.org/abs/2607.04617",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Jizhizi Li",
+      "Amy Shi-Nash"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.04617",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04617",
+    "pdf_url": "https://arxiv.org/pdf/2607.04617",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.05363",
+    "title": "SovereignPA-Bench: Evaluating User-Owned Personal Agents under Evolving Intent, Platform Mediation, and Consent Constraints",
+    "url": "https://arxiv.org/abs/2607.05363",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Dylan Zongmin Liu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "embodied-agent",
+      "memory",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.05363",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05363",
+    "pdf_url": "https://arxiv.org/pdf/2607.05363",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.05458",
+    "title": "Learning to Control LLM Agent Harnesses with Offline Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2607.05458",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Haiwen Yi",
+      "Xinyuan Song"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.05458",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05458",
+    "pdf_url": "https://arxiv.org/pdf/2607.05458",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.04219",
+    "title": "Agentic IoT: Architectures, Applications, and Challenges Toward the Internet of Agents",
+    "url": "https://arxiv.org/abs/2607.04219",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Rümeysa Hilal Sevinç",
+      "Bahaeddin Türkoğlu",
+      "İbrahim Kök"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA",
+      "cs.NI"
+    ],
+    "topics": [
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.04219",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04219",
+    "pdf_url": "https://arxiv.org/pdf/2607.04219",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.04212",
+    "title": "An Evaluation of Role-Based Multi-Agent Code Generation on Repository-Scale Problems",
+    "url": "https://arxiv.org/abs/2607.04212",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Benedetta Donato",
+      "Noah Hagar-Dent",
+      "Aaron Worsnop",
+      "Leonardo Mariani",
+      "Valerio Terragni"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.04212",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04212",
+    "pdf_url": "https://arxiv.org/pdf/2607.04212",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.04149",
+    "title": "Beyond Scene Priors: Fine-Grained Traffic Scene Reasoning with Benchmarking and Query-Guided Small-Object Focus",
+    "url": "https://arxiv.org/abs/2607.04149",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Waikit Xiu",
+      "Qiang Lu",
+      "Zian Wang",
+      "Xinjie Yang",
+      "Zhiwei Chen",
+      "Chen Sun",
+      "Xiying Li"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "multi-agent",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.04149",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04149",
+    "pdf_url": "https://arxiv.org/pdf/2607.04149",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.04034",
+    "title": "The \"I Don't Know\" Filter: Enhancing Agentic Reliability in Function Calling",
+    "url": "https://arxiv.org/abs/2607.04034",
+    "published": "2026-07-04",
+    "updated": "2026-07-04",
+    "authors": [
+      "Stefan Broecker",
+      "Mason del Rosario",
+      "Boris Selitser",
+      "Thomas Strohmer"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "function-calling"
+    ],
+    "arxiv_id": "2607.04034",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04034",
+    "pdf_url": "https://arxiv.org/pdf/2607.04034",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.03691",
+    "title": "Don't Blame the Large Language Model: How Scaffolding Evolution Shapes Coding Agent Quality",
+    "url": "https://arxiv.org/abs/2607.03691",
+    "published": "2026-07-04",
+    "updated": "2026-07-04",
+    "authors": [
+      "Oussama Ben Sghaier",
+      "Hao Li",
+      "Bram Adams",
+      "Ahmed E. Hassan"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.03691",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03691",
+    "pdf_url": "https://arxiv.org/pdf/2607.03691",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.02927",
+    "title": "VideoSearcher: Empowering Video Deep Research with Multi-Tool Agentic Reasoning via Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2607.02927",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Zhenkun Gao",
+      "Yicheng Bao",
+      "Jinlong Peng",
+      "Xueheng Li",
+      "Theo Huang",
+      "Bangwei Liu",
+      "Kunquan Li",
+      "Zhenye Gan",
+      "Tao Hu",
+      "Chengjun Xie",
+      "Mingqian Yang",
+      "Xuanhua He",
+      "Zhizhong Zhang",
+      "Xin Tan",
+      "Chengjie Wang",
+      "Yuan Xie"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2607.02927",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02927",
+    "pdf_url": "https://arxiv.org/pdf/2607.02927",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2607.03525",
+    "title": "GameEngineBench: Evaluating Coding Agents on Real C++ Runtime Environments",
+    "url": "https://arxiv.org/abs/2607.03525",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Brian La",
+      "Sejoon Chang",
+      "Ben Kim",
+      "Junyoung Bae",
+      "Aamish Ahmad Beg",
+      "Sei Chang",
+      "Gonzalo Gonzalez-Pumariega"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "embodied-agent",
+      "memory",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.03525",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03525",
+    "pdf_url": "https://arxiv.org/pdf/2607.03525",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.02689",
+    "title": "S-EMBER: A Large-Scale Benchmark for Streaming Egocentric Memory Retrieval",
+    "url": "https://arxiv.org/abs/2607.02689",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Xiaodong Wang",
+      "Xuanyi Zhao",
+      "Pedro Rodriguez",
+      "Devendra Singh Sachan",
+      "Barlas Oguz",
+      "Seungwhan Moon",
+      "Shang-Wen Li",
+      "Gargi Ghosh",
+      "Xin Dong",
+      "Wen-Tau Yih"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.02689",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02689",
+    "pdf_url": "https://arxiv.org/pdf/2607.02689",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.01788",
+    "title": "KRCA: An Efficient Root Cause Analysis System in Hyper-Scale Microservice Systems via Agentic AI",
+    "url": "https://arxiv.org/abs/2607.01788",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Jiamin Jiang",
+      "Jingfei Feng",
+      "Yu Luo",
+      "Qingliang Zhang",
+      "Yongqian Su",
+      "Wenwei Gu",
+      "Shenglin Zhang",
+      "Tianyu Cui",
+      "Yao Wu",
+      "Jielong Huang",
+      "Nan Qi",
+      "Dan Pei"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.01788",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01788",
+    "pdf_url": "https://arxiv.org/pdf/2607.01788",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.02716",
+    "title": "Evaluating Large Language Models for Decision-Making in Agent-Based Urban Mobility Simulations",
+    "url": "https://arxiv.org/abs/2607.02716",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Bruno Cascaes Alves",
+      "Míriam Blank Born",
+      "Ulisses Gilioli Francescatto Júnior",
+      "Felipe Moura Goulart",
+      "Letícia Brandão Caldas",
+      "Marilton Sanchotene de Aguiar"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "multi-agent",
+      "planning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.02716",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02716",
+    "pdf_url": "https://arxiv.org/pdf/2607.02716",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.02186",
+    "title": "UA-ChatDev: Uncertainty-Aware Multi-Agent Collaboration for Reliable Software Development",
+    "url": "https://arxiv.org/abs/2607.02186",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Temitayo Olamilekan Ogunsusi",
+      "Lijun Qian",
+      "Xishuang Dong"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.02186",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02186",
+    "pdf_url": "https://arxiv.org/pdf/2607.02186",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.01084",
+    "title": "Can Agents Generalize to the Open World? Unveiling the Fragility of Static Training in Tool Use",
+    "url": "https://arxiv.org/abs/2607.01084",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Song-Lin Lv",
+      "Weiming Wu",
+      "Rui Zhu",
+      "Zi-Jian Cheng",
+      "Lan-Zhe Guo"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.01084",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01084",
+    "pdf_url": "https://arxiv.org/pdf/2607.01084",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.02599",
+    "title": "AgentLTL: A Trace-Verification Framework for Measuring, Enforcing, and Training Procedural Compliance in Tool-Using LLM Agents",
+    "url": "https://arxiv.org/abs/2607.02599",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Laïla Elkoussy",
+      "Julien Perez"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.LO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.02599",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02599",
+    "pdf_url": "https://arxiv.org/pdf/2607.02599",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.00555",
+    "title": "Rise From The Ashes: LLM-based Static Analysis for Deep Learning Framework Bugs",
+    "url": "https://arxiv.org/abs/2607.00555",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Shaoyu Yang",
+      "Haifeng Lin",
+      "Chunrong Fang",
+      "Xiang Chen",
+      "Wei Cheng",
+      "Jiawei Liu",
+      "Yiyu Zhang",
+      "Hongyu Liu",
+      "Zhenyu Chen"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "multi-agent",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.00555",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00555",
+    "pdf_url": "https://arxiv.org/pdf/2607.00555",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.00436",
+    "title": "PHREEQC-MCQ-200: A Diagnostic Benchmark for Tool-Augmented Scientific Simulator Agents",
+    "url": "https://arxiv.org/abs/2607.00436",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Ke Zhang",
+      "Sahchit Chundur",
+      "Mohammad Javad Qomi",
+      "Maziar Raissi"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2607.00436",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00436",
+    "pdf_url": "https://arxiv.org/pdf/2607.00436",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2607.00502",
+    "title": "A Task-State Representation for Long-Horizon Mobile GUI Agents",
+    "url": "https://arxiv.org/abs/2607.00502",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Yujie Zheng",
+      "Zikang Liu",
+      "Xin Zhao",
+      "Ji-Rong Wen"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2607.00502",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00502",
+    "pdf_url": "https://arxiv.org/pdf/2607.00502",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2607.00440",
+    "title": "Minos: A Multi-Agent Collaborative Framework for Provenance-Based Backward Tracking",
+    "url": "https://arxiv.org/abs/2607.00440",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Jiahui Wang",
+      "Zhenyuan Li",
+      "Zhengkai Wang",
+      "Xiangmin Shen",
+      "Fan Zhang"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "rag",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.00440",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00440",
+    "pdf_url": "https://arxiv.org/pdf/2607.00440",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.00233",
+    "title": "From Signals to Structure: How Memory Architecture Drives Language Emergence in LLM Agents",
+    "url": "https://arxiv.org/abs/2607.00233",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Yashar Talebirad",
+      "Eden Redman",
+      "Ali Parsaee",
+      "Osmar R. Zaiane"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.IT",
+      "cs.MA"
+    ],
+    "topics": [
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.00233",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00233",
+    "pdf_url": "https://arxiv.org/pdf/2607.00233",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.32034",
+    "title": "QVal: Cheaply Evaluating Dense Supervision Signals for Long-Horizon LLM Agents",
+    "url": "https://arxiv.org/abs/2606.32034",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Sergio Hernández-Gutiérrez",
+      "Matteo Merler",
+      "Ilze Amanda Auzina",
+      "Joschka Strüber",
+      "Ameya Prabhu",
+      "Matthias Bethge"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.32034",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.32034",
+    "pdf_url": "https://arxiv.org/pdf/2606.32034",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.02579",
+    "title": "When Not to Write Memory: Governing False Promotion from Correlated Agent Traces",
+    "url": "https://arxiv.org/abs/2607.02579",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Yijiashun Qi",
+      "Xiang Xu",
+      "Yuxuan Li"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "language-agent"
+    ],
+    "arxiv_id": "2607.02579",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02579",
+    "pdf_url": "https://arxiv.org/pdf/2607.02579",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.31339",
+    "title": "Verification-Gated Agentic Mission-State Governance for Intelligent Industrial Multi-Robot Systems",
+    "url": "https://arxiv.org/abs/2606.31339",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Guoqin Tang",
+      "Qingxuan Jia",
+      "Yichen Tan",
+      "Zeyuan Huang",
+      "Ning Ji",
+      "Gang Chen"
+    ],
+    "categories": [
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "embodied-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.31339",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31339",
+    "pdf_url": "https://arxiv.org/pdf/2606.31339",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.31980",
+    "title": "DigitalCoach: Communication and Grounding Gaps in Human and Agentic Computer Use Coaching",
+    "url": "https://arxiv.org/abs/2606.31980",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Meng Chen",
+      "Anya Ji",
+      "Tsung-Han Wu",
+      "Tobias Maringgele",
+      "David M. Chan",
+      "Alane Suhr",
+      "Amy Pavel"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "rag"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.31980",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31980",
+    "pdf_url": "https://arxiv.org/pdf/2606.31980",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.31134",
+    "title": "Beyond the Library: An Agentic Framework for Autoformalizing Research Mathematics",
+    "url": "https://arxiv.org/abs/2606.31134",
+    "published": "2026-06-30",
+    "updated": "2026-07-01",
+    "authors": [
+      "Arshia Soltani Moakhar",
+      "Iman Gholami",
+      "Max Springer",
+      "Mahdi JafariRaviz",
+      "MohammadTaghi Hajiaghayi"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "rag",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.31134",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31134",
+    "pdf_url": "https://arxiv.org/pdf/2606.31134",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.31200",
+    "title": "Agentic RAG-VLM: Affordance-Aware Retrieval-Augmented Generation with Self-Reflective Planning for Robotic Grasping",
+    "url": "https://arxiv.org/abs/2606.31200",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Tao Chen",
+      "Lizheng Liu",
+      "Jiaxu Wang",
+      "Ziyue Jiang",
+      "Ruiqi Tian",
+      "JiGuang Huo",
+      "Zhongxue Gan"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.31200",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31200",
+    "pdf_url": "https://arxiv.org/pdf/2606.31200",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.30251",
+    "title": "TACO: Tool-Augmented Credit Optimization for Agentic Tool Use",
+    "url": "https://arxiv.org/abs/2606.30251",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Mingkuan Feng",
+      "Jinyang Wu",
+      "Hao Gu",
+      "Fangrui Lv",
+      "Ruihan Jin",
+      "Chuyuan Zhang",
+      "Zhengqi Wen",
+      "Jianhua Tao"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.30251",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30251",
+    "pdf_url": "https://arxiv.org/pdf/2606.30251",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.30294",
+    "title": "Rehearsed Multi-Agent Live Product Demonstrations with Real-Time Voice Question Answering",
+    "url": "https://arxiv.org/abs/2606.30294",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Rahul Khedar",
+      "Mayank Malhotra",
+      "Avinash Karn",
+      "Mouli V",
+      "Prakhar Mehrotra"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.HC",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "multi-agent",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.30294",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30294",
+    "pdf_url": "https://arxiv.org/pdf/2606.30294",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.29354",
+    "title": "When LLMs Develop Languages: Symbolic Communication for Efficient Multi-Agent Reasoning",
+    "url": "https://arxiv.org/abs/2606.29354",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Zhengqi Pei",
+      "Qingming Huang",
+      "Shuhui Wang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.NE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.29354",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29354",
+    "pdf_url": "https://arxiv.org/pdf/2606.29354",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.29315",
+    "title": "Hierarchical Experimentalist Agents",
+    "url": "https://arxiv.org/abs/2606.29315",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Abhranil Chandra",
+      "Sankaran Vaidyanathan",
+      "Utsav Dhanuka",
+      "Varun Gandhi",
+      "Scott Niekum"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.29315",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29315",
+    "pdf_url": "https://arxiv.org/pdf/2606.29315",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.29445",
+    "title": "Bridging VideoQA and Video-Guided Agentic Tasks via Generalized Keyframe Extraction",
+    "url": "https://arxiv.org/abs/2606.29445",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Sunqi Fan",
+      "Qingle Liu",
+      "Runqi Yin",
+      "Meng-Hao Guo",
+      "Shuojin Yang"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.29445",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29445",
+    "pdf_url": "https://arxiv.org/pdf/2606.29445",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.28841",
+    "title": "LAMP: Lean-based Agentic framework with MCP and Proof Repair",
+    "url": "https://arxiv.org/abs/2606.28841",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Santhana Srinivasan R",
+      "Maithilee Patawar"
+    ],
+    "categories": [
+      "cs.LO",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "coding-agent",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.28841",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28841",
+    "pdf_url": "https://arxiv.org/pdf/2606.28841",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.28450",
+    "title": "LLM agents security duality: a comprehensive survey of self-security and empowered cybersecurity",
+    "url": "https://arxiv.org/abs/2606.28450",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Yiwei Xu",
+      "Yong Zhuang",
+      "Xuanming Liu",
+      "Tian Zhang",
+      "Bowen Xiao",
+      "Xiaoyang Xu",
+      "Delong Jiang",
+      "Juan Wang",
+      "Hongxin Hu"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "llm-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.28450",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28450",
+    "pdf_url": "https://arxiv.org/pdf/2606.28450",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.28480",
+    "title": "TUA-Bench: A Benchmark for General-Purpose Terminal-Use Agents",
+    "url": "https://arxiv.org/abs/2606.28480",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Shoufa Chen",
+      "Luyuan Wang",
+      "Xuan Yang",
+      "Zhiheng Liu",
+      "Yuren Cong",
+      "Yuanfeng Ji",
+      "Feiyan Zhou",
+      "Xiaohui Zhang",
+      "Fanny Yang",
+      "Belinda Zeng"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.28480",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28480",
+    "pdf_url": "https://arxiv.org/pdf/2606.28480",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.27350",
+    "title": "CHIA: An open-source framework for principled, agentic AI-driven hardware/software co-design research",
+    "url": "https://arxiv.org/abs/2606.27350",
+    "published": "2026-06-25",
+    "updated": "2026-06-27",
+    "authors": [
+      "Angela Cui",
+      "Ferran Hermida-Rivera",
+      "Jack Toubes",
+      "Raghav Gupta",
+      "Jim Fang",
+      "Chengyi Lux Zhang",
+      "Ella Schwarz",
+      "Junha Kim",
+      "Yakun Sophia Shao",
+      "Borivoje Nikolic",
+      "Christopher W. Fletcher",
+      "Sagar Karandikar"
+    ],
+    "categories": [
+      "cs.AR"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.27350",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27350",
+    "pdf_url": "https://arxiv.org/pdf/2606.27350",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.26721",
+    "title": "Knowledge-Based Pull Requests: A Trusted Workflow for Agent-Mediated Knowledge Collaboration",
+    "url": "https://arxiv.org/abs/2606.26721",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Xinyu Zhang",
+      "Weiwei Sun"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "memory",
+      "multi-agent",
+      "planning",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.26721",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26721",
+    "pdf_url": "https://arxiv.org/pdf/2606.26721",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.27330",
+    "title": "Empowering GUI Agents via Autonomous Experience Exploration and Hindsight Experience Utilization for Task Planning",
+    "url": "https://arxiv.org/abs/2606.27330",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Tianyi Men",
+      "Zhuoran Jin",
+      "Pengfei Cao",
+      "Yubo Chen",
+      "Kang Liu",
+      "Jun Zhao"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.CV",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.27330",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27330",
+    "pdf_url": "https://arxiv.org/pdf/2606.27330",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.27009",
+    "title": "Semantic Early-Stopping for Iterative LLM Agent Loops",
+    "url": "https://arxiv.org/abs/2606.27009",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Sahil Shrivastava"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.27009",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27009",
+    "pdf_url": "https://arxiv.org/pdf/2606.27009",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.27483",
+    "title": "Internalizing the Future: A Unified Agentic Training Paradigm for World Model Planning",
+    "url": "https://arxiv.org/abs/2606.27483",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Xuan Zhang",
+      "Zhijian Zhou",
+      "Lingfeng Qiao",
+      "Yulei Qin",
+      "Ke Li",
+      "Xing Sun",
+      "Xiaoyu Tan",
+      "Chao Qu",
+      "Yuan Qi"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "world-model"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.27483",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27483",
+    "pdf_url": "https://arxiv.org/pdf/2606.27483",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.26216",
+    "title": "CyberChainBench: Can AI Agents Secure Smart Contracts Against Real-World On-Chain Vulnerabilities?",
+    "url": "https://arxiv.org/abs/2606.26216",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Jintao Huang",
+      "Fengqing Jiang",
+      "Radha Poovendran",
+      "Zhiqiang Lin"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.26216",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26216",
+    "pdf_url": "https://arxiv.org/pdf/2606.26216",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.26203",
+    "title": "Agentic Analysis for Agentic Infrastructure: An LLM-Powered Pipeline for Comparative Governance of DAO and Corporate AI Protocols",
+    "url": "https://arxiv.org/abs/2606.26203",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Yutian Wang",
+      "Luyao Zhang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CY",
+      "cs.MA",
+      "cs.SI"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.26203",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26203",
+    "pdf_url": "https://arxiv.org/pdf/2606.26203",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.25760",
+    "title": "Uncertainty Quantification for Computer-Use Agents: A Benchmark across Vision-Language Models and GUI Grounding Datasets",
+    "url": "https://arxiv.org/abs/2606.25760",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Divake Kumar",
+      "Sina Tayebati",
+      "Devashri Naik",
+      "Amanda Sofie Rios",
+      "Nilesh Ahuja",
+      "Omesh Tickoo",
+      "Ranganath Krishnan",
+      "Amit Ranjan Trivedi"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL",
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "rag"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.25760",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25760",
+    "pdf_url": "https://arxiv.org/pdf/2606.25760",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.25651",
+    "title": "MedGuards: Multi-Agent System for Reliable Medical Error Detection and Correction",
+    "url": "https://arxiv.org/abs/2606.25651",
+    "published": "2026-06-24",
+    "updated": "2026-06-26",
+    "authors": [
+      "Congbo Ma",
+      "Hu Wang",
+      "Yichun Zhang",
+      "Farah E. Shamout"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.25651",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25651",
+    "pdf_url": "https://arxiv.org/pdf/2606.25651",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.24595",
+    "title": "MEMPROBE: Probing Long-Term Agent Memory via Hidden User-State Recovery",
+    "url": "https://arxiv.org/abs/2606.24595",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Enze Ma",
+      "Yufan Zhou",
+      "Wei-Chieh Huang",
+      "Jie Yang",
+      "Huanhuan Ma",
+      "Zixuan Wang",
+      "Chengze Li",
+      "Chunyu Miao",
+      "Philip S. Yu",
+      "Zhen Wang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.24595",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24595",
+    "pdf_url": "https://arxiv.org/pdf/2606.24595",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.24453",
+    "title": "Bayesian control for coding agents",
+    "url": "https://arxiv.org/abs/2606.24453",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Theodore Papamarkou",
+      "Vladislav Smirnov",
+      "Viktor Mazanov",
+      "Artem Vazhentsev",
+      "Preslav Nakov",
+      "Timothy Baldwin",
+      "Artem Shelmanov"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.24453",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24453",
+    "pdf_url": "https://arxiv.org/pdf/2606.24453",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.24525",
+    "title": "VisCritic: Visual State Comparison as Process Reward for GUI Agents",
+    "url": "https://arxiv.org/abs/2606.24525",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Jiachen Qian"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.24525",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24525",
+    "pdf_url": "https://arxiv.org/pdf/2606.24525",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.24689",
+    "title": "Automated Summarization of Software Documents: An LLM-based Multi-Agent Approach",
+    "url": "https://arxiv.org/abs/2606.24689",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Duc S. H. Nguyen",
+      "Minh T. Nguyen",
+      "Phuong T. Nguyen",
+      "Juri Di Rocco",
+      "Davide Di Ruscio"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.24689",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24689",
+    "pdf_url": "https://arxiv.org/pdf/2606.24689",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.24623",
+    "title": "Privacy-Preserving RAG via Multi-Agent Semantic Rewriting: Achieving Confidentiality Without Compromising Contextual Fidelity",
+    "url": "https://arxiv.org/abs/2606.24623",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Yuanhe Zhao",
+      "Tianyu Zhang",
+      "Huafei Xing",
+      "Derek F. Wong",
+      "Jianbin Li",
+      "Tao Fang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.24623",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24623",
+    "pdf_url": "https://arxiv.org/pdf/2606.24623",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.23664",
+    "title": "MAS-PromptBench: When Does Prompt Optimization Improve Multi-Agent LLM Systems?",
+    "url": "https://arxiv.org/abs/2606.23664",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Juyang Bai",
+      "Laixi Shi"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.23664",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23664",
+    "pdf_url": "https://arxiv.org/pdf/2606.23664",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.23032",
+    "title": "IPO Finance Agent: Benchmark of LLM Financial Analysts Beyond Finance Agent v2, with Automated Rubric Generation, on the SpaceX (SPCX) IPO",
+    "url": "https://arxiv.org/abs/2606.23032",
+    "published": "2026-06-22",
+    "updated": "2026-06-30",
+    "authors": [
+      "Mostapha Benhenda"
+    ],
+    "categories": [
+      "cs.AI",
+      "q-fin.GN"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.23032",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23032",
+    "pdf_url": "https://arxiv.org/pdf/2606.23032",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.22741",
+    "title": "GRADE: Graph Representation of LLM Agent Dependency and Execution",
+    "url": "https://arxiv.org/abs/2606.22741",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Yue Zhao"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "coding-agent",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.22741",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22741",
+    "pdf_url": "https://arxiv.org/pdf/2606.22741",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.23752",
+    "title": "ESAA-Conversational: An Event-Sourced Memory Layer for Continuity, Handoff, and Curation Across Heterogeneous LLM Coding Agents",
+    "url": "https://arxiv.org/abs/2606.23752",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Elzo Brito dos Santos Filho"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "coding-agent",
+      "memory",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.23752",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23752",
+    "pdf_url": "https://arxiv.org/pdf/2606.23752",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.23327",
+    "title": "VideoAgent: All-in-One Framework for Video Understanding and Editing",
+    "url": "https://arxiv.org/abs/2606.23327",
+    "published": "2026-06-22",
+    "updated": "2026-07-03",
+    "authors": [
+      "Hengji Zhou",
+      "Lingxuan Huang",
+      "Jian Wang",
+      "Bing Zhou",
+      "Si Wu",
+      "Lianghao Xia",
+      "Chao Huang"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.23327",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23327",
+    "pdf_url": "https://arxiv.org/pdf/2606.23327",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.22110",
+    "title": "TraceView: Interactive Visualization of Agentic Program Repair Trajectories",
+    "url": "https://arxiv.org/abs/2606.22110",
+    "published": "2026-06-20",
+    "updated": "2026-06-20",
+    "authors": [
+      "Amirali Sajadi",
+      "Tu Nguyen",
+      "Kimmie Huynh",
+      "Esteban Parra",
+      "Preetha Chatterjee"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.22110",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22110",
+    "pdf_url": "https://arxiv.org/pdf/2606.22110",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.22082",
+    "title": "CodeTeam: An LLM-Powered Multi-Agent Framework for Repository-Level Code Generation",
+    "url": "https://arxiv.org/abs/2606.22082",
+    "published": "2026-06-20",
+    "updated": "2026-06-20",
+    "authors": [
+      "Yifei Wang",
+      "Ruiyin Li",
+      "Peng Liang",
+      "Qiong Feng",
+      "Zengyang Li",
+      "Mojtaba Shahin",
+      "Arif Ali Khan"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "planning",
+      "rag"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.22082",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22082",
+    "pdf_url": "https://arxiv.org/pdf/2606.22082",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.21963",
+    "title": "Holmes: Multimodal Agentic Diagnosis for Mixed-Language Mobile Crashes at Industrial Scale",
+    "url": "https://arxiv.org/abs/2606.21963",
+    "published": "2026-06-20",
+    "updated": "2026-06-20",
+    "authors": [
+      "Jia Li",
+      "Wenyuan Ma",
+      "Ting Peng",
+      "Haibin Zheng",
+      "Yuetang Deng"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.21963",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21963",
+    "pdf_url": "https://arxiv.org/pdf/2606.21963",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.21445",
+    "title": "AutoRAS: Learning Robust Agentic Systems with Primitive Representations",
+    "url": "https://arxiv.org/abs/2606.21445",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Yang Yue",
+      "Xuancheng Zhu",
+      "Yuyang Ma",
+      "Guoshun Nan",
+      "Zihan Dou",
+      "Jingru Shan",
+      "Congyu Guo",
+      "Ji Zhang",
+      "Hua Wang",
+      "Jingfeng Zhang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "multi-agent",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.21445",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21445",
+    "pdf_url": "https://arxiv.org/pdf/2606.21445",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.20954",
+    "title": "Learning What Not to Forget: Long-Horizon Agent Memory from a Few Kilobytes of Learning",
+    "url": "https://arxiv.org/abs/2606.20954",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Nusrat Jahan Lia",
+      "Aritra Mazumder"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.20954",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20954",
+    "pdf_url": "https://arxiv.org/pdf/2606.20954",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.19980",
+    "title": "ENPIRE: Agentic Robot Policy Self-Improvement in the Real World",
+    "url": "https://arxiv.org/abs/2606.19980",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Wenli Xiao",
+      "Jia Xie",
+      "Tonghe Zhang",
+      "Haotian Lin",
+      "Letian \"Max\" Fu",
+      "Haoru Xue",
+      "Jalen Lu",
+      "Yi Yang",
+      "Cunxi Dai",
+      "Zi Wang",
+      "Jimmy Wu",
+      "Guanzhi Wang",
+      "S. Shankar Sastry",
+      "Ken Goldberg",
+      "Linxi \"Jim\" Fan",
+      "Yuke Zhu",
+      "Guanya Shi"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "embodied-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.19980",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19980",
+    "pdf_url": "https://arxiv.org/pdf/2606.19980",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.19245",
+    "title": "TxBench-PP: Analyzing AI Agent Performance on Small-Molecule Preclinical Pharmacology",
+    "url": "https://arxiv.org/abs/2606.19245",
+    "published": "2026-06-17",
+    "updated": "2026-06-18",
+    "authors": [
+      "Hannah Le",
+      "Ramesh Ramasamy",
+      "Alex Urrutia",
+      "Mahsa Yazdani",
+      "Tim Proctor",
+      "Kenny Workman"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.19245",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19245",
+    "pdf_url": "https://arxiv.org/pdf/2606.19245",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.18619",
+    "title": "Code-Augur: Agentic Vulnerability Detection via Specification Inference",
+    "url": "https://arxiv.org/abs/2606.18619",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Zhengxiong Luo",
+      "Mehtab Zafar",
+      "Dylan Wolff",
+      "Abhik Roychoudhury"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "rag",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.18619",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18619",
+    "pdf_url": "https://arxiv.org/pdf/2606.18619",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.19464",
+    "title": "Deontic Policies for Runtime Governance of Agentic AI Systems",
+    "url": "https://arxiv.org/abs/2606.19464",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Anupam Joshi",
+      "Tim Finin",
+      "Karuna Pande Joshi",
+      "Lalana Kagal"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.19464",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19464",
+    "pdf_url": "https://arxiv.org/pdf/2606.19464",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.18502",
+    "title": "Towards Scalable Customization and Deployment of Multi-Agent Systems for Enterprise Applications",
+    "url": "https://arxiv.org/abs/2606.18502",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Paresh Dashore",
+      "Shreyas Kulkarni",
+      "Uttam Gurram",
+      "Nadia Bathaee",
+      "Kartik Balasubramaniam",
+      "Genta Indra Winata",
+      "Sambit Sahu",
+      "Shi-Xiong Zhang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "coding-agent",
+      "multi-agent",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.18502",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18502",
+    "pdf_url": "https://arxiv.org/pdf/2606.18502",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.18037",
+    "title": "ProvenanceGuard: Source-Aware Factuality Verification for MCP-Based LLM Agents",
+    "url": "https://arxiv.org/abs/2606.18037",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Ander Alvarez",
+      "Santhiya Rajan",
+      "Samuel Mugel",
+      "Román Orús"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.18037",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18037",
+    "pdf_url": "https://arxiv.org/pdf/2606.18037",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.17573",
+    "title": "Cordon: Semantic Transactions for Tool-Using LLM Agents",
+    "url": "https://arxiv.org/abs/2606.17573",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Zheng Chen",
+      "Hanqing Liu",
+      "Duling Xu",
+      "Dong Dong",
+      "Jialin Li",
+      "Bangzheng Pu",
+      "Jidong Zhai"
+    ],
+    "categories": [
+      "cs.OS",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.17573",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17573",
+    "pdf_url": "https://arxiv.org/pdf/2606.17573",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.18051",
+    "title": "Compositional Skill Routing for LLM Agents: Decompose, Retrieve, and Compose",
+    "url": "https://arxiv.org/abs/2606.18051",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Xueping Gao"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.18051",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18051",
+    "pdf_url": "https://arxiv.org/pdf/2606.18051",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.16295",
+    "title": "VisualClaw: A Real-Time, Personalized Agent for the Physical World",
+    "url": "https://arxiv.org/abs/2606.16295",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Haoqin Tu",
+      "Jianwen Chen",
+      "Zijun Wang",
+      "Siwei Han",
+      "Juncheng Wu",
+      "Hardy Chen",
+      "Haonian Ji",
+      "Kaiwen Xiong",
+      "Jiaqi Liu",
+      "Peng Xia",
+      "Jieru Mei",
+      "Hongliang Fei",
+      "Jason Eshraghian",
+      "Zeyu Zheng",
+      "Yuyin Zhou",
+      "Huaxiu Yao",
+      "Cihang Xie"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "tool-use",
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.16295",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16295",
+    "pdf_url": "https://arxiv.org/pdf/2606.16295",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.16534",
+    "title": "Generated, Parallel, Scalable? A Study of Agentic AI-Generated Julia Code on Supercomputers",
+    "url": "https://arxiv.org/abs/2606.16534",
+    "published": "2026-06-15",
+    "updated": "2026-06-16",
+    "authors": [
+      "Linus Bantel",
+      "Anna-Lena Roth",
+      "Jonas Posner",
+      "Dirk Pflüger"
+    ],
+    "categories": [
+      "cs.DC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.16534",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16534",
+    "pdf_url": "https://arxiv.org/pdf/2606.16534",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.15376",
+    "title": "CoAgent: Concurrency Control for Multi-Agent Systems",
+    "url": "https://arxiv.org/abs/2606.15376",
+    "published": "2026-06-13",
+    "updated": "2026-06-13",
+    "authors": [
+      "Hongtao Lyu",
+      "Dingyan Zhang",
+      "Mingyu Wu",
+      "Xingda Wei",
+      "Haibo Chen"
+    ],
+    "categories": [
+      "cs.DC",
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "coding-agent",
+      "multi-agent",
+      "planning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.15376",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15376",
+    "pdf_url": "https://arxiv.org/pdf/2606.15376",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.14517",
+    "title": "From Shield to Target: Denial-of-Service Attacks on LLM-Based Agent Guardrails",
+    "url": "https://arxiv.org/abs/2606.14517",
+    "published": "2026-06-12",
+    "updated": "2026-06-16",
+    "authors": [
+      "Yuguang Zhou",
+      "Xunguang Wang",
+      "Pingchuan Ma",
+      "Zhantong Xue",
+      "Zhaoyu Wang",
+      "Shuai Wang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.14517",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.14517",
+    "pdf_url": "https://arxiv.org/pdf/2606.14517",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.14470",
+    "title": "GitOfThoughts: Version-Controlled Reasoning and Agent Memory You Can Replay, Diff, and Merge",
+    "url": "https://arxiv.org/abs/2606.14470",
+    "published": "2026-06-12",
+    "updated": "2026-06-22",
+    "authors": [
+      "Pavan C Shekar",
+      "Abhishek H S",
+      "Aswanth Krishnan"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "memory",
+      "rag",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.14470",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.14470",
+    "pdf_url": "https://arxiv.org/pdf/2606.14470",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.14106",
+    "title": "Naive Visual Memory is Not Enough: A Failure-Mode Study of GUI Agents",
+    "url": "https://arxiv.org/abs/2606.14106",
+    "published": "2026-06-12",
+    "updated": "2026-06-12",
+    "authors": [
+      "Seoyoung Choi",
+      "Minseok Ko",
+      "Hyunseok Lee",
+      "Kunwoong Kim",
+      "Woomin Song",
+      "Chanseok Jeon",
+      "Jinwoo Shin"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.CV"
+    ],
+    "topics": [
+      "computer-use",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.14106",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.14106",
+    "pdf_url": "https://arxiv.org/pdf/2606.14106",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.13608",
+    "title": "AgentBeats: Agentifying Agent Assessment for Openness, Standardization, and Reproducibility",
+    "url": "https://arxiv.org/abs/2606.13608",
+    "published": "2026-06-11",
+    "updated": "2026-06-14",
+    "authors": [
+      "Xiaoyuan Liu",
+      "Jianhong Tu",
+      "Yuqi Chen",
+      "Siyuan Xie",
+      "Sihan Ren",
+      "Tianneng Shi",
+      "Gal Gantar",
+      "Evan Sandoval",
+      "Donghyun Lee",
+      "Daniel Miao",
+      "Peter J. Gilbert",
+      "Nick Hynes",
+      "Mauro Staver",
+      "Warren He",
+      "David Marn",
+      "Andrew Low",
+      "Xi Zhang",
+      "Elron Bandel",
+      "Michal Shmueli-Scheuer",
+      "Siva Reddy",
+      "Alexandre Drouin",
+      "Alexandre Lacoste",
+      "Ramayya Krishnan",
+      "Elham Tabassi",
+      "Yu Su",
+      "Victor Barres",
+      "Chenguang Wang",
+      "Wenbo Guo",
+      "Dawn Song"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.13608",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13608",
+    "pdf_url": "https://arxiv.org/pdf/2606.13608",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.13177",
+    "title": "MemRefine: LLM-Guided Compression for Long-Term Agent Memory",
+    "url": "https://arxiv.org/abs/2606.13177",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Minjae Kim",
+      "Jinheon Baek",
+      "Soyeong Jeong",
+      "Sung Ju Hwang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.13177",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13177",
+    "pdf_url": "https://arxiv.org/pdf/2606.13177",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.13148",
+    "title": "TerraBench: Can Agents Reason Over Heterogeneous Earth-System Data?",
+    "url": "https://arxiv.org/abs/2606.13148",
+    "published": "2026-06-11",
+    "updated": "2026-07-01",
+    "authors": [
+      "Dat Tien Nguyen",
+      "Thao Nguyen",
+      "Fadillah Adamsyah Maani",
+      "Huy M. Le",
+      "Muhammad Umer Sheikh",
+      "Numan Saeed",
+      "Muhammad Haris Khan",
+      "Salman Khan"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.13148",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13148",
+    "pdf_url": "https://arxiv.org/pdf/2606.13148",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.28360",
+    "title": "Carolina Guide: A Multi-Agent RAG System with Institutional Guardrails for Academic Policy Assistance",
+    "url": "https://arxiv.org/abs/2606.28360",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Ben Torsion",
+      "Jun Zhou"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "rag"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.28360",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28360",
+    "pdf_url": "https://arxiv.org/pdf/2606.28360",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.12344",
+    "title": "Claw-SWE-Bench: A Benchmark for Evaluating OpenClaw-style Agent Harnesses on Coding Tasks",
+    "url": "https://arxiv.org/abs/2606.12344",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Mengyu Zheng",
+      "Kai Han",
+      "Boxun Li",
+      "Haiyang Xu",
+      "Yuchuan Tian",
+      "Wei He",
+      "Hang Zhou",
+      "Jianyuan Guo",
+      "Hailin Hu",
+      "Lin Ma",
+      "Chao Xu",
+      "Guohao Dai",
+      "Lixue Xia",
+      "Yunchao Wei",
+      "Yunhe Wang",
+      "Yu Wang"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.12344",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12344",
+    "pdf_url": "https://arxiv.org/pdf/2606.12344",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.11702",
+    "title": "MedCTA: A Benchmark for Clinical Tool Agents",
+    "url": "https://arxiv.org/abs/2606.11702",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Tajamul Ashraf",
+      "Hyewon Jeong",
+      "Fida Mohammad Thoker",
+      "Bernard Ghanem"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.11702",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.11702",
+    "pdf_url": "https://arxiv.org/pdf/2606.11702",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.11119",
+    "title": "TRACE: A Unified Rollout Budget Allocation Framework for Efficient Agentic Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2606.11119",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Heming Zou",
+      "Qi Wang",
+      "Yun Qu",
+      "Yuhang Jiang",
+      "Lizhou Cai",
+      "Yixiu Mao",
+      "Ru Peng",
+      "Xin Xu",
+      "Weijie Liu",
+      "Kai Yang",
+      "Saiyong Yang",
+      "Xiangyang Ji"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.11119",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.11119",
+    "pdf_url": "https://arxiv.org/pdf/2606.11119",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.10921",
+    "title": "Trace Only What You Need: Structure-Aware On-Demand Hypergraph Memory for Long-Document Question Answering",
+    "url": "https://arxiv.org/abs/2606.10921",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Xiangjun Zai",
+      "Xingyu Tan",
+      "Chen Chen",
+      "Xiaoyang Wang",
+      "Wenjie Zhang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.10921",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10921",
+    "pdf_url": "https://arxiv.org/pdf/2606.10921",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.10316",
+    "title": "TabClaw: An Interactive and Self-Evolving Agent for Spreadsheet Manipulation and Table Reasoning",
+    "url": "https://arxiv.org/abs/2606.10316",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Mingyue Cheng",
+      "Shuo Yu",
+      "Daoyu Wang",
+      "Qingchuan Li",
+      "Xiaoyu Tao",
+      "Qingyang Mao",
+      "Yitong Zhou",
+      "Qi Liu"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.10316",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10316",
+    "pdf_url": "https://arxiv.org/pdf/2606.10316",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.09037",
+    "title": "A Multi-Agent System for IPMSM Design Optimization via an FEA-AI Hybrid Approach",
+    "url": "https://arxiv.org/abs/2606.09037",
+    "published": "2026-06-08",
+    "updated": "2026-06-08",
+    "authors": [
+      "Jinseong Han",
+      "Sunwoong Yang",
+      "Namwoo Kang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.09037",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09037",
+    "pdf_url": "https://arxiv.org/pdf/2606.09037",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.09738",
+    "title": "HDSL: A Hierarchical Domain-Specific Language for Structured 3D Indoor Scene Generation and Localized Editing with LLM Agents",
+    "url": "https://arxiv.org/abs/2606.09738",
+    "published": "2026-06-08",
+    "updated": "2026-06-08",
+    "authors": [
+      "Letian Li",
+      "Chao Shen",
+      "Shuzhao Xie",
+      "Chenghao Gu",
+      "ZhengXiao He",
+      "Yu Meng",
+      "Xin Yang",
+      "Wenyuan Jiang",
+      "Zhi Wang"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "planning",
+      "rag"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.09738",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09738",
+    "pdf_url": "https://arxiv.org/pdf/2606.09738",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.10209",
+    "title": "Less Context, Better Agents: Efficient Context Engineering for Long-Horizon Tool-Using LLM Agents",
+    "url": "https://arxiv.org/abs/2606.10209",
+    "published": "2026-06-08",
+    "updated": "2026-06-08",
+    "authors": [
+      "Abhilasha Lodha",
+      "Mahsa Pahlavikhah Varnosfaderani",
+      "Abir Chakraborty",
+      "Abhinav Mithal"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.10209",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10209",
+    "pdf_url": "https://arxiv.org/pdf/2606.10209",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.05711",
+    "title": "Beyond tokens: a unified framework for latent communication in LLM-based multi-agent systems",
+    "url": "https://arxiv.org/abs/2606.05711",
+    "published": "2026-06-04",
+    "updated": "2026-06-05",
+    "authors": [
+      "Yingzhuo Liu"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.05711",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.05711",
+    "pdf_url": "https://arxiv.org/pdf/2606.05711",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.06090",
+    "title": "Beyond Semantic Organization: Memory as Execution State Management for Long-Horizon Agents",
+    "url": "https://arxiv.org/abs/2606.06090",
+    "published": "2026-06-04",
+    "updated": "2026-06-04",
+    "authors": [
+      "Yaoqi Chen",
+      "Haibin Lai",
+      "Yuru Feng",
+      "Chuyu Han",
+      "Qianxi Zhang",
+      "Baotong Lu",
+      "Menghao Li",
+      "Xinjiang Wang",
+      "Zhirui Wang",
+      "Shusen Xu",
+      "Zengzhong Li",
+      "Zewen Jin",
+      "Hao Wu",
+      "Cheng Li",
+      "Qi Chen"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "computer-use",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.06090",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.06090",
+    "pdf_url": "https://arxiv.org/pdf/2606.06090",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.06388",
+    "title": "Humans' ALMANAC: A Human Collaboration Dataset of Action-Level Mental Model Annotations for Agent Collaboration",
+    "url": "https://arxiv.org/abs/2606.06388",
+    "published": "2026-06-04",
+    "updated": "2026-06-06",
+    "authors": [
+      "Jiaju Chen",
+      "Yuxuan Lu",
+      "Jiayi Su",
+      "Chaoran Chen",
+      "Songlin Xiao",
+      "Zheng Zhang",
+      "Yun Wang",
+      "Yunyao Li",
+      "Jian Zhao",
+      "Tongshuang Wu",
+      "Toby Jia-Jun Li",
+      "Dakuo Wang",
+      "Bingsheng Yao"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.06388",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.06388",
+    "pdf_url": "https://arxiv.org/pdf/2606.06388",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.05241",
+    "title": "Search-Time Contamination in Deep Research Agents: Measuring Performance Inflation in Public Benchmark Evaluation",
+    "url": "https://arxiv.org/abs/2606.05241",
+    "published": "2026-06-03",
+    "updated": "2026-06-03",
+    "authors": [
+      "Yongjie Wang",
+      "Xinyue Zhang",
+      "Kunhong Yao",
+      "Zhiwei Zeng",
+      "Kaisong Song",
+      "Jun Lin",
+      "Zhiqi Shen"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.05241",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.05241",
+    "pdf_url": "https://arxiv.org/pdf/2606.05241",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.03197",
+    "title": "MemTrain: Self-Supervised Context Memory Training",
+    "url": "https://arxiv.org/abs/2606.03197",
+    "published": "2026-06-02",
+    "updated": "2026-06-02",
+    "authors": [
+      "Ziheng Li",
+      "Xingrun Xing",
+      "Haoqing Wang",
+      "Zhi-Hong Deng",
+      "Yehui Tang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.03197",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.03197",
+    "pdf_url": "https://arxiv.org/pdf/2606.03197",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.02497",
+    "title": "Bridging the Last Mile of Time Series Forecasting with LLM Agents",
+    "url": "https://arxiv.org/abs/2606.02497",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Yuhua Liao",
+      "Zetian Wang",
+      "Qiangqiang Nie",
+      "Zhenhua Zhang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.02497",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.02497",
+    "pdf_url": "https://arxiv.org/pdf/2606.02497",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.01222",
+    "title": "RAG-driven Multi-Agent LLM Framework with Task Decomposition for Beyond 5G Auto-Configuration",
+    "url": "https://arxiv.org/abs/2606.01222",
+    "published": "2026-05-31",
+    "updated": "2026-05-31",
+    "authors": [
+      "İrşat Emin Sarıdaş",
+      "Onur Salan",
+      "Ali Görçin",
+      "Ibrahim Hokelek",
+      "Hakan Ali Çırpan"
+    ],
+    "categories": [
+      "eess.SP"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.01222",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.01222",
+    "pdf_url": "https://arxiv.org/pdf/2606.01222",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.00619",
+    "title": "MemPro: Agentic Memory Systems as Evolvable Programs",
+    "url": "https://arxiv.org/abs/2606.00619",
+    "published": "2026-05-30",
+    "updated": "2026-05-30",
+    "authors": [
+      "Qingshan Liu",
+      "Guoqing Wang",
+      "Wen Wu",
+      "Jingqi Huang",
+      "Xinqi Tao",
+      "Dejia Song",
+      "Jie Zhou",
+      "Liang He"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "rag"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.00619",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.00619",
+    "pdf_url": "https://arxiv.org/pdf/2606.00619",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.00922",
+    "title": "A Machine-to-Machine Knowledge-Guided LLM Agent for Generalizable Radiotherapy Treatment Planning",
+    "url": "https://arxiv.org/abs/2606.00922",
+    "published": "2026-05-30",
+    "updated": "2026-05-30",
+    "authors": [
+      "Md Mainul Abrar",
+      "Xun Jia",
+      "Yujie Chi"
+    ],
+    "categories": [
+      "physics.med-ph",
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "planning",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.00922",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.00922",
+    "pdf_url": "https://arxiv.org/pdf/2606.00922",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.07595",
+    "title": "VisualLeakBench: Reproducible Action-Boundary Propagation Failures in Vision-Language Agents",
+    "url": "https://arxiv.org/abs/2606.07595",
+    "published": "2026-05-29",
+    "updated": "2026-05-29",
+    "authors": [
+      "Youting Wang",
+      "Yuan Tang",
+      "Yitian Qian",
+      "Chen Zhao"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.07595",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.07595",
+    "pdf_url": "https://arxiv.org/pdf/2606.07595",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.00341",
+    "title": "ROGUE: Misaligned Agent Behavior Arising from Ordinary Computer Use",
+    "url": "https://arxiv.org/abs/2606.00341",
+    "published": "2026-05-29",
+    "updated": "2026-05-29",
+    "authors": [
+      "Jeremy Tien",
+      "Abishek Anand",
+      "Yu-Rou Tuan",
+      "Yuchen Shen",
+      "J. Zico Kolter",
+      "Aran Nayebi"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2606.00341",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.00341",
+    "pdf_url": "https://arxiv.org/pdf/2606.00341",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.31377",
+    "title": "DynaTree: Dynamic Agentic Retrieval Tree for Time-Sensitive News Retrieval",
+    "url": "https://arxiv.org/abs/2605.31377",
+    "published": "2026-05-29",
+    "updated": "2026-05-29",
+    "authors": [
+      "Siyuan Qi",
+      "Xinyuan Wang",
+      "Yingxuan Yang",
+      "Haochuan Guo",
+      "Jianghao Lin",
+      "Weiwen Liu",
+      "Yong Yu",
+      "Weinan Zhang"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.31377",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.31377",
+    "pdf_url": "https://arxiv.org/pdf/2605.31377",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.30947",
+    "title": "Extending AI for Research to the Humanities: A Multi-Agent Framework for Evidence-Grounded Scholarship",
+    "url": "https://arxiv.org/abs/2605.30947",
+    "published": "2026-05-29",
+    "updated": "2026-06-03",
+    "authors": [
+      "Yating Pan",
+      "Jiajun Zhang",
+      "Jun Wang",
+      "Qi Su"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.30947",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.30947",
+    "pdf_url": "https://arxiv.org/pdf/2605.30947",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.29960",
+    "title": "Hijacking Agent Memory: Stealthy Trojan Attacks Through Conversational Interaction",
+    "url": "https://arxiv.org/abs/2605.29960",
+    "published": "2026-05-28",
+    "updated": "2026-05-28",
+    "authors": [
+      "Hongtao Wang",
+      "Se Yang",
+      "Yu Chen",
+      "Puzhuo Liu"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.29960",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.29960",
+    "pdf_url": "https://arxiv.org/pdf/2605.29960",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.25435",
+    "title": "Security of OpenClaw Agents: Fundamentals, Attacks, and Countermeasures",
+    "url": "https://arxiv.org/abs/2605.25435",
+    "published": "2026-05-25",
+    "updated": "2026-05-25",
+    "authors": [
+      "Yuntao Wang",
+      "Jianle Ba",
+      "Han Liu",
+      "Yanghe Pan",
+      "Jintao Wei",
+      "Zhou Su",
+      "Tom H. Luan",
+      "Linkang Du"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "memory",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.25435",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.25435",
+    "pdf_url": "https://arxiv.org/pdf/2605.25435",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.24812",
+    "title": "CoRe-Code: Collaborative Reinforcement Learning for Code Generation",
+    "url": "https://arxiv.org/abs/2605.24812",
+    "published": "2026-05-24",
+    "updated": "2026-05-24",
+    "authors": [
+      "Zhihao Dou",
+      "Qinjian Zhao",
+      "Zhongwei Wan",
+      "Xiaoyu Xia",
+      "Sumon Biswas"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.24812",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.24812",
+    "pdf_url": "https://arxiv.org/pdf/2605.24812",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.23067",
+    "title": "What Training Data Teaches RL Memory Agents: An Empirical Study of Curriculum Effects in Memory-Augmented QA",
+    "url": "https://arxiv.org/abs/2605.23067",
+    "published": "2026-05-21",
+    "updated": "2026-05-21",
+    "authors": [
+      "Xinjie He",
+      "Zhiyuan Lin",
+      "Su Liu",
+      "Jialun Wu",
+      "Qiyang Xie",
+      "Weikai Zhou",
+      "Shuai Xiao"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.23067",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.23067",
+    "pdf_url": "https://arxiv.org/pdf/2605.23067",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.20616",
+    "title": "Auto-Dreamer: Learning Offline Memory Consolidation for Language Agents",
+    "url": "https://arxiv.org/abs/2605.20616",
+    "published": "2026-05-20",
+    "updated": "2026-05-20",
+    "authors": [
+      "Chongrui Ye",
+      "Yuxiang Liu",
+      "Yu Wang",
+      "Haofei Yu",
+      "Yining Zhao",
+      "Ge Liu",
+      "Julian McAuley",
+      "Jiaxuan You"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "language-agent"
+    ],
+    "arxiv_id": "2605.20616",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.20616",
+    "pdf_url": "https://arxiv.org/pdf/2605.20616",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.16233",
+    "title": "FORGE: Self-Evolving Agent Memory With No Weight Updates via Population Broadcast",
+    "url": "https://arxiv.org/abs/2605.16233",
+    "published": "2026-05-15",
+    "updated": "2026-05-15",
+    "authors": [
+      "Igor Bogdanov",
+      "Chung-Horng Lung",
+      "Thomas Kunz",
+      "Jie Gao",
+      "Adrian Taylor",
+      "Marzia Zaman"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.LG",
+      "cs.MA",
+      "eess.SY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.16233",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.16233",
+    "pdf_url": "https://arxiv.org/pdf/2605.16233",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.15759",
+    "title": "DimMem: Dimensional Structuring for Efficient Long-Term Agent Memory",
+    "url": "https://arxiv.org/abs/2605.15759",
+    "published": "2026-05-15",
+    "updated": "2026-05-24",
+    "authors": [
+      "Wentao Qiu",
+      "Haotian Hu",
+      "Fanyi Wang",
+      "Jinwei Kong",
+      "Yu Zhang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.15759",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.15759",
+    "pdf_url": "https://arxiv.org/pdf/2605.15759",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.15710",
+    "title": "SMMBench: A Benchmark for Source-Distributed Multimodal Agent Memory",
+    "url": "https://arxiv.org/abs/2605.15710",
+    "published": "2026-05-15",
+    "updated": "2026-05-15",
+    "authors": [
+      "Huacan Chai",
+      "Yukai Wang",
+      "Yingxuan Yang",
+      "Dan Peng",
+      "Yuanyi Song",
+      "Zhihui Fu",
+      "Weiwen Liu",
+      "Jianghao Lin",
+      "Jun Wang",
+      "Weinan Zhang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.15710",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.15710",
+    "pdf_url": "https://arxiv.org/pdf/2605.15710",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.15128",
+    "title": "MemEye: A Visual-Centric Evaluation Framework for Multimodal Agent Memory",
+    "url": "https://arxiv.org/abs/2605.15128",
+    "published": "2026-05-14",
+    "updated": "2026-05-14",
+    "authors": [
+      "Minghao Guo",
+      "Qingyue Jiao",
+      "Zeru Shi",
+      "Yihao Quan",
+      "Boxuan Zhang",
+      "Danrui Li",
+      "Liwei Che",
+      "Wujiang Xu",
+      "Shilong Liu",
+      "Zirui Liu",
+      "Mubbasir Kapadia",
+      "Vladimir Pavlovic",
+      "Jiang Liu",
+      "Mengdi Wang",
+      "Yiyu Shi",
+      "Dimitris N. Metaxas",
+      "Ruixiang Tang"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.CL",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.15128",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.15128",
+    "pdf_url": "https://arxiv.org/pdf/2605.15128",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.14906",
+    "title": "MemLens: Benchmarking Multimodal Long-Term Memory in Large Vision-Language Models",
+    "url": "https://arxiv.org/abs/2605.14906",
+    "published": "2026-05-14",
+    "updated": "2026-05-14",
+    "authors": [
+      "Xiyu Ren",
+      "Zhaowei Wang",
+      "Yiming Du",
+      "Zhongwei Xie",
+      "Chi Liu",
+      "Xinlin Yang",
+      "Haoyue Feng",
+      "Wenjun Pan",
+      "Tianshi Zheng",
+      "Baixuan Xu",
+      "Zhengnan Li",
+      "Yangqiu Song",
+      "Ginny Wong",
+      "Simon See"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.14906",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.14906",
+    "pdf_url": "https://arxiv.org/pdf/2605.14906",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.14932",
+    "title": "Toward Securing AI Agents Like Operating Systems",
+    "url": "https://arxiv.org/abs/2605.14932",
+    "published": "2026-05-14",
+    "updated": "2026-05-14",
+    "authors": [
+      "Lukas Pirch",
+      "Micha Horlboge",
+      "Patrick Großmann",
+      "Syeda Mahnur Asif",
+      "Klim Kireev",
+      "Thorsten Holz",
+      "Konrad Rieck"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.14932",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.14932",
+    "pdf_url": "https://arxiv.org/pdf/2605.14932",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.11946",
+    "title": "Counterfactual Trace Auditing of LLM Agent Skills",
+    "url": "https://arxiv.org/abs/2605.11946",
+    "published": "2026-05-12",
+    "updated": "2026-05-28",
+    "authors": [
+      "Xiaolin Zhou",
+      "Jinbo Liu",
+      "Li Li",
+      "Ryan A. Rossi",
+      "Xiyang Hu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "rag"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.11946",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.11946",
+    "pdf_url": "https://arxiv.org/pdf/2605.11946",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.08964",
+    "title": "Trustworthy AI: Ensuring Reliability and Accountability from Models to Agents",
+    "url": "https://arxiv.org/abs/2605.08964",
+    "published": "2026-05-09",
+    "updated": "2026-05-09",
+    "authors": [
+      "Carol Xuan Long"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.08964",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.08964",
+    "pdf_url": "https://arxiv.org/pdf/2605.08964",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.07830",
+    "title": "CyBiasBench: Benchmarking Bias in LLM Agents for Cyber-Attack Scenarios",
+    "url": "https://arxiv.org/abs/2605.07830",
+    "published": "2026-05-08",
+    "updated": "2026-05-08",
+    "authors": [
+      "Taein Lim",
+      "Seongyong Ju",
+      "Munhyeok Kim",
+      "Hyunjun Kim",
+      "Hoki Kim"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.07830",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.07830",
+    "pdf_url": "https://arxiv.org/pdf/2605.07830",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.04808",
+    "title": "DecodingTrust-Agent Platform (DTap): A Controllable and Interactive Red-Teaming Platform for AI Agents",
+    "url": "https://arxiv.org/abs/2605.04808",
+    "published": "2026-05-06",
+    "updated": "2026-05-06",
+    "authors": [
+      "Zhaorun Chen",
+      "Xun Liu",
+      "Haibo Tong",
+      "Chengquan Guo",
+      "Yuzhou Nie",
+      "Jiawei Zhang",
+      "Mintong Kang",
+      "Chejian Xu",
+      "Qichang Liu",
+      "Xiaogeng Liu",
+      "Tianneng Shi",
+      "Chaowei Xiao",
+      "Sanmi Koyejo",
+      "Percy Liang",
+      "Wenbo Guo",
+      "Dawn Song",
+      "Bo Li"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "planning",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.04808",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.04808",
+    "pdf_url": "https://arxiv.org/pdf/2605.04808",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.01644",
+    "title": "Toward a Principled Framework for Agent Safety Measurement",
+    "url": "https://arxiv.org/abs/2605.01644",
+    "published": "2026-05-02",
+    "updated": "2026-05-02",
+    "authors": [
+      "Shuyi Lin",
+      "Anshuman Suri",
+      "Alina Oprea",
+      "Cheng Tan"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.01644",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.01644",
+    "pdf_url": "https://arxiv.org/pdf/2605.01644",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.00741",
+    "title": "Self-Adaptive Multi-Agent LLM-Based Security Pattern Selection for IoT Systems",
+    "url": "https://arxiv.org/abs/2605.00741",
+    "published": "2026-05-01",
+    "updated": "2026-05-01",
+    "authors": [
+      "Saeid Jamshidi",
+      "Foutse Khomh",
+      "Carol Fung",
+      "Kawser Wazed Nafi"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.00741",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.00741",
+    "pdf_url": "https://arxiv.org/pdf/2605.00741",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.24212",
+    "title": "Empowering Autonomous Debugging Agents with Efficient Dynamic Analysis",
+    "url": "https://arxiv.org/abs/2604.24212",
+    "published": "2026-04-27",
+    "updated": "2026-04-27",
+    "authors": [
+      "Jiahong Xiang",
+      "Xiaoyang Xu",
+      "Xiaopan Chu",
+      "Hongliang Tian",
+      "Yuqun Zhang"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "embodied-agent",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2604.24212",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.24212",
+    "pdf_url": "https://arxiv.org/pdf/2604.24212",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2604.23210",
+    "title": "Discovering Agentic Safety Specifications from 1-Bit Danger Signals",
+    "url": "https://arxiv.org/abs/2604.23210",
+    "published": "2026-04-25",
+    "updated": "2026-04-25",
+    "authors": [
+      "Víctor Gallego"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.23210",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.23210",
+    "pdf_url": "https://arxiv.org/pdf/2604.23210",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.13954",
+    "title": "HINTBench: Horizon-agent Intrinsic Non-attack Trajectory Benchmark",
+    "url": "https://arxiv.org/abs/2604.13954",
+    "published": "2026-04-15",
+    "updated": "2026-04-15",
+    "authors": [
+      "Jiacheng Wang",
+      "Jinchang Hou",
+      "Fabian Wang",
+      "Ping Jian",
+      "Chenfu Bao",
+      "Zhonghou Lv"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.13954",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.13954",
+    "pdf_url": "https://arxiv.org/pdf/2604.13954",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.13536",
+    "title": "Don't Let AI Agents YOLO Your Files: Shifting Information and Control to Filesystems for Agent Safety and Autonomy",
+    "url": "https://arxiv.org/abs/2604.13536",
+    "published": "2026-04-15",
+    "updated": "2026-04-16",
+    "authors": [
+      "Shawn Wanxiang Zhong",
+      "Junxuan Liao",
+      "Jing Liu",
+      "Mai Zheng",
+      "Andrea C. Arpaci-Dusseau",
+      "Remzi H. Arpaci-Dusseau"
+    ],
+    "categories": [
+      "cs.OS"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.13536",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.13536",
+    "pdf_url": "https://arxiv.org/pdf/2604.13536",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.13298",
+    "title": "Can Agents Secure Hardware? Evaluating Agentic LLM-Driven Obfuscation for IP Protection",
+    "url": "https://arxiv.org/abs/2604.13298",
+    "published": "2026-04-14",
+    "updated": "2026-04-14",
+    "authors": [
+      "Sujan Ghimire",
+      "Parsa Mirfasihi",
+      "Muhtasim Alam Chowdhury",
+      "Veeramani Pugazhenthi",
+      "Harish Kumar Dharavath",
+      "Farshad Firouzi",
+      "Rozhin Yasaei",
+      "Pratik Satam",
+      "Soheil Salehi"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.13298",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.13298",
+    "pdf_url": "https://arxiv.org/pdf/2604.13298",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.28835",
+    "title": "GenesisFunc: Multi-Agent Data Generation for Accurate and Generalizable Function-Calling",
+    "url": "https://arxiv.org/abs/2605.28835",
+    "published": "2026-04-10",
+    "updated": "2026-04-10",
+    "authors": [
+      "Hao-Xiang Xu",
+      "Chong Deng",
+      "Jiaqing Liu",
+      "Wen Wang",
+      "Qian Chen",
+      "Lujia Bao",
+      "Xiangang Li",
+      "Zhen-Hua Ling"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2605.28835",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.28835",
+    "pdf_url": "https://arxiv.org/pdf/2605.28835",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2605.00845",
+    "title": "Graph Query Generation with Constraint-guided Large Language Agents",
+    "url": "https://arxiv.org/abs/2605.00845",
+    "published": "2026-04-09",
+    "updated": "2026-04-09",
+    "authors": [
+      "Mengying Wang",
+      "Nicolaas Jedema",
+      "Rahul Pandey",
+      "RaviKiran Krishnan",
+      "Jens Lehmann",
+      "Yinghui Wu"
+    ],
+    "categories": [
+      "cs.DB",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.00845",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.00845",
+    "pdf_url": "https://arxiv.org/pdf/2605.00845",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2604.26959",
+    "title": "CareGuardAI: Context-Aware Multi-Agent Guardrails for Clinical Safety & Hallucination Mitigation in Patient-Facing LLMs",
+    "url": "https://arxiv.org/abs/2604.26959",
+    "published": "2026-04-07",
+    "updated": "2026-04-07",
+    "authors": [
+      "Elham Nasarian",
+      "Abhilash Neog",
+      "Kwok-Leung Tsui",
+      "Niyousha HosseiniChimeh"
+    ],
+    "categories": [
+      "cs.CY",
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.26959",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.26959",
+    "pdf_url": "https://arxiv.org/pdf/2604.26959",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.04131",
+    "title": "Profile-Then-Reason: Bounded Semantic Complexity for Tool-Augmented Language Agents",
+    "url": "https://arxiv.org/abs/2604.04131",
+    "published": "2026-04-05",
+    "updated": "2026-04-05",
+    "authors": [
+      "Paulo Akira F. Enabe"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2604.04131",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.04131",
+    "pdf_url": "https://arxiv.org/pdf/2604.04131",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.28428",
+    "title": "Synergy: A Next-Generation General-Purpose Agent for Open Agentic Web",
+    "url": "https://arxiv.org/abs/2603.28428",
+    "published": "2026-03-30",
+    "updated": "2026-03-30",
+    "authors": [
+      "Xiaohang Nie",
+      "Zihan Guo",
+      "Kezhuo Yang",
+      "Zhichong Zheng",
+      "Bochen Ge",
+      "Shuai Pan",
+      "Zeyi Chen",
+      "Youling Xiang",
+      "Yu Zhang",
+      "Weiwen Liu",
+      "Yuanjian Zhou",
+      "Weinan Zhang"
+    ],
+    "categories": [
+      "cs.CY",
+      "cs.MA"
+    ],
+    "topics": [
+      "coding-agent",
+      "embodied-agent",
+      "memory",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2603.28428",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.28428",
+    "pdf_url": "https://arxiv.org/pdf/2603.28428",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2603.28166",
+    "title": "Evaluating Privilege Usage of Agents with Real-World Tools",
+    "url": "https://arxiv.org/abs/2603.28166",
+    "published": "2026-03-30",
+    "updated": "2026-04-20",
+    "authors": [
+      "Quan Zhang",
+      "Lianhang Fu",
+      "Lvsi Lian",
+      "Gwihwan Go",
+      "Yujue Wang",
+      "Chijin Zhou",
+      "Yu Jiang",
+      "Geguang Pu"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.28166",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.28166",
+    "pdf_url": "https://arxiv.org/pdf/2603.28166",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.21564",
+    "title": "Toward a Theory of Hierarchical Memory for Language Agents",
+    "url": "https://arxiv.org/abs/2603.21564",
+    "published": "2026-03-23",
+    "updated": "2026-03-23",
+    "authors": [
+      "Yashar Talebirad",
+      "Ali Parsaee",
+      "Csongor Y. Szepesvari",
+      "Amirhossein Nadiri",
+      "Osmar Zaiane"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI",
+      "cs.IT",
+      "cs.SI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.21564",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.21564",
+    "pdf_url": "https://arxiv.org/pdf/2603.21564",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.21357",
+    "title": "AgentHER: Hindsight Experience Replay for LLM Agent Trajectory Relabeling",
+    "url": "https://arxiv.org/abs/2603.21357",
+    "published": "2026-03-22",
+    "updated": "2026-05-10",
+    "authors": [
+      "Liang Ding"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "computer-use",
+      "memory",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.21357",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.21357",
+    "pdf_url": "https://arxiv.org/pdf/2603.21357",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.05553",
+    "title": "EigenData: A Self-Evolving Multi-Agent Platform for Function-Calling Data Synthesis, Auditing, and Repair",
+    "url": "https://arxiv.org/abs/2603.05553",
+    "published": "2026-03-05",
+    "updated": "2026-03-05",
+    "authors": [
+      "Jiaao Chen",
+      "Jingyuan Qi",
+      "Mingye Gao",
+      "Wei-Chen Wang",
+      "Hanrui Wang",
+      "Di Jin"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2603.05553",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.05553",
+    "pdf_url": "https://arxiv.org/pdf/2603.05553",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2602.07652",
+    "title": "Agent-Fence: Mapping Security Vulnerabilities Across Deep Research Agents",
+    "url": "https://arxiv.org/abs/2602.07652",
+    "published": "2026-02-07",
+    "updated": "2026-02-07",
+    "authors": [
+      "Sai Puppala",
+      "Ismail Hossain",
+      "Md Jahangir Alam",
+      "Yoonpyo Lee",
+      "Jay Yoo",
+      "Tanzim Ahad",
+      "Syed Bahauddin Alam",
+      "Sajedul Talukder"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.07652",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.07652",
+    "pdf_url": "https://arxiv.org/pdf/2602.07652",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2602.05115",
+    "title": "SocialVeil: Probing Social Intelligence of Language Agents under Communication Barriers",
+    "url": "https://arxiv.org/abs/2602.05115",
+    "published": "2026-02-04",
+    "updated": "2026-02-04",
+    "authors": [
+      "Keyang Xuan",
+      "Pengda Wang",
+      "Chongrui Ye",
+      "Haofei Yu",
+      "Tal August",
+      "Jiaxuan You"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2602.05115",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.05115",
+    "pdf_url": "https://arxiv.org/pdf/2602.05115",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2602.03786",
+    "title": "AOrchestra: Automating Sub-Agent Creation for Agentic Orchestration",
+    "url": "https://arxiv.org/abs/2602.03786",
+    "published": "2026-02-03",
+    "updated": "2026-02-07",
+    "authors": [
+      "Jianhao Ruan",
+      "Zhihao Xu",
+      "Yiran Peng",
+      "Fashen Ren",
+      "Zhaoyang Yu",
+      "Xinbing Liang",
+      "Jinyu Xiang",
+      "Yongru Chen",
+      "Bang Liu",
+      "Chenglin Wu",
+      "Yuyu Luo",
+      "Jiayi Zhang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2602.03786",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.03786",
+    "pdf_url": "https://arxiv.org/pdf/2602.03786",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2601.00268",
+    "title": "Beyond Perfect APIs: A Comprehensive Evaluation of LLM Agents Under Real-World API Complexity",
+    "url": "https://arxiv.org/abs/2601.00268",
+    "published": "2026-01-01",
+    "updated": "2026-01-01",
+    "authors": [
+      "Doyoung Kim",
+      "Zhiwei Ren",
+      "Jie Hao",
+      "Zhongkai Sun",
+      "Lichao Wang",
+      "Xiyao Ma",
+      "Zack Ye",
+      "Xu Han",
+      "Jun Yin",
+      "Heng Ji",
+      "Wei Shen",
+      "Xing Fan",
+      "Benjamin Yao",
+      "Chenlei Guo"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2601.00268",
+    "source": "arxiv",
+    "source_id": "arxiv:2601.00268",
+    "pdf_url": "https://arxiv.org/pdf/2601.00268",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2511.22138",
+    "title": "TinyLLM: Evaluation and Optimization of Small Language Models for Agentic Tasks on Edge Devices",
+    "url": "https://arxiv.org/abs/2511.22138",
+    "published": "2025-11-27",
+    "updated": "2025-11-27",
+    "authors": [
+      "Mohd Ariful Haque",
+      "Fahad Rahman",
+      "Kishor Datta Gupta",
+      "Khalil Shujaee",
+      "Roy George"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2511.22138",
+    "source": "arxiv",
+    "source_id": "arxiv:2511.22138",
+    "pdf_url": "https://arxiv.org/pdf/2511.22138",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2511.11169",
+    "title": "Refine and Align: Confidence Calibration through Multi-Agent Interaction in VQA",
+    "url": "https://arxiv.org/abs/2511.11169",
+    "published": "2025-11-14",
+    "updated": "2025-11-14",
+    "authors": [
+      "Ayush Pandey",
+      "Jai Bardhan",
+      "Ishita Jain",
+      "Ramya S Hebbalaguppe",
+      "Rohan Raju Dhanakshirur",
+      "Lovekesh Vig"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2511.11169",
+    "source": "arxiv",
+    "source_id": "arxiv:2511.11169",
+    "pdf_url": "https://arxiv.org/pdf/2511.11169",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2510.21524",
+    "title": "EU-Agent-Bench: Measuring Illegal Behavior of LLM Agents Under EU Law",
+    "url": "https://arxiv.org/abs/2510.21524",
+    "published": "2025-10-24",
+    "updated": "2025-10-24",
+    "authors": [
+      "Ilija Lichkovski",
+      "Alexander Müller",
+      "Mariam Ibrahim",
+      "Tiwai Mhundwa"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2510.21524",
+    "source": "arxiv",
+    "source_id": "arxiv:2510.21524",
+    "pdf_url": "https://arxiv.org/pdf/2510.21524",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2509.08863",
+    "title": "GeoJSON Agents:A Multi-Agent LLM Architecture for Geospatial Analysis-Function Calling vs Code Generation",
+    "url": "https://arxiv.org/abs/2509.08863",
+    "published": "2025-09-10",
+    "updated": "2025-12-03",
+    "authors": [
+      "Qianqian Luo",
+      "Qingming Lin",
+      "Liuchang Xu",
+      "Sensen Wu",
+      "Ruichen Mao",
+      "Chao Wang",
+      "Hailin Feng",
+      "Bo Huang",
+      "Zhenhong Du"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2509.08863",
+    "source": "arxiv",
+    "source_id": "arxiv:2509.08863",
+    "pdf_url": "https://arxiv.org/pdf/2509.08863",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2508.11027",
+    "title": "Hell or High Water: Evaluating Agentic Recovery from External Failures",
+    "url": "https://arxiv.org/abs/2508.11027",
+    "published": "2025-08-14",
+    "updated": "2025-08-14",
+    "authors": [
+      "Andrew Wang",
+      "Sophia Hager",
+      "Adi Asija",
+      "Daniel Khashabi",
+      "Nicholas Andrews"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2508.11027",
+    "source": "arxiv",
+    "source_id": "arxiv:2508.11027",
+    "pdf_url": "https://arxiv.org/pdf/2508.11027",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2607.06223",
+    "title": "Information Gain-based Rollout Policy Optimization: An Adaptive Tree-Structured Rollout Approach for Multi-Turn LLM Agents",
+    "url": "https://arxiv.org/abs/2607.06223",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Yijun Zhang",
+      "Fan Xu",
+      "Jiaxin Ding",
+      "Yule Xie",
+      "Shiqing Gao",
+      "Xin Ding",
+      "Haoxiang Zhang",
+      "Luoyi Fu",
+      "Xinbing Wang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "rag"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.06223",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06223",
+    "pdf_url": "https://arxiv.org/pdf/2607.06223",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.05915",
+    "title": "PCBWorld: A Benchmark Environment for Engine-Grounded PCB Design Automation",
+    "url": "https://arxiv.org/abs/2607.05915",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Hyungseok Song",
+      "Junseok Park",
+      "Won-Seok Choi",
+      "Seohui Bae",
+      "Han-Seul Jeong",
+      "Youngjoon Park",
+      "Soonyoung Lee"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "llm-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.05915",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05915",
+    "pdf_url": "https://arxiv.org/pdf/2607.05915",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.05805",
+    "title": "Onnes: A Physics-Grounded Multi-Agent LLM Simulator for Cryogenic Fault Diagnosis in Quantum Computing Infrastructure",
+    "url": "https://arxiv.org/abs/2607.05805",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Praneeth Narisetty",
+      "Uday Kumar Reddy Kattamanchi",
+      "Shiva Nagendra Babu Kore"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG",
+      "quant-ph"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.05805",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05805",
+    "pdf_url": "https://arxiv.org/pdf/2607.05805",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.05743",
+    "title": "The Balkanization of Execution-Security Research for AI Coding Agents: Isolation, Access Control, and Time-of-Check-to-Time-of-Use Vulnerabilities",
+    "url": "https://arxiv.org/abs/2607.05743",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Mohammadreza Rashidi"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.05743",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05743",
+    "pdf_url": "https://arxiv.org/pdf/2607.05743",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.05690",
+    "title": "Memory in the Loop: In-Process Retrieval as ExtendedWorking Memory for Language Agents",
+    "url": "https://arxiv.org/abs/2607.05690",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Yusuf Khan",
+      "Carlo Lipizzi"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2607.05690",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05690",
+    "pdf_url": "https://arxiv.org/pdf/2607.05690",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2607.05055",
+    "title": "Toward Trustworthy Large Language Model Agents in Healthcare",
+    "url": "https://arxiv.org/abs/2607.05055",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Hadi Hasan",
+      "Safaa Salman",
+      "Adam Tai Abou Dargham",
+      "Ammar Mohanna",
+      "Ali Chehab"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling",
+      "rag-agent"
+    ],
+    "arxiv_id": "2607.05055",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05055",
+    "pdf_url": "https://arxiv.org/pdf/2607.05055",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2607.04089",
+    "title": "PLACEMEM: Toward a Compute-Aware Memory Plane for Lifelong Agents",
+    "url": "https://arxiv.org/abs/2607.04089",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Sukanta Ganguly"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2607.04089",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04089",
+    "pdf_url": "https://arxiv.org/pdf/2607.04089",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2607.03702",
+    "title": "Agent Reinforcement Learning via Pivotal-Aware Self-Feedback Retry",
+    "url": "https://arxiv.org/abs/2607.03702",
+    "published": "2026-07-04",
+    "updated": "2026-07-04",
+    "authors": [
+      "Weiyang Guo",
+      "Zesheng Shi",
+      "Longhui Zhang",
+      "Zeen Zhu",
+      "Min Zhang",
+      "Jing Li"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.03702",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03702",
+    "pdf_url": "https://arxiv.org/pdf/2607.03702",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.03695",
+    "title": "Social Networks of LLM Agents",
+    "url": "https://arxiv.org/abs/2607.03695",
+    "published": "2026-07-04",
+    "updated": "2026-07-04",
+    "authors": [
+      "Kaixuan Liu",
+      "Guojun Xiong",
+      "Weinan Zhang",
+      "Shengpu Tang"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.03695",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03695",
+    "pdf_url": "https://arxiv.org/pdf/2607.03695",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.03821",
+    "title": "DualView: Preventing Indirect Prompt Injection in Personal AI Agents",
+    "url": "https://arxiv.org/abs/2607.03821",
+    "published": "2026-07-04",
+    "updated": "2026-07-04",
+    "authors": [
+      "Juhee Kim",
+      "Woohyuk Choi",
+      "Taehyun Kang",
+      "Youngmin Kim",
+      "Byoungyoung Lee"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.03821",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03821",
+    "pdf_url": "https://arxiv.org/pdf/2607.03821",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.04009",
+    "title": "PhysMiner: An Agentic AI Framework for Discovering Turbulence Physics",
+    "url": "https://arxiv.org/abs/2607.04009",
+    "published": "2026-07-04",
+    "updated": "2026-07-04",
+    "authors": [
+      "Jiawei Chen",
+      "Han Gao",
+      "Ping He"
+    ],
+    "categories": [
+      "physics.flu-dyn"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.04009",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04009",
+    "pdf_url": "https://arxiv.org/pdf/2607.04009",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.03968",
+    "title": "Refused in Chat, Written in Code: Workflow-Level Jailbreak Construction in IDE Coding Agents",
+    "url": "https://arxiv.org/abs/2607.03968",
+    "published": "2026-07-04",
+    "updated": "2026-07-04",
+    "authors": [
+      "Abhishek Kumar",
+      "Carsten Maple"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.03968",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03968",
+    "pdf_url": "https://arxiv.org/pdf/2607.03968",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.02846",
+    "title": "Object-Centric Environment Modeling for Agentic Tasks",
+    "url": "https://arxiv.org/abs/2607.02846",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Yiyang Li",
+      "Tianyi Ma",
+      "Zehong Wang",
+      "Yijun Ma",
+      "Yanfang Ye"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.02846",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02846",
+    "pdf_url": "https://arxiv.org/pdf/2607.02846",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.03269",
+    "title": "Agentic-SecPBFT: Agentic AI-Driven Proactive Security Framework for Wireless PBFT Consensus in Mobile Ad-Hoc Networks",
+    "url": "https://arxiv.org/abs/2607.03269",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Haoxiang Luo",
+      "Yinqiu Liu",
+      "Ruichen Zhang",
+      "Guangyuan Liu",
+      "Gang Sun",
+      "Hongfang Yu",
+      "Zhu Han",
+      "Dong In Kim"
+    ],
+    "categories": [
+      "cs.NI"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "rag",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.03269",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03269",
+    "pdf_url": "https://arxiv.org/pdf/2607.03269",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.02911",
+    "title": "CoACT: Action-Preserving Observation Compression for Coding Agents",
+    "url": "https://arxiv.org/abs/2607.02911",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Haorui Chen",
+      "Yuancheng Zhu",
+      "Yitong Zhang",
+      "Jia Li"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "coding-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.02911",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02911",
+    "pdf_url": "https://arxiv.org/pdf/2607.02911",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.01767",
+    "title": "Repair the Amplifier, Not the Symptom: Stable World-Model Correction for Agent Rollouts",
+    "url": "https://arxiv.org/abs/2607.01767",
+    "published": "2026-07-02",
+    "updated": "2026-07-05",
+    "authors": [
+      "Xinyuan Song",
+      "Zekun Cai"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2607.01767",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01767",
+    "pdf_url": "https://arxiv.org/pdf/2607.01767",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2607.01709",
+    "title": "COMFYCLAW: Self-Evolving Skill Harnesses for Image Generation Workflows",
+    "url": "https://arxiv.org/abs/2607.01709",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Zongxia Li",
+      "Dawei Liu",
+      "Fuxiao Liu",
+      "Yuhang Zhou",
+      "Xiyang Wu",
+      "Jingxi Chen",
+      "Jing Xie",
+      "Xiaomin Wu",
+      "Lichao Sun"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2607.01709",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01709",
+    "pdf_url": "https://arxiv.org/pdf/2607.01709",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2607.02807",
+    "title": "SwarmResearch: Orchestrating Coding Agents for Open-Ended Discovery",
+    "url": "https://arxiv.org/abs/2607.02807",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Yuvraj Virk",
+      "Zack Edds",
+      "Chunqiu Steven Xia",
+      "Lingming Zhang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "coding-agent",
+      "computer-use",
+      "multi-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.02807",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02807",
+    "pdf_url": "https://arxiv.org/pdf/2607.02807",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.02448",
+    "title": "AgentsCAD: Automated Design for Manufacturing of FDM Parts via Multi-Agent LLM Reasoning and Geometric Feature Recognition",
+    "url": "https://arxiv.org/abs/2607.02448",
+    "published": "2026-07-02",
+    "updated": "2026-07-07",
+    "authors": [
+      "Emmanuel George",
+      "Christopher Keefe",
+      "Peter Pak",
+      "Amir Barati Farimani"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.02448",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02448",
+    "pdf_url": "https://arxiv.org/pdf/2607.02448",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.01213",
+    "title": "RepoRescue: An Empirical Study of LLM Agents on Whole-Repository Compatibility Rescue",
+    "url": "https://arxiv.org/abs/2607.01213",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Zhihao Lin",
+      "Mingyi Zhou",
+      "Zhensu Sun",
+      "Yizhuo Yang",
+      "Renyu Yang",
+      "David Lo",
+      "Li Li"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.01213",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01213",
+    "pdf_url": "https://arxiv.org/pdf/2607.01213",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.01120",
+    "title": "Next-Generation Agentic Reinforcement Learning Systems Enable Self-Evolving Agents",
+    "url": "https://arxiv.org/abs/2607.01120",
+    "published": "2026-07-01",
+    "updated": "2026-07-02",
+    "authors": [
+      "Ran Yan",
+      "Wei Fu",
+      "Jiale Li",
+      "Shusheng Xu",
+      "Zhiyu Mei",
+      "Jiaxuan Gao",
+      "Jiarui Zhang",
+      "Wentai Zhang",
+      "Hao Dai",
+      "Xujie Shen",
+      "Chuyi He",
+      "Zhen Pu",
+      "Jun Mei",
+      "Zhiyao Lin",
+      "Haitao Wang",
+      "Zhiqiang Ding",
+      "Jiawei Zhang",
+      "Huaijie Wang",
+      "Ruida Xu",
+      "Honghua Dong",
+      "Youhe Jiang",
+      "Yi Wu",
+      "Tongkai Yang",
+      "Binhang Yuan"
+    ],
+    "categories": [
+      "cs.DC"
+    ],
+    "topics": [
+      "coding-agent",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.01120",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01120",
+    "pdf_url": "https://arxiv.org/pdf/2607.01120",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.00345",
+    "title": "Registry-Governed Agent Lifecycle:Completing EDDOps with Evaluation-DrivenRegistration, Promotion, and Retirement on AWS AgentCore",
+    "url": "https://arxiv.org/abs/2607.00345",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Richard Kang",
+      "Vincent Wang"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.00345",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00345",
+    "pdf_url": "https://arxiv.org/pdf/2607.00345",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.00911",
+    "title": "From Registry to Repository: How AI Agent Skills Are Written, Adapted, and Maintained",
+    "url": "https://arxiv.org/abs/2607.00911",
+    "published": "2026-07-01",
+    "updated": "2026-07-06",
+    "authors": [
+      "Haoyu Gao",
+      "Jai Lal Lulla",
+      "Hong Yi Lin",
+      "Sebastian Baltes",
+      "Christoph Treude",
+      "Mansooreh Zahedi"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "coding-agent",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.00911",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00911",
+    "pdf_url": "https://arxiv.org/pdf/2607.00911",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.01421",
+    "title": "Risk Architecture for AI-Native Engineering Teams: An Organizational Framework for Agentic System Governance",
+    "url": "https://arxiv.org/abs/2607.01421",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Laxmipriya Ganesh Iyer"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.01421",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01421",
+    "pdf_url": "https://arxiv.org/pdf/2607.01421",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.01061",
+    "title": "Agentic generation of verifiable rules for deterministic, self-expanding reaction classification",
+    "url": "https://arxiv.org/abs/2607.01061",
+    "published": "2026-07-01",
+    "updated": "2026-07-05",
+    "authors": [
+      "Daniel Armstrong",
+      "Maarten Dobbelaere",
+      "Valentas Olikauskas",
+      "Helena Avila",
+      "Octavian Susanu",
+      "Jérôme Waser",
+      "Philippe Schwaller"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "coding-agent",
+      "multi-agent",
+      "planning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.01061",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01061",
+    "pdf_url": "https://arxiv.org/pdf/2607.01061",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.31744",
+    "title": "A Conversational Agentic Interface to Physics-Based Household Digital Twins for Residential Energy Decision Support",
+    "url": "https://arxiv.org/abs/2606.31744",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Costas Mylonas",
+      "Titos Georgoulakis",
+      "Magda Foti"
+    ],
+    "categories": [
+      "eess.SY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.31744",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31744",
+    "pdf_url": "https://arxiv.org/pdf/2606.31744",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.31209",
+    "title": "Long-term Traffic Simulation via Structured Autoregressive Modeling",
+    "url": "https://arxiv.org/abs/2606.31209",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Lingyu Xiao",
+      "Zexin Feng",
+      "Xintao Yan"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.31209",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31209",
+    "pdf_url": "https://arxiv.org/pdf/2606.31209",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.31085",
+    "title": "DDIAgents: Mechanism-Conditioned Context Flow for Drug-Drug Interaction Prediction",
+    "url": "https://arxiv.org/abs/2606.31085",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Zhenqian Shen",
+      "Yu Liu",
+      "Xiaoyi Fu",
+      "Quanming Yao"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.31085",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31085",
+    "pdf_url": "https://arxiv.org/pdf/2606.31085",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.30383",
+    "title": "Whose Side Is Your Agent On? Multi-Party Principal Loyalty in LLM Agents",
+    "url": "https://arxiv.org/abs/2606.30383",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Bojie Li",
+      "Noah Shi"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.30383",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30383",
+    "pdf_url": "https://arxiv.org/pdf/2606.30383",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.30697",
+    "title": "LUMOS: A Semantic Operating-System Layer for Accessibility-Grounded AI Agents",
+    "url": "https://arxiv.org/abs/2606.30697",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Yogeswar Reddy Thota"
+    ],
+    "categories": [
+      "cs.OS",
+      "cs.AI",
+      "cs.CV"
+    ],
+    "topics": [
+      "computer-use",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.30697",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30697",
+    "pdf_url": "https://arxiv.org/pdf/2606.30697",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.30877",
+    "title": "A Systematic Approach to Multi-Agent AI from Advanced Regulatory Control Theory: Safe and Auditable LLM Operator Agents for Process Control",
+    "url": "https://arxiv.org/abs/2606.30877",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Idelfonso B. R. Nogueira",
+      "Sigurd Skogestad"
+    ],
+    "categories": [
+      "eess.SY",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.30877",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30877",
+    "pdf_url": "https://arxiv.org/pdf/2606.30877",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.29957",
+    "title": "SWE-Together: Evaluating Coding Agents in Interactive User Sessions",
+    "url": "https://arxiv.org/abs/2606.29957",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Yifan Wu",
+      "Zhuokai Zhao",
+      "Songlin Li",
+      "Ho Hin Lee",
+      "Jiacheng Zhu",
+      "Shirley Wu",
+      "Tianhe Yu",
+      "Serena Li",
+      "Lizhu Zhang",
+      "Xiangjun Fan",
+      "Shengzhi Li"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.29957",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29957",
+    "pdf_url": "https://arxiv.org/pdf/2606.29957",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.29778",
+    "title": "Mandol: An Agglomerative Agent Memory System for Long-Term Conversations",
+    "url": "https://arxiv.org/abs/2606.29778",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Yuhan Zhang",
+      "Zhiyuan Guo",
+      "Ziheng Zeng",
+      "Wei Wang",
+      "Wentao Wu",
+      "Lijie Xu"
+    ],
+    "categories": [
+      "cs.DB",
+      "cs.AI",
+      "cs.CL",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.29778",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29778",
+    "pdf_url": "https://arxiv.org/pdf/2606.29778",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.30573",
+    "title": "SWE-INTERACT: Reimagining SWE Benchmarks as User-Driven Long-Horizon Coding Sessions",
+    "url": "https://arxiv.org/abs/2606.30573",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Mohit Raghavendra",
+      "Anisha Gunjal",
+      "Aakash Sabharwal",
+      "Yunzhong He"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.30573",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30573",
+    "pdf_url": "https://arxiv.org/pdf/2606.30573",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.30560",
+    "title": "TraceLab: Characterizing Coding Agent Workloads for LLM Serving",
+    "url": "https://arxiv.org/abs/2606.30560",
+    "published": "2026-06-29",
+    "updated": "2026-06-30",
+    "authors": [
+      "Kan Zhu",
+      "Mathew Jacob",
+      "Chenxi Ma",
+      "Yi Pan",
+      "Stephanie Wang",
+      "Arvind Krishnamurthy",
+      "Baris Kasikci"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.PF"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.30560",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30560",
+    "pdf_url": "https://arxiv.org/pdf/2606.30560",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.30887",
+    "title": "Training Therapeutic Judges and Multi-Agent Systems for Human-Aligned Mental Health Support",
+    "url": "https://arxiv.org/abs/2606.30887",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Mizanur Rahman",
+      "Abeer Badawi",
+      "Elahe Rahimi",
+      "Laleh Seyyed-Kalantari",
+      "Frank Rudzicz",
+      "Enamul Hoque",
+      "Elham Dolatabadi"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.30887",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30887",
+    "pdf_url": "https://arxiv.org/pdf/2606.30887",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.00038",
+    "title": "Stop Hand-Holding Your Coding Agent: Engineering the Loops that Replace Step-by-Step Prompting",
+    "url": "https://arxiv.org/abs/2607.00038",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Sandeco Macedo"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.00038",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00038",
+    "pdf_url": "https://arxiv.org/pdf/2607.00038",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.29014",
+    "title": "Customized Generative AI Agent for Transportation Engineering Practice: A Development and Continued Pre-training Guideline",
+    "url": "https://arxiv.org/abs/2606.29014",
+    "published": "2026-06-27",
+    "updated": "2026-07-03",
+    "authors": [
+      "Dianwei Chen",
+      "Yuan-Zheng Lei",
+      "Zifan Zhang",
+      "Yuchen Liu",
+      "Xianfeng Yang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.DL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "planning",
+      "reasoning"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.29014",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29014",
+    "pdf_url": "https://arxiv.org/pdf/2606.29014",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.28781",
+    "title": "HyphaeDB: A Living Knowledge Topology for Agent-First Memory",
+    "url": "https://arxiv.org/abs/2606.28781",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Krishna Halaharvi"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "coding-agent",
+      "memory",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.28781",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28781",
+    "pdf_url": "https://arxiv.org/pdf/2606.28781",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.28839",
+    "title": "The Contagion Tensor: A Framework for Measuring Output-Distribution Coupling in Multi-Agent LLM Systems -- and Auditing the Claims It Enables",
+    "url": "https://arxiv.org/abs/2606.28839",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Zewen Liu"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.28839",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28839",
+    "pdf_url": "https://arxiv.org/pdf/2606.28839",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.28270",
+    "title": "Agent-Native Immune System: Architecture, Taxonomy, and Engineering",
+    "url": "https://arxiv.org/abs/2606.28270",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Bo Shen",
+      "Lifeng Chang",
+      "Tianyuan Wei",
+      "Yunpeng Li",
+      "Feng Shi",
+      "Yichen Han",
+      "Peijie Gao",
+      "Shiyi Kuang",
+      "Xin Chang",
+      "Dehui Li"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "multi-agent",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.28270",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28270",
+    "pdf_url": "https://arxiv.org/pdf/2606.28270",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.28436",
+    "title": "Dockerless: Environment-Free Program Verifier for Coding Agents",
+    "url": "https://arxiv.org/abs/2606.28436",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Wenhao Zeng",
+      "Yuling Shi",
+      "Xiaodong Gu",
+      "Chao Hu",
+      "Chaofan Wang",
+      "Yuhao Cui",
+      "Hongting Zhou",
+      "Mengnan Qi",
+      "Jianqiao Wangni",
+      "Zhaojian Yu",
+      "Shuzheng Gao",
+      "Kai Cai",
+      "Shilin He"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "reasoning"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.28436",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28436",
+    "pdf_url": "https://arxiv.org/pdf/2606.28436",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.26918",
+    "title": "Diagnosing Task Insensitivity in Language Agents",
+    "url": "https://arxiv.org/abs/2606.26918",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Jingyu Liu",
+      "Xiaopeng Wu",
+      "Kehan Chen",
+      "Chuan Yu",
+      "Yong Liu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.26918",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26918",
+    "pdf_url": "https://arxiv.org/pdf/2606.26918",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.26524",
+    "title": "VIGIL: Runtime Enforcement of Behavioral Specifications in AI Agent Skills",
+    "url": "https://arxiv.org/abs/2606.26524",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Ying Li",
+      "Yanju Chen",
+      "Hongbo Wen",
+      "Bosi Zhang",
+      "Hanzhi Liu",
+      "Peiran Wang",
+      "Yu Feng",
+      "Yuan Tian"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.26524",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26524",
+    "pdf_url": "https://arxiv.org/pdf/2606.26524",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.27499",
+    "title": "DMV-Bench: Diagnosing Long-Horizon Multimodal Agents' Visual Memory with Incidental Cue Injection",
+    "url": "https://arxiv.org/abs/2606.27499",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Yujin Tang",
+      "Chenming Shang",
+      "Ruize Xu",
+      "Nikhil Singh"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.27499",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27499",
+    "pdf_url": "https://arxiv.org/pdf/2606.27499",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.27243",
+    "title": "NOVA: A Verification-Aware Agent Harness for Architecture Evolution in Industrial Recommender Systems",
+    "url": "https://arxiv.org/abs/2606.27243",
+    "published": "2026-06-25",
+    "updated": "2026-06-26",
+    "authors": [
+      "Shaohua Liu",
+      "Liang Fang",
+      "Yilong Sun",
+      "Shudong Huang",
+      "Qingsong Luo",
+      "Shaoxin Liu",
+      "Xiaoyang Chen",
+      "Dongqiang Liu",
+      "Chuangang Ma",
+      "Zhenzhen Chai",
+      "Henghuan Wang",
+      "Shijie Quan",
+      "Changyuan Cui",
+      "Zhangbin Zhu",
+      "Peng Chen",
+      "Wei Xu",
+      "Lei Xiao",
+      "Haijie Gu",
+      "Jie Jiang"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "memory",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.27243",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27243",
+    "pdf_url": "https://arxiv.org/pdf/2606.27243",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.26924",
+    "title": "A Deterministic Control Plane for LLM Coding Agents",
+    "url": "https://arxiv.org/abs/2606.26924",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Padmaraj Madatha"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.26924",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26924",
+    "pdf_url": "https://arxiv.org/pdf/2606.26924",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.26883",
+    "title": "EconSimulacra: A Digital Twin Platform of Socio-Economic Systems Powered by LLM Agents",
+    "url": "https://arxiv.org/abs/2606.26883",
+    "published": "2026-06-25",
+    "updated": "2026-06-30",
+    "authors": [
+      "Ryuji Hashimoto",
+      "Masahiro Kaneko",
+      "Kentaro Ueda",
+      "Takehiro Takayanagi",
+      "Kiyoshi Izumi"
+    ],
+    "categories": [
+      "cs.DL"
+    ],
+    "topics": [
+      "memory",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.26883",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26883",
+    "pdf_url": "https://arxiv.org/pdf/2606.26883",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.28409",
+    "title": "Evidence-Driven LLM Agent for C-to-Synthesizable-C Conversion and Verification",
+    "url": "https://arxiv.org/abs/2606.28409",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Zhe Zhao",
+      "Hongbing Lang",
+      "Zhihan Xiao",
+      "Luke Ztz Hu",
+      "John Imoleayo Adebisi",
+      "Songping Mai"
+    ],
+    "categories": [
+      "cs.AR",
+      "cs.AI"
+    ],
+    "topics": [
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.28409",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28409",
+    "pdf_url": "https://arxiv.org/pdf/2606.28409",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.25819",
+    "title": "Beyond Function Calling: Benchmarking Tool-Using Agents under Tool-Environment Unreliability",
+    "url": "https://arxiv.org/abs/2606.25819",
+    "published": "2026-06-24",
+    "updated": "2026-06-27",
+    "authors": [
+      "Yang Tian",
+      "Zhengpeng Shi",
+      "Yu Zhou",
+      "Bo Zhao"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.25819",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25819",
+    "pdf_url": "https://arxiv.org/pdf/2606.25819",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2606.26453",
+    "title": "Optimizing CUDA like a Human: Micro-Profiling Tools as Expert Surrogates for LLM-Based GPU Kernel Optimization",
+    "url": "https://arxiv.org/abs/2606.26453",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Jiading Gai",
+      "Shuai Zhang",
+      "Kaj Bostrom",
+      "Jin Huang",
+      "Vihang Patil",
+      "Haoyang Fang",
+      "Bernie Wang",
+      "Huzefa Rangwala",
+      "George Karypis"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "memory",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.26453",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26453",
+    "pdf_url": "https://arxiv.org/pdf/2606.26453",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.25361",
+    "title": "Memory Makes the Difference: Evaluating How Different Memory Roles Shape Conversational Agents",
+    "url": "https://arxiv.org/abs/2606.25361",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Yuxin Wang",
+      "Paul Thomas",
+      "Zhiwei Yu",
+      "Yuan Gao",
+      "Saeed Hassanpour",
+      "Soroush Vosoughi",
+      "Robert Sim",
+      "Nick Craswell"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.25361",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25361",
+    "pdf_url": "https://arxiv.org/pdf/2606.25361",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.27397",
+    "title": "SidConArena: An Environment Evaluating Agents in Open-Ended,Positive-Sum Bargaining Game",
+    "url": "https://arxiv.org/abs/2606.27397",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Yeqi Feng",
+      "Yuxin Chen",
+      "Tianxing He"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.AI",
+      "cs.GT"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.27397",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27397",
+    "pdf_url": "https://arxiv.org/pdf/2606.27397",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.25139",
+    "title": "Buildrix: An Open Platform for Sharing and Benchmarking Agentic AI Skills in Building Engineering",
+    "url": "https://arxiv.org/abs/2606.25139",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Zixin Jiang",
+      "Bing Dong"
+    ],
+    "categories": [
+      "eess.SY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.25139",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25139",
+    "pdf_url": "https://arxiv.org/pdf/2606.25139",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.24779",
+    "title": "DeepBD: A Grounded Agentic Workflow for Variant Prioritization and Diagnosis of Genetic Birth Defects",
+    "url": "https://arxiv.org/abs/2606.24779",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Shiyu Li",
+      "Ziqi Yan",
+      "Zhihao Wu",
+      "Jielong Lu",
+      "Weiran Liao",
+      "Jiajun Yu",
+      "Genjie Li",
+      "Zeyu Chu",
+      "Jiajun Bu",
+      "Haishuai Wang"
+    ],
+    "categories": [
+      "q-bio.GN",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.24779",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24779",
+    "pdf_url": "https://arxiv.org/pdf/2606.24779",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.24193",
+    "title": "SkyChain Intelligence: A Blockchain-Secured Multi-Agent DRL Framework for Low-Altitude Embodied Artificial Intelligence",
+    "url": "https://arxiv.org/abs/2606.24193",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Haoxiang Luo",
+      "Tianqi Jiang",
+      "Ruichen Zhang",
+      "Yinqiu Liu",
+      "Gang Sun",
+      "Hongfang Yu",
+      "Abbas Jamalipour",
+      "Dong In Kim"
+    ],
+    "categories": [
+      "cs.NI",
+      "cs.DC"
+    ],
+    "topics": [
+      "agent-safety",
+      "embodied-agent",
+      "multi-agent",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.24193",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24193",
+    "pdf_url": "https://arxiv.org/pdf/2606.24193",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.24597",
+    "title": "Qwen-AgentWorld: Language World Models for General Agents",
+    "url": "https://arxiv.org/abs/2606.24597",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Yuxin Zuo",
+      "Zikai Xiao",
+      "Li Sheng",
+      "Fei Huang",
+      "Jianhong Tu",
+      "Yuxuan Liu",
+      "Tianyi Tang",
+      "Xiaomeng Hu",
+      "Yang Su",
+      "Qingfeng Lan",
+      "Yantao Liu",
+      "Qin Zhu",
+      "Yinger Zhang",
+      "Bowen Yu",
+      "Haiquan Zhao",
+      "Haiyang Xu",
+      "Jianxin Yang",
+      "Jiayang Cheng",
+      "Junyang Wang",
+      "Lianghao Deng",
+      "Mingfeng Xue",
+      "Tianyi Bai",
+      "Yang Fan",
+      "Yubo Ma",
+      "Yucheng Li",
+      "Zeyu Cui",
+      "Zhihai Wang",
+      "Zhihui Xie",
+      "Zhuorui Ye",
+      "An Yang",
+      "Dayiheng Liu",
+      "Jingren Zhou",
+      "Ning Ding"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.24597",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24597",
+    "pdf_url": "https://arxiv.org/pdf/2606.24597",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.25115",
+    "title": "Forget to Improve: On-Device LLM-Agent Continual Learning via Budget-Curated Memory",
+    "url": "https://arxiv.org/abs/2606.25115",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Beining Wu",
+      "Zihao Ding",
+      "Jun Huang",
+      "Yanxiao Zhao"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.NI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.25115",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25115",
+    "pdf_url": "https://arxiv.org/pdf/2606.25115",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.24322",
+    "title": "Securing LLM-Agent Long-Term Memory Against Poisoning: Non-Malleable, Origin-Bound Authority with Machine-Checked Guarantees",
+    "url": "https://arxiv.org/abs/2606.24322",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Yedidel Louck"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.24322",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24322",
+    "pdf_url": "https://arxiv.org/pdf/2606.24322",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.24839",
+    "title": "Grading the Grader: Lessons from Evaluating an Agentic Data Analysis System",
+    "url": "https://arxiv.org/abs/2606.24839",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Tian Zheng",
+      "Kai-Tai Hsu"
+    ],
+    "categories": [
+      "cs.AI",
+      "stat.AP"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.24839",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24839",
+    "pdf_url": "https://arxiv.org/pdf/2606.24839",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.23927",
+    "title": "RIFT-Bench: Dynamic Red-teaming For Agentic AI Systems",
+    "url": "https://arxiv.org/abs/2606.23927",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Yarin Yerushalmi Levi",
+      "Roy Betser",
+      "Amit Giloni",
+      "Lidor Erez",
+      "Itay Gershon",
+      "Oren Rachmil",
+      "Sindhu Padakandla",
+      "Roman Vainshtein"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.23927",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23927",
+    "pdf_url": "https://arxiv.org/pdf/2606.23927",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.24551",
+    "title": "GUI vs. CLI: Execution Bottlenecks in Screen-Only and Skill-Mediated Computer-Use Agents",
+    "url": "https://arxiv.org/abs/2606.24551",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Xiao Zhou",
+      "Siyue Zhang",
+      "Yilun Zhao",
+      "Jinbiao Wei",
+      "Tingyu Song",
+      "Arman Cohan",
+      "Chen Zhao"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.24551",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24551",
+    "pdf_url": "https://arxiv.org/pdf/2606.24551",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.22948",
+    "title": "ENVS: Environment-Native Verified Search for Long-Horizon GUI Agents",
+    "url": "https://arxiv.org/abs/2606.22948",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Yincheng Zhou",
+      "Athena Zhuoming Zhong",
+      "Shijie Zhang",
+      "Kevin Zhang",
+      "Teresa Xiaotao Shang",
+      "Shanghang Zhang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.22948",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22948",
+    "pdf_url": "https://arxiv.org/pdf/2606.22948",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.23764",
+    "title": "Emergent Relational Order in LLM Agent Societies: From Collective Affect to Authority Stratification",
+    "url": "https://arxiv.org/abs/2606.23764",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Zhiyuan Ji",
+      "Xinyu Chen",
+      "Ziqi Dai",
+      "Shiyun Tang",
+      "Chunyu Wei",
+      "Yueguo Chen"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.AI"
+    ],
+    "topics": [
+      "multi-agent",
+      "planning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.23764",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23764",
+    "pdf_url": "https://arxiv.org/pdf/2606.23764",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.23343",
+    "title": "Group Selection Promotes Prosocial Prompts in Populations of LLM Agents",
+    "url": "https://arxiv.org/abs/2606.23343",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Luis Celiktemel",
+      "Edward Eichhorn",
+      "Levin Brinkmann",
+      "Robin Schimmelpfennig",
+      "Aron Vallinder",
+      "Yaomin Jiang",
+      "Edward Hughes",
+      "Iyad Rahwan"
+    ],
+    "categories": [
+      "cs.CY"
+    ],
+    "topics": [
+      "coding-agent",
+      "computer-use",
+      "multi-agent",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.23343",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23343",
+    "pdf_url": "https://arxiv.org/pdf/2606.23343",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.22388",
+    "title": "PlanBench-XL: Evaluating Long-Horizon Planning of LLM Tool-Use Agents in Large-Scale Tool Ecosystems",
+    "url": "https://arxiv.org/abs/2606.22388",
+    "published": "2026-06-21",
+    "updated": "2026-06-21",
+    "authors": [
+      "Jiayu Liu",
+      "Qihan Lin",
+      "Cheng Qian",
+      "Rui Wang",
+      "Emre Can Acikgoz",
+      "Xiaocheng Yang",
+      "Jiateng Liu",
+      "Zhenhailong Wang",
+      "Xiusi Chen",
+      "Heng Ji",
+      "Dilek Hakkani-Tür"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.22388",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22388",
+    "pdf_url": "https://arxiv.org/pdf/2606.22388",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.22610",
+    "title": "PaperClaw: Harnessing Agents for Autonomous Research and Human-in-the-Loop Refinement",
+    "url": "https://arxiv.org/abs/2606.22610",
+    "published": "2026-06-21",
+    "updated": "2026-06-21",
+    "authors": [
+      "Weiwei Ye",
+      "Hangchen Liu",
+      "Dongyuan Li",
+      "Renhe Jiang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.22610",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22610",
+    "pdf_url": "https://arxiv.org/pdf/2606.22610",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.21565",
+    "title": "Composing Verifiable Conceptual Models via Building Blocks: Towards Design-Time Verification of Agentic AI Workflows",
+    "url": "https://arxiv.org/abs/2606.21565",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Noe Y. Flandre",
+      "Alexander C. Nwala",
+      "Philippe J. Giabbanelli"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.21565",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21565",
+    "pdf_url": "https://arxiv.org/pdf/2606.21565",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.21732",
+    "title": "Safe to Check, Unsafe to Use: Relinking at the Compression Boundary of LLM Agents",
+    "url": "https://arxiv.org/abs/2606.21732",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Zesen Liu",
+      "Zihan Zhang",
+      "Dongdong She"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.21732",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21732",
+    "pdf_url": "https://arxiv.org/pdf/2606.21732",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.20479",
+    "title": "GroundControl: Anticipating Navigation Failures in Vision-Language Agents via Trajectory-Consistent Uncertainty Estimates",
+    "url": "https://arxiv.org/abs/2606.20479",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Nastaran Darabi",
+      "Divake Kumar",
+      "Sina Tayebati",
+      "Devashri Naik",
+      "Amit Ranjan Trivedi"
+    ],
+    "categories": [
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "embodied-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.20479",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20479",
+    "pdf_url": "https://arxiv.org/pdf/2606.20479",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.20041",
+    "title": "AI Economist Agent: An Agentic Framework for Model-Grounded Economic Analysis with RAG, Knowledge Graphs, and Large Language Models",
+    "url": "https://arxiv.org/abs/2606.20041",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Masahiro Kato"
+    ],
+    "categories": [
+      "econ.GN",
+      "cs.AI",
+      "cs.LG",
+      "q-fin.GN"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.20041",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20041",
+    "pdf_url": "https://arxiv.org/pdf/2606.20041",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.19852",
+    "title": "Prompt, Plan, Extract: Zero-Shot Agentic LLMs Workflows for Lung Pathology Extraction from Clinical Narratives",
+    "url": "https://arxiv.org/abs/2606.19852",
+    "published": "2026-06-18",
+    "updated": "2026-06-25",
+    "authors": [
+      "Aman Pathak",
+      "Cheng Peng",
+      "Mengxian Lyu",
+      "Ziyi Chen",
+      "Reema Solan",
+      "Sankalp Talankar",
+      "Yasir Khan",
+      "Hiren Mehta",
+      "Aokun Chen",
+      "Yi Guo",
+      "Yonghui Wu"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.19852",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19852",
+    "pdf_url": "https://arxiv.org/pdf/2606.19852",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.20047",
+    "title": "PACMS: Submodular Context Selection as a Pluggable Engine for LLM Agents",
+    "url": "https://arxiv.org/abs/2606.20047",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Manu Ghulyani",
+      "Arunabh Singh",
+      "Karan Bharadwaj",
+      "Ankit Nath",
+      "Suranjan Goswami"
+    ],
+    "categories": [
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.20047",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20047",
+    "pdf_url": "https://arxiv.org/pdf/2606.20047",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.19926",
+    "title": "MemGUI-Agent: An End-to-End Long-Horizon Mobile GUI Agent with Proactive Context Management",
+    "url": "https://arxiv.org/abs/2606.19926",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Guangyi Liu",
+      "Gao Wu",
+      "Congxiao Liu",
+      "Pengxiang Zhao",
+      "Liang Liu",
+      "Mading Li",
+      "Qi Zhang",
+      "Mengyan Wang",
+      "Liang Guo",
+      "Yong Liu"
+    ],
+    "categories": [
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.19926",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19926",
+    "pdf_url": "https://arxiv.org/pdf/2606.19926",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.20922",
+    "title": "Think Twice Before You Act: Protecting LLM Agents Against Tool Description Poisoning via Isolated Planning",
+    "url": "https://arxiv.org/abs/2606.20922",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Shanghao Shi",
+      "Xiao Wang",
+      "Chaoyu Zhang",
+      "Hao Li",
+      "Wenjing Lou",
+      "Thomas Hou",
+      "Yevgeniy Vorobeychik",
+      "Chongjie Zhang",
+      "Ning Zhang"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.20922",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20922",
+    "pdf_url": "https://arxiv.org/pdf/2606.20922",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.19787",
+    "title": "ORAgentBench: Can LLM Agents Solve Challenging Operations Research Tasks End to End?",
+    "url": "https://arxiv.org/abs/2606.19787",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Jiajun Li",
+      "Mingshu Cai",
+      "Yixuan Li",
+      "Yu Ding",
+      "Ran Hou",
+      "Guanyu Nie",
+      "Xiongwei Han",
+      "Wanyuan Wang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.19787",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19787",
+    "pdf_url": "https://arxiv.org/pdf/2606.19787",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.18068",
+    "title": "Agentic AI-based Framework for Mitigating Premature Diagnostic Handoff and Silent Hallucination in Healthcare Applications",
+    "url": "https://arxiv.org/abs/2606.18068",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Divyansh Srivastava",
+      "Shreya Ghosh",
+      "Anshul Verma",
+      "Rajkumar Buyya"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "reasoning"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.18068",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18068",
+    "pdf_url": "https://arxiv.org/pdf/2606.18068",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.17680",
+    "title": "EnvRL: Learn from Environment Dynamics in Agentic Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2606.17680",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Zhitong Wang",
+      "Songze Li",
+      "Hao Peng",
+      "Shuzheng Si",
+      "Yi Wang",
+      "Maosong Sun",
+      "Juanzi Li"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.17680",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17680",
+    "pdf_url": "https://arxiv.org/pdf/2606.17680",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.18406",
+    "title": "CoreMem: Riemannian Retrieval and Fisher-Guided Distillation for Long-Term Memory in Dialogue Agents",
+    "url": "https://arxiv.org/abs/2606.18406",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Jiaqi Chen",
+      "Yongqin Zeng",
+      "Shaoshen Chen",
+      "Yijian Zhang",
+      "Hai-Tao Zheng",
+      "Chunxia Ma",
+      "XiuTeng Zhou"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.18406",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18406",
+    "pdf_url": "https://arxiv.org/pdf/2606.18406",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.17449",
+    "title": "MODE-RAG: Manifold Outlier Diagnosis and Energy-based Retrieval-Augmented Generation Evaluation",
+    "url": "https://arxiv.org/abs/2606.17449",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Zehang Wei",
+      "Jiaxin Dai",
+      "Jiamin Yan",
+      "Xiang Xiang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.CV",
+      "cs.LG",
+      "cs.MM"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "rag",
+      "reasoning"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.17449",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17449",
+    "pdf_url": "https://arxiv.org/pdf/2606.17449",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.16432",
+    "title": "ACCORD: Action-Conditioned Contextual Grounding for Language Agents",
+    "url": "https://arxiv.org/abs/2606.16432",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Lai Jiang",
+      "Cheng Qian",
+      "Zhenhailong Wang",
+      "Pan Lu",
+      "Heng Ji",
+      "Hao Peng"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.16432",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16432",
+    "pdf_url": "https://arxiv.org/pdf/2606.16432",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.15903",
+    "title": "Control-Plane Placement Shapes Forgetting: An Architectural Study of Agent Memory Across Thirteen System Configurations",
+    "url": "https://arxiv.org/abs/2606.15903",
+    "published": "2026-06-14",
+    "updated": "2026-06-16",
+    "authors": [
+      "Dongxu Yang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.15903",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15903",
+    "pdf_url": "https://arxiv.org/pdf/2606.15903",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.15609",
+    "title": "FragFuse: Bypassing Access Control of Large Language Model Agents via Memory-Based Query Fragmentation and Fusion",
+    "url": "https://arxiv.org/abs/2606.15609",
+    "published": "2026-06-14",
+    "updated": "2026-06-14",
+    "authors": [
+      "Zixin Rao",
+      "Wentian Zhu",
+      "Chan Aristella Lu",
+      "Zhaorun Chen",
+      "Wei Niu",
+      "Le Guan",
+      "Bo Li",
+      "Zhen Xiang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.15609",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15609",
+    "pdf_url": "https://arxiv.org/pdf/2606.15609",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.15079",
+    "title": "Ling and Ring 2.6 Technical Report: Efficient and Instant Agentic Intelligence at Trillion-Parameter Scale",
+    "url": "https://arxiv.org/abs/2606.15079",
+    "published": "2026-06-13",
+    "updated": "2026-06-13",
+    "authors": [
+      "Ang Li",
+      "Ben Liu",
+      "Bin Han",
+      "Bin Hu",
+      "Bin Jing",
+      "Binbin Hu",
+      "Bing Li",
+      "Cai Chen",
+      "Caizhi Tang",
+      "Changxin Tian",
+      "Chao Huang",
+      "Chao Zhang",
+      "Chen Liang",
+      "Chen Qian",
+      "Chengfu Tang",
+      "Chengyao Wen",
+      "Chilin Fu",
+      "Chunwei Wu",
+      "Cong Zhang",
+      "Cunyin Peng",
+      "Daixin Wang",
+      "Dalong Zhang",
+      "Deng Zhao",
+      "Dingnan Jin",
+      "Dingyuan Zhu",
+      "Donghao Zhang",
+      "Fan Yuan",
+      "Fangzheng Zhao",
+      "Fanzhuang Meng",
+      "Feifan Wu",
+      "Feng Xu",
+      "Fengbin Fang",
+      "Gangshan Wang",
+      "Guodong Yang",
+      "Hailin Zhao",
+      "Haitao Wang",
+      "Haitao Zhang",
+      "Hanxiao Zhang",
+      "Hanzi Wang",
+      "Hao Dai",
+      "Hao Liu",
+      "Hao Qian",
+      "Hao Wu",
+      "Haoxiong Liu",
+      "Haoyu Xu",
+      "Heng Zhang",
+      "Hong Liu",
+      "Hongliang Zhang",
+      "Hongrui Liu",
+      "Hongxun Li",
+      "Hongzhi Ruan",
+      "Huaidong Xiong",
+      "Huihuang Zheng",
+      "Huikang Tang",
+      "Jia Guo",
+      "Jia Li",
+      "Jia Liu",
+      "Jiameng Wang",
+      "Jiaming Liu",
+      "Jiannan Shi",
+      "Jianping Wei",
+      "Jiaolong Yang",
+      "Jiapeng Wang",
+      "Jie Gao",
+      "Jie Wang",
+      "Jiewei Wu",
+      "Jin Yang",
+      "Jinjin Li",
+      "Jinjing Huang",
+      "Jinquan Sun",
+      "Jinyao Chen",
+      "Juanhui Tu",
+      "Jun Liu",
+      "Jun Mei",
+      "Jun Xu",
+      "Jun Zhou",
+      "Junjie Ou",
+      "Junnan Sipan",
+      "Junpeng Fang",
+      "Kaihong Zhang",
+      "Kaiqin Hu",
+      "Ke Shi",
+      "Kuan Xu",
+      "Kun Tang",
+      "Kunlong Chen",
+      "Lanyin Mei",
+      "Lei Chen",
+      "Lei Liang",
+      "Lei Xu",
+      "Li Tang",
+      "Liang Jiang",
+      "Liangcheng Fu",
+      "Lihui Zhang",
+      "Linfeng Shi",
+      "Lintao Ma",
+      "Liyuan Liu",
+      "Longfei Li",
+      "Longfei Zheng",
+      "Lu Liu",
+      "Lu Yu",
+      "Man Li",
+      "Meiqi Zhu",
+      "Meng Li",
+      "Mengjie Gao",
+      "Mengshu Sun",
+      "Mingming Yin",
+      "Mingyang Zhang",
+      "Mingyuan Fan",
+      "Nuo Xu",
+      "Pan Tang",
+      "Peijie Jiang",
+      "Peilong Zhao",
+      "Peng Lin",
+      "Pingping Liu",
+      "Qi Zuo",
+      "Qian Zhao",
+      "Qiang Cheng",
+      "Qianggang Cao",
+      "Qiaoben Bao",
+      "Qing Cui",
+      "Qingyuan Yang",
+      "Qitao Shi",
+      "Qiyin Huang",
+      "Qizheng Zhou",
+      "Quan Wan",
+      "Runyuan Zhao",
+      "Shaomian Zheng",
+      "Shaowei Wei",
+      "Shengnan Zhang",
+      "Shuaicheng Li",
+      "Shujie Li",
+      "Shuo Zhang",
+      "Sikang Bian",
+      "Tianchu Yao",
+      "Tiange Xu",
+      "Tianshu Wang",
+      "Ting Guo",
+      "Tinghao Wang",
+      "Tingwei Huang",
+      "Tong Zhao",
+      "Tongkai Yang",
+      "Wang Hong",
+      "Wanli Gu",
+      "Wei Lu",
+      "Weichang Wu",
+      "Weiguang Han",
+      "Weiquan Li",
+      "Wenbo Shen",
+      "Wenjing Fang",
+      "Wenzhi Tang",
+      "Xiang Shu",
+      "Xiao Shi",
+      "Xiaodong Yan",
+      "Xiaolu Zhang",
+      "Xiaopei Wan",
+      "Xiaqing Sun",
+      "Xin Zhao",
+      "Xingyu Lu",
+      "Xinxing Yang",
+      "Xinyao Tang",
+      "Xinyu Kong",
+      "Xinyu Liu",
+      "Xiong Xu",
+      "Xuan Sun",
+      "Xudong Han",
+      "Xudong Wang",
+      "Xujie Shen",
+      "Yalin Zhang",
+      "Yangyang Hou",
+      "Yankun Ren",
+      "Yao Zhao",
+      "Ye Chen",
+      "Yeyang Chen",
+      "Yibo Cao",
+      "Yifan Zuo",
+      "Yijie Chen",
+      "Ying Li",
+      "Yingjie Song",
+      "Yingxue Li",
+      "Yiqi Wang",
+      "Yixuan Sun",
+      "Yizhu Xiao",
+      "Yongfei Xu",
+      "Yu Liu",
+      "Yuchen Fang",
+      "Yue Gao",
+      "Yue Yu",
+      "Yue Zhang",
+      "Yuqi Zhang",
+      "Yuxiao He",
+      "Yuxiao Lu",
+      "Yuxin Tian",
+      "Yuxuan Li",
+      "Yuzhuo Fu",
+      "Zhankai Xu",
+      "Zhaoxin Huan",
+      "Zhenduo Zhang",
+      "Zhengke Gui",
+      "Zhengyu Huang",
+      "Zhenjun Ma",
+      "Zhenxuan Pan",
+      "Zheping Qu",
+      "Zhibo Zhu",
+      "Zhidong Fan",
+      "Zhigang Huangfu",
+      "Zhihao Wang",
+      "Zhiqiang Zhang",
+      "Zhizhen Liu",
+      "Zhuyan Zhou",
+      "Zibin Lin",
+      "Zihang Zeng",
+      "Zihao Wang",
+      "Zilong Wang",
+      "Ziqi Liu",
+      "Zitao Xuan",
+      "Zixuan Cheng",
+      "Zujie Wen",
+      "Zuoli Tang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.15079",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15079",
+    "pdf_url": "https://arxiv.org/pdf/2606.15079",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.14571",
+    "title": "StreamMemBench: Streaming Evaluation of Agent Memory for Future-Oriented Assistance",
+    "url": "https://arxiv.org/abs/2606.14571",
+    "published": "2026-06-12",
+    "updated": "2026-06-12",
+    "authors": [
+      "Guanming Liu",
+      "Yuqi Ren",
+      "Hansu Gu",
+      "Peng Zhang",
+      "Weihang Wang",
+      "Jiahao Liu",
+      "Ning Gu",
+      "Tun Lu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.14571",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.14571",
+    "pdf_url": "https://arxiv.org/pdf/2606.14571",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.13643",
+    "title": "Recursive Agent Harnesses",
+    "url": "https://arxiv.org/abs/2606.13643",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Elias Lumer",
+      "Sahil Sen",
+      "Kevin Paul",
+      "Vamse Kumar Subbiah"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2606.13643",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13643",
+    "pdf_url": "https://arxiv.org/pdf/2606.13643",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2606.13385",
+    "title": "Who Pays the Price? Stakeholder-Centric Prompt Injection Benchmarking for Real-world Web Agents",
+    "url": "https://arxiv.org/abs/2606.13385",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Zihao Wang",
+      "Yiming Li",
+      "Yutong Wu",
+      "Zheyu Liu",
+      "Kangjie Chen",
+      "Fok Kar Wai",
+      "Pin-Yu Chen",
+      "Vrizlynn L. L. Thing",
+      "Bo Li",
+      "Dacheng Tao",
+      "Tianwei Zhang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.CY",
+      "cs.HC",
+      "cs.MM"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.13385",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13385",
+    "pdf_url": "https://arxiv.org/pdf/2606.13385",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.13904",
+    "title": "SANA: What Matters for QA Agents over Massive Data Lakes?",
+    "url": "https://arxiv.org/abs/2606.13904",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Austin Senna Wijaya",
+      "Jiaxiang Liu",
+      "Haonan Wang",
+      "Eugene Wu"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.DB"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "embodied-agent",
+      "planning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.13904",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13904",
+    "pdf_url": "https://arxiv.org/pdf/2606.13904",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.12341",
+    "title": "OCELOT: Inference-Leakage Budgets for Privacy-Preserving LLM Agents",
+    "url": "https://arxiv.org/abs/2606.12341",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Jin Xie",
+      "Songze Li"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.12341",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12341",
+    "pdf_url": "https://arxiv.org/pdf/2606.12341",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.12195",
+    "title": "InternVideo3: Agentify Foundation Models with Multimodal Contextual Reasoning",
+    "url": "https://arxiv.org/abs/2606.12195",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Ziang Yan",
+      "Sheng Xia",
+      "Jiashuo Yu",
+      "Yue Wu",
+      "Tianxiang Jiang",
+      "Songze Li",
+      "Kanghui Tian",
+      "Yicheng Xu",
+      "Yinan He",
+      "Kai Chen",
+      "Limin Wang",
+      "Yu Qiao",
+      "Yi Wang"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.12195",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12195",
+    "pdf_url": "https://arxiv.org/pdf/2606.12195",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.11869",
+    "title": "Agents All the Way Down; A Methodology for Building Custom AI Agents from Substrate to Production",
+    "url": "https://arxiv.org/abs/2606.11869",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Marc Alier Forment",
+      "Juanan Pereira",
+      "Francisco José García-Peñalvo",
+      "María José Casañ Guerrero"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2606.11869",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.11869",
+    "pdf_url": "https://arxiv.org/pdf/2606.11869",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2606.17076",
+    "title": "CMIP-Forge: An Agentic System that Retrieves, Computes, and Self-Reviews Climate Science",
+    "url": "https://arxiv.org/abs/2606.17076",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Dmitrii Pantiukhin",
+      "Boris Shapkin",
+      "Ivan Kuznetsov",
+      "Thomas Jung",
+      "Nikolay Koldunov"
+    ],
+    "categories": [
+      "physics.ao-ph",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.17076",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17076",
+    "pdf_url": "https://arxiv.org/pdf/2606.17076",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.11349",
+    "title": "Knowing When to Ask: Self-Gated Clarification for Hierarchical Language Agents",
+    "url": "https://arxiv.org/abs/2606.11349",
+    "published": "2026-06-09",
+    "updated": "2026-06-12",
+    "authors": [
+      "Aijing Gao",
+      "Yiming Kang",
+      "Mengdie Flora Wang",
+      "Jae Oh Woo"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.11349",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.11349",
+    "pdf_url": "https://arxiv.org/pdf/2606.11349",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.11078",
+    "title": "A History-Aware Visually Grounded Critic for Computer Use Agents",
+    "url": "https://arxiv.org/abs/2606.11078",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Jaewoo Lee",
+      "Zaid Khan",
+      "Archiki Prasad",
+      "Justin Chih-Yao Chen",
+      "Supriyo Chakraborty",
+      "Kartik Balasubramaniam",
+      "Sambit Sahu",
+      "Elias Stengel-Eskin",
+      "Hyunji Lee",
+      "Mohit Bansal"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.11078",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.11078",
+    "pdf_url": "https://arxiv.org/pdf/2606.11078",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.10423",
+    "title": "WebChallenger: A Reliable and Efficient Generalist Web Agent",
+    "url": "https://arxiv.org/abs/2606.10423",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Jayoo Hwang",
+      "Xiaowen Zhang",
+      "Vedant Padwal"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "computer-use",
+      "embodied-agent",
+      "memory",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.10423",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10423",
+    "pdf_url": "https://arxiv.org/pdf/2606.10423",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.10381",
+    "title": "Agentic Hybrid RAG for Evidence-Grounded Muon Collider Analysis",
+    "url": "https://arxiv.org/abs/2606.10381",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Ruobing Jiang",
+      "Dawei Fu",
+      "Cheng Jiang",
+      "Tianyi Yang",
+      "Zijian Wang",
+      "Youpeng Wu",
+      "Yong Ban",
+      "Yajun Mao",
+      "Qiang Li"
+    ],
+    "categories": [
+      "hep-ex",
+      "cs.AI",
+      "cs.CL",
+      "cs.IR",
+      "physics.ins-det"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.10381",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10381",
+    "pdf_url": "https://arxiv.org/pdf/2606.10381",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.09764",
+    "title": "iOSWorld: A Benchmark for Personally Intelligent Phone Agents",
+    "url": "https://arxiv.org/abs/2606.09764",
+    "published": "2026-06-08",
+    "updated": "2026-06-08",
+    "authors": [
+      "Lawrence Keunho Jang",
+      "Mareks Woodside",
+      "Geronimo Carom",
+      "Andrew Keunwoo Jang",
+      "Jing Yu Koh",
+      "Ruslan Salakhutdinov"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.09764",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09764",
+    "pdf_url": "https://arxiv.org/pdf/2606.09764",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.09399",
+    "title": "RunAgent SuperBrowser: A Theory of Autonomous Web Navigation Grounded in Human Browsing Behaviour",
+    "url": "https://arxiv.org/abs/2606.09399",
+    "published": "2026-06-08",
+    "updated": "2026-06-08",
+    "authors": [
+      "Radeen Mostafa",
+      "Sawradip Saha"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.09399",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09399",
+    "pdf_url": "https://arxiv.org/pdf/2606.09399",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.09549",
+    "title": "SecureClaw: Clawing Back Control of LLM Agents",
+    "url": "https://arxiv.org/abs/2606.09549",
+    "published": "2026-06-08",
+    "updated": "2026-06-08",
+    "authors": [
+      "Yuhan Ma",
+      "Stefan Schmid"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.09549",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09549",
+    "pdf_url": "https://arxiv.org/pdf/2606.09549",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.09071",
+    "title": "REFLECT: Intervention-Supported Error Attribution for Silent Failures in LLM Agent Traces",
+    "url": "https://arxiv.org/abs/2606.09071",
+    "published": "2026-06-08",
+    "updated": "2026-06-08",
+    "authors": [
+      "Xiaofeng Lin",
+      "Yingxu Wang",
+      "Tung Sum Thomas Kwok",
+      "Daniel Guo",
+      "Sahil Arun Nale",
+      "Charles Fleming",
+      "Guang Cheng"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.09071",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09071",
+    "pdf_url": "https://arxiv.org/pdf/2606.09071",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.05463",
+    "title": "PSEBench: A Controllable and Verifiable Benchmark for Evaluating LLMs in Patient Safety Event Triage",
+    "url": "https://arxiv.org/abs/2606.05463",
+    "published": "2026-06-03",
+    "updated": "2026-06-09",
+    "authors": [
+      "Keqi Han",
+      "Ryan Young",
+      "Annabel Strauss",
+      "Lindsey Hughes",
+      "Katharine M. Nesbitt",
+      "Nicole Schueler",
+      "Che Ngufor",
+      "Carl Yang",
+      "Yuan Xue",
+      "Zhijun Yin"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.05463",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.05463",
+    "pdf_url": "https://arxiv.org/pdf/2606.05463",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.03135",
+    "title": "Uncertainty-Aware Clarification in LLM Agents with Information Gain",
+    "url": "https://arxiv.org/abs/2606.03135",
+    "published": "2026-06-02",
+    "updated": "2026-06-02",
+    "authors": [
+      "Mengyi Deng",
+      "Zhiwei Li",
+      "Xin Li",
+      "Tingyu Zhu",
+      "Ying Zhao",
+      "Zhijiang Guo",
+      "Wei Wang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.03135",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.03135",
+    "pdf_url": "https://arxiv.org/pdf/2606.03135",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.04120",
+    "title": "SaliMory: Orchestrating Cognitive Memory for Conversational Agents",
+    "url": "https://arxiv.org/abs/2606.04120",
+    "published": "2026-06-02",
+    "updated": "2026-06-02",
+    "authors": [
+      "Kai Zhang",
+      "Xinyuan Zhang",
+      "Hongda Jiang",
+      "Shiun-Zu Kuo",
+      "Hyokun Yun",
+      "Ejaz Ahmed",
+      "Shereen Oraby",
+      "Ziyun Li",
+      "Sanat Sharma",
+      "Ann Lee",
+      "Ahmed A Aly",
+      "Anuj Kumar",
+      "Raffay Hamid",
+      "Xin Luna Dong"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.04120",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.04120",
+    "pdf_url": "https://arxiv.org/pdf/2606.04120",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.04296",
+    "title": "The Saturation Trap and the Subjectivity of Intervention Timing: Why Affect-Based Triggers and LLM Judges Fail to Time Interventions on Autonomous Agents",
+    "url": "https://arxiv.org/abs/2606.04296",
+    "published": "2026-06-02",
+    "updated": "2026-06-02",
+    "authors": [
+      "Manvendra Modgil"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.04296",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.04296",
+    "pdf_url": "https://arxiv.org/pdf/2606.04296",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.03108",
+    "title": "EvoTrainer: Co-Evolving LLM Policies and Training Harnesses for Autonomous Agentic Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2606.03108",
+    "published": "2026-06-02",
+    "updated": "2026-06-12",
+    "authors": [
+      "Guhong Chen",
+      "Yingcheng Shi",
+      "Yongbin Li",
+      "Binhua Li",
+      "Xander Xu",
+      "Hu Wei",
+      "Shiwen Ni",
+      "Min Yang",
+      "Jieping Ye"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "reasoning"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.03108",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.03108",
+    "pdf_url": "https://arxiv.org/pdf/2606.03108",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.02388",
+    "title": "Policy and World Modeling Co-Training for Language Agents",
+    "url": "https://arxiv.org/abs/2606.02388",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Ning Lu",
+      "Baijiong Lin",
+      "Shengcai Liu",
+      "Jiahao Wu",
+      "Haoze Lv",
+      "Yanbin Wei",
+      "Lingting Zhu",
+      "Shengju Qian",
+      "Xin Wang",
+      "Ying-Cong Chen",
+      "Qi Wang",
+      "Ke Tang"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.02388",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.02388",
+    "pdf_url": "https://arxiv.org/pdf/2606.02388",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.01815",
+    "title": "CRAB-Bench: Evaluating LLM Agents under Complex Task Dependencies and Human-aligned User Simulation",
+    "url": "https://arxiv.org/abs/2606.01815",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Danqing Wang",
+      "Akshay Sivaraman",
+      "Lei Li"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.01815",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.01815",
+    "pdf_url": "https://arxiv.org/pdf/2606.01815",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.02380",
+    "title": "SPADE-Bench: Evaluating Spontaneous Strategic Deception in Agents via Plan-Action Divergence",
+    "url": "https://arxiv.org/abs/2606.02380",
+    "published": "2026-06-01",
+    "updated": "2026-06-28",
+    "authors": [
+      "Yuyan Bu",
+      "Haowei Li",
+      "Qirui Zheng",
+      "Bowen Dong",
+      "Kaiyue Yang",
+      "Jiaming Ji",
+      "Yingshui Tan",
+      "Wenxin Li",
+      "Yaodong Yang",
+      "Juntao Dai"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2606.02380",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.02380",
+    "pdf_url": "https://arxiv.org/pdf/2606.02380",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.00914",
+    "title": "Adversarial Feeds Steer LLM Agent Decisions Against Their Defaults",
+    "url": "https://arxiv.org/abs/2606.00914",
+    "published": "2026-05-30",
+    "updated": "2026-05-30",
+    "authors": [
+      "Rana Muhammad Usman"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.00914",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.00914",
+    "pdf_url": "https://arxiv.org/pdf/2606.00914",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.00915",
+    "title": "Autonomous agentic design for photonics",
+    "url": "https://arxiv.org/abs/2606.00915",
+    "published": "2026-05-30",
+    "updated": "2026-05-30",
+    "authors": [
+      "Prashanta Kharel",
+      "Amin Khavasi",
+      "Xinzhong Chen",
+      "Tyler W. Hughes"
+    ],
+    "categories": [
+      "physics.optics"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.00915",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.00915",
+    "pdf_url": "https://arxiv.org/pdf/2606.00915",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.31278",
+    "title": "Industrializing Prediction-Powered Inference: The GLIDE Library for Reliable GenAI and Agentic Systems Evaluation",
+    "url": "https://arxiv.org/abs/2605.31278",
+    "published": "2026-05-29",
+    "updated": "2026-06-04",
+    "authors": [
+      "Grégoire Martinon",
+      "Ibrahim Merad",
+      "Mohammed Raki"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG",
+      "stat.ME"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2605.31278",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.31278",
+    "pdf_url": "https://arxiv.org/pdf/2605.31278",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2605.29676",
+    "title": "Notation Matters: A Benchmark Study of Token-Optimized Formats in Agentic AI Systems",
+    "url": "https://arxiv.org/abs/2605.29676",
+    "published": "2026-05-28",
+    "updated": "2026-06-17",
+    "authors": [
+      "Lorenz Kutschka",
+      "Bernhard Geiger"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2605.29676",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.29676",
+    "pdf_url": "https://arxiv.org/pdf/2605.29676",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2605.28046",
+    "title": "MemCog: From Memory-as-Tool to Memory-as-Cognition in Conversational Agents",
+    "url": "https://arxiv.org/abs/2605.28046",
+    "published": "2026-05-27",
+    "updated": "2026-05-27",
+    "authors": [
+      "Zihan Li",
+      "Xingyu Fan",
+      "Feifei Li",
+      "Wenhui Que"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.28046",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.28046",
+    "pdf_url": "https://arxiv.org/pdf/2605.28046",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.28607",
+    "title": "Adaptive Multimodal Agents-Based Framework for Automatic Workflow Execution",
+    "url": "https://arxiv.org/abs/2605.28607",
+    "published": "2026-05-27",
+    "updated": "2026-05-27",
+    "authors": [
+      "Susanna Cifani",
+      "Mario Luca Bernardi",
+      "Marta Cimitile"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "embodied-agent",
+      "multi-agent",
+      "planning",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.28607",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.28607",
+    "pdf_url": "https://arxiv.org/pdf/2605.28607",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.28120",
+    "title": "LegalGraphRAG: Multi-Agent Graph Retrieval-Augmented Generation for Reliable Legal Reasoning",
+    "url": "https://arxiv.org/abs/2605.28120",
+    "published": "2026-05-27",
+    "updated": "2026-05-27",
+    "authors": [
+      "Zerui Chen",
+      "Qinggang Zhang",
+      "Zhishang Xiang",
+      "Zhimin Wei",
+      "Linfeng Gao",
+      "Xiao Huang",
+      "Zhihong Zhang",
+      "Jinsong Su"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.28120",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.28120",
+    "pdf_url": "https://arxiv.org/pdf/2605.28120",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.28787",
+    "title": "Do Agents Need Semantic Metadata? A Comparative Study in Agentic Data Retrieval",
+    "url": "https://arxiv.org/abs/2605.28787",
+    "published": "2026-05-27",
+    "updated": "2026-05-27",
+    "authors": [
+      "Shiyu Chen",
+      "Tarfah Alrashed",
+      "Alon Halevy",
+      "Natasha Noy"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.28787",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.28787",
+    "pdf_url": "https://arxiv.org/pdf/2605.28787",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.27366",
+    "title": "MUSE-Autoskill: Self-Evolving Agents via Skill Creation, Memory, Management, and Evaluation",
+    "url": "https://arxiv.org/abs/2605.27366",
+    "published": "2026-05-26",
+    "updated": "2026-07-03",
+    "authors": [
+      "Huawei Lin",
+      "Peng Li",
+      "Jie Song",
+      "Fuxin Jiang",
+      "Tieying Zhang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.LG",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.27366",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.27366",
+    "pdf_url": "https://arxiv.org/pdf/2605.27366",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.26926",
+    "title": "From Norms to Indicators (N2I-RAG): An Agentic Retrieval-Augmented Generation Framework for Legal Indicator Computation",
+    "url": "https://arxiv.org/abs/2605.26926",
+    "published": "2026-05-26",
+    "updated": "2026-05-26",
+    "authors": [
+      "Youssef Al Mouatamid",
+      "Marie Bonnin",
+      "Jihad Zahir"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.26926",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.26926",
+    "pdf_url": "https://arxiv.org/pdf/2605.26926",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.27333",
+    "title": "FinHarness: An Inline Lifecycle Safety Harness for Finance LLM Agents",
+    "url": "https://arxiv.org/abs/2605.27333",
+    "published": "2026-05-26",
+    "updated": "2026-05-26",
+    "authors": [
+      "Haoxuan Jia",
+      "Yang Liu",
+      "Bin Chong",
+      "Yingguang Yang",
+      "Yancheng Chen",
+      "Jiayu Liang",
+      "Qian Li",
+      "Hanning Lu",
+      "Kefu Xu",
+      "Hao Zheng",
+      "Chongyang Zhang",
+      "Hao Peng",
+      "Philip S. Yu"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.27333",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.27333",
+    "pdf_url": "https://arxiv.org/pdf/2605.27333",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.26720",
+    "title": "Towards Feedback-to-Plan Decisions for Self-Evolving LLM Agents in CUDA Kernel Generation",
+    "url": "https://arxiv.org/abs/2605.26720",
+    "published": "2026-05-26",
+    "updated": "2026-05-26",
+    "authors": [
+      "Yee Hin Chong",
+      "Jiaming Wu",
+      "Youhui Zhang",
+      "Peng Qu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.26720",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.26720",
+    "pdf_url": "https://arxiv.org/pdf/2605.26720",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.26252",
+    "title": "Is Agent Memory a Database? Rethinking Data Foundations for Long-Term AI Agent Memory",
+    "url": "https://arxiv.org/abs/2605.26252",
+    "published": "2026-05-25",
+    "updated": "2026-05-25",
+    "authors": [
+      "Abdelghny Orogat",
+      "Essam Mansour"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.DB"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.26252",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.26252",
+    "pdf_url": "https://arxiv.org/pdf/2605.26252",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.26305",
+    "title": "Experiments in Agentic AI for Science",
+    "url": "https://arxiv.org/abs/2605.26305",
+    "published": "2026-05-25",
+    "updated": "2026-05-29",
+    "authors": [
+      "Judy Fox",
+      "Geoffrey Fox"
+    ],
+    "categories": [
+      "cs.AI",
+      "eess.SY",
+      "hep-ph"
+    ],
+    "topics": [
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm",
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.26305",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.26305",
+    "pdf_url": "https://arxiv.org/pdf/2605.26305",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.23636",
+    "title": "RF Instrument Agent (RFIA): Empowering RF Instruments with Natural Language Understanding, Scheduling and Execution of Complex Tasks",
+    "url": "https://arxiv.org/abs/2605.23636",
+    "published": "2026-05-22",
+    "updated": "2026-05-22",
+    "authors": [
+      "Chunhui Li",
+      "Wei Fan"
+    ],
+    "categories": [
+      "eess.SY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.23636",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.23636",
+    "pdf_url": "https://arxiv.org/pdf/2605.23636",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.22321",
+    "title": "Benchmarking Autonomous Agents against Temporal, Spatial, and Semantic Evasions",
+    "url": "https://arxiv.org/abs/2605.22321",
+    "published": "2026-05-21",
+    "updated": "2026-05-21",
+    "authors": [
+      "Jianan Ma",
+      "Xiaohu Du",
+      "Ruixiao Lin",
+      "Yaoxiang Bian",
+      "Jialuo Chen",
+      "Jingyi Wang",
+      "Xiaofang Yang",
+      "Shiwen Cui",
+      "Changhua Meng",
+      "Xinhao Deng",
+      "Zhen Wang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.22321",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.22321",
+    "pdf_url": "https://arxiv.org/pdf/2605.22321",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.21740",
+    "title": "SMDD-Bench: Can LLMs Solve Real-World Small Molecule Drug Design Tasks?",
+    "url": "https://arxiv.org/abs/2605.21740",
+    "published": "2026-05-20",
+    "updated": "2026-05-24",
+    "authors": [
+      "Kevin Han",
+      "Renfei Zhang",
+      "Kathy Wei",
+      "Hamed Mahdavi",
+      "Niloofar Mireshghallah",
+      "Amir Barati Farimani"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.21740",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.21740",
+    "pdf_url": "https://arxiv.org/pdf/2605.21740",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.20874",
+    "title": "Governance by Construction for Generalist Agents",
+    "url": "https://arxiv.org/abs/2605.20874",
+    "published": "2026-05-20",
+    "updated": "2026-05-20",
+    "authors": [
+      "Segev Shlomov",
+      "Iftach Shoham",
+      "Alon Oved",
+      "Ido Levy",
+      "Sami Marreed",
+      "Harold Ship",
+      "Offer Akrabi",
+      "Sergey Zeltyn",
+      "Avi Yaeli",
+      "Nir Mashkif"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.20874",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.20874",
+    "pdf_url": "https://arxiv.org/pdf/2605.20874",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.20306",
+    "title": "WildRoadBench: A Wild Aerial Road-Damage Grounding Benchmark for Vision-Language Models and Autonomous Agents",
+    "url": "https://arxiv.org/abs/2605.20306",
+    "published": "2026-05-19",
+    "updated": "2026-06-02",
+    "authors": [
+      "Bingnan Liu",
+      "Chenhang Cui",
+      "Rui Huang",
+      "Jiani Luo",
+      "Zhirong Shen",
+      "Tinghao Wang",
+      "Xiande Huang",
+      "Lingbei Meng",
+      "Fei Shen",
+      "An Zhang"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.20306",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.20306",
+    "pdf_url": "https://arxiv.org/pdf/2605.20306",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.18672",
+    "title": "Position: A Three-Layer Probabilistic Assume-Guarantee Architecture Is Structurally Required for Safe LLM Agent Deployment",
+    "url": "https://arxiv.org/abs/2605.18672",
+    "published": "2026-05-18",
+    "updated": "2026-05-18",
+    "authors": [
+      "S. Bensalem",
+      "Y. Dong",
+      "M. Franzle",
+      "X. Huang",
+      "J. Kroger",
+      "D. Nickovic",
+      "A. Nouri",
+      "R. Roy",
+      "C. Wu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.18672",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.18672",
+    "pdf_url": "https://arxiv.org/pdf/2605.18672",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.17348",
+    "title": "Taming \"Zombie'' Agents: A Markov State-Aware Framework for Resilient Multi-Agent Evolution",
+    "url": "https://arxiv.org/abs/2605.17348",
+    "published": "2026-05-17",
+    "updated": "2026-05-17",
+    "authors": [
+      "Taolin Zhang",
+      "Pukun Zhao",
+      "Qizhou Chen",
+      "Jiuheng Wan",
+      "Chen Chen",
+      "Xiaofeng He",
+      "Chengyu Wang",
+      "Richang Hong"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-safety",
+      "memory",
+      "multi-agent",
+      "reasoning"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.17348",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.17348",
+    "pdf_url": "https://arxiv.org/pdf/2605.17348",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.17453",
+    "title": "Trust No Tool: Evaluating and Defending LLM Agents under Untrusted Tool Feedback",
+    "url": "https://arxiv.org/abs/2605.17453",
+    "published": "2026-05-17",
+    "updated": "2026-05-17",
+    "authors": [
+      "Lecheng Yan",
+      "Ruizhe Li",
+      "Xicheng Han",
+      "Wenxi Li",
+      "Binwu Wang",
+      "Longyue Wang",
+      "Chenyang Lyu",
+      "Guanhua Chen"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.17453",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.17453",
+    "pdf_url": "https://arxiv.org/pdf/2605.17453",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.23986",
+    "title": "MemForest: An Efficient Agent Memory System with Hierarchical Temporal Indexing",
+    "url": "https://arxiv.org/abs/2605.23986",
+    "published": "2026-05-16",
+    "updated": "2026-05-16",
+    "authors": [
+      "Han Chen",
+      "Zining Zhang",
+      "Wenqi Pei",
+      "Bingsheng He",
+      "Ming Wu",
+      "Jason Zeng",
+      "Michael Heinrich",
+      "Wei Wu",
+      "Hongbao Zhang"
+    ],
+    "categories": [
+      "cs.DB",
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.23986",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.23986",
+    "pdf_url": "https://arxiv.org/pdf/2605.23986",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.28850",
+    "title": "Representation Signatures and Risk-Feedback Alignment in LLM Trading Agents",
+    "url": "https://arxiv.org/abs/2605.28850",
+    "published": "2026-05-16",
+    "updated": "2026-05-30",
+    "authors": [
+      "Weicheng Xue"
+    ],
+    "categories": [
+      "cs.LG",
+      "q-fin.CP"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.28850",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.28850",
+    "pdf_url": "https://arxiv.org/pdf/2605.28850",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.14460",
+    "title": "Exploiting LLM Agent Supply Chains via Payload-less Skills",
+    "url": "https://arxiv.org/abs/2605.14460",
+    "published": "2026-05-14",
+    "updated": "2026-05-14",
+    "authors": [
+      "Xinyu Liu",
+      "Yukai Zhao",
+      "Xing Hu",
+      "Xin Xia"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.14460",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.14460",
+    "pdf_url": "https://arxiv.org/pdf/2605.14460",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.14126",
+    "title": "Reinforcement Learning for Tool-Calling Agents in Fast Healthcare Interoperability Resources (FHIR)",
+    "url": "https://arxiv.org/abs/2605.14126",
+    "published": "2026-05-13",
+    "updated": "2026-05-13",
+    "authors": [
+      "Marius S. Knorr",
+      "Robert Müller",
+      "Jan P. Bremer",
+      "Nils Schweingruber"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.14126",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.14126",
+    "pdf_url": "https://arxiv.org/pdf/2605.14126",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.11882",
+    "title": "On-Policy Self-Evolution via Failure Trajectories for Agentic Safety Alignment",
+    "url": "https://arxiv.org/abs/2605.11882",
+    "published": "2026-05-12",
+    "updated": "2026-05-12",
+    "authors": [
+      "Bo Yin",
+      "Qi Li",
+      "Xinchao Wang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.11882",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.11882",
+    "pdf_url": "https://arxiv.org/pdf/2605.11882",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.11534",
+    "title": "PRISM: : Planning and Reasoning with Intent in Simulated Embodied Environments",
+    "url": "https://arxiv.org/abs/2605.11534",
+    "published": "2026-05-12",
+    "updated": "2026-05-12",
+    "authors": [
+      "Yunn Kang Lim",
+      "Pengzhan Sun",
+      "Ziyi Bai",
+      "Xun Xu",
+      "Angela Yao",
+      "Xulei Yang",
+      "Shijie Li"
+    ],
+    "categories": [
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.11534",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.11534",
+    "pdf_url": "https://arxiv.org/pdf/2605.11534",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.11388",
+    "title": "Deep Reasoning in General Purpose Agents via Structured Meta-Cognition",
+    "url": "https://arxiv.org/abs/2605.11388",
+    "published": "2026-05-12",
+    "updated": "2026-05-12",
+    "authors": [
+      "Dean Light",
+      "Michael Theologitis",
+      "Kshitish Ghate",
+      "Shuyue Stella Li",
+      "Benjamin Newman",
+      "Chirag Shah",
+      "Aylin Caliskan",
+      "Pang Wei Koh",
+      "Dan Suciu",
+      "Yulia Tsvetkov"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.11388",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.11388",
+    "pdf_url": "https://arxiv.org/pdf/2605.11388",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.11039",
+    "title": "The Granularity Mismatch in Agent Security: Argument-Level Provenance Solves Enforcement and Isolates the LLM Reasoning Bottleneck",
+    "url": "https://arxiv.org/abs/2605.11039",
+    "published": "2026-05-11",
+    "updated": "2026-05-11",
+    "authors": [
+      "Linfeng Fan",
+      "Ziwei Li",
+      "Yuan Tian",
+      "Yichen Wang",
+      "Rongsheng Li",
+      "Xiong Wang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.11039",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.11039",
+    "pdf_url": "https://arxiv.org/pdf/2605.11039",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.10763",
+    "title": "MATRA: Modeling the Attack Surface of Agentic AI Systems -- OpenClaw Case Study",
+    "url": "https://arxiv.org/abs/2605.10763",
+    "published": "2026-05-11",
+    "updated": "2026-05-11",
+    "authors": [
+      "Tim Van hamme",
+      "Thomas Vissers",
+      "Javier Carnerero-Cano",
+      "Mario Fritz",
+      "Emil C. Lupu",
+      "Lieven Desmet",
+      "Dinil Mon Divakaran"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.10763",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.10763",
+    "pdf_url": "https://arxiv.org/pdf/2605.10763",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.10365",
+    "title": "Agent-ValueBench: A Comprehensive Benchmark for Evaluating Agent Values",
+    "url": "https://arxiv.org/abs/2605.10365",
+    "published": "2026-05-11",
+    "updated": "2026-05-11",
+    "authors": [
+      "Haonan Dong",
+      "Qiguan Feng",
+      "Kehan Jiang",
+      "Haoran Ye",
+      "Xin Zhang",
+      "Guojie Song"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.10365",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.10365",
+    "pdf_url": "https://arxiv.org/pdf/2605.10365",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.09168",
+    "title": "CIVeX: Causal Intervention Verification for Language Agents",
+    "url": "https://arxiv.org/abs/2605.09168",
+    "published": "2026-05-09",
+    "updated": "2026-05-09",
+    "authors": [
+      "Fabio Rovai"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-safety",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.09168",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.09168",
+    "pdf_url": "https://arxiv.org/pdf/2605.09168",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.08763",
+    "title": "When LLMs Team Up: A Coordinated Attack Framework for Automated Cyber Intrusions",
+    "url": "https://arxiv.org/abs/2605.08763",
+    "published": "2026-05-09",
+    "updated": "2026-05-09",
+    "authors": [
+      "Minfeng Qi",
+      "Tianqing Zhu",
+      "Zijie Xu",
+      "Congcong Zhu",
+      "Qin Wang",
+      "Wanlei Zhou"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.08763",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.08763",
+    "pdf_url": "https://arxiv.org/pdf/2605.08763",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.06078",
+    "title": "Milestone-Guided Policy Learning for Long-Horizon Language Agents",
+    "url": "https://arxiv.org/abs/2605.06078",
+    "published": "2026-05-07",
+    "updated": "2026-05-07",
+    "authors": [
+      "Zixuan Wang",
+      "Yuchen Yan",
+      "Hongxing Li",
+      "Teng Pan",
+      "Dingming Li",
+      "Ruiqing Zhang",
+      "Weiming Lu",
+      "Jun Xiao",
+      "Yueting Zhuang",
+      "Yongliang Shen"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.06078",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.06078",
+    "pdf_url": "https://arxiv.org/pdf/2605.06078",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.03505",
+    "title": "LATS-RCA: Language Agent Tree Search for Root Cause Analysis in Microservices",
+    "url": "https://arxiv.org/abs/2605.03505",
+    "published": "2026-05-05",
+    "updated": "2026-06-12",
+    "authors": [
+      "Alexander Naakka",
+      "Yuqing Wang",
+      "Mika V Mäntylä"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "reasoning"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.03505",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.03505",
+    "pdf_url": "https://arxiv.org/pdf/2605.03505",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.04107",
+    "title": "TSCG: Deterministic Tool-Schema Compilation for Agentic LLM Deployments",
+    "url": "https://arxiv.org/abs/2605.04107",
+    "published": "2026-05-04",
+    "updated": "2026-05-04",
+    "authors": [
+      "Furkan Sakizli"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2605.04107",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.04107",
+    "pdf_url": "https://arxiv.org/pdf/2605.04107",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2605.05242",
+    "title": "Beyond Semantic Similarity: Rethinking Retrieval for Agentic Search via Direct Corpus Interaction",
+    "url": "https://arxiv.org/abs/2605.05242",
+    "published": "2026-05-03",
+    "updated": "2026-05-03",
+    "authors": [
+      "Zhuofeng Li",
+      "Haoxiang Zhang",
+      "Cong Wei",
+      "Pan Lu",
+      "Ping Nie",
+      "Yi Lu",
+      "Yuyang Bai",
+      "Shangbin Feng",
+      "Hangxiao Zhu",
+      "Ming Zhong",
+      "Yuyu Zhang",
+      "Jianwen Xie",
+      "Yejin Choi",
+      "James Zou",
+      "Jiawei Han",
+      "Wenhu Chen",
+      "Jimmy Lin",
+      "Dongfu Jiang",
+      "Yu Zhang"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.05242",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.05242",
+    "pdf_url": "https://arxiv.org/pdf/2605.05242",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.00081",
+    "title": "Alignment Contracts for Agentic Security Systems",
+    "url": "https://arxiv.org/abs/2605.00081",
+    "published": "2026-04-30",
+    "updated": "2026-04-30",
+    "authors": [
+      "Isaac David",
+      "Marco Guarnieri",
+      "Arthur Gervais"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.LO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.00081",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.00081",
+    "pdf_url": "https://arxiv.org/pdf/2605.00081",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.27699",
+    "title": "Bridging Values and Behavior: A Hierarchical Framework for Proactive Embodied Agents",
+    "url": "https://arxiv.org/abs/2604.27699",
+    "published": "2026-04-30",
+    "updated": "2026-04-30",
+    "authors": [
+      "Chunhui Zhang",
+      "Yuxuan Wang",
+      "Aoyang Qin",
+      "Yi-Long Lu",
+      "Kunlun Wu",
+      "Yizhou Wang",
+      "Wei Wang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2604.27699",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.27699",
+    "pdf_url": "https://arxiv.org/pdf/2604.27699",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2604.26274",
+    "title": "Enforcing Benign Trajectories: A Behavioral Firewall for Structured-Workflow AI Agents",
+    "url": "https://arxiv.org/abs/2604.26274",
+    "published": "2026-04-29",
+    "updated": "2026-04-29",
+    "authors": [
+      "Hung Dang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.26274",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.26274",
+    "pdf_url": "https://arxiv.org/pdf/2604.26274",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.24826",
+    "title": "A Comparative Evaluation of AI Agent Security Guardrails",
+    "url": "https://arxiv.org/abs/2604.24826",
+    "published": "2026-04-27",
+    "updated": "2026-04-27",
+    "authors": [
+      "Qi Li",
+      "Jiu Li",
+      "Pingtao Wei",
+      "Jianjun Xu",
+      "Xueyi Wei",
+      "Jiwei Shi",
+      "Xuan Zhang",
+      "Yanhui Yang",
+      "Xiaodong Hui",
+      "Peng Xu",
+      "Lingquan Zhou"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.24826",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.24826",
+    "pdf_url": "https://arxiv.org/pdf/2604.24826",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.13686",
+    "title": "Benchmarking Web Agent Safety under E-commerce Deceptive Interfaces",
+    "url": "https://arxiv.org/abs/2606.13686",
+    "published": "2026-04-26",
+    "updated": "2026-04-26",
+    "authors": [
+      "Zijing Shi",
+      "Meng Fang",
+      "Ling Chen"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.CY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2606.13686",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13686",
+    "pdf_url": "https://arxiv.org/pdf/2606.13686",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.19821",
+    "title": "JTPRO: A Joint Tool-Prompt Reflective Optimization Framework for Language Agents",
+    "url": "https://arxiv.org/abs/2604.19821",
+    "published": "2026-04-20",
+    "updated": "2026-04-20",
+    "authors": [
+      "Sandip Ghoshal",
+      "Anshul Mittal",
+      "Jyotika Singh",
+      "Miguel Ballesteros",
+      "Weiyi Sun",
+      "Fang Tu",
+      "Shailender Singh",
+      "Yassine Benajiba",
+      "Fahad Shah",
+      "Sujeeth Bharadwaj",
+      "Sujith Ravi",
+      "Dan Roth"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2604.19821",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.19821",
+    "pdf_url": "https://arxiv.org/pdf/2604.19821",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2604.18718",
+    "title": "Towards Optimal Agentic Architectures for Offensive Security Tasks",
+    "url": "https://arxiv.org/abs/2604.18718",
+    "published": "2026-04-20",
+    "updated": "2026-04-20",
+    "authors": [
+      "Isaac David",
+      "Arthur Gervais"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.18718",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.18718",
+    "pdf_url": "https://arxiv.org/pdf/2604.18718",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.12986",
+    "title": "Parallax: Why AI Agents That Think Must Never Act",
+    "url": "https://arxiv.org/abs/2604.12986",
+    "published": "2026-04-14",
+    "updated": "2026-04-14",
+    "authors": [
+      "Joel Fokou"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.12986",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.12986",
+    "pdf_url": "https://arxiv.org/pdf/2604.12986",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.06972",
+    "title": "Differentiable Environment-Trajectory Co-Optimization for Safe Multi-Agent Navigation",
+    "url": "https://arxiv.org/abs/2604.06972",
+    "published": "2026-04-08",
+    "updated": "2026-04-08",
+    "authors": [
+      "Zhan Gao",
+      "Gabriele Fadini",
+      "Stelian Coros",
+      "Amanda Prorok"
+    ],
+    "categories": [
+      "cs.RO",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "embodied-agent",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.06972",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.06972",
+    "pdf_url": "https://arxiv.org/pdf/2604.06972",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.04426",
+    "title": "ShieldNet: Network-Level Guardrails against Emerging Supply-Chain Injections in Agentic Systems",
+    "url": "https://arxiv.org/abs/2604.04426",
+    "published": "2026-04-06",
+    "updated": "2026-04-06",
+    "authors": [
+      "Zhuowen Yuan",
+      "Zhaorun Chen",
+      "Zhen Xiang",
+      "Nathaniel D. Bastian",
+      "Seyyed Hadi Hashemi",
+      "Chaowei Xiao",
+      "Wenbo Guo",
+      "Bo Li"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.04426",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.04426",
+    "pdf_url": "https://arxiv.org/pdf/2604.04426",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.03098",
+    "title": "Co-Evolution of Policy and Internal Reward for Language Agents",
+    "url": "https://arxiv.org/abs/2604.03098",
+    "published": "2026-04-03",
+    "updated": "2026-04-03",
+    "authors": [
+      "Xinyu Wang",
+      "Hanwei Wu",
+      "Jingwei Song",
+      "Shuyuan Zhang",
+      "Jiayi Zhang",
+      "Fanqi Kong",
+      "Tung Sum Thomas Kwok",
+      "Xiao-Wen Chang",
+      "Yuyu Luo",
+      "Chenglin Wu",
+      "Bang Liu"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2604.03098",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.03098",
+    "pdf_url": "https://arxiv.org/pdf/2604.03098",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.15309",
+    "title": "CCTU: A Benchmark for Tool Use under Complex Constraints",
+    "url": "https://arxiv.org/abs/2603.15309",
+    "published": "2026-03-16",
+    "updated": "2026-03-16",
+    "authors": [
+      "Junjie Ye",
+      "Guoqiang Zhang",
+      "Wenjie Fu",
+      "Tao Gui",
+      "Qi Zhang",
+      "Xuanjing Huang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2603.15309",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.15309",
+    "pdf_url": "https://arxiv.org/pdf/2603.15309",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2603.11890",
+    "title": "QUARE: Quality-Aware Requirements Analysis through Multi-Agent Dialectical Negotiation",
+    "url": "https://arxiv.org/abs/2603.11890",
+    "published": "2026-03-12",
+    "updated": "2026-06-05",
+    "authors": [
+      "Haowei Cheng",
+      "Milhan Kim",
+      "Foutse Khomh",
+      "Teeradaj Racharak",
+      "Nobukazu Yoshioka",
+      "Naoyasu Ubayashi",
+      "Hironori Washizaki"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "rag"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.11890",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.11890",
+    "pdf_url": "https://arxiv.org/pdf/2603.11890",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.07557",
+    "title": "AgentRaft: Automated Detection of Data Over-Exposure in LLM Agents",
+    "url": "https://arxiv.org/abs/2603.07557",
+    "published": "2026-03-08",
+    "updated": "2026-03-08",
+    "authors": [
+      "Yixi Lin",
+      "Jiangrong Wu",
+      "Yuhong Nan",
+      "Xueqiang Wang",
+      "Xinyuan Zhang",
+      "Zibin Zheng"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2603.07557",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.07557",
+    "pdf_url": "https://arxiv.org/pdf/2603.07557",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2603.05578",
+    "title": "Tool-Genesis: A Task-Driven Tool Creation Benchmark for Self-Evolving Language Agent",
+    "url": "https://arxiv.org/abs/2603.05578",
+    "published": "2026-03-05",
+    "updated": "2026-03-05",
+    "authors": [
+      "Bowei Xia",
+      "Mengkang Hu",
+      "Shijian Wang",
+      "Jiarui Jin",
+      "Wenxiang Jiao",
+      "Yuan Lu",
+      "Kexin Li",
+      "Ping Luo"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.05578",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.05578",
+    "pdf_url": "https://arxiv.org/pdf/2603.05578",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.01712",
+    "title": "FT-Dojo: Towards Autonomous LLM Fine-Tuning with Language Agents",
+    "url": "https://arxiv.org/abs/2603.01712",
+    "published": "2026-03-02",
+    "updated": "2026-05-20",
+    "authors": [
+      "Qizheng Li",
+      "Yifei Zhang",
+      "Xiao Yang",
+      "Xu Yang",
+      "Zhuo Wang",
+      "Weiqing Liu",
+      "Jiang Bian"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.01712",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.01712",
+    "pdf_url": "https://arxiv.org/pdf/2603.01712",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2602.13379",
+    "title": "Unsafer in Many Turns: Benchmarking and Defending Multi-Turn Safety Risks in Tool-Using Agents",
+    "url": "https://arxiv.org/abs/2602.13379",
+    "published": "2026-02-13",
+    "updated": "2026-06-10",
+    "authors": [
+      "Xu Li",
+      "Simon Yu",
+      "Minzhou Pan",
+      "Yiyou Sun",
+      "Bo Li",
+      "Dawn Song",
+      "Xue Lin",
+      "Weiyan Shi"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.CL",
+      "cs.LG",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.13379",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.13379",
+    "pdf_url": "https://arxiv.org/pdf/2602.13379",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2602.11749",
+    "title": "AIR: Improving Agent Safety through Incident Response",
+    "url": "https://arxiv.org/abs/2602.11749",
+    "published": "2026-02-12",
+    "updated": "2026-06-20",
+    "authors": [
+      "Zibo Xiao",
+      "Jun Sun",
+      "Junjie Chen"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.11749",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.11749",
+    "pdf_url": "https://arxiv.org/pdf/2602.11749",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2602.18456",
+    "title": "Beyond single-channel agentic benchmarking",
+    "url": "https://arxiv.org/abs/2602.18456",
+    "published": "2026-02-05",
+    "updated": "2026-02-05",
+    "authors": [
+      "Nelu D. Radpour"
+    ],
+    "categories": [
+      "cs.CY",
+      "cs.AI",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.18456",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.18456",
+    "pdf_url": "https://arxiv.org/pdf/2602.18456",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2601.14652",
+    "title": "MAS-Orchestra: Understanding and Improving Multi-Agent Reasoning Through Holistic Orchestration and Controlled Benchmarks",
+    "url": "https://arxiv.org/abs/2601.14652",
+    "published": "2026-01-21",
+    "updated": "2026-05-21",
+    "authors": [
+      "Zixuan Ke",
+      "Yifei Ming",
+      "Austin Xu",
+      "Ryan Chin",
+      "Xuan-Phi Nguyen",
+      "Prathyusha Jwalapuram",
+      "Jiayu Wang",
+      "Semih Yavuz",
+      "Caiming Xiong",
+      "Shafiq Joty"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent",
+      "reasoning"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2601.14652",
+    "source": "arxiv",
+    "source_id": "arxiv:2601.14652",
+    "pdf_url": "https://arxiv.org/pdf/2601.14652",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2512.23647",
+    "title": "Nested Browser-Use Learning for Agentic Information Seeking",
+    "url": "https://arxiv.org/abs/2512.23647",
+    "published": "2025-12-29",
+    "updated": "2025-12-29",
+    "authors": [
+      "Baixuan Li",
+      "Jialong Wu",
+      "Wenbiao Yin",
+      "Kuan Li",
+      "Zhongwang Zhang",
+      "Huifeng Yin",
+      "Zhengwei Tao",
+      "Liwen Zhang",
+      "Pengjun Xie",
+      "Jingren Zhou",
+      "Yong Jiang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.IR",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2512.23647",
+    "source": "arxiv",
+    "source_id": "arxiv:2512.23647",
+    "pdf_url": "https://arxiv.org/pdf/2512.23647",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2512.23611",
+    "title": "Close the Loop: Synthesizing Infinite Tool-Use Data via Multi-Agent Role-Playing",
+    "url": "https://arxiv.org/abs/2512.23611",
+    "published": "2025-12-29",
+    "updated": "2025-12-29",
+    "authors": [
+      "Yuwen Li",
+      "Wei Zhang",
+      "Zelong Huang",
+      "Mason Yang",
+      "Jiajun Wu",
+      "Shawn Guo",
+      "Huahao Hu",
+      "Lingyi Sun",
+      "Jian Yang",
+      "Mingjie Tang",
+      "Byran Dai"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2512.23611",
+    "source": "arxiv",
+    "source_id": "arxiv:2512.23611",
+    "pdf_url": "https://arxiv.org/pdf/2512.23611",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2512.02605",
+    "title": "IACT: A Self-Organizing Recursive Model for General AI Agents: A Technical White Paper on the Architecture Behind kragent.ai",
+    "url": "https://arxiv.org/abs/2512.02605",
+    "published": "2025-12-02",
+    "updated": "2025-12-02",
+    "authors": [
+      "Pengju Lu"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2512.02605",
+    "source": "arxiv",
+    "source_id": "arxiv:2512.02605",
+    "pdf_url": "https://arxiv.org/pdf/2512.02605",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2510.14548",
+    "title": "LLM Agents Beyond Utility: An Open-Ended Perspective",
+    "url": "https://arxiv.org/abs/2510.14548",
+    "published": "2025-10-16",
+    "updated": "2025-10-16",
+    "authors": [
+      "Asen Nachkov",
+      "Xi Wang",
+      "Luc Van Gool"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2510.14548",
+    "source": "arxiv",
+    "source_id": "arxiv:2510.14548",
+    "pdf_url": "https://arxiv.org/pdf/2510.14548",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2509.26553",
+    "title": "Towards Reliable Benchmarking: A Contamination Free, Controllable Evaluation Framework for Multi-step LLM Function Calling",
+    "url": "https://arxiv.org/abs/2509.26553",
+    "published": "2025-09-30",
+    "updated": "2026-02-06",
+    "authors": [
+      "Seiji Maekawa",
+      "Jackson Hassell",
+      "Pouya Pezeshkpour",
+      "Tom Mitchell",
+      "Estevam Hruschka"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.PL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2509.26553",
+    "source": "arxiv",
+    "source_id": "arxiv:2509.26553",
+    "pdf_url": "https://arxiv.org/pdf/2509.26553",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2509.14477",
+    "title": "Ticket-Bench: A Kickoff for Multilingual and Regionalized Agent Evaluation",
+    "url": "https://arxiv.org/abs/2509.14477",
+    "published": "2025-09-17",
+    "updated": "2025-09-17",
+    "authors": [
+      "Thales Sales Almeida",
+      "João Guilherme Alves Santos",
+      "Thiago Laitz",
+      "Giovana Kerche Bonás"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "reasoning"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2509.14477",
+    "source": "arxiv",
+    "source_id": "arxiv:2509.14477",
+    "pdf_url": "https://arxiv.org/pdf/2509.14477",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2509.02444",
+    "title": "AppCopilot: Toward General, Accurate, Long-Horizon, and Efficient Mobile Agent",
+    "url": "https://arxiv.org/abs/2509.02444",
+    "published": "2025-09-02",
+    "updated": "2025-10-17",
+    "authors": [
+      "Jingru Fan",
+      "Yufan Dang",
+      "Jingyao Wu",
+      "Huatao Li",
+      "Runde Yang",
+      "Xiyuan Yang",
+      "Yuheng Wang",
+      "Chen Qian"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.CV",
+      "cs.HC"
+    ],
+    "topics": [
+      "computer-use",
+      "memory",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2509.02444",
+    "source": "arxiv",
+    "source_id": "arxiv:2509.02444",
+    "pdf_url": "https://arxiv.org/pdf/2509.02444",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2607.06140",
+    "title": "CurateEvo: Data-Curation Evolving for Agentic Post-Training",
+    "url": "https://arxiv.org/abs/2607.06140",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Dingzirui Wang",
+      "Xuanliang Zhang",
+      "Keyan Xu",
+      "Qingfu Zhu",
+      "Wanxiang Che"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "memory",
+      "planning",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.06140",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06140",
+    "pdf_url": "https://arxiv.org/pdf/2607.06140",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.06452",
+    "title": "From Voting to Agent Collaboration: Answer-Type-Aware LLM Pipelines for BioASQ 14b",
+    "url": "https://arxiv.org/abs/2607.06452",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Taeyun Roh",
+      "Eunha Lee",
+      "Wonjune Jang",
+      "Sohyun Chung",
+      "Junha Jung",
+      "Jaewoo Kang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.06452",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06452",
+    "pdf_url": "https://arxiv.org/pdf/2607.06452",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.05772",
+    "title": "Detecting Vulnerability-Inducing Commits via Multi-Stage Reasoning with LLM-Based Agents",
+    "url": "https://arxiv.org/abs/2607.05772",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Liyou Chen",
+      "Hailong Sun",
+      "Xiang Gao",
+      "Yue Pan"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-safety",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.05772",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05772",
+    "pdf_url": "https://arxiv.org/pdf/2607.05772",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.05378",
+    "title": "CompactionRL: Reinforcement Learning with Context Compaction for Long-Horizon Agents",
+    "url": "https://arxiv.org/abs/2607.05378",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Yujiang Li",
+      "Zhenyu Hou",
+      "Yi Jing",
+      "Jie Tang",
+      "Yuxiao Dong"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "coding-agent",
+      "planning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent",
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.05378",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05378",
+    "pdf_url": "https://arxiv.org/pdf/2607.05378",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.05132",
+    "title": "When Agents Lie: Premeditation, Persistence, and Exploitation in Repeated Games",
+    "url": "https://arxiv.org/abs/2607.05132",
+    "published": "2026-07-06",
+    "updated": "2026-07-07",
+    "authors": [
+      "Jerick Shi",
+      "Terry Jingcheng Zhang",
+      "Bernhard Schölkopf",
+      "Vincent Conitzer",
+      "Zhijing Jin"
+    ],
+    "categories": [
+      "cs.CY",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm",
+      "llm-agent",
+      "planning-agent"
+    ],
+    "arxiv_id": "2607.05132",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05132",
+    "pdf_url": "https://arxiv.org/pdf/2607.05132",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2607.04963",
+    "title": "STAPO: Selective Trajectory-Aware Policy Optimization for LLM Agent Training",
+    "url": "https://arxiv.org/abs/2607.04963",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Qiuyi Qi",
+      "Tian Liang",
+      "Mutian Bao",
+      "Jinjian Zhang",
+      "Dongnan Liu",
+      "Wei Zhou",
+      "Linjian Mo",
+      "Ming Kong",
+      "Jie Liu",
+      "Feng Zhang",
+      "Qiang Zhu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.04963",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04963",
+    "pdf_url": "https://arxiv.org/pdf/2607.04963",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.05677",
+    "title": "From Conversation to Contribution: Characterizing Coding Agent in Open-Source Software",
+    "url": "https://arxiv.org/abs/2607.05677",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Zihan Fang",
+      "Yueke Zhang",
+      "Ningzhi Tang",
+      "Collin McMillan",
+      "Toby Jia-Jun Li",
+      "Yu Huang"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "multi-agent",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.05677",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05677",
+    "pdf_url": "https://arxiv.org/pdf/2607.05677",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.05188",
+    "title": "Latent Programming Horizons in Coding Agents",
+    "url": "https://arxiv.org/abs/2607.05188",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "André Silva",
+      "Han Tu",
+      "Martin Monperrus"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "reasoning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.05188",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05188",
+    "pdf_url": "https://arxiv.org/pdf/2607.05188",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.04623",
+    "title": "Can LLMs Really Recover Microservice Failures? A Recovery-Aware Evaluation of Diagnosis-to-Action Reasoning",
+    "url": "https://arxiv.org/abs/2607.04623",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Jiaxing Qi",
+      "Zhongzhi Luan",
+      "Hongyu Zhang",
+      "Shaohan Huang",
+      "Carol Fung",
+      "Yongxin Tong",
+      "Hailong Yang",
+      "Depei Qian"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.DC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2607.04623",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04623",
+    "pdf_url": "https://arxiv.org/pdf/2607.04623",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2607.04470",
+    "title": "Regime-Conditional Stabilisation of LLM-Augmented Cooperative Multi-Agent Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2607.04470",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Faid Keddouri",
+      "Sohaib Houhou",
+      "Aissa Boulmerka",
+      "Nadir Farhi"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "math.OC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.04470",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04470",
+    "pdf_url": "https://arxiv.org/pdf/2607.04470",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.03853",
+    "title": "CogRad: A Cognitively-Inspired Multi-Agent Framework for Radiology Report Generation",
+    "url": "https://arxiv.org/abs/2607.03853",
+    "published": "2026-07-04",
+    "updated": "2026-07-04",
+    "authors": [
+      "Saif Ur Rehman Khan",
+      "Hasaan Maqsood",
+      "Sebastian Vollmer",
+      "Andreas Dengel",
+      "Muhammad Nabeel Asim"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "reasoning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.03853",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03853",
+    "pdf_url": "https://arxiv.org/pdf/2607.03853",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.02879",
+    "title": "MedCalc-Pro: Solving Complex Medical Calculations with LLM Agents",
+    "url": "https://arxiv.org/abs/2607.02879",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Siran Zhao",
+      "Ruihui Hou",
+      "Ziyue Huai",
+      "Chennuo Zhang",
+      "Tong Ruan"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.02879",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02879",
+    "pdf_url": "https://arxiv.org/pdf/2607.02879",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.03423",
+    "title": "Securing Multi-Tool AI Agent Chains With Dynamic, Real-Time Compositional Policies",
+    "url": "https://arxiv.org/abs/2607.03423",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Chris Schneider",
+      "Kriti Faujdar",
+      "Philipp Schoenegger",
+      "Ben Bariach"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.03423",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03423",
+    "pdf_url": "https://arxiv.org/pdf/2607.03423",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.03162",
+    "title": "APeB: Benchmarking Personalization Ability of Large Language Model Agents",
+    "url": "https://arxiv.org/abs/2607.03162",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Garry Yang",
+      "Zizhe Chen",
+      "Xinru Chen",
+      "Yongqiang Chen",
+      "Jianxiang Wang",
+      "Deyu Zou",
+      "Linyi Ding",
+      "Jialiang Wu",
+      "Yunzhong He",
+      "Yu Gong",
+      "James Cheng",
+      "Huaixiao Tou"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.03162",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03162",
+    "pdf_url": "https://arxiv.org/pdf/2607.03162",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.02882",
+    "title": "Diagnosis-Driven Automatic Repair for Agentic Workflow via Symbolic Inference",
+    "url": "https://arxiv.org/abs/2607.02882",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Xuyan Ma",
+      "Yawen Wang",
+      "Junjie Wang",
+      "Xiaofei Xie",
+      "Boyu Wu",
+      "Mingyang Li",
+      "Dandan Wang",
+      "Qing Wang"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.02882",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02882",
+    "pdf_url": "https://arxiv.org/pdf/2607.02882",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.03628",
+    "title": "Swarm-Driven Multi-Agent Reasoning for Smart City Security",
+    "url": "https://arxiv.org/abs/2607.03628",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Saeid Jamshidi",
+      "Kawser Wazed Nafi",
+      "Carol Fung",
+      "Foutse Khomh"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-safety",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.03628",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03628",
+    "pdf_url": "https://arxiv.org/pdf/2607.03628",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.01600",
+    "title": "BOUNDARY_SYNC: Measuring Communication-Induced Representational Coupling in Multi-Agent LLM Systems",
+    "url": "https://arxiv.org/abs/2607.01600",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Zewen Liu"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.CL"
+    ],
+    "topics": [
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.01600",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01600",
+    "pdf_url": "https://arxiv.org/pdf/2607.01600",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.02210",
+    "title": "Criticality-Based Guard Rail Validation for AI Agent Decisions in Autonomous Telecom Networks",
+    "url": "https://arxiv.org/abs/2607.02210",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Ravi Kant Sharma"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.NI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.02210",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02210",
+    "pdf_url": "https://arxiv.org/pdf/2607.02210",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.01661",
+    "title": "Diverse Evidence, Better Forecasts: Multi-Agent Deliberation Under Information Asymmetry",
+    "url": "https://arxiv.org/abs/2607.01661",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Yuante Li",
+      "Yicheng Tao",
+      "Kate Zhang",
+      "Taozhi Wang",
+      "Gefei Gu",
+      "Yaxin Zhou"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "reasoning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.01661",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01661",
+    "pdf_url": "https://arxiv.org/pdf/2607.01661",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.00692",
+    "title": "Self-GC: Self-Governing Context for Long-Horizon LLM Agents",
+    "url": "https://arxiv.org/abs/2607.00692",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Xubin Hao",
+      "Hongjin Meng",
+      "Xin Yin",
+      "Jiawei Zhu",
+      "Chenpeng Cao"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "planning-agent"
+    ],
+    "arxiv_id": "2607.00692",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00692",
+    "pdf_url": "https://arxiv.org/pdf/2607.00692",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.00297",
+    "title": "EPC: A Standardized Protocol for Measuring Evaluator Preference Dynamics in LLM Agent Systems",
+    "url": "https://arxiv.org/abs/2607.00297",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Zewen Liu"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.00297",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00297",
+    "pdf_url": "https://arxiv.org/pdf/2607.00297",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.01523",
+    "title": "Multi-Head Recurrent Memory Agents",
+    "url": "https://arxiv.org/abs/2607.01523",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Jiatong Li",
+      "Samuel Yeh",
+      "Sharon Li"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2607.01523",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01523",
+    "pdf_url": "https://arxiv.org/pdf/2607.01523",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2607.01211",
+    "title": "Are Performance-Optimization Benchmarks Reliably Measuring Coding Agents?",
+    "url": "https://arxiv.org/abs/2607.01211",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Zhi Chen",
+      "Zhensu Sun",
+      "Yuling Shi",
+      "David Lo",
+      "Lingxiao Jiang"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.01211",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01211",
+    "pdf_url": "https://arxiv.org/pdf/2607.01211",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.00918",
+    "title": "From Personas to Plot: Character-Grounded Multi-Agent Story Generation for Long-Form Narratives",
+    "url": "https://arxiv.org/abs/2607.00918",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Aayush Aluru",
+      "Chloe Ho",
+      "Muhammad Hammouri",
+      "Kerry Luo",
+      "Myra Malik",
+      "Ryan Lagasse",
+      "Arjun Bahuguna",
+      "Vasu Sharma"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.00918",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00918",
+    "pdf_url": "https://arxiv.org/pdf/2607.00918",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.00604",
+    "title": "Vehicle Routing Problem Meets Large Language Models: An Overview and Perspectives",
+    "url": "https://arxiv.org/abs/2607.00604",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Xianchao Xiu",
+      "Chong Shen",
+      "Yanjiao Zhu",
+      "Wanquan Liu"
+    ],
+    "categories": [
+      "math.OC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.00604",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00604",
+    "pdf_url": "https://arxiv.org/pdf/2607.00604",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.00972",
+    "title": "Bayesian Uncertainty Propagation for Agentic RAG Pipelines: A Proof-of-Concept Study on Multi-Hop Question Answering",
+    "url": "https://arxiv.org/abs/2607.00972",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Louis Donaldson",
+      "Connor Walker",
+      "Koorosh Aslansefat",
+      "Yiannis Papadopoulos"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2607.00972",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00972",
+    "pdf_url": "https://arxiv.org/pdf/2607.00972",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2607.00422",
+    "title": "KidnapRAG: A Black-Box Attack for Hijacking Reasoning in Agentic Retrieval-Augmented Generation Systems",
+    "url": "https://arxiv.org/abs/2607.00422",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Chanwoo Choi",
+      "Euntae Kim",
+      "Kyuho Lee",
+      "Youngsam Chun",
+      "Jinhee Jeong",
+      "Eunmi Kim",
+      "Myunggyo Oh",
+      "Junseo Jang",
+      "Buru Chang"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2607.00422",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00422",
+    "pdf_url": "https://arxiv.org/pdf/2607.00422",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.31635",
+    "title": "A Tutorial on Autonomous Fault-Tolerant Control Using Knowledge-Grounded LLM Agents",
+    "url": "https://arxiv.org/abs/2606.31635",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Javal Vyas",
+      "Milapji Singh Gill",
+      "Artan Markaj",
+      "Felix Gehlhoff",
+      "Mehmet Mercangöz"
+    ],
+    "categories": [
+      "eess.SY",
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-safety",
+      "planning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.31635",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31635",
+    "pdf_url": "https://arxiv.org/pdf/2606.31635",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.31227",
+    "title": "Securing the AI Agent: A Unified Framework for Multi-Layer Agent Red Teaming",
+    "url": "https://arxiv.org/abs/2606.31227",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Yong Yang",
+      "Xing Zheng",
+      "Huiyu Wu",
+      "Huangsheng Cheng",
+      "Xiaorong Shi",
+      "Jing Guo",
+      "Bo Yang",
+      "Yi Zhou",
+      "Xiangfan Wu",
+      "Zonghao Ying"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.31227",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31227",
+    "pdf_url": "https://arxiv.org/pdf/2606.31227",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2607.00255",
+    "title": "SLM, LLM or Agentic AI? Toward Intelligent UAV-Enabled WPT Systems in Low-Altitude Economy Networks",
+    "url": "https://arxiv.org/abs/2607.00255",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Feibo Jiang",
+      "Li Dong",
+      "Lei Mao",
+      "Kezhi Wang",
+      "Xianbin Wang",
+      "Abbas Jamalipour"
+    ],
+    "categories": [
+      "cs.IT"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.00255",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00255",
+    "pdf_url": "https://arxiv.org/pdf/2607.00255",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.31648",
+    "title": "Think in English, Answer in Korean: Efficient Adaptation of Multilingual Tool-Using Agents",
+    "url": "https://arxiv.org/abs/2606.31648",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Utsav Garg",
+      "Sungjin Hong",
+      "Jason Jung",
+      "Justin Lee",
+      "Shaan Desai",
+      "Joon Hee Kim",
+      "Anirudh Shrinivason",
+      "Edmond Wen",
+      "Susie Park"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "memory",
+      "multi-agent",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "function-calling",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.31648",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31648",
+    "pdf_url": "https://arxiv.org/pdf/2606.31648",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.02577",
+    "title": "Benchmarking the Benchmarks: A Validity Audit of Tool-Calling Evaluation",
+    "url": "https://arxiv.org/abs/2607.02577",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Vishvesh Bhat",
+      "Jay Vaghasiya",
+      "Muhammad Ahmed Mohsin",
+      "Asad Aali"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2607.02577",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02577",
+    "pdf_url": "https://arxiv.org/pdf/2607.02577",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.31314",
+    "title": "A Novel Method for Differential-Algebraic Dynamic Model Discovery in Power Systems: An LLM-Based Multi-Agent Collaborative Framework",
+    "url": "https://arxiv.org/abs/2606.31314",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Xinming Wang",
+      "Fan Tang",
+      "Yingli Wei",
+      "Yakun He",
+      "Zhe Liu",
+      "Ping Jiang",
+      "Haoyu Wu",
+      "Zihan Guo",
+      "Chao Shen"
+    ],
+    "categories": [
+      "eess.SY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.31314",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31314",
+    "pdf_url": "https://arxiv.org/pdf/2606.31314",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.30840",
+    "title": "Contrastive Reflection for Iterative Prompt Optimization",
+    "url": "https://arxiv.org/abs/2606.30840",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Derek Koh",
+      "Jinghui Mo",
+      "Benjamin H. Le",
+      "Jiening Zhan",
+      "Baofen Zheng",
+      "Kevin Bevis",
+      "Nathaniel C. Owen",
+      "Lauren Elizabeth Charney",
+      "Wenqiong Liu",
+      "Jingwei Wu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.30840",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30840",
+    "pdf_url": "https://arxiv.org/pdf/2606.30840",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.30454",
+    "title": "Collective cooperation without individual fidelity in LLM agents",
+    "url": "https://arxiv.org/abs/2606.30454",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Henrique Ferraz de Arruda",
+      "Carlos Gracia Lázaro",
+      "Alberto Aleta",
+      "Yamir Moreno"
+    ],
+    "categories": [
+      "physics.soc-ph",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.30454",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30454",
+    "pdf_url": "https://arxiv.org/pdf/2606.30454",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.30005",
+    "title": "LLM Agents Are Latent Context Managers: Eliciting Self-Managed Context via a Proprioceptive Dashboard",
+    "url": "https://arxiv.org/abs/2606.30005",
+    "published": "2026-06-29",
+    "updated": "2026-07-05",
+    "authors": [
+      "Binyan Xu",
+      "Haitao Li",
+      "Kehuan Zhang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "memory",
+      "planning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.30005",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30005",
+    "pdf_url": "https://arxiv.org/pdf/2606.30005",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.29762",
+    "title": "Do Recommendation Algorithms Work When Users Are LLM Agents? A Case Study on Moltbook",
+    "url": "https://arxiv.org/abs/2606.29762",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Daming Li",
+      "Simeng Han",
+      "Jialu Zhang"
+    ],
+    "categories": [
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.29762",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29762",
+    "pdf_url": "https://arxiv.org/pdf/2606.29762",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.30266",
+    "title": "Towards Continual Motion-Language Agents: LoRA Variants for Incremental Motion Understanding and Generation",
+    "url": "https://arxiv.org/abs/2606.30266",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Bertram Taetz",
+      "Hugo Albuquerque Cosme da Silva",
+      "Gabriele Bleser-Taetz"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm",
+      "language-agent"
+    ],
+    "arxiv_id": "2606.30266",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30266",
+    "pdf_url": "https://arxiv.org/pdf/2606.30266",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.30185",
+    "title": "Dynamo: Dynamic Skill-Tool Evolution for Vision-Language Agents",
+    "url": "https://arxiv.org/abs/2606.30185",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Yutao Sun",
+      "Yanting Miao",
+      "Hao-Xuan Ma",
+      "Mengyu Zhou",
+      "Mingshuai Chen",
+      "Tiancheng Zhao",
+      "Dexin Wang",
+      "Lei Lv",
+      "Li Xu",
+      "Xiaoxi Jiang",
+      "Guanjun Jiang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.30185",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30185",
+    "pdf_url": "https://arxiv.org/pdf/2606.30185",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.30755",
+    "title": "Understanding and Evaluating Claw-like Agent Security Through a Computer-Systems Lens",
+    "url": "https://arxiv.org/abs/2606.30755",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Peizhi Niu",
+      "Wenjie Qu",
+      "Shangding Gu",
+      "Tianneng Shi",
+      "Yuankai Li",
+      "Ahmad Tawaha",
+      "Hend Alzahrani",
+      "Vincent Siu",
+      "Boyi Li",
+      "Chenguang Wang",
+      "Jiaheng Zhang",
+      "Basel Alomair",
+      "Ming Jin",
+      "Muhao Chen",
+      "Chi Wang",
+      "Costas Spanos",
+      "Dawn Song"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.30755",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30755",
+    "pdf_url": "https://arxiv.org/pdf/2606.30755",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.29894",
+    "title": "SABER-Math: Automated Benchmark for Information Retrieval Evaluation in Mathematics",
+    "url": "https://arxiv.org/abs/2606.29894",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Nikolay Georgiev",
+      "Maria Drencheva",
+      "Kseniia Ibragimova",
+      "Ivo Petrov",
+      "Dimitar I. Dimitrov",
+      "Martin Vechev"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI",
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.29894",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29894",
+    "pdf_url": "https://arxiv.org/pdf/2606.29894",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.29961",
+    "title": "DuoMem: Towards Capable On-Device Memory Agents via Dual-Space Distillation",
+    "url": "https://arxiv.org/abs/2606.29961",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Peyman Hosseini",
+      "Ondrej Bohdal",
+      "Ahmed Alajrami",
+      "Andrea Maracani",
+      "Ignacio Castro",
+      "Matthew Purver",
+      "Mete Ozay",
+      "Savas Ozkan",
+      "Taha Ceritli"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.29961",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29961",
+    "pdf_url": "https://arxiv.org/pdf/2606.29961",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.30119",
+    "title": "On the Internet, Nobody Knows You're an LLM Bot: Unmasking Web Agents with Multi-Layer Fingerprinting",
+    "url": "https://arxiv.org/abs/2606.30119",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Iliana Fayolle",
+      "Sihem Bouhenniche",
+      "Samuel Pélissier",
+      "Pierre Laperdrix",
+      "Clémentine Maurice",
+      "Walter Rudametkin"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "embodied-agent",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.30119",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30119",
+    "pdf_url": "https://arxiv.org/pdf/2606.30119",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.30602",
+    "title": "MESA: Prioritizing Vulnerable Communication Channels for Securing Multi-Agent Systems",
+    "url": "https://arxiv.org/abs/2606.30602",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Kunyang Li",
+      "Kyle Domico",
+      "Jonathan Gregory",
+      "Patrick McDaniel"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.30602",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30602",
+    "pdf_url": "https://arxiv.org/pdf/2606.30602",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.29932",
+    "title": "SAGA: Scene-Aware, Goal-Evolving Agents for Long-Horizon CivRealm Strategy Planning",
+    "url": "https://arxiv.org/abs/2606.29932",
+    "published": "2026-06-29",
+    "updated": "2026-07-02",
+    "authors": [
+      "Tianyu Jin",
+      "Shuo Chen",
+      "Yida Wang",
+      "Liuyu Xiang",
+      "Yingzhuo Liu",
+      "Zhiyao Jiang",
+      "Yexin Li",
+      "Zhaofeng He"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.29932",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29932",
+    "pdf_url": "https://arxiv.org/pdf/2606.29932",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.29746",
+    "title": "DEEPMED Search: An Open-Source Agentic Platform for Medical Deep Research with Introspective Verification",
+    "url": "https://arxiv.org/abs/2606.29746",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Maolin Liu",
+      "Fanyu Xu",
+      "Ruoqing Xu",
+      "Jiahang Zhang",
+      "Hao Wang",
+      "Rui Wang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.HC"
+    ],
+    "topics": [
+      "computer-use",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.29746",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29746",
+    "pdf_url": "https://arxiv.org/pdf/2606.29746",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.29225",
+    "title": "PolicyGuard: A Dialogue-Grounded Sub-Agent Verifier for Policy Adherence in LLM Agents",
+    "url": "https://arxiv.org/abs/2606.29225",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Seongjae Kang",
+      "Taehyung Yu",
+      "Sung Ju Hwang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "computer-use",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.29225",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29225",
+    "pdf_url": "https://arxiv.org/pdf/2606.29225",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.29142",
+    "title": "Agent Security Meets Regulatory Reality -- A Practitioner Systematization of Autonomous-Agent Threats and Controls in Regulated Financial Systems",
+    "url": "https://arxiv.org/abs/2606.29142",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Krishna Mohan",
+      "Guda Nagavenkata Srinivasa"
+    ],
+    "categories": [
+      "cs.CY",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "autonomous-agent-llm",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.29142",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29142",
+    "pdf_url": "https://arxiv.org/pdf/2606.29142",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.28733",
+    "title": "Agentic Abstention: Do Agents Know When to Stop Instead of Act?",
+    "url": "https://arxiv.org/abs/2606.28733",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Han Luo",
+      "Bingbing Wen",
+      "Lucy Lu Wang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.28733",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28733",
+    "pdf_url": "https://arxiv.org/pdf/2606.28733",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.28679",
+    "title": "Capability Gates Are Not Authorization: Confused-Deputy Failures in LLM Agent Frameworks",
+    "url": "https://arxiv.org/abs/2606.28679",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "David Mellafe Zuvic"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.28679",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28679",
+    "pdf_url": "https://arxiv.org/pdf/2606.28679",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.29026",
+    "title": "Preventing Error Propagation in Multi-Agent AI through Runtime Monitoring",
+    "url": "https://arxiv.org/abs/2606.29026",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Shahnewaz Karim Sakib",
+      "Anindya Bijoy Das"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.ET"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "reasoning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.29026",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29026",
+    "pdf_url": "https://arxiv.org/pdf/2606.29026",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.28739",
+    "title": "Agent Safety Is Action Alignment",
+    "url": "https://arxiv.org/abs/2606.28739",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Shawn Li",
+      "Yue Zhao"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2606.28739",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28739",
+    "pdf_url": "https://arxiv.org/pdf/2606.28739",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.27632",
+    "title": "Yuvion LLM: An Adversarially-Aware Large Language Model for Content And AI Safety",
+    "url": "https://arxiv.org/abs/2606.27632",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Ting Ma",
+      "Xiufeng Huang",
+      "Benlei Cui",
+      "Xiaowen Xu",
+      "Shikai Qiu",
+      "Ruijie Jian",
+      "Hongxing Li",
+      "Guanghui Wang",
+      "Longtao Huang",
+      "Haiwen Hong",
+      "Haolei Xu",
+      "Wenjing Jiang",
+      "Ziwen Xu",
+      "Zhaoyu Fan",
+      "Shaoxuan He",
+      "Chuxi Xiao",
+      "Yujian Li",
+      "Xinyue Chen",
+      "Chunyang Chai",
+      "Wenxuan Liu",
+      "Ziheng Wang",
+      "Dongjie Zhang",
+      "Yangfan Zhou",
+      "Libin Dong",
+      "Yupeng Cao",
+      "Xiaoqian Xia",
+      "Jing Wang",
+      "Zhe Jiang",
+      "Zhenan Ye",
+      "Guang Yang",
+      "Bin Liu",
+      "Wei Peng",
+      "Ziqiang Zhu",
+      "Meihui Lian",
+      "Kaiwen Lv Kacuila",
+      "Haidong Ding",
+      "Bingyu Zhu",
+      "Yan Wang",
+      "Hai Zhao",
+      "Xuan Jin",
+      "Wei Zhao",
+      "Pengfei Sun",
+      "Wei Wang",
+      "Huiming Zhang",
+      "Bin Li",
+      "Hui Xue"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.27632",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27632",
+    "pdf_url": "https://arxiv.org/pdf/2606.27632",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.26806",
+    "title": "Memory Depth, Not Memory Access: Selective Parametric Consolidation for Long-Running Language Agents",
+    "url": "https://arxiv.org/abs/2606.26806",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Haoliang Han"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.26806",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26806",
+    "pdf_url": "https://arxiv.org/pdf/2606.26806",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.27154",
+    "title": "OpenRCA 2.0: From Outcome Labels to Causal Process Supervision",
+    "url": "https://arxiv.org/abs/2606.27154",
+    "published": "2026-06-25",
+    "updated": "2026-06-30",
+    "authors": [
+      "Aoyang Fang",
+      "Yifan Yang",
+      "Jin'ao Shang",
+      "Qisheng Lu",
+      "Junjielung Xu",
+      "Rui Wang",
+      "Songhan Zhang",
+      "Yuzhong Zhang",
+      "Boxi Yu",
+      "Pinjia He"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.27154",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27154",
+    "pdf_url": "https://arxiv.org/pdf/2606.27154",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.27492",
+    "title": "QueenBee Planner: Skill-Evolving Communication Topologies for Token-Efficient LLM Multi-Agent Systems",
+    "url": "https://arxiv.org/abs/2606.27492",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Congjia Tian",
+      "Yuhang Yao",
+      "Jiaming Cui"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.27492",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27492",
+    "pdf_url": "https://arxiv.org/pdf/2606.27492",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.26758",
+    "title": "EGG: An Expert-Guided Agent Framework for Kernel Generation",
+    "url": "https://arxiv.org/abs/2606.26758",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Yaochen Han",
+      "Ke Fan",
+      "Hongxu Jiang",
+      "Wanqi Xu",
+      "Weiyu Xie",
+      "Runhua Zhang",
+      "Chenhui Zhu",
+      "Yixiang Zhang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "computer-use",
+      "memory",
+      "multi-agent",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.26758",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26758",
+    "pdf_url": "https://arxiv.org/pdf/2606.26758",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.26205",
+    "title": "Knowledge-augmented Agentic AI for Mental Health Medication Information Seeking",
+    "url": "https://arxiv.org/abs/2606.26205",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Huizi Yu",
+      "Jian Liu",
+      "Wenkong Wang",
+      "Lingyao Li",
+      "Jiayan Zhou",
+      "Zhaoqian Xue",
+      "Xiang Li",
+      "Xinxin Lin",
+      "Zhiying Liang",
+      "Zhuoru Wu",
+      "Siyuan Ma",
+      "Xin Ma",
+      "Lizhou Fan"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.26205",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26205",
+    "pdf_url": "https://arxiv.org/pdf/2606.26205",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.25899",
+    "title": "Manipulation Is Task-Dependent: A Multi-Axis, Multi-Environment Evaluation of Frontier LLMs",
+    "url": "https://arxiv.org/abs/2606.25899",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Adeeb Zaman",
+      "Erik Nordby",
+      "Fred Heiding"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.25899",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25899",
+    "pdf_url": "https://arxiv.org/pdf/2606.25899",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.25484",
+    "title": "From Causal Discovery to Implementation: An Agentic AI Framework for E-Scooter Mobility Hub Planning Across 29 German Cities",
+    "url": "https://arxiv.org/abs/2606.25484",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Meng Jin",
+      "Melanie Handrich",
+      "Simone Martinenz",
+      "Nicholas Hoeser",
+      "Ziyue Li"
+    ],
+    "categories": [
+      "cs.CY",
+      "econ.GN",
+      "stat.AP"
+    ],
+    "topics": [
+      "coding-agent",
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.25484",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25484",
+    "pdf_url": "https://arxiv.org/pdf/2606.25484",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.26300",
+    "title": "The Verification Horizon: No Silver Bullet for Coding Agent Rewards",
+    "url": "https://arxiv.org/abs/2606.26300",
+    "published": "2026-06-24",
+    "updated": "2026-06-29",
+    "authors": [
+      "Binghai Wang",
+      "Chenlong Zhang",
+      "Dayiheng Liu",
+      "Jiajun Zhang",
+      "Jiawei Chen",
+      "Mingze Li",
+      "Mouxiang Chen",
+      "Rongyao Fang",
+      "Siyuan Zhang",
+      "Xuwu Wang",
+      "Yuheng Jing",
+      "Zeyao Ma",
+      "Zeyu Cui"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.26300",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26300",
+    "pdf_url": "https://arxiv.org/pdf/2606.26300",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.25705",
+    "title": "GUI agent: Guided Exploration of User-Sensitive Screens",
+    "url": "https://arxiv.org/abs/2606.25705",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Aradhana Nayak",
+      "Mussadiq Nazeer",
+      "Wang Peng",
+      "Feng Liu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.25705",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25705",
+    "pdf_url": "https://arxiv.org/pdf/2606.25705",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.25656",
+    "title": "Is GraphRAG Needed? From Basic RAG to Graph-/Agentic Solutions with Context Optimization",
+    "url": "https://arxiv.org/abs/2606.25656",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Long Chen",
+      "Ryan Razkenari",
+      "Yuxuan Zhou",
+      "Yuan Tian",
+      "Rahul Ghosh",
+      "Venkatesh Pappakrishnan",
+      "Disha Ahuja",
+      "Vidya Sagar Ravipati"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.25656",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25656",
+    "pdf_url": "https://arxiv.org/pdf/2606.25656",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.25189",
+    "title": "ActPlane: Programmable OS-Level Policy Enforcement for Agent Harnesses",
+    "url": "https://arxiv.org/abs/2606.25189",
+    "published": "2026-06-23",
+    "updated": "2026-06-30",
+    "authors": [
+      "Yusheng Zheng",
+      "Tianyuan Wu",
+      "Quanzhi Fu",
+      "Tong Yu",
+      "Wenan Mao",
+      "Tao Ma",
+      "Dan Williams",
+      "Wei Wang",
+      "Andi Quinn"
+    ],
+    "categories": [
+      "cs.OS"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "planning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.25189",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25189",
+    "pdf_url": "https://arxiv.org/pdf/2606.25189",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.24402",
+    "title": "Poisoned Playbooks: Demystifying Knowledge Poisoning Effects on AI Security Agents",
+    "url": "https://arxiv.org/abs/2606.24402",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Juho Park",
+      "Hyunmin Choi",
+      "Kevin Nam"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.24402",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24402",
+    "pdf_url": "https://arxiv.org/pdf/2606.24402",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.24235",
+    "title": "SP-Mind: An Autonomous Reasoning Agent for Spatial Proteomics Analysis",
+    "url": "https://arxiv.org/abs/2606.24235",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Yucheng Yuan",
+      "Yuanfeng Ji",
+      "Zhongxiao Li",
+      "Ruijiang Li"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.24235",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24235",
+    "pdf_url": "https://arxiv.org/pdf/2606.24235",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.25206",
+    "title": "RAVEN: Long-Horizon Reasoning & Navigation with a Visuo-Spatio-Temporal Memory",
+    "url": "https://arxiv.org/abs/2606.25206",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Yixun Hu",
+      "Zhicheng Zheng",
+      "Lihan Zha",
+      "Chunwei Xing",
+      "Rajdeep Singh",
+      "Omar Hossain",
+      "Antonio Loquercio",
+      "Dhruv Shah"
+    ],
+    "categories": [
+      "cs.RO",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.25206",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25206",
+    "pdf_url": "https://arxiv.org/pdf/2606.25206",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.24515",
+    "title": "Reinforcement Learning for Computer-Use Agents with Autonomous Evaluation",
+    "url": "https://arxiv.org/abs/2606.24515",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Marta Sumyk",
+      "Oleksandr Kosovan"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.24515",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24515",
+    "pdf_url": "https://arxiv.org/pdf/2606.24515",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.24694",
+    "title": "SupplyNet: Supporting Visual Exploratory Learning in Supply Chain via Contextual Multi-Agent Simulation",
+    "url": "https://arxiv.org/abs/2606.24694",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Yanjia Li",
+      "Kelcy Kexin Han",
+      "Tianrui Hu",
+      "Yi-Fan Cao",
+      "Huamin Qu",
+      "Sicheng Song"
+    ],
+    "categories": [
+      "cs.HC"
+    ],
+    "topics": [
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.24694",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24694",
+    "pdf_url": "https://arxiv.org/pdf/2606.24694",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.24976",
+    "title": "Diagnosing and Mitigating Compounding Failures in Agentic Persuasion via Taxonomic Strategy Retrieval",
+    "url": "https://arxiv.org/abs/2606.24976",
+    "published": "2026-06-23",
+    "updated": "2026-07-04",
+    "authors": [
+      "Sana Ayromlou",
+      "Purvi Sehgal",
+      "Pradyumna Narayana"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.24976",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24976",
+    "pdf_url": "https://arxiv.org/pdf/2606.24976",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.22737",
+    "title": "GroundEval: A Deterministic Replacement for LLM-as-Judge in Stateful Agent Evaluation",
+    "url": "https://arxiv.org/abs/2606.22737",
+    "published": "2026-06-22",
+    "updated": "2026-07-02",
+    "authors": [
+      "Jeffrey Flynt"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.22737",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22737",
+    "pdf_url": "https://arxiv.org/pdf/2606.22737",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.23283",
+    "title": "Towards Root Memories: Benchmarking and Enhancing Implicit Logical Memory Retrieval for Personalized LLMs",
+    "url": "https://arxiv.org/abs/2606.23283",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Hongxun Ding",
+      "Xiang Yu",
+      "Chengbing Wang",
+      "Jianfei Xiao",
+      "Keqin Bao",
+      "Wenjie Wang",
+      "Xiangnan He"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.23283",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23283",
+    "pdf_url": "https://arxiv.org/pdf/2606.23283",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.23195",
+    "title": "Memory Contagion: Cross-Temporal Propagation of Evaluator Bias via Agent Memory",
+    "url": "https://arxiv.org/abs/2606.23195",
+    "published": "2026-06-22",
+    "updated": "2026-06-24",
+    "authors": [
+      "Zewen Liu"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.23195",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23195",
+    "pdf_url": "https://arxiv.org/pdf/2606.23195",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.22864",
+    "title": "When AUC 0.998 Is Not Enough: A Candidate Evaluation Protocol for Hidden-State Probes of Indirect Prompt Injection in Multimodal Computer-Use Agents",
+    "url": "https://arxiv.org/abs/2606.22864",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Yanhang Li",
+      "Zhichao Fan",
+      "Zexin Zhuang"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.22864",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22864",
+    "pdf_url": "https://arxiv.org/pdf/2606.22864",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.22495",
+    "title": "Grounded Scaling: Why Agentic AI Needs Deterministic Environments",
+    "url": "https://arxiv.org/abs/2606.22495",
+    "published": "2026-06-21",
+    "updated": "2026-06-21",
+    "authors": [
+      "Liang Ding",
+      "Xintong Wang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "embodied-agent",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.22495",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22495",
+    "pdf_url": "https://arxiv.org/pdf/2606.22495",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.22484",
+    "title": "Governed AI-Assisted Engineering: Graduated Human Oversight for Agentic Code Generation in Regulated Domains",
+    "url": "https://arxiv.org/abs/2606.22484",
+    "published": "2026-06-21",
+    "updated": "2026-07-04",
+    "authors": [
+      "Richard Kang"
+    ],
+    "categories": [
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.22484",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22484",
+    "pdf_url": "https://arxiv.org/pdf/2606.22484",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.22030",
+    "title": "Nous: A Predictive World Model for Long-Term Agent Memory",
+    "url": "https://arxiv.org/abs/2606.22030",
+    "published": "2026-06-20",
+    "updated": "2026-06-20",
+    "authors": [
+      "Pranav Singh"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.IR",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.22030",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22030",
+    "pdf_url": "https://arxiv.org/pdf/2606.22030",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.22151",
+    "title": "Novelty-Aware Agentic Retrieval: Comparing Research Contributions Through Structured Multi-Step Reasoning",
+    "url": "https://arxiv.org/abs/2606.22151",
+    "published": "2026-06-20",
+    "updated": "2026-06-20",
+    "authors": [
+      "Shou-Tzu Han"
+    ],
+    "categories": [
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.22151",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22151",
+    "pdf_url": "https://arxiv.org/pdf/2606.22151",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.21842",
+    "title": "Agent-Assisted Side-Channel Attacks on Non-Prefix KV Cache in RAG",
+    "url": "https://arxiv.org/abs/2606.21842",
+    "published": "2026-06-20",
+    "updated": "2026-06-20",
+    "authors": [
+      "He Sun",
+      "Shinan Liu",
+      "Siyuan Ma",
+      "Junhao Li",
+      "Mingjun Xiao",
+      "Wenhao Jiang"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.21842",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21842",
+    "pdf_url": "https://arxiv.org/pdf/2606.21842",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.21409",
+    "title": "Don't Blindly Trust It: How Unreliable Feedback Breaks Tool-Using LLM Agents",
+    "url": "https://arxiv.org/abs/2606.21409",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Chubin Zhang",
+      "Zhenglin Wan",
+      "Xingrui Yu",
+      "Pengfei Zhou",
+      "Wangbo Zhao",
+      "Jingxuan Wu",
+      "Yaxin Zhou",
+      "Ivor Tsang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.21409",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21409",
+    "pdf_url": "https://arxiv.org/pdf/2606.21409",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.21553",
+    "title": "Dissecting Agentic RAG: A Component Ablation for Multi-Hop QA with a Local 7B Model",
+    "url": "https://arxiv.org/abs/2606.21553",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Sheroz Shaikh"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.21553",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21553",
+    "pdf_url": "https://arxiv.org/pdf/2606.21553",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.20470",
+    "title": "Analyzing Defensive Misdirection Against Model-Guided Automated Attacks on Agentic AI Systems",
+    "url": "https://arxiv.org/abs/2606.20470",
+    "published": "2026-06-18",
+    "updated": "2026-06-26",
+    "authors": [
+      "Reza Soosahabi",
+      "Vivek Namsani"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.20470",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20470",
+    "pdf_url": "https://arxiv.org/pdf/2606.20470",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.19812",
+    "title": "Human-on-the-Loop Orchestration for AI-Assisted Legal Discovery",
+    "url": "https://arxiv.org/abs/2606.19812",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Anushree Sinha",
+      "Srivaths Ranganathan",
+      "Abhishek Dharmaratnakar",
+      "Debanshu Das"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag",
+      "reasoning",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.19812",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19812",
+    "pdf_url": "https://arxiv.org/pdf/2606.19812",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.20515",
+    "title": "S-Agent: Spatial Tool-Use Elicits Reasoning for Spatial Intelligence",
+    "url": "https://arxiv.org/abs/2606.20515",
+    "published": "2026-06-18",
+    "updated": "2026-06-28",
+    "authors": [
+      "Yalun Dai",
+      "Hao Li",
+      "Shulin Tian",
+      "Runmao Yao",
+      "Yuhao Dong",
+      "Fangzhou Hong",
+      "Zhaoxi Chen",
+      "Fangfu Liu",
+      "Baoliang Tian",
+      "Dingwen Zhang",
+      "Tao Wang",
+      "Kim-Hui Yap",
+      "Ziwei Liu"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.20515",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20515",
+    "pdf_url": "https://arxiv.org/pdf/2606.20515",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.20023",
+    "title": "When Lower Privileges Suffice: Investigating Over-Privileged Tool Selection in LLM Agents",
+    "url": "https://arxiv.org/abs/2606.20023",
+    "published": "2026-06-18",
+    "updated": "2026-07-07",
+    "authors": [
+      "Kaiyue Yang",
+      "Yuyan Bu",
+      "Jingwei Yi",
+      "Yuchi Wang",
+      "Biyu Zhou",
+      "Juntao Dai",
+      "Songlin Hu",
+      "Yaodong Yang"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.20023",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20023",
+    "pdf_url": "https://arxiv.org/pdf/2606.20023",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.20785",
+    "title": "Fara-1.5: Scalable Learning Environments for Computer Use Agents",
+    "url": "https://arxiv.org/abs/2606.20785",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Ahmed Awadallah",
+      "Sahil Gupta",
+      "Yash Lara",
+      "Yadong Lu",
+      "Hussein Mozannar",
+      "Akshay Nambi",
+      "Zach Nussbaum",
+      "Yash Pandya",
+      "Aravind Rajeswaran",
+      "Corby Rosset",
+      "Alexey Taymanov",
+      "Luiz do Valle",
+      "Vibhav Vineet",
+      "Spencer Whitehead",
+      "Andrew Zhao"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.20785",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20785",
+    "pdf_url": "https://arxiv.org/pdf/2606.20785",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.19930",
+    "title": "MobileForge: Annotation-Free Adaptation for Mobile GUI Agents with Hierarchical Feedback-Guided Policy Optimization",
+    "url": "https://arxiv.org/abs/2606.19930",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Guangyi Liu",
+      "Pengxiang Zhao",
+      "Gao Wu",
+      "Yiwen Yin",
+      "Mading Li",
+      "Liang Liu",
+      "Congxiao Liu",
+      "Zhang Qi",
+      "Mengyan Wang",
+      "Liang Guo",
+      "Yong Liu"
+    ],
+    "categories": [
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.19930",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19930",
+    "pdf_url": "https://arxiv.org/pdf/2606.19930",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.18671",
+    "title": "HANSEL: Extracting Breadcrumbs from Web Agent Trajectories for Interactive Verification",
+    "url": "https://arxiv.org/abs/2606.18671",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Yujin Zhang",
+      "Daye Nam"
+    ],
+    "categories": [
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "embodied-agent",
+      "planning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.18671",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18671",
+    "pdf_url": "https://arxiv.org/pdf/2606.18671",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.19063",
+    "title": "PYPILINE: Malicious PyPI Package Detection via Suspicious API Knowledge and Agent Workflow",
+    "url": "https://arxiv.org/abs/2606.19063",
+    "published": "2026-06-17",
+    "updated": "2026-06-26",
+    "authors": [
+      "Siyuan Pang",
+      "Yepeng Yao",
+      "Zhengwei Jiang",
+      "Zijing Fan",
+      "Haozhe Li",
+      "Baoxu Liu"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.19063",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19063",
+    "pdf_url": "https://arxiv.org/pdf/2606.19063",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.19409",
+    "title": "OpenRath: Session-Centered Runtime State for Agent Systems",
+    "url": "https://arxiv.org/abs/2606.19409",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Fukang Wen",
+      "Zhijie Wang",
+      "Ruilin Xu"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.PL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.19409",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19409",
+    "pdf_url": "https://arxiv.org/pdf/2606.19409",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.19613",
+    "title": "StaminaBench: Stress-Testing Coding Agents over 100 Interaction Turns",
+    "url": "https://arxiv.org/abs/2606.19613",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Vlad Sobal",
+      "Shuo Yang",
+      "Yuting Zhang",
+      "Wei Xia",
+      "Stefano Soatto"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.19613",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19613",
+    "pdf_url": "https://arxiv.org/pdf/2606.19613",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.20717",
+    "title": "MIRAGE: Stealthy Visual Prompt Injection for Vulnerability Detection in Web Agents",
+    "url": "https://arxiv.org/abs/2606.20717",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Xuelong Dai",
+      "Jianyu Ma",
+      "Boyang Ma",
+      "Biwei Yan",
+      "Yijun Yang",
+      "Yue Zhang"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.20717",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20717",
+    "pdf_url": "https://arxiv.org/pdf/2606.20717",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.16111",
+    "title": "Towards Pareto-Optimal Tool-Integrated Agents with Pareto Ranking Policy Optimization",
+    "url": "https://arxiv.org/abs/2606.16111",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Junyi Li",
+      "Xiaowei Qian",
+      "Yingyi Zhang",
+      "Wenlin Zhang",
+      "Guojing Li",
+      "Sheng Zhang",
+      "Xiao Han",
+      "Yichao Wang",
+      "Xiangyu Zhao"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.16111",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16111",
+    "pdf_url": "https://arxiv.org/pdf/2606.16111",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.16748",
+    "title": "MyPCBench: A Benchmark for Personally Intelligent Computer-Use Agents",
+    "url": "https://arxiv.org/abs/2606.16748",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Lawrence Keunho Jang",
+      "Andrew Keunwoo Jang",
+      "Jing Yu Koh",
+      "Ruslan Salakhutdinov"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.16748",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16748",
+    "pdf_url": "https://arxiv.org/pdf/2606.16748",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.15591",
+    "title": "Agentic Retrieval and Reinforcement Learned Equation Chains: A Controlled Generation Framework for Complex and Novel Physics Word Problems",
+    "url": "https://arxiv.org/abs/2606.15591",
+    "published": "2026-06-14",
+    "updated": "2026-06-14",
+    "authors": [
+      "Tirthankar Mittra"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.15591",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15591",
+    "pdf_url": "https://arxiv.org/pdf/2606.15591",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.15152",
+    "title": "Can Agents Read the Room? Benchmarking Visual Social Intelligence in Multimodal Simulation",
+    "url": "https://arxiv.org/abs/2606.15152",
+    "published": "2026-06-13",
+    "updated": "2026-06-13",
+    "authors": [
+      "Shijun Wan",
+      "Xuehai Wu",
+      "Jiwen Zhang",
+      "Siyuan Wang",
+      "Zhongyu Wei"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.15152",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15152",
+    "pdf_url": "https://arxiv.org/pdf/2606.15152",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.15242",
+    "title": "Benign in Isolation, Harmful in Composition: Security Risks in Agent Skill Ecosystems",
+    "url": "https://arxiv.org/abs/2606.15242",
+    "published": "2026-06-13",
+    "updated": "2026-06-13",
+    "authors": [
+      "Yi Xie",
+      "Jiawei Du",
+      "Yu Cheng",
+      "Jiuan Zhou",
+      "Zhaoxia Yin"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.15242",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15242",
+    "pdf_url": "https://arxiv.org/pdf/2606.15242",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.14502",
+    "title": "From Chatbot to Digital Colleague: The Paradigm Shift Toward Persistent Autonomous AI",
+    "url": "https://arxiv.org/abs/2606.14502",
+    "published": "2026-06-12",
+    "updated": "2026-06-12",
+    "authors": [
+      "Yongheng Zhang",
+      "Ziang Liu",
+      "Jiaxuan Zhu",
+      "Shuai Wang",
+      "Xiangqi Chen",
+      "Haojing Huang",
+      "Jiayi Kuang",
+      "Siyu Chen",
+      "Ao Shen",
+      "Hao Wu",
+      "Qiufeng Wang",
+      "Qian-Wen Zhang",
+      "Junnan Dong",
+      "Wenhao Jiang",
+      "Ying Shen",
+      "Hai-Tao Zheng",
+      "Yinghui Li",
+      "Di Yin",
+      "Xing Sun",
+      "Philip S. Yu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.14502",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.14502",
+    "pdf_url": "https://arxiv.org/pdf/2606.14502",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.15017",
+    "title": "Are Online Skill and Memory Modules Always Worth Their Tokens? A Budget-Constrained Study of Web Agents",
+    "url": "https://arxiv.org/abs/2606.15017",
+    "published": "2026-06-12",
+    "updated": "2026-06-12",
+    "authors": [
+      "Sina Hajimiri",
+      "Masih Aminbeidokhti",
+      "Jose Dolz",
+      "Ismail Ben Ayed",
+      "Issam H. Laradji",
+      "Spandana Gella",
+      "Nicolas Gontier"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.15017",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15017",
+    "pdf_url": "https://arxiv.org/pdf/2606.15017",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.14574",
+    "title": "SIMMER: Benchmarking Latent Failures in LLM Executable Planning with a World Model",
+    "url": "https://arxiv.org/abs/2606.14574",
+    "published": "2026-06-12",
+    "updated": "2026-06-12",
+    "authors": [
+      "Xiaoxin Lu",
+      "Ranran Haoran Zhang",
+      "Rui Zhang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.14574",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.14574",
+    "pdf_url": "https://arxiv.org/pdf/2606.14574",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.13317",
+    "title": "SkillCAT: Contrastive Assessment and Topology-Aware Skill Self-Evolution for LLM Agents",
+    "url": "https://arxiv.org/abs/2606.13317",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Kunfeng Chen",
+      "Qihuang Zhong",
+      "Juhua Liu",
+      "Bo Du"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.13317",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13317",
+    "pdf_url": "https://arxiv.org/pdf/2606.13317",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.14805",
+    "title": "Knowledge-Based Zero-Replay Debugging of Multi-Agent LLM Traces",
+    "url": "https://arxiv.org/abs/2606.14805",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Dong Ho Kang",
+      "Hyeonjeong Cha",
+      "Daein Weon"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "memory",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.14805",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.14805",
+    "pdf_url": "https://arxiv.org/pdf/2606.14805",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.13602",
+    "title": "EpiBench: Verifiable Evaluation of AI Agents on Epigenomics Analysis",
+    "url": "https://arxiv.org/abs/2606.13602",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Harihara Muralidharan",
+      "Reema Baskar",
+      "Soo Hee Lee",
+      "Tim Proctor",
+      "Kenny Workman"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.13602",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13602",
+    "pdf_url": "https://arxiv.org/pdf/2606.13602",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.13192",
+    "title": "Reasoning for Mobile User Experience with Multimodal LLMs: Task, Benchmark, and Approach",
+    "url": "https://arxiv.org/abs/2606.13192",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Ruichao Mao",
+      "Zhou Fang",
+      "Teng Guo",
+      "Hao Yang",
+      "Yaping Li",
+      "Shaohua Peng",
+      "Maji Huang",
+      "Xiaoyu Lin",
+      "Shuoyang Liu",
+      "Xuepeng Li",
+      "Yuyu Zhang",
+      "Hai Rao"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.13192",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13192",
+    "pdf_url": "https://arxiv.org/pdf/2606.13192",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.12674",
+    "title": "Evoflux: Inference-Time Evolution of Executable Tool Workflows for Compact Agents",
+    "url": "https://arxiv.org/abs/2606.12674",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Kushal Raj Bhandari",
+      "Ling Yue",
+      "Ching-Yun Ko",
+      "Dhaval Patel",
+      "Shaowu Pan",
+      "Pin-Yu Chen",
+      "Jianxi Gao"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.12674",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12674",
+    "pdf_url": "https://arxiv.org/pdf/2606.12674",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2606.12384",
+    "title": "APPO: Agentic Procedural Policy Optimization",
+    "url": "https://arxiv.org/abs/2606.12384",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Xucong Wang",
+      "Ziyu Ma",
+      "Yong Wang",
+      "Yuxiang Ji",
+      "Shidong Yang",
+      "Guanhua Chen",
+      "Pengkun Wang",
+      "Xiangxiang Chu"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.12384",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12384",
+    "pdf_url": "https://arxiv.org/pdf/2606.12384",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.12563",
+    "title": "Arbor: Tree Search as a Cognition Layer for Autonomous Agents",
+    "url": "https://arxiv.org/abs/2606.12563",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Neha Prakriya",
+      "Chaojun Hou",
+      "Zheng Gong",
+      "Huasha Zhao",
+      "Xi Zhao",
+      "Mou Li",
+      "Zhenyu Gu",
+      "Emad Barsoum"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.12563",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12563",
+    "pdf_url": "https://arxiv.org/pdf/2606.12563",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.11079",
+    "title": "VISTA: A Versatile Interactive User Simulation Toolkit for Agent Evaluation",
+    "url": "https://arxiv.org/abs/2606.11079",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Yunan Lu",
+      "Ryan Shea",
+      "Yusen Zhang",
+      "Zhou Yu"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.11079",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.11079",
+    "pdf_url": "https://arxiv.org/pdf/2606.11079",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.10742",
+    "title": "MemVenom: Triggered Poisoning of Multimodal Memories in Web Agents",
+    "url": "https://arxiv.org/abs/2606.10742",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Yv Zhang",
+      "Hao Sun",
+      "Hao Fang",
+      "Kuofeng Gao",
+      "Fan Mo",
+      "Bin Chen",
+      "Shu-Tao Xia",
+      "Yaowei Wang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.10742",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10742",
+    "pdf_url": "https://arxiv.org/pdf/2606.10742",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.09774",
+    "title": "Auto-Configuring Scientific Simulators with Lightweight Coding-Agent Adapters",
+    "url": "https://arxiv.org/abs/2606.09774",
+    "published": "2026-06-08",
+    "updated": "2026-06-25",
+    "authors": [
+      "Matthew Ho",
+      "Brian Liu",
+      "Jixuan Chen",
+      "Audrey Wang",
+      "Lianhui Qin"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "memory",
+      "rag",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.09774",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09774",
+    "pdf_url": "https://arxiv.org/pdf/2606.09774",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.09198",
+    "title": "MASS: Deep Research for Social Sciences with Memory-Augmented Social Simulation",
+    "url": "https://arxiv.org/abs/2606.09198",
+    "published": "2026-06-08",
+    "updated": "2026-06-08",
+    "authors": [
+      "Yongrui Liu",
+      "Deyi Xiong"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "rag",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.09198",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09198",
+    "pdf_url": "https://arxiv.org/pdf/2606.09198",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.08790",
+    "title": "RAILS: Verification-Native Clearing For Agentic Commerce",
+    "url": "https://arxiv.org/abs/2606.08790",
+    "published": "2026-06-07",
+    "updated": "2026-06-07",
+    "authors": [
+      "Adrian de Valois-Franklin",
+      "Alex Bogdan"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CR",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.08790",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.08790",
+    "pdf_url": "https://arxiv.org/pdf/2606.08790",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.08625",
+    "title": "From Holistic Evaluation to Structured Criteria: Rubrics Across the Evolving LLM Landscape",
+    "url": "https://arxiv.org/abs/2606.08625",
+    "published": "2026-06-07",
+    "updated": "2026-07-01",
+    "authors": [
+      "Hao Chen",
+      "Ziyu Han",
+      "Yukun Yan",
+      "Qingfu Zhu",
+      "Maosong Sun",
+      "Wanxiang Che"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.08625",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.08625",
+    "pdf_url": "https://arxiv.org/pdf/2606.08625",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.07379",
+    "title": "Do Coding Agents Deceive Us? Detecting and Preventing Cheating via Capped Evaluation with Randomized Tests",
+    "url": "https://arxiv.org/abs/2606.07379",
+    "published": "2026-06-05",
+    "updated": "2026-06-08",
+    "authors": [
+      "Thanawat Lodkaew",
+      "Johannes Ackermann",
+      "Soichiro Nishimori",
+      "Nontawat Charoenphakdee",
+      "Masashi Sugiyama",
+      "Takashi Ishida"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL",
+      "stat.ME"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.07379",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.07379",
+    "pdf_url": "https://arxiv.org/pdf/2606.07379",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.05548",
+    "title": "ADK Arena: Evaluating Agent Development Kits via LLM-as-a-Developer",
+    "url": "https://arxiv.org/abs/2606.05548",
+    "published": "2026-06-04",
+    "updated": "2026-06-04",
+    "authors": [
+      "Jintao Huang",
+      "Xiaomin Li",
+      "Gaurav Mittal",
+      "Yu Hu"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.05548",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.05548",
+    "pdf_url": "https://arxiv.org/pdf/2606.05548",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.06473",
+    "title": "MLEvolve: A Self-Evolving Framework for Automated Machine Learning Algorithm Discovery",
+    "url": "https://arxiv.org/abs/2606.06473",
+    "published": "2026-06-04",
+    "updated": "2026-06-04",
+    "authors": [
+      "Shangheng Du",
+      "Xiangchao Yan",
+      "Jinxin Shi",
+      "Zongsheng Cao",
+      "Shiyang Feng",
+      "Zichen Liang",
+      "Boyuan Sun",
+      "Tianshuo Peng",
+      "Yifan Zhou",
+      "Xin Li",
+      "Jie Zhou",
+      "Liang He",
+      "Bo Zhang",
+      "Lei Bai"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.06473",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.06473",
+    "pdf_url": "https://arxiv.org/pdf/2606.06473",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.06462",
+    "title": "Benchmark Everything Everywhere All at Once",
+    "url": "https://arxiv.org/abs/2606.06462",
+    "published": "2026-06-04",
+    "updated": "2026-06-04",
+    "authors": [
+      "Shiyun Xiong",
+      "Dongming Wu",
+      "Peiwen Sun",
+      "Yuang Ai",
+      "Bokang Yang",
+      "Wencheng Han",
+      "Xiao-Hui Li",
+      "Xiangyu Yue"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.06462",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.06462",
+    "pdf_url": "https://arxiv.org/pdf/2606.06462",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.05263",
+    "title": "Policy-Conditioned Counterfactual Credit for Verifiable Reinforcement Learning of Long-Horizon Language Agents",
+    "url": "https://arxiv.org/abs/2606.05263",
+    "published": "2026-06-03",
+    "updated": "2026-06-03",
+    "authors": [
+      "Renwei Meng"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.05263",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.05263",
+    "pdf_url": "https://arxiv.org/pdf/2606.05263",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.04628",
+    "title": "RAMPART: Registry-based Agentic Memory with Priority-Aware Runtime Transformation",
+    "url": "https://arxiv.org/abs/2606.04628",
+    "published": "2026-06-03",
+    "updated": "2026-06-03",
+    "authors": [
+      "Nikodem Tomczak"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.MA"
+    ],
+    "topics": [
+      "memory"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.04628",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.04628",
+    "pdf_url": "https://arxiv.org/pdf/2606.04628",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.05414",
+    "title": "When Evidence is Sparse: Weakly Supervised Early Failure Alerting in Dialogs and LLM-Agent Trajectories",
+    "url": "https://arxiv.org/abs/2606.05414",
+    "published": "2026-06-03",
+    "updated": "2026-06-03",
+    "authors": [
+      "Avinash Baidya",
+      "Xinran Liang",
+      "Ruocheng Guo",
+      "Xiang Gao",
+      "Kamalika Das"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.HC",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.05414",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.05414",
+    "pdf_url": "https://arxiv.org/pdf/2606.05414",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.03329",
+    "title": "InfoMem: Training Long-Context Memory Agents with Answer-Conditioned Information Gain",
+    "url": "https://arxiv.org/abs/2606.03329",
+    "published": "2026-06-02",
+    "updated": "2026-06-02",
+    "authors": [
+      "Tiancheng Han",
+      "Yong Li",
+      "Wuzhou Yu",
+      "Qiaosheng Zhang",
+      "Wenqi Shao"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.03329",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.03329",
+    "pdf_url": "https://arxiv.org/pdf/2606.03329",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.02965",
+    "title": "What Benchmarks Don't Measure: The Case for Evaluating Abstention Competence in Autonomous Agents",
+    "url": "https://arxiv.org/abs/2606.02965",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Victor Ojewale",
+      "Suresh Venkatasubramanian"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.02965",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.02965",
+    "pdf_url": "https://arxiv.org/pdf/2606.02965",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.02404",
+    "title": "K-BrowseComp: A Web Browsing Agent Benchmark Grounded in Korean Contexts",
+    "url": "https://arxiv.org/abs/2606.02404",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Nahyun Lee",
+      "Dongkeun Yoon",
+      "Guijin Son",
+      "Geewook Kim",
+      "Dayoon Ko",
+      "Jeonghun Park",
+      "Haneul Yoo",
+      "Jaewon Cho",
+      "Junghun Park",
+      "Changyoon Lee",
+      "Kyochul Jang",
+      "Jaeyeon Kim",
+      "Eunsu Kim",
+      "Woojin Cho",
+      "Seungone Kim"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.02404",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.02404",
+    "pdf_url": "https://arxiv.org/pdf/2606.02404",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.09863",
+    "title": "From Confident Closing to Silent Failure: Characterizing False Success in LLM Agents",
+    "url": "https://arxiv.org/abs/2606.09863",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Laksh Advani"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.09863",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09863",
+    "pdf_url": "https://arxiv.org/pdf/2606.09863",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.00611",
+    "title": "TRACE: Trajectory Risk-Aware Compression for Long-Horizon Agent Safety",
+    "url": "https://arxiv.org/abs/2606.00611",
+    "published": "2026-05-30",
+    "updated": "2026-05-30",
+    "authors": [
+      "Zhepei Hong",
+      "Lin Wang",
+      "Liting Li",
+      "Haokai Ma",
+      "Junfeng Fang",
+      "Fei Shen",
+      "Dan Zhang",
+      "Xiang Wang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2606.00611",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.00611",
+    "pdf_url": "https://arxiv.org/pdf/2606.00611",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.00198",
+    "title": "BAGEN: Are LLM Agents Budget-Aware?",
+    "url": "https://arxiv.org/abs/2606.00198",
+    "published": "2026-05-29",
+    "updated": "2026-05-29",
+    "authors": [
+      "Yuxiang Lin",
+      "Zihan Wang",
+      "Mengyang Liu",
+      "Yuxuan Shan",
+      "Longju Bai",
+      "Junyao Zhang",
+      "Xing Jin",
+      "Boshan Chen",
+      "Jinyan Su",
+      "Xingyao Wang",
+      "Jiaxin Pei",
+      "Manling Li"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.00198",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.00198",
+    "pdf_url": "https://arxiv.org/pdf/2606.00198",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.29790",
+    "title": "Evolve as a Team: Collaborative Self-Evolution for LLM-based Multi-Agent Systems",
+    "url": "https://arxiv.org/abs/2605.29790",
+    "published": "2026-05-28",
+    "updated": "2026-05-28",
+    "authors": [
+      "Zhezheng Hao",
+      "Tianfu Wang",
+      "Huanshuo Dong",
+      "Ziyan Liu",
+      "Hong Wang",
+      "Xiankun Lin",
+      "Qiang Lin",
+      "Can Wang",
+      "Hande Dong",
+      "Jiawei Chen"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2605.29790",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.29790",
+    "pdf_url": "https://arxiv.org/pdf/2605.29790",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2605.29640",
+    "title": "VikingMem: A Memory Base Management System for Stateful LLM-based Applications",
+    "url": "https://arxiv.org/abs/2605.29640",
+    "published": "2026-05-28",
+    "updated": "2026-06-12",
+    "authors": [
+      "Jiajie Fu",
+      "Junwen Chen",
+      "Mengzhao Wang",
+      "Aoxiang He",
+      "Maojia Sheng",
+      "Xiangyu Ke",
+      "Yifan Zhu",
+      "Yunjun Gao"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.29640",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.29640",
+    "pdf_url": "https://arxiv.org/pdf/2605.29640",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.29801",
+    "title": "AgentDoG 1.5: A Lightweight and Scalable Alignment Framework for AI Agent Safety and Security",
+    "url": "https://arxiv.org/abs/2605.29801",
+    "published": "2026-05-28",
+    "updated": "2026-05-28",
+    "authors": [
+      "Dongrui Liu",
+      "Yu Li",
+      "Zhonghao Yang",
+      "Peng Wang",
+      "Guanxu Chen",
+      "Yuejin Xie",
+      "Qinghua Mao",
+      "Wanying Qu",
+      "Yanxu Zhu",
+      "Tianyi Zhou",
+      "Leitao Yuan",
+      "Zhijie Zheng",
+      "Qihao Lin",
+      "Yimin Wang",
+      "Haoyu Luo",
+      "Shuai Shao",
+      "Chen Qian",
+      "Qingyu Liu",
+      "Ling Tang",
+      "Ruiyang Qin",
+      "Qihan Ren",
+      "Junxiao Yang",
+      "Kun Wang",
+      "Zhiheng Xi",
+      "Linfeng Zhang",
+      "Ranjie Duan",
+      "Bo Zhang",
+      "Wenjie Wang",
+      "Wen Shen",
+      "Qiaosheng Zhang",
+      "Yan Teng",
+      "Chaochao Lu",
+      "Rui Mei",
+      "Man Li",
+      "Jialing Tao",
+      "Xi Lin",
+      "Tianhang Zheng",
+      "Yong Liu",
+      "Quanshi Zhang",
+      "Lei Zhu",
+      "Xingjun Ma",
+      "Junhua Liu",
+      "Hui Xue",
+      "Xiaoxiang Zuo",
+      "Xiangnan He",
+      "Chao Shen",
+      "Xianglong Liu",
+      "Minlie Huang",
+      "Jing Shao",
+      "Xia Hu"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.CR",
+      "cs.CV",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.29801",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.29801",
+    "pdf_url": "https://arxiv.org/pdf/2605.29801",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.30407",
+    "title": "Exploring Autonomous Agentic Data Engineering for Model Specialization",
+    "url": "https://arxiv.org/abs/2605.30407",
+    "published": "2026-05-28",
+    "updated": "2026-06-08",
+    "authors": [
+      "Yujie Luo",
+      "Xiangyuan Ru",
+      "Jingsheng Zheng",
+      "Jingjing Wang",
+      "Yuqi Zhu",
+      "Jintian Zhang",
+      "Runnan Fang",
+      "Kewei Xu",
+      "Ye Liu",
+      "Zheng Wei",
+      "Jiang Bian",
+      "Zang Li",
+      "Shumin Deng"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.IR",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.30407",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.30407",
+    "pdf_url": "https://arxiv.org/pdf/2605.30407",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.27825",
+    "title": "MRMMIA: Membership Inference Attacks on Memory in Chat Agents",
+    "url": "https://arxiv.org/abs/2605.27825",
+    "published": "2026-05-27",
+    "updated": "2026-05-27",
+    "authors": [
+      "Kai Chen",
+      "Yan Pang",
+      "Tianhao Wang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.27825",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.27825",
+    "pdf_url": "https://arxiv.org/pdf/2605.27825",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.28175",
+    "title": "Mixture-of-Experts Knowledge Graph Retrieval-Augmented Generation for Multi-Agent LLM-based Recommendation",
+    "url": "https://arxiv.org/abs/2605.28175",
+    "published": "2026-05-27",
+    "updated": "2026-05-29",
+    "authors": [
+      "Shijie Wang",
+      "Chengyi Liu",
+      "Yujuan Ding",
+      "Shanru Lin",
+      "See-Kiong Ng",
+      "Xu Xin",
+      "Wenqi Fan"
+    ],
+    "categories": [
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.28175",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.28175",
+    "pdf_url": "https://arxiv.org/pdf/2605.28175",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.27935",
+    "title": "Do Agents Think Deeper? A Mechanistic Investigation of Layer-Wise Dynamics in Sequential Planning",
+    "url": "https://arxiv.org/abs/2605.27935",
+    "published": "2026-05-27",
+    "updated": "2026-05-27",
+    "authors": [
+      "Zhenyu Cui",
+      "Xiangzhong Luo"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm",
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.27935",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.27935",
+    "pdf_url": "https://arxiv.org/pdf/2605.27935",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.25920",
+    "title": "Can LLMs Time Travel? Enhancing Temporal Consistency in Legal Agentic Search through Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2605.25920",
+    "published": "2026-05-25",
+    "updated": "2026-05-25",
+    "authors": [
+      "Wei Fan",
+      "Yining Zhou",
+      "Mufan Zhang",
+      "Yanbing Weng",
+      "Yiran HU",
+      "Tianshi Zheng",
+      "Baixuan Xu",
+      "Chunyang Li",
+      "Jianhui Yang",
+      "Haoran Li",
+      "Yangqiu Song"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.25920",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.25920",
+    "pdf_url": "https://arxiv.org/pdf/2605.25920",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.25393",
+    "title": "Decision-Making with Lightweight Confidence-Aware Language Model for Autonomous Driving",
+    "url": "https://arxiv.org/abs/2605.25393",
+    "published": "2026-05-25",
+    "updated": "2026-05-25",
+    "authors": [
+      "Ruoyu Yao",
+      "Ruiguo Zhong",
+      "Pei Liu",
+      "Mingxing Peng",
+      "Rui Yang",
+      "Jun Ma"
+    ],
+    "categories": [
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.25393",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.25393",
+    "pdf_url": "https://arxiv.org/pdf/2605.25393",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.25310",
+    "title": "Tool-Call Dependency Structure is Linearly Decodable in LLM Agent Residual Streams",
+    "url": "https://arxiv.org/abs/2605.25310",
+    "published": "2026-05-25",
+    "updated": "2026-05-25",
+    "authors": [
+      "Tianda Sun",
+      "Dimitar Kazakov"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.25310",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.25310",
+    "pdf_url": "https://arxiv.org/pdf/2605.25310",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.24309",
+    "title": "Reframing LLM Agent Security as an Agent-Human Interaction Problem",
+    "url": "https://arxiv.org/abs/2605.24309",
+    "published": "2026-05-23",
+    "updated": "2026-05-23",
+    "authors": [
+      "Peiran Wang",
+      "Ying Li",
+      "Yuan Tian"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.24309",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.24309",
+    "pdf_url": "https://arxiv.org/pdf/2605.24309",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.19604",
+    "title": "Formal Skill: Programmable Runtime Skills for Efficient and Accurate LLM Agents",
+    "url": "https://arxiv.org/abs/2605.19604",
+    "published": "2026-05-19",
+    "updated": "2026-05-19",
+    "authors": [
+      "Xi Zhang",
+      "Meijun Gao",
+      "Yuntian Zhao",
+      "Xinyu Tan",
+      "Yilun Yao",
+      "Feiyu Wang",
+      "Yanshu Wang",
+      "Dingsiyi",
+      "Tong Yang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2605.19604",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.19604",
+    "pdf_url": "https://arxiv.org/pdf/2605.19604",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2605.18502",
+    "title": "The distance-based formation controller design for multi-agent systems in port-Hamiltonian form",
+    "url": "https://arxiv.org/abs/2605.18502",
+    "published": "2026-05-18",
+    "updated": "2026-05-18",
+    "authors": [
+      "Jingyi Zhao",
+      "Yongxin Wu",
+      "Héctor García de Marina",
+      "Yuhu Wu",
+      "Yann Le Gorrec"
+    ],
+    "categories": [
+      "math.OC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.18502",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.18502",
+    "pdf_url": "https://arxiv.org/pdf/2605.18502",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.15701",
+    "title": "H-Mem: A Novel Memory Mechanism for Evolving and Retrieving Agent Memory via a Hybrid Structure",
+    "url": "https://arxiv.org/abs/2605.15701",
+    "published": "2026-05-15",
+    "updated": "2026-05-15",
+    "authors": [
+      "Jiawei Yu",
+      "Yixiang Fang",
+      "Xilin Liu",
+      "Yuchi Ma"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.15701",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.15701",
+    "pdf_url": "https://arxiv.org/pdf/2605.15701",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.15625",
+    "title": "ColPackAgent: Agent-Skill-Guided Hard-Particle Monte Carlo Workflows for Colloidal Packing",
+    "url": "https://arxiv.org/abs/2605.15625",
+    "published": "2026-05-15",
+    "updated": "2026-05-15",
+    "authors": [
+      "Lijie Ding",
+      "Changwoo Do"
+    ],
+    "categories": [
+      "cs.AI",
+      "cond-mat.soft"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.15625",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.15625",
+    "pdf_url": "https://arxiv.org/pdf/2605.15625",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.14290",
+    "title": "Web Agents Should Adopt the Plan-Then-Execute Paradigm",
+    "url": "https://arxiv.org/abs/2605.14290",
+    "published": "2026-05-14",
+    "updated": "2026-05-14",
+    "authors": [
+      "Julien Piet",
+      "Annabella Chow",
+      "Yiwei Hou",
+      "Muxi Lyu",
+      "Sylvie Venuto",
+      "Jinhao Zhu",
+      "Raluca Ada Popa",
+      "David Wagner"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.CL",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.14290",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.14290",
+    "pdf_url": "https://arxiv.org/pdf/2605.14290",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.13716",
+    "title": "SkillOps: Managing LLM Agent Skill Libraries as Self-Maintaining Software Ecosystems",
+    "url": "https://arxiv.org/abs/2605.13716",
+    "published": "2026-05-13",
+    "updated": "2026-05-13",
+    "authors": [
+      "Hongji Pu",
+      "Xinyuan Song",
+      "Liang Zhao"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.13716",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.13716",
+    "pdf_url": "https://arxiv.org/pdf/2605.13716",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.13618",
+    "title": "OpenAaaS: An Open Agent-as-a-Service Framework for Distributed Materials-Informatics Research",
+    "url": "https://arxiv.org/abs/2605.13618",
+    "published": "2026-05-13",
+    "updated": "2026-05-13",
+    "authors": [
+      "Peng Kang",
+      "Bixuan Li",
+      "Xiaoya Huang",
+      "Shuo Shi",
+      "Weiqiao Zhou",
+      "Zhen Li",
+      "Yu Liu",
+      "Lei Zheng"
+    ],
+    "categories": [
+      "cond-mat.mtrl-sci",
+      "cs.AI"
+    ],
+    "topics": [
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.13618",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.13618",
+    "pdf_url": "https://arxiv.org/pdf/2605.13618",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.07112",
+    "title": "Switchcraft: AI Model Router for Agentic Tool Calling",
+    "url": "https://arxiv.org/abs/2605.07112",
+    "published": "2026-05-08",
+    "updated": "2026-05-08",
+    "authors": [
+      "Sharad Agarwal",
+      "Pooria Namyar",
+      "Alec Wolman",
+      "Rahul Ambavat",
+      "Ankur Gupta",
+      "Qizheng Zhang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2605.07112",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.07112",
+    "pdf_url": "https://arxiv.org/pdf/2605.07112",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2605.06992",
+    "title": "Why Does Agentic Safety Fail to Generalize Across Tasks?",
+    "url": "https://arxiv.org/abs/2605.06992",
+    "published": "2026-05-07",
+    "updated": "2026-05-07",
+    "authors": [
+      "Yonatan Slutzky",
+      "Yotam Alexander",
+      "Tomer Slor",
+      "Yoav Nagel",
+      "Nadav Cohen"
+    ],
+    "categories": [
+      "cs.LG",
+      "stat.ML"
+    ],
+    "topics": [
+      "agent-safety",
+      "embodied-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.06992",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.06992",
+    "pdf_url": "https://arxiv.org/pdf/2605.06992",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.06957",
+    "title": "Learning and Reusing Policy Decompositions for Hierarchical Generalized Planning with LLM Agents",
+    "url": "https://arxiv.org/abs/2605.06957",
+    "published": "2026-05-07",
+    "updated": "2026-05-07",
+    "authors": [
+      "Shirin Sohrabi",
+      "Haritha Ananthakrishnan",
+      "Harsha Kokel",
+      "Kavitha Srinivas",
+      "Michael Katz"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.06957",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.06957",
+    "pdf_url": "https://arxiv.org/pdf/2605.06957",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.06737",
+    "title": "A Self-Healing Framework for Reliable LLM-Based Autonomous Agents",
+    "url": "https://arxiv.org/abs/2605.06737",
+    "published": "2026-05-07",
+    "updated": "2026-05-07",
+    "authors": [
+      "Cheonsu Jeong",
+      "Younggun Shin"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.06737",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.06737",
+    "pdf_url": "https://arxiv.org/pdf/2605.06737",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2604.27464",
+    "title": "Security Attack and Defense Strategies for Autonomous Agent Frameworks: A Layered Review with OpenClaw as a Case Study",
+    "url": "https://arxiv.org/abs/2604.27464",
+    "published": "2026-04-30",
+    "updated": "2026-04-30",
+    "authors": [
+      "Luyao Xu",
+      "Xiang Chen"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2604.27464",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.27464",
+    "pdf_url": "https://arxiv.org/pdf/2604.27464",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.28157",
+    "title": "FlashRT: Towards Computationally and Memory Efficient Red-Teaming for Prompt Injection and Knowledge Corruption",
+    "url": "https://arxiv.org/abs/2604.28157",
+    "published": "2026-04-30",
+    "updated": "2026-04-30",
+    "authors": [
+      "Yanting Wang",
+      "Chenlong Yin",
+      "Ying Chen",
+      "Jinyuan Jia"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2604.28157",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.28157",
+    "pdf_url": "https://arxiv.org/pdf/2604.28157",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2604.27859",
+    "title": "Rethinking Agentic Reinforcement Learning In Large Language Models",
+    "url": "https://arxiv.org/abs/2604.27859",
+    "published": "2026-04-30",
+    "updated": "2026-05-15",
+    "authors": [
+      "Fangming Cui",
+      "Ruixiao Zhu",
+      "Cheng Fang",
+      "Sunan Li",
+      "Jiahong Li"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.ET"
+    ],
+    "topics": [
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2604.27859",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.27859",
+    "pdf_url": "https://arxiv.org/pdf/2604.27859",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.20573",
+    "title": "AONA: A Comprehensive Architecture and Workflow Design for Global Agentic Collaboration",
+    "url": "https://arxiv.org/abs/2606.20573",
+    "published": "2026-04-30",
+    "updated": "2026-04-30",
+    "authors": [
+      "Jinliang Xu",
+      "Runkai Zhu",
+      "Bingqi Li",
+      "Fanjie Nie",
+      "Jin Li",
+      "Jiagui Xie"
+    ],
+    "categories": [
+      "cs.NI",
+      "cs.MA"
+    ],
+    "topics": [
+      "multi-agent",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.20573",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20573",
+    "pdf_url": "https://arxiv.org/pdf/2606.20573",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2604.25555",
+    "title": "From CRUD to Autonomous Agents: Formal Validation and Zero-Trust Security for Semantic Gateways in AI-Native Enterprise Systems",
+    "url": "https://arxiv.org/abs/2604.25555",
+    "published": "2026-04-28",
+    "updated": "2026-04-28",
+    "authors": [
+      "Ignacio Peyrano"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2604.25555",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.25555",
+    "pdf_url": "https://arxiv.org/pdf/2604.25555",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2604.21190",
+    "title": "SpatiO: Adaptive Test-Time Orchestration of Vision-Language Agents for Spatial Reasoning",
+    "url": "https://arxiv.org/abs/2604.21190",
+    "published": "2026-04-23",
+    "updated": "2026-06-27",
+    "authors": [
+      "Chan Yeong Hwang",
+      "Miso Choi",
+      "Sunghyun On",
+      "Jinkyu Kim",
+      "Jungbeom Lee"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "reasoning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2604.21190",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.21190",
+    "pdf_url": "https://arxiv.org/pdf/2604.21190",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2604.16706",
+    "title": "Evaluating Tool-Using Language Agents: Judge Reliability, Propagation Cascades, and Runtime Mitigation in AgentProp-Bench",
+    "url": "https://arxiv.org/abs/2604.16706",
+    "published": "2026-04-17",
+    "updated": "2026-04-17",
+    "authors": [
+      "Bhaskar Gurram"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2604.16706",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.16706",
+    "pdf_url": "https://arxiv.org/pdf/2604.16706",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2604.15579",
+    "title": "Don't Make Models Guess Security and Safety: Symbolic Guardrails for Domain-Specific AI Agents",
+    "url": "https://arxiv.org/abs/2604.15579",
+    "published": "2026-04-16",
+    "updated": "2026-07-05",
+    "authors": [
+      "Yining Hong",
+      "Yining She",
+      "Eunsuk Kang",
+      "Christopher S. Timperley",
+      "Christian Kästner"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.15579",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.15579",
+    "pdf_url": "https://arxiv.org/pdf/2604.15579",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.15415",
+    "title": "HarmfulSkillBench: How Do Harmful Skills Weaponize Your Agents?",
+    "url": "https://arxiv.org/abs/2604.15415",
+    "published": "2026-04-16",
+    "updated": "2026-04-16",
+    "authors": [
+      "Yukun Jiang",
+      "Yage Zhang",
+      "Michael Backes",
+      "Xinyue Shen",
+      "Yang Zhang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.15415",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.15415",
+    "pdf_url": "https://arxiv.org/pdf/2604.15415",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.14399",
+    "title": "SpaceMind: A Modular and Self-Evolving Embodied Vision-Language Agent Framework for Autonomous On-orbit Servicing",
+    "url": "https://arxiv.org/abs/2604.14399",
+    "published": "2026-04-15",
+    "updated": "2026-04-15",
+    "authors": [
+      "Aodi Wu",
+      "Haodong Han",
+      "Xubo Luo",
+      "Ruisuo Wang",
+      "Shan He",
+      "Xue Wan"
+    ],
+    "categories": [
+      "cs.RO",
+      "cs.AI",
+      "eess.SY"
+    ],
+    "topics": [
+      "embodied-agent",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2604.14399",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.14399",
+    "pdf_url": "https://arxiv.org/pdf/2604.14399",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2604.08388",
+    "title": "Awakening the Sleeping Agent: Lean-Specific Agentic Data Reactivates General Tool Use in Goedel Prover",
+    "url": "https://arxiv.org/abs/2604.08388",
+    "published": "2026-04-09",
+    "updated": "2026-04-09",
+    "authors": [
+      "Jui-Hui Chung",
+      "Hongzhou Lin",
+      "Lai Jiang",
+      "Shange Tang",
+      "Chi Jin"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2604.08388",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.08388",
+    "pdf_url": "https://arxiv.org/pdf/2604.08388",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2604.06762",
+    "title": "ARuleCon: Agentic Security Rule Conversion",
+    "url": "https://arxiv.org/abs/2604.06762",
+    "published": "2026-04-08",
+    "updated": "2026-04-08",
+    "authors": [
+      "Ming Xu",
+      "Hongtai Wang",
+      "Yanpei Guo",
+      "Zhengmin Yu",
+      "Weili Han",
+      "Hoon Wei Lim",
+      "Jin Song Dong",
+      "Jiaheng Zhang"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.06762",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.06762",
+    "pdf_url": "https://arxiv.org/pdf/2604.06762",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.02155",
+    "title": "Brief Is Better: Non-Monotonic Chain-of-Thought Budget Effects in Function-Calling Language Agents",
+    "url": "https://arxiv.org/abs/2604.02155",
+    "published": "2026-04-02",
+    "updated": "2026-04-02",
+    "authors": [
+      "Xuan Qi"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling",
+      "language-agent"
+    ],
+    "arxiv_id": "2604.02155",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.02155",
+    "pdf_url": "https://arxiv.org/pdf/2604.02155",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2603.27148",
+    "title": "SafetyDrift: Predicting When AI Agents Cross the Line Before They Actually Do",
+    "url": "https://arxiv.org/abs/2603.27148",
+    "published": "2026-03-28",
+    "updated": "2026-03-28",
+    "authors": [
+      "Aditya Dhodapkar",
+      "Farhaan Pishori"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.27148",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.27148",
+    "pdf_url": "https://arxiv.org/pdf/2603.27148",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.19469",
+    "title": "A Framework for Formalizing LLM Agent Security",
+    "url": "https://arxiv.org/abs/2603.19469",
+    "published": "2026-03-19",
+    "updated": "2026-03-19",
+    "authors": [
+      "Vincent Siu",
+      "Jingxuan He",
+      "Kyle Montgomery",
+      "Zhun Wang",
+      "Neil Gong",
+      "Chenguang Wang",
+      "Dawn Song"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "memory",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.19469",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.19469",
+    "pdf_url": "https://arxiv.org/pdf/2603.19469",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.11088",
+    "title": "The Attack and Defense Landscape of Agentic AI: A Comprehensive Survey",
+    "url": "https://arxiv.org/abs/2603.11088",
+    "published": "2026-03-11",
+    "updated": "2026-03-11",
+    "authors": [
+      "Juhee Kim",
+      "Xiaoyuan Liu",
+      "Zhun Wang",
+      "Shi Qiu",
+      "Bo Li",
+      "Wenbo Guo",
+      "Dawn Song"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.11088",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.11088",
+    "pdf_url": "https://arxiv.org/pdf/2603.11088",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.01438",
+    "title": "Enhancing Persona Following at Decoding Time via Dynamic Importance Estimation for Role-Playing Agents",
+    "url": "https://arxiv.org/abs/2603.01438",
+    "published": "2026-03-02",
+    "updated": "2026-03-02",
+    "authors": [
+      "Yuxin Liu",
+      "Mingye Zhu",
+      "Siyuan Liu",
+      "Bo Hu",
+      "Lei Zhang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "planning",
+      "rag",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.01438",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.01438",
+    "pdf_url": "https://arxiv.org/pdf/2603.01438",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2602.23320",
+    "title": "ParamMem: Augmenting Language Agents with Parametric Reflective Memory",
+    "url": "https://arxiv.org/abs/2602.23320",
+    "published": "2026-02-26",
+    "updated": "2026-02-27",
+    "authors": [
+      "Tianjun Yao",
+      "Yongqiang Chen",
+      "Yujia Zheng",
+      "Pan Li",
+      "Zhiqiang Shen",
+      "Kun Zhang"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "reasoning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2602.23320",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.23320",
+    "pdf_url": "https://arxiv.org/pdf/2602.23320",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2602.16931",
+    "title": "Narrow Fine-Tuning Erodes Safety Alignment in Vision-Language Agents",
+    "url": "https://arxiv.org/abs/2602.16931",
+    "published": "2026-02-18",
+    "updated": "2026-03-15",
+    "authors": [
+      "Idhant Gulati",
+      "Shivam Raval"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2602.16931",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.16931",
+    "pdf_url": "https://arxiv.org/pdf/2602.16931",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2602.07391",
+    "title": "NAAMSE: Framework for Evolutionary Security Evaluation of Agents",
+    "url": "https://arxiv.org/abs/2602.07391",
+    "published": "2026-02-07",
+    "updated": "2026-03-08",
+    "authors": [
+      "Kunal Pai",
+      "Parth Shah",
+      "Harshil Patel"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.07391",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.07391",
+    "pdf_url": "https://arxiv.org/pdf/2602.07391",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2601.05467",
+    "title": "STELP: Secure Transpilation and Execution of LLM-Generated Programs",
+    "url": "https://arxiv.org/abs/2601.05467",
+    "published": "2026-01-09",
+    "updated": "2026-01-15",
+    "authors": [
+      "Swapnil Shinde",
+      "Sahil Wadhwa",
+      "Andy Luo",
+      "Akshay Gupta",
+      "Mohammad Shahed Sorower"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2601.05467",
+    "source": "arxiv",
+    "source_id": "arxiv:2601.05467",
+    "pdf_url": "https://arxiv.org/pdf/2601.05467",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2510.26167",
+    "title": "ToolRM: Towards Agentic Tool-Use Reward Modeling",
+    "url": "https://arxiv.org/abs/2510.26167",
+    "published": "2025-10-30",
+    "updated": "2026-01-13",
+    "authors": [
+      "Renhao Li",
+      "Jianhong Tu",
+      "Yang Su",
+      "Yantao Liu",
+      "Fei Huang",
+      "Hamid Alinejad-Rokny",
+      "Derek F. Wong",
+      "Junyang Lin",
+      "Min Yang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2510.26167",
+    "source": "arxiv",
+    "source_id": "arxiv:2510.26167",
+    "pdf_url": "https://arxiv.org/pdf/2510.26167",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2510.22768",
+    "title": "Seeing is Believing? Evaluating Vision-Language Model Susceptibility in Agent-to-Agent Multimodal Persuasion",
+    "url": "https://arxiv.org/abs/2510.22768",
+    "published": "2025-10-26",
+    "updated": "2026-06-02",
+    "authors": [
+      "Haoyi Qiu",
+      "Yilun Zhou",
+      "Pranav Narayanan Venkit",
+      "Kung-Hsiang Huang",
+      "Jiaxin Zhang",
+      "Nanyun Peng",
+      "Chien-Sheng Wu"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2510.22768",
+    "source": "arxiv",
+    "source_id": "arxiv:2510.22768",
+    "pdf_url": "https://arxiv.org/pdf/2510.22768",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2607.05794",
+    "title": "From Passive Retrieval to Active Memory Navigation: Learning to Use Memory as a Structured Action Space",
+    "url": "https://arxiv.org/abs/2607.05794",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Yue Xu",
+      "Yutao Sun",
+      "Yihao Liu",
+      "Mengyu Zhou",
+      "Jiayi Qiao",
+      "Lu Ma",
+      "Kai Tang",
+      "Wenjie Wang",
+      "Xiaoxi Jiang",
+      "Guanjun Jiang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2607.05794",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05794",
+    "pdf_url": "https://arxiv.org/pdf/2607.05794",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2607.06341",
+    "title": "Harnessing Code Agents for Automatic Software Verification",
+    "url": "https://arxiv.org/abs/2607.06341",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Shuangxiang Kan",
+      "Shuanglong Kan",
+      "Sebastian Ertel"
+    ],
+    "categories": [
+      "cs.FL",
+      "cs.AI",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.06341",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06341",
+    "pdf_url": "https://arxiv.org/pdf/2607.06341",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.05001",
+    "title": "TACTIC-KG: Toward Small Agent Teams for Cyber Threat Intelligence Knowledge Graph Construction",
+    "url": "https://arxiv.org/abs/2607.05001",
+    "published": "2026-07-06",
+    "updated": "2026-07-07",
+    "authors": [
+      "Mouhamed Amine Bouchiha",
+      "Gregory Blanc"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.LG",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.05001",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05001",
+    "pdf_url": "https://arxiv.org/pdf/2607.05001",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.05666",
+    "title": "What Do AI Agents Actually Change? An Empirical Taxonomy of Mutation Patterns in Performance-Improving Pull Requests",
+    "url": "https://arxiv.org/abs/2607.05666",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Illia Dovhoshliubnyi",
+      "Nima Soroush",
+      "Ashkan Sami",
+      "Alexander Brownlee"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "coding-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.05666",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05666",
+    "pdf_url": "https://arxiv.org/pdf/2607.05666",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.05518",
+    "title": "aiAuthZ: Off-Host, Identity-Bound Authorization for AI Agents",
+    "url": "https://arxiv.org/abs/2607.05518",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Sai Varun Kodathala"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.05518",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05518",
+    "pdf_url": "https://arxiv.org/pdf/2607.05518",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.04697",
+    "title": "AI Agent Pull Requests on GitHub: Frequency, Structure, and Merge Conflict Rates",
+    "url": "https://arxiv.org/abs/2607.04697",
+    "published": "2026-07-06",
+    "updated": "2026-07-07",
+    "authors": [
+      "George Xu",
+      "Arjun Subramanian",
+      "Nithilan Karthik"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.04697",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04697",
+    "pdf_url": "https://arxiv.org/pdf/2607.04697",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.05391",
+    "title": "LLM-as-a-Verifier: A General-Purpose Verification Framework",
+    "url": "https://arxiv.org/abs/2607.05391",
+    "published": "2026-07-06",
+    "updated": "2026-07-07",
+    "authors": [
+      "Jacky Kwok",
+      "Shulu Li",
+      "Pranav Atreya",
+      "Yuejiang Liu",
+      "Yixing Jiang",
+      "Chelsea Finn",
+      "Marco Pavone",
+      "Ion Stoica",
+      "Azalia Mirhoseini"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.LG",
+      "cs.MA",
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "embodied-agent",
+      "reasoning"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.05391",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05391",
+    "pdf_url": "https://arxiv.org/pdf/2607.05391",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.04334",
+    "title": "Do GUI Agents Believe Their Eyes? Diagnosing State-Belief Reliance on Pixels versus Structure",
+    "url": "https://arxiv.org/abs/2607.04334",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Guijia Zhang",
+      "Harry Yang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2607.04334",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04334",
+    "pdf_url": "https://arxiv.org/pdf/2607.04334",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2607.04394",
+    "title": "MechMath Agent Team: LLM Driven Agents for Mathematical Research",
+    "url": "https://arxiv.org/abs/2607.04394",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Yichuan Cao",
+      "Ruichen Qiu",
+      "Junqi Liu",
+      "Jiaqi Wang",
+      "Dakai Guo",
+      "Ruyong Feng",
+      "Lihong Zhi",
+      "Xiao-Shan Gao"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.SC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.04394",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04394",
+    "pdf_url": "https://arxiv.org/pdf/2607.04394",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.02942",
+    "title": "A Workflow-Aware Serving Layer for Agentic Applications",
+    "url": "https://arxiv.org/abs/2607.02942",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Jiayi Qian",
+      "Zishen Wan",
+      "Hanchen Yang",
+      "Chun Tao",
+      "Souvik Kundu",
+      "Tushar Krishna"
+    ],
+    "categories": [
+      "cs.DC",
+      "cs.MA"
+    ],
+    "topics": [
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.02942",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02942",
+    "pdf_url": "https://arxiv.org/pdf/2607.02942",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.03105",
+    "title": "ORBIT-Q: Dual-axis benchmarking of autonomous agents in scientific quantum programming",
+    "url": "https://arxiv.org/abs/2607.03105",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Shi-Xin Zhang",
+      "Yu-Qin Chen"
+    ],
+    "categories": [
+      "quant-ph"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.03105",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03105",
+    "pdf_url": "https://arxiv.org/pdf/2607.03105",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.03316",
+    "title": "Is Agentic Code Review Helpful? Mining Developers' Feedback to CodeRabbit Reviews in the Wild",
+    "url": "https://arxiv.org/abs/2607.03316",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Hong Yi Lin",
+      "Mingzhao Liang",
+      "Kla Tantithamthavorn",
+      "Patanamon Thongtanunam"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2607.03316",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03316",
+    "pdf_url": "https://arxiv.org/pdf/2607.03316",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2607.03220",
+    "title": "CONTRA: Red-Teaming Configurations of Personalizable Agents",
+    "url": "https://arxiv.org/abs/2607.03220",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Jonathan Nöther",
+      "Adish Singla",
+      "Goran Radanovic"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2607.03220",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03220",
+    "pdf_url": "https://arxiv.org/pdf/2607.03220",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2607.01812",
+    "title": "TO-Master: an LLM-agent framework for automated topology optimization",
+    "url": "https://arxiv.org/abs/2607.01812",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Haoju Lin",
+      "Wenchang Zhang",
+      "Weipeng Xu",
+      "Xiang Li",
+      "Tian Xu",
+      "Tianju Xue"
+    ],
+    "categories": [
+      "cs.CE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.01812",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01812",
+    "pdf_url": "https://arxiv.org/pdf/2607.01812",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.02453",
+    "title": "Adoption and Ecosystem Health: A Longitudinal Analysis of Open-Source Multi-Agent Frameworks",
+    "url": "https://arxiv.org/abs/2607.02453",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Xi Zhang",
+      "Papi Menon",
+      "Vivian Chu",
+      "Koray Cosguner"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.02453",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02453",
+    "pdf_url": "https://arxiv.org/pdf/2607.02453",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.02245",
+    "title": "Copewell: A Multi-Agent Swarm Architecture for Equitable Mental Wellness Support",
+    "url": "https://arxiv.org/abs/2607.02245",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Seren Yenikent",
+      "Jack Vinijtrongjit",
+      "Katherine Ng"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CY",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.02245",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02245",
+    "pdf_url": "https://arxiv.org/pdf/2607.02245",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.02381",
+    "title": "HULAT2 at MER-TRANS 2026: Governed Multi-Agent Simplification for Spanish Easy-to-Read Generation",
+    "url": "https://arxiv.org/abs/2607.02381",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Lourdes Moreno",
+      "Paloma Martínez",
+      "Marco Antonio Sanchez-Escudero",
+      "Miguel Domínguez-Gómez"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.02381",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02381",
+    "pdf_url": "https://arxiv.org/pdf/2607.02381",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.01531",
+    "title": "OPINE-World: Programmatic World Modeling with Ontology-error-Prioritized Interactive Exploration",
+    "url": "https://arxiv.org/abs/2607.01531",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "David Courtis",
+      "Wenhao Li",
+      "Scott Sanner"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "planning-agent"
+    ],
+    "arxiv_id": "2607.01531",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01531",
+    "pdf_url": "https://arxiv.org/pdf/2607.01531",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.01047",
+    "title": "Conversable Complexity: Agentic LLM Collectives as Interpretable Substrates",
+    "url": "https://arxiv.org/abs/2607.01047",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Elias Najarro",
+      "Ane Espeseth",
+      "Eleni Nisioti",
+      "Sebastian Risi",
+      "Stefano Nichele"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "memory",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.01047",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01047",
+    "pdf_url": "https://arxiv.org/pdf/2607.01047",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.01510",
+    "title": "Janus: a Playground for User-Involved Agentic Permission Management",
+    "url": "https://arxiv.org/abs/2607.01510",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Natalie Grace Brigham",
+      "Eugene Bagdasarian",
+      "Tadayoshi Kohno",
+      "Franziska Roesner"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.01510",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01510",
+    "pdf_url": "https://arxiv.org/pdf/2607.01510",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.00407",
+    "title": "Personalization as Inverse Planning: Learning Latent Design Intents for Agentic Slide Generation via Structural Denoising",
+    "url": "https://arxiv.org/abs/2607.00407",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Tianci Liu",
+      "Zihan Dong",
+      "Linjun Zhang",
+      "Haoyu Wang",
+      "jing Gao",
+      "Emre Kiciman",
+      "Ranveer Chandra",
+      "Wei-Ting Chen"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "multi-agent",
+      "planning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.00407",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00407",
+    "pdf_url": "https://arxiv.org/pdf/2607.00407",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.01366",
+    "title": "Auto-FL-Research: Agentic Search for Federated Learning Algorithms",
+    "url": "https://arxiv.org/abs/2607.01366",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Holger R. Roth",
+      "Ziyue Xu",
+      "Chester Chen",
+      "Daguang Xu",
+      "Peter Cnudde",
+      "Andrew Feng"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.01366",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01366",
+    "pdf_url": "https://arxiv.org/pdf/2607.01366",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.00990",
+    "title": "SWE-Doctor: Guiding Software Engineering Agents with Runtime Diagnosis from Multi-Faceted Bug Reproduction Tests",
+    "url": "https://arxiv.org/abs/2607.00990",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Yaoqi Guo",
+      "Yang Liu",
+      "Jie M. Zhang",
+      "Yun Ma",
+      "Yiling Lou",
+      "Zhenpeng Chen"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "rag"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.00990",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00990",
+    "pdf_url": "https://arxiv.org/pdf/2607.00990",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.31471",
+    "title": "Think While You Map: Asynchronous Vision-Language Agents for Incremental 3D Scene Graphs",
+    "url": "https://arxiv.org/abs/2606.31471",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Deniz Bickici",
+      "Michael Pabst",
+      "Shohei Mori",
+      "Dieter Schmalstieg"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.31471",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31471",
+    "pdf_url": "https://arxiv.org/pdf/2606.31471",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.31831",
+    "title": "An Agentic AI Framework to Accelerate Scientific Discovery in Plant Phenotyping",
+    "url": "https://arxiv.org/abs/2606.31831",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Renan Souza",
+      "Daniel Rosendo",
+      "Kelsey Carter",
+      "John Lagergren",
+      "Frédéric Suter",
+      "Shelaine L. Curd",
+      "Gerald A. Tuskan",
+      "Rafael Ferreira da Silva",
+      "David Weston"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "planning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.31831",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31831",
+    "pdf_url": "https://arxiv.org/pdf/2606.31831",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.31916",
+    "title": "Theory of Mind and Persuasion Beyond Conversation: Assessing the Capacity of LLMs to Induce Belief States via Planning and Action",
+    "url": "https://arxiv.org/abs/2606.31916",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Ben Slater",
+      "Matteo G. Mecattaf",
+      "Lucy G. Cheke",
+      "John Burden",
+      "Winnie Street"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.31916",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31916",
+    "pdf_url": "https://arxiv.org/pdf/2606.31916",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.31767",
+    "title": "JETO-Bench: A Reproducible Benchmark for Execution Time Improvement Patches in Java",
+    "url": "https://arxiv.org/abs/2606.31767",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Khashayar Etemadi",
+      "Zhendong Su"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.31767",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31767",
+    "pdf_url": "https://arxiv.org/pdf/2606.31767",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.31665",
+    "title": "ForecastAgentSearch: Towards a Multi-Expert Agent Search System for Geopolitical Event Forecasting",
+    "url": "https://arxiv.org/abs/2606.31665",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Miaomiao Cai",
+      "He Chang",
+      "Yunshan Ma",
+      "See-kiong Ng"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "rag"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.31665",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31665",
+    "pdf_url": "https://arxiv.org/pdf/2606.31665",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.29719",
+    "title": "A Diagnostic Framework and Multi-Evaluator Audit of Evaluator-Driven Preference Dynamics in Self-Adapting LLM Agents",
+    "url": "https://arxiv.org/abs/2606.29719",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Liu Zewen"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.29719",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29719",
+    "pdf_url": "https://arxiv.org/pdf/2606.29719",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.29745",
+    "title": "ECHO: Learning Epistemically Adaptive Language Agents with Turn-Level Credit",
+    "url": "https://arxiv.org/abs/2606.29745",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Abhijnan Nath",
+      "Nikhil Krishnaswamy"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.29745",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29745",
+    "pdf_url": "https://arxiv.org/pdf/2606.29745",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.30970",
+    "title": "Behavioral Governance for Autonomous AI Agents: The AgentBound Framework",
+    "url": "https://arxiv.org/abs/2606.30970",
+    "published": "2026-06-29",
+    "updated": "2026-07-01",
+    "authors": [
+      "Anuj Kaul",
+      "Qianlong Lan",
+      "Pranay Gupta"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.30970",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30970",
+    "pdf_url": "https://arxiv.org/pdf/2606.30970",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.29788",
+    "title": "MemLeak: Diagnosing Information Leaks in Multimodal Agent Memory",
+    "url": "https://arxiv.org/abs/2606.29788",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Kuan Wang",
+      "Chao Zhang"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.29788",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29788",
+    "pdf_url": "https://arxiv.org/pdf/2606.29788",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.30616",
+    "title": "Scaling the Horizon, Not the Parameters: Reaching Trillion-Parameter Performance with a 35B Agent",
+    "url": "https://arxiv.org/abs/2606.30616",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Lei Bai",
+      "Zongsheng Cao",
+      "Yang Chen",
+      "Zhiyao Cui",
+      "Shangheng Du",
+      "Yue Fan",
+      "Shiyang Feng",
+      "Zijie Guo",
+      "Haonan He",
+      "Liang He",
+      "Xiaohan He",
+      "Shuyue Hu",
+      "Yusong Hu",
+      "Songtao Huang",
+      "Yichen Jiang",
+      "Hao Li",
+      "Xin Li",
+      "Dahua Lin",
+      "Weihao Lin",
+      "Fenghua Ling",
+      "Dongrui Liu",
+      "Zhuo Liu",
+      "Runmin Ma",
+      "Chunjiang Mu",
+      "Haoyang Peng",
+      "Tianshuo Peng",
+      "Jinxin Shi",
+      "Luohe Shi",
+      "Boyuan Sun",
+      "Zelin Tan",
+      "Shengji Tang",
+      "Qianyi Wang",
+      "Yiming Wu",
+      "Yi Xie",
+      "Xiangchao Yan",
+      "Jingqi Ye",
+      "Peng Ye",
+      "Fangchen Yu",
+      "Jiakang Yuan",
+      "Bihao Zhan",
+      "Bo Zhang",
+      "Chen Zhang",
+      "Shufei Zhang",
+      "Shuaiyu Zhang",
+      "Wenlong Zhang",
+      "Yiqun Zhang",
+      "Junpeng Zhao",
+      "Zhijie Zhong",
+      "Bowen Zhou",
+      "Yuhao Zhou"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.30616",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30616",
+    "pdf_url": "https://arxiv.org/pdf/2606.30616",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.30111",
+    "title": "Automating the Design of Embodied Agent Architectures",
+    "url": "https://arxiv.org/abs/2606.30111",
+    "published": "2026-06-29",
+    "updated": "2026-07-03",
+    "authors": [
+      "Jian Zhou",
+      "Sihao Lin",
+      "Jin Li",
+      "Shuai Fu",
+      "Gengze Zhou",
+      "Qi Wu"
+    ],
+    "categories": [
+      "cs.RO",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.30111",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30111",
+    "pdf_url": "https://arxiv.org/pdf/2606.30111",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.29495",
+    "title": "Cognitive World Models for Process-Level Social Influence Evaluation",
+    "url": "https://arxiv.org/abs/2606.29495",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Minghui Ma",
+      "Bin Guo",
+      "Han Wang",
+      "Mengqi Chen",
+      "Jingqi Liu",
+      "Yan Liu",
+      "Zhiwen Yu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent",
+      "world-model"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.29495",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29495",
+    "pdf_url": "https://arxiv.org/pdf/2606.29495",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.28896",
+    "title": "A Task-Driven and Quality-Assured Agent Framework for SAR Data Generation",
+    "url": "https://arxiv.org/abs/2606.28896",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Xuanting Wu",
+      "Fan Zhanga",
+      "Fei Ma",
+      "Ling Guan",
+      "Guochun Ma",
+      "Yongsheng Zhou"
+    ],
+    "categories": [
+      "eess.IV",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.28896",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28896",
+    "pdf_url": "https://arxiv.org/pdf/2606.28896",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.28279",
+    "title": "Agentic Hardware Design as Repository-Level Code Evolution",
+    "url": "https://arxiv.org/abs/2606.28279",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Cunxi Yu",
+      "Chenhui Deng",
+      "Nathaniel Pinckney",
+      "Brucek Khailany"
+    ],
+    "categories": [
+      "cs.AR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.28279",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28279",
+    "pdf_url": "https://arxiv.org/pdf/2606.28279",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.28430",
+    "title": "Building to the Test: Coding Agents Deliver What You Check, Not What You Requested",
+    "url": "https://arxiv.org/abs/2606.28430",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Yanuo Ma",
+      "Ben Kereopa-Yorke",
+      "Ben Schultz"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.28430",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28430",
+    "pdf_url": "https://arxiv.org/pdf/2606.28430",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.28187",
+    "title": "GBC: Gradient-Based Connections for Optimizing Multi-Agent Systems",
+    "url": "https://arxiv.org/abs/2606.28187",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Xiaocheng Yang",
+      "Abdulrahman Alrabah",
+      "Dilek Hakkani-Tür",
+      "Gokhan Tur"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.28187",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28187",
+    "pdf_url": "https://arxiv.org/pdf/2606.28187",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.27416",
+    "title": "Glite ARF: Verifier-Driven Research with Parallel LLM Coding Agents",
+    "url": "https://arxiv.org/abs/2606.27416",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Vassili Philippov",
+      "Pavel Katunin",
+      "Dmitry Andreev",
+      "Igor Ostanin",
+      "Anton Nikolaev"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.SE"
+    ],
+    "topics": [
+      "coding-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.27416",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27416",
+    "pdf_url": "https://arxiv.org/pdf/2606.27416",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.26649",
+    "title": "Autoformalization of Agent Instructions into Policy-as-Code",
+    "url": "https://arxiv.org/abs/2606.26649",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Adam Mondl",
+      "Matthew Maisel",
+      "John H. Brock"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2606.26649",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26649",
+    "pdf_url": "https://arxiv.org/pdf/2606.26649",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.26057",
+    "title": "The Unfireable Safety Kernel: Execution-Time AI Alignment for AI Agents and Other Escapable AI Systems",
+    "url": "https://arxiv.org/abs/2606.26057",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Seth Dobrin",
+      "Łukasz Chmiel"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CR",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.26057",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26057",
+    "pdf_url": "https://arxiv.org/pdf/2606.26057",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.26356",
+    "title": "Instruction Bleed: Cross-Module Interference in Prompt-Composed Agentic Systems",
+    "url": "https://arxiv.org/abs/2606.26356",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Ching-Yu Lin",
+      "Yifan Liu"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.IR",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.26356",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26356",
+    "pdf_url": "https://arxiv.org/pdf/2606.26356",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.25519",
+    "title": "Quantization Inflates Reasoning: Token Inflation as a Hidden Cost of Low-Bit Reasoning Models",
+    "url": "https://arxiv.org/abs/2606.25519",
+    "published": "2026-06-24",
+    "updated": "2026-06-29",
+    "authors": [
+      "Xinyu Lian",
+      "Walid Krichene",
+      "Beichen Huang",
+      "Masahiro Tanaka",
+      "Olatunji Ruwase",
+      "Li Zhang",
+      "Minjia Zhang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.25519",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25519",
+    "pdf_url": "https://arxiv.org/pdf/2606.25519",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.25195",
+    "title": "SoK: AI Secure Code Generation: Progress, Pitfalls, and Paths Forward",
+    "url": "https://arxiv.org/abs/2606.25195",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Rupam Patir",
+      "Keyan Guo",
+      "Haipeng Cai",
+      "Hongxin Hu"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.25195",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25195",
+    "pdf_url": "https://arxiv.org/pdf/2606.25195",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.24416",
+    "title": "Agentic AI for Bilevel Long-Term Optimization of Policy-Driven Physical Layer Systems",
+    "url": "https://arxiv.org/abs/2606.24416",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Bingnan Xiao",
+      "Chenhao Yang",
+      "Wei Ni",
+      "Xin Wang",
+      "Tony Q. S. Quek"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "rag"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.24416",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24416",
+    "pdf_url": "https://arxiv.org/pdf/2606.24416",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.24855",
+    "title": "OpenThoughts-Agent: Data Recipes for Agentic Models",
+    "url": "https://arxiv.org/abs/2606.24855",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Negin Raoof",
+      "Richard Zhuang",
+      "Marianna Nezhurina",
+      "Etash Guha",
+      "Atula Tejaswi",
+      "Ryan Marten",
+      "Charlie F. Ruan",
+      "Tyler Griggs",
+      "Alexander Glenn Shaw",
+      "Hritik Bansal",
+      "E. Kelly Buchanan",
+      "Artem Gazizov",
+      "Reinhard Heckel",
+      "Chinmay Hegde",
+      "Sankalp Jajee",
+      "Daanish Khazi",
+      "Emmanouil Koukoumidis",
+      "Xiangyi Li",
+      "Hange Liu",
+      "Shlok Natarajan",
+      "Harsh Raj",
+      "Nicholas Roberts",
+      "Ethan Shen",
+      "Nishad Singhi",
+      "Michael Siu",
+      "Ashima Suvarna",
+      "Hanwen Xing",
+      "Patrick Yubeaton",
+      "Robert Zhang",
+      "Leon Liangyu Chen",
+      "Xiaokun Chen",
+      "Steven Dillmann",
+      "Saadia Gabriel",
+      "Xunyi Jiang",
+      "Anurag Kashyap",
+      "Boxuan Li",
+      "Yein Park",
+      "Minh Pham",
+      "Sujay Sanghavi",
+      "Lin Shi",
+      "Ke Sun",
+      "Yixin Wang",
+      "Zhiwei Xu",
+      "Erica Zhang",
+      "Siyan Zhao",
+      "Wanjia Zhao",
+      "Jenia Jitsev",
+      "Alex Dimakis",
+      "Benjamin Feuer",
+      "Ludwig Schmidt"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.24855",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24855",
+    "pdf_url": "https://arxiv.org/pdf/2606.24855",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.25191",
+    "title": "To Isolate or to Score? Model-Adaptive Assessment for Cost-Efficient Multi-Agent RAG",
+    "url": "https://arxiv.org/abs/2606.25191",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Jungseob Lee",
+      "Chanjun Park",
+      "Heuiseok Lim"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "reasoning"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.25191",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25191",
+    "pdf_url": "https://arxiv.org/pdf/2606.25191",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.23130",
+    "title": "Understanding the (In)Security of Vibe-Coded Applications",
+    "url": "https://arxiv.org/abs/2606.23130",
+    "published": "2026-06-22",
+    "updated": "2026-06-23",
+    "authors": [
+      "Junquan Deng",
+      "Zhiyu Fan",
+      "Ruijie Meng"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "memory",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.23130",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23130",
+    "pdf_url": "https://arxiv.org/pdf/2606.23130",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.22953",
+    "title": "Plans Don't Persist: Why Context Management Is Load Bearing for LLM Agents",
+    "url": "https://arxiv.org/abs/2606.22953",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Aman Mehta",
+      "Anupam Datta"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.22953",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22953",
+    "pdf_url": "https://arxiv.org/pdf/2606.22953",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.21836",
+    "title": "AgentDSE: Reasoning-Augmented Architectural Design Space Exploration",
+    "url": "https://arxiv.org/abs/2606.21836",
+    "published": "2026-06-20",
+    "updated": "2026-06-20",
+    "authors": [
+      "Chenyu Wang",
+      "Jiahe Caroline Shi",
+      "David Kong",
+      "Duane S. Boning",
+      "Zishen Wan",
+      "Yilun Du",
+      "Vijay Janapa Reddi"
+    ],
+    "categories": [
+      "cs.AR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "reasoning"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.21836",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21836",
+    "pdf_url": "https://arxiv.org/pdf/2606.21836",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.21401",
+    "title": "SwarmX: Agentic Scheduling for Low-Latency Agentic Systems",
+    "url": "https://arxiv.org/abs/2606.21401",
+    "published": "2026-06-19",
+    "updated": "2026-06-28",
+    "authors": [
+      "Yeqi Huang",
+      "Yanwei Ye",
+      "Guomin Chen",
+      "Wenhao Su",
+      "Bin Gong",
+      "Jialian Li",
+      "Zhan Lu",
+      "Yangshen Deng",
+      "Xuan Sun",
+      "Le Xu",
+      "Luo Mai"
+    ],
+    "categories": [
+      "cs.DC",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.21401",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21401",
+    "pdf_url": "https://arxiv.org/pdf/2606.21401",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.21228",
+    "title": "Sakana Fugu Technical Report",
+    "url": "https://arxiv.org/abs/2606.21228",
+    "published": "2026-06-19",
+    "updated": "2026-06-23",
+    "authors": [
+      "Yujin Tang",
+      "Edoardo Cetin",
+      "Jinglue Xu",
+      "Qi Sun",
+      "Stefan Nielsen",
+      "Vincent Richard",
+      "Haruto Goda",
+      "Iaroslav Tymchenko",
+      "Nhan Nguyen",
+      "Hyunin Lee",
+      "Mari Ashiga",
+      "Shashank Kotyan",
+      "So Kuroki",
+      "Tarin Clanuwat"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "coding-agent",
+      "multi-agent",
+      "rag",
+      "reasoning"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.21228",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21228",
+    "pdf_url": "https://arxiv.org/pdf/2606.21228",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.20510",
+    "title": "Efficient and Sound Probabilistic Verification for AI Agents",
+    "url": "https://arxiv.org/abs/2606.20510",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Alaia Solko-Breslin",
+      "Pramod Kaushik Mudrakarta",
+      "Mihai Christodorescu",
+      "Somesh Jha",
+      "Krishnamurthy Dj Dvijotham"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.20510",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20510",
+    "pdf_url": "https://arxiv.org/pdf/2606.20510",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.19242",
+    "title": "Runtime Compliance Verification for AI Agents",
+    "url": "https://arxiv.org/abs/2606.19242",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Nafiseh Kahani",
+      "Masoud Barati",
+      "Diana Addae"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "function-calling",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.19242",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19242",
+    "pdf_url": "https://arxiv.org/pdf/2606.19242",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.28374",
+    "title": "Recursive Self-Evolving Agents via Held-Out Selection",
+    "url": "https://arxiv.org/abs/2606.28374",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Michael Nguyen",
+      "Quoc Nguyen",
+      "Paul Vuong"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.28374",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28374",
+    "pdf_url": "https://arxiv.org/pdf/2606.28374",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.18363",
+    "title": "Guava: An Effective and Universal Harness for Embodied Manipulation",
+    "url": "https://arxiv.org/abs/2606.18363",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Haowen Liu",
+      "Xirui Li",
+      "Shaoxiong Yao",
+      "Peng Shi",
+      "Tianyi Zhou",
+      "Jia-Bin Huang",
+      "Furong Huang",
+      "Jiayuan Mao"
+    ],
+    "categories": [
+      "cs.RO",
+      "cs.AI"
+    ],
+    "topics": [
+      "embodied-agent",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.18363",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18363",
+    "pdf_url": "https://arxiv.org/pdf/2606.18363",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.17453",
+    "title": "MapSatisfyBench: Benchmarking Satisfaction-Aware Map Agents through Behavior-Grounded Implicit Decision Factors",
+    "url": "https://arxiv.org/abs/2606.17453",
+    "published": "2026-06-16",
+    "updated": "2026-06-17",
+    "authors": [
+      "Lubin Bai",
+      "Mengyu Cao",
+      "Sixue Wang",
+      "Zhongwei Wan",
+      "Yue Pan",
+      "Jiale Hou",
+      "Xiang Li",
+      "Xiuyuan Zhang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.17453",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17453",
+    "pdf_url": "https://arxiv.org/pdf/2606.17453",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.18023",
+    "title": "LoopCoder-v2: Only Loop Once for Efficient Test-Time Computation Scaling",
+    "url": "https://arxiv.org/abs/2606.18023",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Jian Yang",
+      "Shawn Guo",
+      "Wei Zhang",
+      "Tianyu Zheng",
+      "Yaxin Du",
+      "Haau-Sing Li",
+      "Jiajun Wu",
+      "Yue Song",
+      "Yan Xing",
+      "Qingsong Cai",
+      "Zelong Huang",
+      "Chuan Hao",
+      "Ran Tao",
+      "Xianglong Liu",
+      "Wayne Xin Zhao",
+      "Mingjie Tang",
+      "Weifeng Lv",
+      "Ming Zhou",
+      "Bryan Dai"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "memory",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.18023",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18023",
+    "pdf_url": "https://arxiv.org/pdf/2606.18023",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.17383",
+    "title": "Model Validation of Agentic AI Systems: A POMDP-Based Framework for Belief-State, Forecast, and Policy Validation",
+    "url": "https://arxiv.org/abs/2606.17383",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Matthew Francis Dixon"
+    ],
+    "categories": [
+      "q-fin.RM",
+      "cs.AI",
+      "cs.LG",
+      "stat.ML"
+    ],
+    "topics": [
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.17383",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17383",
+    "pdf_url": "https://arxiv.org/pdf/2606.17383",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.16813",
+    "title": "GIST-CMTF: Goal-State Inference for Causal Minimal Tool Filtering in LLM Agents",
+    "url": "https://arxiv.org/abs/2606.16813",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Rahul Suresh Babu",
+      "Rohit Shukla"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.16813",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16813",
+    "pdf_url": "https://arxiv.org/pdf/2606.16813",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.16839",
+    "title": "Towards LLM Accelerated Rapid Reviews for Software Tool Discovery -- Case for Log Anomaly Detection",
+    "url": "https://arxiv.org/abs/2606.16839",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Jesse Nyyssölä",
+      "Hamza Bin Mazhar",
+      "Alexander Bakhtin",
+      "Matteo Esposito",
+      "Nana Reinikainen",
+      "Yuqing Wang",
+      "Ying Song",
+      "Davide Taibi",
+      "Mika Mäntylä"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.16839",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16839",
+    "pdf_url": "https://arxiv.org/pdf/2606.16839",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.16481",
+    "title": "Steering Emotional Dynamics for Art Therapy: Controllable Narrative Script Generation through Hierarchically Guided LLM Agents",
+    "url": "https://arxiv.org/abs/2606.16481",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Suqing Wang",
+      "Qinghai Miao",
+      "Chao Guo",
+      "Yisheng Lv"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.16481",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16481",
+    "pdf_url": "https://arxiv.org/pdf/2606.16481",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.17368",
+    "title": "Distributed General-Purpose Agent Networks: Architecture, Key Mechanisms, and Prototypes",
+    "url": "https://arxiv.org/abs/2606.17368",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Shengli Zhang",
+      "Deen Ma",
+      "Zibin Lin",
+      "Taotao Wang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.NI"
+    ],
+    "topics": [
+      "computer-use",
+      "multi-agent",
+      "planning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.17368",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17368",
+    "pdf_url": "https://arxiv.org/pdf/2606.17368",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.15709",
+    "title": "AI-Driven Framework for Adaptive Water Network Management with Proof-of-Concept Implementation: Addressing Non-Revenue Water in Jordan",
+    "url": "https://arxiv.org/abs/2606.15709",
+    "published": "2026-06-14",
+    "updated": "2026-06-14",
+    "authors": [
+      "Mohammed Fasha",
+      "Nahel Al-Maayta",
+      "Bilal Sowan",
+      "Mohammad Athamneh",
+      "Husam Barham"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.15709",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15709",
+    "pdf_url": "https://arxiv.org/pdf/2606.15709",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2606.15874",
+    "title": "LLM-as-Code: Agentic Programming for Agent Harness",
+    "url": "https://arxiv.org/abs/2606.15874",
+    "published": "2026-06-14",
+    "updated": "2026-06-22",
+    "authors": [
+      "Junjia Qi",
+      "Zichuan Fu",
+      "Jingtong Gao",
+      "Wenlin Zhang",
+      "Hanyu Yan",
+      "Xian Wu",
+      "Xiangyu Zhao"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.SE"
+    ],
+    "topics": [
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.15874",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15874",
+    "pdf_url": "https://arxiv.org/pdf/2606.15874",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.15906",
+    "title": "MAGE-RAG: Multigranular Adaptive Graph Evidence for Agentic Multimodal RAG in Long-Document QA",
+    "url": "https://arxiv.org/abs/2606.15906",
+    "published": "2026-06-14",
+    "updated": "2026-06-14",
+    "authors": [
+      "Yilong Zuo",
+      "Xunkai Li",
+      "Jing Yuan",
+      "Qiangqiang Dai",
+      "Hongchao Qin",
+      "Ronghua Li"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI",
+      "cs.CL",
+      "cs.DB",
+      "cs.MM"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "rag"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.15906",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15906",
+    "pdf_url": "https://arxiv.org/pdf/2606.15906",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.15994",
+    "title": "Agentic Framework for Deep Learning workload migration via In-Context Learning",
+    "url": "https://arxiv.org/abs/2606.15994",
+    "published": "2026-06-14",
+    "updated": "2026-06-14",
+    "authors": [
+      "Qiyue Liang",
+      "Steven Ingram",
+      "George Vanica",
+      "Andi Gavrilescu",
+      "Newfel Harrat",
+      "Hassan Sipra",
+      "Sethuraman Sankaran"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.15994",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15994",
+    "pdf_url": "https://arxiv.org/pdf/2606.15994",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.15034",
+    "title": "OSGuard: A Benchmark for Safety in Computer-Use Agents",
+    "url": "https://arxiv.org/abs/2606.15034",
+    "published": "2026-06-13",
+    "updated": "2026-06-13",
+    "authors": [
+      "Mina Mohammadmirzaei",
+      "Jeffrey Flanigan"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.15034",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15034",
+    "pdf_url": "https://arxiv.org/pdf/2606.15034",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.12837",
+    "title": "LoHoSearch: Benchmarking Long-Horizon Search Agents Beyond the Human Difficulty Ceiling",
+    "url": "https://arxiv.org/abs/2606.12837",
+    "published": "2026-06-11",
+    "updated": "2026-06-17",
+    "authors": [
+      "Jiarui Zhao",
+      "Rongzhi Zhang",
+      "Lingchuan Liu",
+      "Hao Yang",
+      "Xunliang Cai",
+      "Xi Su"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.12837",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12837",
+    "pdf_url": "https://arxiv.org/pdf/2606.12837",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.13663",
+    "title": "HyperTool: Beyond Step-Wise Tool Calls for Tool-Augmented Agents",
+    "url": "https://arxiv.org/abs/2606.13663",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Yaxin Du",
+      "Yifan Zhou",
+      "Yujie Ge",
+      "Jiajun Wang",
+      "Xianghe Pang",
+      "Shuo Tang",
+      "Tuney Zheng",
+      "Bryan Dai",
+      "Jian Yang",
+      "Siheng Chen"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.13663",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13663",
+    "pdf_url": "https://arxiv.org/pdf/2606.13663",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.12634",
+    "title": "Keep Policy Gradient in Charge: Sibling-Guided Credit Distillation for Long-Horizon Tool-Use Agents",
+    "url": "https://arxiv.org/abs/2606.12634",
+    "published": "2026-06-10",
+    "updated": "2026-06-29",
+    "authors": [
+      "Tianyu Ding",
+      "Jianhong Xin",
+      "Juan Pablo De la Cruz Weinstein"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.12634",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12634",
+    "pdf_url": "https://arxiv.org/pdf/2606.12634",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.11688",
+    "title": "Goal-Autopilot: A Verifiable Anti-Fabrication Firewall for Unattended Long-Horizon Agents",
+    "url": "https://arxiv.org/abs/2606.11688",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Youwang Deng"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "rag"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.11688",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.11688",
+    "pdf_url": "https://arxiv.org/pdf/2606.11688",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.10394",
+    "title": "STAGE-Claw: Automated State-based Agent Benchmarking for Realistic Scenarios",
+    "url": "https://arxiv.org/abs/2606.10394",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Sirui Liang",
+      "Bohan Yu",
+      "Peiyu Wang",
+      "Shiguang Guo",
+      "Wenxing Hu",
+      "Pengfei Cao",
+      "Jian Zhao",
+      "Cao Liu",
+      "Ke Zeng",
+      "Xunliang Cai",
+      "Kang Liu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.10394",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10394",
+    "pdf_url": "https://arxiv.org/pdf/2606.10394",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.10532",
+    "title": "ActiveMem: Distributed Active Memory for Long-Horizon LLM Reasoning",
+    "url": "https://arxiv.org/abs/2606.10532",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Yunhan Jiang",
+      "Wenbin Duan",
+      "Shasha Guo",
+      "Liang Pang",
+      "Xiaoqian Sun",
+      "Huawei Shen"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.10532",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10532",
+    "pdf_url": "https://arxiv.org/pdf/2606.10532",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.11176",
+    "title": "Data Journalist Agent: Transforming Data into Verifiable Multimodal Stories",
+    "url": "https://arxiv.org/abs/2606.11176",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Kevin Qinghong Lin",
+      "Batu EI",
+      "Yuhong Shi",
+      "Pan Lu",
+      "Philip Torr",
+      "James Zou"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.CL",
+      "cs.CY",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.11176",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.11176",
+    "pdf_url": "https://arxiv.org/pdf/2606.11176",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.08960",
+    "title": "Hardening Agent Benchmarks with Adversarial Hacker-Fixer Loops",
+    "url": "https://arxiv.org/abs/2606.08960",
+    "published": "2026-06-08",
+    "updated": "2026-06-08",
+    "authors": [
+      "Ziqian Zhong",
+      "Ivgeni Segal",
+      "Ivan Bercovich",
+      "Shashwat Saxena",
+      "Kexun Zhang",
+      "Aditi Raghunathan"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.LG",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.08960",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.08960",
+    "pdf_url": "https://arxiv.org/pdf/2606.08960",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.09447",
+    "title": "AliyunConsoleAgent: Training Web Agents in Real-World Cloud Environments via Distillation and Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2606.09447",
+    "published": "2026-06-08",
+    "updated": "2026-06-08",
+    "authors": [
+      "Bojie Rong",
+      "Zheyu Shen",
+      "Qiaoping Wang",
+      "Pengfei Kang",
+      "Yang Xu",
+      "Yawen Wei",
+      "Hanyu Wu",
+      "Zhi Zhao",
+      "Leihao Pei",
+      "Linquan Jiang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.09447",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09447",
+    "pdf_url": "https://arxiv.org/pdf/2606.09447",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.09426",
+    "title": "WeaveBench: A Long-Horizon, Real-World Benchmark for Computer-Use Agents with Hybrid Interfaces",
+    "url": "https://arxiv.org/abs/2606.09426",
+    "published": "2026-06-08",
+    "updated": "2026-07-06",
+    "authors": [
+      "Wanli Li",
+      "Bowen Zhou",
+      "Yunyao Yu",
+      "Zhou Xu",
+      "Yifan Yang",
+      "Dongsheng Li",
+      "Caihua Shan"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.09426",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09426",
+    "pdf_url": "https://arxiv.org/pdf/2606.09426",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.09316",
+    "title": "Anything2Skill: Compiling External Knowledge into Reusable Skills for Agents",
+    "url": "https://arxiv.org/abs/2606.09316",
+    "published": "2026-06-08",
+    "updated": "2026-06-19",
+    "authors": [
+      "Qianjun Pan",
+      "Yutao Yang",
+      "Junsong Li",
+      "Jie Zhou",
+      "Kai Chen",
+      "Xin Li",
+      "Qin Chen",
+      "Liang He"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.09316",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09316",
+    "pdf_url": "https://arxiv.org/pdf/2606.09316",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.09961",
+    "title": "3SPO: State-Score-Supervised Policy Optimization for LLM Agents",
+    "url": "https://arxiv.org/abs/2606.09961",
+    "published": "2026-06-08",
+    "updated": "2026-06-08",
+    "authors": [
+      "Yu Han",
+      "Kailing Li",
+      "Yang Jiao",
+      "Yulin Dai",
+      "Yuqian Fu",
+      "Linhai Zhuo",
+      "Tianwen Qian"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.09961",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09961",
+    "pdf_url": "https://arxiv.org/pdf/2606.09961",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.08172",
+    "title": "The Governance of Human-LLM Interaction: Safety Gating, Civility Steering, and Affective Default Lock-In",
+    "url": "https://arxiv.org/abs/2606.08172",
+    "published": "2026-06-06",
+    "updated": "2026-06-06",
+    "authors": [
+      "Manuele Reani",
+      "Hongjian Zhang",
+      "Hongyu Tian"
+    ],
+    "categories": [
+      "cs.HC",
+      "cs.AI",
+      "cs.CY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "planning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.08172",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.08172",
+    "pdf_url": "https://arxiv.org/pdf/2606.08172",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.08162",
+    "title": "Silent Failure in LLM Agent Systems: The Entropy Principle and the Inevitable Disorder of Autonomous Agents",
+    "url": "https://arxiv.org/abs/2606.08162",
+    "published": "2026-06-06",
+    "updated": "2026-06-06",
+    "authors": [
+      "Dexing Liu"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "memory",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.08162",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.08162",
+    "pdf_url": "https://arxiv.org/pdf/2606.08162",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.07836",
+    "title": "Agentic multi-fidelity learning of quasiparticle and excitonic properties",
+    "url": "https://arxiv.org/abs/2606.07836",
+    "published": "2026-06-05",
+    "updated": "2026-06-05",
+    "authors": [
+      "Arnab Neogi",
+      "Aaron Forde",
+      "Christopher A. Lane",
+      "Sergei Tretiak",
+      "Jian-Xin Zhu"
+    ],
+    "categories": [
+      "cond-mat.mtrl-sci",
+      "cond-mat.stat-mech",
+      "cs.AI",
+      "physics.comp-ph",
+      "quant-ph"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.07836",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.07836",
+    "pdf_url": "https://arxiv.org/pdf/2606.07836",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.18272",
+    "title": "Mitigating Anchoring Bias in LLM-Based Agents for Energy-Efficient 6G Autonomous Networks",
+    "url": "https://arxiv.org/abs/2606.18272",
+    "published": "2026-06-05",
+    "updated": "2026-06-18",
+    "authors": [
+      "Hatim Chergui",
+      "Claudia Carballo González",
+      "Farhad Rezazadeh",
+      "Merouane Debbah"
+    ],
+    "categories": [
+      "cs.NI",
+      "cs.AI",
+      "eess.SY"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "reasoning"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.18272",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18272",
+    "pdf_url": "https://arxiv.org/pdf/2606.18272",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.05658",
+    "title": "Agent-Orchestrated Adaptive RAG: A Comparative Study on Structured and Multi-Hop Retrieval",
+    "url": "https://arxiv.org/abs/2606.05658",
+    "published": "2026-06-04",
+    "updated": "2026-06-04",
+    "authors": [
+      "Anuj Maharjan",
+      "Devinder Kaur",
+      "Richard Molyet"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.05658",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.05658",
+    "pdf_url": "https://arxiv.org/pdf/2606.05658",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.05622",
+    "title": "AdaPlanBench: Evaluating Adaptive Planning in Large Language Model Agents under World and User Constraints",
+    "url": "https://arxiv.org/abs/2606.05622",
+    "published": "2026-06-04",
+    "updated": "2026-06-04",
+    "authors": [
+      "Jiayu Liu",
+      "Cheng Qian",
+      "Zhenhailong Wang",
+      "Bingxuan Li",
+      "Jiateng Liu",
+      "Heng Wang",
+      "Jeonghwan Kim",
+      "Yumeng Wang",
+      "Xiusi Chen",
+      "Yi R. Fung",
+      "Heng Ji"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.05622",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.05622",
+    "pdf_url": "https://arxiv.org/pdf/2606.05622",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.05436",
+    "title": "Ten Headache Specialists versus Artificial Intelligence for Clinical Literature Summarization: A Critical Evaluation and Comparison",
+    "url": "https://arxiv.org/abs/2606.05436",
+    "published": "2026-06-03",
+    "updated": "2026-06-03",
+    "authors": [
+      "Alejandro Lozano",
+      "Keiko Ihara",
+      "Ping-Hao Yang",
+      "Carrie E. Robertson",
+      "Jennifer Stern",
+      "Allan Purdy",
+      "Hsiangkuo Yuan",
+      "Pengfei Zhang",
+      "Yulia Orlova",
+      "Olga Fermo",
+      "Jennifer Hranilovich",
+      "Fred Cohen",
+      "Todd J. Schwedt",
+      "Jenelle A. Jindal",
+      "Serena Yeung-Levy",
+      "Chia-Chun Chiang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.05436",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.05436",
+    "pdf_url": "https://arxiv.org/pdf/2606.05436",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.03544",
+    "title": "SAGE: A Quantitative Evaluation of Socialized Evolution in Agent Ecosystems",
+    "url": "https://arxiv.org/abs/2606.03544",
+    "published": "2026-06-02",
+    "updated": "2026-06-02",
+    "authors": [
+      "Linyue Pan",
+      "Yaoming Zhu",
+      "Lin Qiu",
+      "Xuezhi Cao",
+      "Xunliang Cai"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.03544",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.03544",
+    "pdf_url": "https://arxiv.org/pdf/2606.03544",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.03157",
+    "title": "ClinicalMC: A Benchmark for Multi-Course Clinical Decision-Making with Large Language Models",
+    "url": "https://arxiv.org/abs/2606.03157",
+    "published": "2026-06-02",
+    "updated": "2026-06-02",
+    "authors": [
+      "Ruihui Hou",
+      "Siyi Zhu",
+      "Ziyue Huai",
+      "Guangya Yu",
+      "Yongqi Fan",
+      "Chunming Wang",
+      "Tong Ruan"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.03157",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.03157",
+    "pdf_url": "https://arxiv.org/pdf/2606.03157",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.01961",
+    "title": "AutoMedBench: Towards Medical AutoResearch with Agentic AI Models",
+    "url": "https://arxiv.org/abs/2606.01961",
+    "published": "2026-06-01",
+    "updated": "2026-06-03",
+    "authors": [
+      "Junqi Liu",
+      "Selena Song",
+      "Yuhan Wang",
+      "Jiawei Mao",
+      "Hardy Chen",
+      "Xiaoke Huang",
+      "Tianhao Qi",
+      "Pengfei Guo",
+      "Yucheng Tang",
+      "Yufan He",
+      "Can Zhao",
+      "Andriy Myronenko",
+      "Dong Yang",
+      "Daguang Xu",
+      "Yuyin Zhou"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.01961",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.01961",
+    "pdf_url": "https://arxiv.org/pdf/2606.01961",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.01185",
+    "title": "\"Skill issues'': data-centric optimization of lakehouse agents",
+    "url": "https://arxiv.org/abs/2606.01185",
+    "published": "2026-05-31",
+    "updated": "2026-05-31",
+    "authors": [
+      "Nicole Rose Schneider",
+      "Davide Ghilardi",
+      "Giacomo Piccinini",
+      "Jacopo Tagliabue"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.01185",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.01185",
+    "pdf_url": "https://arxiv.org/pdf/2606.01185",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.01138",
+    "title": "memorywire: A Vendor-Neutral Wire Format for Agent Memory Operations",
+    "url": "https://arxiv.org/abs/2606.01138",
+    "published": "2026-05-31",
+    "updated": "2026-06-03",
+    "authors": [
+      "Thamilvendhan Munirathinam"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.DC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.01138",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.01138",
+    "pdf_url": "https://arxiv.org/pdf/2606.01138",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.01166",
+    "title": "BraveGuard: From Open-World Threats to Safer Computer-Use Agents",
+    "url": "https://arxiv.org/abs/2606.01166",
+    "published": "2026-05-31",
+    "updated": "2026-06-02",
+    "authors": [
+      "Yunhao Feng",
+      "Xiaohu Du",
+      "Xinhao Deng",
+      "Yifan Ding",
+      "Ming Wen",
+      "Yixu Wang",
+      "Yuxiang Xie",
+      "Baihui Zheng",
+      "Yingshui Tan",
+      "Yige Li",
+      "Yutao Wu",
+      "Kerui Cao",
+      "Wenke Huang",
+      "Yanming Guo",
+      "Xingjun Ma",
+      "Yu-Gang Jiang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2606.01166",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.01166",
+    "pdf_url": "https://arxiv.org/pdf/2606.01166",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.00644",
+    "title": "ForeSci: Evaluating LLM Agents for Forward-Looking AI Research Judgment",
+    "url": "https://arxiv.org/abs/2606.00644",
+    "published": "2026-05-30",
+    "updated": "2026-06-04",
+    "authors": [
+      "Qiuyu Tian",
+      "Haojie Yin",
+      "Yingce Xia",
+      "Youyong Kong",
+      "Zequn Liu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.00644",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.00644",
+    "pdf_url": "https://arxiv.org/pdf/2606.00644",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.31308",
+    "title": "TraceGraph: Shared Decision Landscapes for Diagnosing and Improving Agent Trajectories",
+    "url": "https://arxiv.org/abs/2605.31308",
+    "published": "2026-05-29",
+    "updated": "2026-05-29",
+    "authors": [
+      "Junjie Nian",
+      "Kang Chen",
+      "Ge Zhang",
+      "Yixin Cao",
+      "Yugang Jiang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "embodied-agent",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2605.31308",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.31308",
+    "pdf_url": "https://arxiv.org/pdf/2605.31308",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2605.31075",
+    "title": "Task-Focused Memorization for Multimodal Agents",
+    "url": "https://arxiv.org/abs/2605.31075",
+    "published": "2026-05-29",
+    "updated": "2026-05-29",
+    "authors": [
+      "Tao Zou",
+      "Yichen He",
+      "Tian Qiu",
+      "Yuan Lin",
+      "Hang Li"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "embodied-agent",
+      "memory"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.31075",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.31075",
+    "pdf_url": "https://arxiv.org/pdf/2605.31075",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.31268",
+    "title": "Mellum2 Technical Report",
+    "url": "https://arxiv.org/abs/2605.31268",
+    "published": "2026-05-29",
+    "updated": "2026-05-29",
+    "authors": [
+      "Marko Kojic",
+      "Ivan Bondyrev",
+      "Aral de Moor",
+      "Joseph Shtok",
+      "Petr Borovlev",
+      "Kseniia Lysaniuk",
+      "Madeeswaran Kannan",
+      "Ivan Dolgov",
+      "Nikita Pavlichenko"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2605.31268",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.31268",
+    "pdf_url": "https://arxiv.org/pdf/2605.31268",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2605.29653",
+    "title": "PTCG-Bench: Can LLM Agents Master Pokémon Trading Card Game?",
+    "url": "https://arxiv.org/abs/2605.29653",
+    "published": "2026-05-28",
+    "updated": "2026-05-28",
+    "authors": [
+      "Dongdong Hua",
+      "Yifei Sun",
+      "Renhong Huang",
+      "Feng Gao",
+      "Chunping Wang",
+      "Yang Yang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.29653",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.29653",
+    "pdf_url": "https://arxiv.org/pdf/2605.29653",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2605.29630",
+    "title": "Entity-Collision: A Stratified Protocol for Attributing Retrieval Lift in Agent Memory",
+    "url": "https://arxiv.org/abs/2605.29630",
+    "published": "2026-05-28",
+    "updated": "2026-05-28",
+    "authors": [
+      "Youwang Deng"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.29630",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.29630",
+    "pdf_url": "https://arxiv.org/pdf/2605.29630",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.07591",
+    "title": "ResearchClawBench: A Benchmark for End-to-End Autonomous Scientific Research",
+    "url": "https://arxiv.org/abs/2606.07591",
+    "published": "2026-05-28",
+    "updated": "2026-07-03",
+    "authors": [
+      "Wanghan Xu",
+      "Shuo Li",
+      "Tianlin Ye",
+      "Qinglong Cao",
+      "Yixin Chen",
+      "Hengjian Gao",
+      "Yiheng Wang",
+      "Qi Li",
+      "Kun Li",
+      "Sheng Xu",
+      "Shengdu Chai",
+      "Fangchen Yu",
+      "Xiangyu Zhao",
+      "Zhangrui Zhao",
+      "Weijie Ma",
+      "Zijie Guo",
+      "Koutian Wu",
+      "Haoyu Zhou",
+      "Haoxiang Yin",
+      "Lixue Cheng",
+      "Chaofan Hu",
+      "Haoxuan Li",
+      "Lu Mi",
+      "Xuxuan Xie",
+      "Yifan Zhou",
+      "Ruizhe Chen",
+      "Zhiwang Zhou",
+      "Xingjian Guo",
+      "Yuhao Zhou",
+      "Xuming He",
+      "Shengyuan Xu",
+      "Xinyu Gu",
+      "Jiamin Wu",
+      "Mianxin Liu",
+      "Chunfeng Song",
+      "Fenghua Ling",
+      "Dongzhan Zhou",
+      "Shixiang Tang",
+      "Yuqiang Li",
+      "Mao Su",
+      "Peng Ye",
+      "Siqi Sun",
+      "Bin Wang",
+      "Xue Yang",
+      "Zhenfei Yin",
+      "Tianfan Fu",
+      "Guangtao Zhai",
+      "Wanli Ouyang",
+      "Bo Zhang",
+      "Lei Bai",
+      "Wenlong Zhang"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "rag"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.07591",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.07591",
+    "pdf_url": "https://arxiv.org/pdf/2606.07591",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.28617",
+    "title": "LACUNA: Safe Agents as Recursive Program Holes",
+    "url": "https://arxiv.org/abs/2605.28617",
+    "published": "2026-05-27",
+    "updated": "2026-05-27",
+    "authors": [
+      "Yaoyu Zhao",
+      "Yichen Xu",
+      "Oliver Bračevac",
+      "Cao Nguyen Pham",
+      "Frank Zhengqing Wu",
+      "Martin Odersky"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.PL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.28617",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.28617",
+    "pdf_url": "https://arxiv.org/pdf/2605.28617",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.28424",
+    "title": "Skill0.5: Joint Skill Internalization and Utilization for Out-of-Distribution Generalization in Agentic Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2605.28424",
+    "published": "2026-05-27",
+    "updated": "2026-05-27",
+    "authors": [
+      "Jiapeng Zhu",
+      "Jianxiang Yu",
+      "Yibo Zhao",
+      "Chengcheng Han",
+      "Qi Gu",
+      "Xunliang Cai",
+      "Xiang Li",
+      "Weining Qian"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-safety",
+      "memory"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.28424",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.28424",
+    "pdf_url": "https://arxiv.org/pdf/2605.28424",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.26497",
+    "title": "Aligning Provenance with Authorization: A Dual-Graph Defense for LLM Agents",
+    "url": "https://arxiv.org/abs/2605.26497",
+    "published": "2026-05-26",
+    "updated": "2026-05-26",
+    "authors": [
+      "Peiran Wang",
+      "Ying Li",
+      "Yuan Tian"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-safety",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.26497",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.26497",
+    "pdf_url": "https://arxiv.org/pdf/2605.26497",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.27123",
+    "title": "Rethinking Agentic RAG: Toward LLM-Driven Logical Retrieval Beyond Embeddings",
+    "url": "https://arxiv.org/abs/2605.27123",
+    "published": "2026-05-26",
+    "updated": "2026-05-26",
+    "authors": [
+      "Yuqi Zeng",
+      "Qixiang Deng",
+      "Yulei Wan",
+      "Ruiquan Jiang",
+      "Xiaoqing Zheng",
+      "Xuanjing Huang"
+    ],
+    "categories": [
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.27123",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.27123",
+    "pdf_url": "https://arxiv.org/pdf/2605.27123",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.26165",
+    "title": "Tool-Schema Compression Enables Agentic RAG Under Constrained Context Budgets",
+    "url": "https://arxiv.org/abs/2605.26165",
+    "published": "2026-05-24",
+    "updated": "2026-05-24",
+    "authors": [
+      "Furkan Sakizli"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.26165",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.26165",
+    "pdf_url": "https://arxiv.org/pdf/2605.26165",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.23899",
+    "title": "From Raw Experience to Skill Consumption: A Systematic Study of Model-Generated Agent Skills",
+    "url": "https://arxiv.org/abs/2605.23899",
+    "published": "2026-05-22",
+    "updated": "2026-05-22",
+    "authors": [
+      "Zisu Huang",
+      "Jingwen Xu",
+      "Yifan Yang",
+      "Ziyang Gong",
+      "Qihao Yang",
+      "Muzhao Tian",
+      "Xiaohua Wang",
+      "Changze Lv",
+      "Xuemei Gao",
+      "Qi Dai",
+      "Bei Liu",
+      "Kai Qiu",
+      "Xue Yang",
+      "Dongdong Chen",
+      "Xiaoqing Zheng",
+      "Chong Luo"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.23899",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.23899",
+    "pdf_url": "https://arxiv.org/pdf/2605.23899",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.17075",
+    "title": "A Red Teaming Framework for Evaluating Robustness of AI-enabled Security Orchestration, Automation, and Response Systems",
+    "url": "https://arxiv.org/abs/2605.17075",
+    "published": "2026-05-16",
+    "updated": "2026-05-16",
+    "authors": [
+      "Ayan Javeed Shaikh",
+      "Nathaniel D. Bastian",
+      "Ankit Shah"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.17075",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.17075",
+    "pdf_url": "https://arxiv.org/pdf/2605.17075",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.14322",
+    "title": "Are Agents Ready to Teach? A Multi-Stage Benchmark for Real-World Teaching Workflows",
+    "url": "https://arxiv.org/abs/2605.14322",
+    "published": "2026-05-14",
+    "updated": "2026-05-20",
+    "authors": [
+      "Zixin Chen",
+      "Peng Liu",
+      "Rui Sheng",
+      "Haobo Li",
+      "Jianhong Tu",
+      "Xiaodong Deng",
+      "Kashun Shum",
+      "Dayiheng Liu",
+      "Huamin Qu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.14322",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.14322",
+    "pdf_url": "https://arxiv.org/pdf/2605.14322",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.11928",
+    "title": "When Simulation Lies: A Sim-to-Real Benchmark and Domain-Randomized RL Recipe for Tool-Use Agents",
+    "url": "https://arxiv.org/abs/2605.11928",
+    "published": "2026-05-12",
+    "updated": "2026-05-12",
+    "authors": [
+      "Xiaolin Zhou",
+      "Aojie Yuan",
+      "Zheng Luo",
+      "Zipeng Ling",
+      "Xixiao Pan",
+      "Yicheng Gao",
+      "Haiyue Zhang",
+      "Jiate Li",
+      "Shuli Jiang",
+      "Prince Zizhuang Wang",
+      "Zixuan Zhu",
+      "Jinbo Liu",
+      "Ryan A. Rossi",
+      "Hua Wei",
+      "Xiyang Hu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling",
+      "language-agent"
+    ],
+    "arxiv_id": "2605.11928",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.11928",
+    "pdf_url": "https://arxiv.org/pdf/2605.11928",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2605.10870",
+    "title": "Remember the Decision, Not the Description: A Rate-Distortion Framework for Agent Memory",
+    "url": "https://arxiv.org/abs/2605.10870",
+    "published": "2026-05-11",
+    "updated": "2026-05-11",
+    "authors": [
+      "Mingxi Zou",
+      "Zhihan Guo",
+      "Langzhang Liang",
+      "Zhuo Wang",
+      "Qifan Wang",
+      "Qingsong Wen",
+      "Irwin King",
+      "Lizhen Qu",
+      "Zenglin Xu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.10870",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.10870",
+    "pdf_url": "https://arxiv.org/pdf/2605.10870",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.08876",
+    "title": "OTora: A Unified Red Teaming Framework for Reasoning-Level Denial-of-Service in LLM Agents",
+    "url": "https://arxiv.org/abs/2605.08876",
+    "published": "2026-05-09",
+    "updated": "2026-06-07",
+    "authors": [
+      "Xinyu Li",
+      "Ronghui Mu",
+      "Lin Li",
+      "Tianjin Huang",
+      "Gaojie Jin"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "computer-use",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.08876",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.08876",
+    "pdf_url": "https://arxiv.org/pdf/2605.08876",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.03328",
+    "title": "LLM-ADAM: A Generalizable LLM Agent Framework for Pre-Print Anomaly Detection in Additive Manufacturing",
+    "url": "https://arxiv.org/abs/2605.03328",
+    "published": "2026-05-05",
+    "updated": "2026-05-05",
+    "authors": [
+      "Ahmadreza Eslaminia",
+      "Chuhan Cai",
+      "Cameron Smith",
+      "Ruo-Syuan Mei",
+      "Shichen Li",
+      "Rajiv Malhotra",
+      "Klara Nahrstedt",
+      "Chenhui Shao"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.03328",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.03328",
+    "pdf_url": "https://arxiv.org/pdf/2605.03328",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2604.27092",
+    "title": "End-to-end autonomous scientific discovery on a real optical platform",
+    "url": "https://arxiv.org/abs/2604.27092",
+    "published": "2026-04-29",
+    "updated": "2026-04-29",
+    "authors": [
+      "Shuxing Yang",
+      "Fujia Chen",
+      "Rui Zhao",
+      "Junyao Wu",
+      "Yize Wang",
+      "Haiyao Luo",
+      "Ning Han",
+      "Qiaolu Chen",
+      "Yuze Hu",
+      "Wenhao Li",
+      "Mingzhu Li",
+      "Hongsheng Chen",
+      "Yihao Yang"
+    ],
+    "categories": [
+      "cs.AI",
+      "physics.optics"
+    ],
+    "topics": [
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2604.27092",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.27092",
+    "pdf_url": "https://arxiv.org/pdf/2604.27092",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2604.20994",
+    "title": "Breaking MCP with Function Hijacking Attacks: Novel Threats for Function Calling and Agentic Models",
+    "url": "https://arxiv.org/abs/2604.20994",
+    "published": "2026-04-22",
+    "updated": "2026-04-22",
+    "authors": [
+      "Yannis Belkhiter",
+      "Giulio Zizzo",
+      "Sergio Maffeis",
+      "Seshu Tirupathi",
+      "John D. Kelleher"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-safety",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2604.20994",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.20994",
+    "pdf_url": "https://arxiv.org/pdf/2604.20994",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2603.28900",
+    "title": "Robust Multi-Agent Reinforcement Learning for Small UAS Separation Assurance under GPS Degradation and Spoofing",
+    "url": "https://arxiv.org/abs/2603.28900",
+    "published": "2026-03-30",
+    "updated": "2026-03-30",
+    "authors": [
+      "Alex Zongo",
+      "Filippos Fotiadis",
+      "Ufuk Topcu",
+      "Peng Wei"
+    ],
+    "categories": [
+      "cs.RO",
+      "cs.AI",
+      "cs.LG",
+      "eess.SY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "world-model"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.28900",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.28900",
+    "pdf_url": "https://arxiv.org/pdf/2603.28900",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.25353",
+    "title": "SafeGuard ASF: SR Agentic Humanoid Robot System for Autonomous Industrial Safety",
+    "url": "https://arxiv.org/abs/2603.25353",
+    "published": "2026-03-26",
+    "updated": "2026-03-26",
+    "authors": [
+      "Thanh Nguyen Canh",
+      "Thang Tran Viet",
+      "Thanh Tuan Tran",
+      "Ben Wei Lim"
+    ],
+    "categories": [
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-safety",
+      "embodied-agent",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.25353",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.25353",
+    "pdf_url": "https://arxiv.org/pdf/2603.25353",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.19684",
+    "title": "TSegAgent: Zero-Shot Tooth Segmentation via Geometry-Aware Vision-Language Agents",
+    "url": "https://arxiv.org/abs/2603.19684",
+    "published": "2026-03-20",
+    "updated": "2026-06-23",
+    "authors": [
+      "Shaojie Zhuang",
+      "Lu Yin",
+      "Guangshun Wei",
+      "Yunpeng Li",
+      "Xilu Wang",
+      "Yuanfeng Zhou"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.19684",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.19684",
+    "pdf_url": "https://arxiv.org/pdf/2603.19684",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.17392",
+    "title": "Agentic Cognitive Profiling: Realigning Automated Alzheimer's Disease Detection with Clinical Construct Validity",
+    "url": "https://arxiv.org/abs/2603.17392",
+    "published": "2026-03-18",
+    "updated": "2026-03-18",
+    "authors": [
+      "Jiawen Kang",
+      "Kun Li",
+      "Dongrui Han",
+      "Jinchao Li",
+      "Junan Li",
+      "Lingwei Meng",
+      "Xixin Wu",
+      "Helen Meng"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.IR",
+      "q-bio.NC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2603.17392",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.17392",
+    "pdf_url": "https://arxiv.org/pdf/2603.17392",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2603.15666",
+    "title": "Compiled Memory: Not More Information, but More Precise Instructions for Language Agents",
+    "url": "https://arxiv.org/abs/2603.15666",
+    "published": "2026-03-12",
+    "updated": "2026-03-12",
+    "authors": [
+      "James Rhodes",
+      "George Kang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "memory",
+      "rag"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.15666",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.15666",
+    "pdf_url": "https://arxiv.org/pdf/2603.15666",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.00801",
+    "title": "The Synthetic Web: Adversarially-Curated Mini-Internets for Diagnosing Epistemic Weaknesses of Language Agents",
+    "url": "https://arxiv.org/abs/2603.00801",
+    "published": "2026-02-28",
+    "updated": "2026-02-28",
+    "authors": [
+      "Shrey Shah",
+      "Levent Ozgur"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.00801",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.00801",
+    "pdf_url": "https://arxiv.org/pdf/2603.00801",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2602.21127",
+    "title": "\"Are You Sure?\": An Empirical Study of Human Perception Vulnerability in LLM-Driven Agentic Systems",
+    "url": "https://arxiv.org/abs/2602.21127",
+    "published": "2026-02-24",
+    "updated": "2026-02-24",
+    "authors": [
+      "Xinfeng Li",
+      "Shenyu Dai",
+      "Kelong Zheng",
+      "Yue Xiao",
+      "Gelei Deng",
+      "Wei Dong",
+      "Xiaofeng Wang"
+    ],
+    "categories": [
+      "cs.HC",
+      "cs.AI",
+      "cs.CR",
+      "cs.SI"
+    ],
+    "topics": [
+      "agent-safety",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.21127",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.21127",
+    "pdf_url": "https://arxiv.org/pdf/2602.21127",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.00131",
+    "title": "Thought Virus: Viral Misalignment via Subliminal Prompting in Multi-Agent Systems",
+    "url": "https://arxiv.org/abs/2603.00131",
+    "published": "2026-02-23",
+    "updated": "2026-02-23",
+    "authors": [
+      "Moritz Weckbecker",
+      "Jonas Müller",
+      "Ben Hagag",
+      "Michael Mulet"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.00131",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.00131",
+    "pdf_url": "https://arxiv.org/pdf/2603.00131",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2602.19008",
+    "title": "Capable but Unreliable: Canonical Path Deviation as a Causal Mechanism of Agent Failure in Long-Horizon Tasks",
+    "url": "https://arxiv.org/abs/2602.19008",
+    "published": "2026-02-22",
+    "updated": "2026-02-22",
+    "authors": [
+      "Wilson Y. Lee"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2602.19008",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.19008",
+    "pdf_url": "https://arxiv.org/pdf/2602.19008",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2602.14234",
+    "title": "REDSearcher: A Scalable and Cost-Efficient Framework for Long-Horizon Search Agents",
+    "url": "https://arxiv.org/abs/2602.14234",
+    "published": "2026-02-15",
+    "updated": "2026-02-15",
+    "authors": [
+      "Zheng Chu",
+      "Xiao Wang",
+      "Jack Hong",
+      "Huiming Fan",
+      "Yuqi Huang",
+      "Yue Yang",
+      "Guohai Xu",
+      "Chenxiao Zhao",
+      "Cheng Xiang",
+      "Shengchao Hu",
+      "Dongdong Kuang",
+      "Ming Liu",
+      "Bing Qin",
+      "Xing Yu"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2602.14234",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.14234",
+    "pdf_url": "https://arxiv.org/pdf/2602.14234",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2602.14281",
+    "title": "MCPShield: A Security Cognition Layer for Adaptive Trust Calibration in Model Context Protocol Agents",
+    "url": "https://arxiv.org/abs/2602.14281",
+    "published": "2026-02-15",
+    "updated": "2026-02-24",
+    "authors": [
+      "Zhenhong Zhou",
+      "Yuanhe Zhang",
+      "Hongwei Cai",
+      "Moayad Aloqaily",
+      "Ouns Bouachir",
+      "Linsey Pang",
+      "Prakhar Mehrotra",
+      "Kun Wang",
+      "Qingsong Wen"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.14281",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.14281",
+    "pdf_url": "https://arxiv.org/pdf/2602.14281",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2602.13665",
+    "title": "HyFunc: Accelerating LLM-based Function Calls for Agentic AI through Hybrid-Model Cascade and Dynamic Templating",
+    "url": "https://arxiv.org/abs/2602.13665",
+    "published": "2026-02-14",
+    "updated": "2026-02-14",
+    "authors": [
+      "Weibin Liao",
+      "Jian-guang Lou",
+      "Haoyi Xiong"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2602.13665",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.13665",
+    "pdf_url": "https://arxiv.org/pdf/2602.13665",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2602.08082",
+    "title": "Spectral Guardrails for Agents in the Wild: Detecting Tool Use Hallucinations via Attention Topology",
+    "url": "https://arxiv.org/abs/2602.08082",
+    "published": "2026-02-08",
+    "updated": "2026-02-08",
+    "authors": [
+      "Valentin Noël"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "eess.SP"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.08082",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.08082",
+    "pdf_url": "https://arxiv.org/pdf/2602.08082",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2602.10133",
+    "title": "AgentTrace: A Structured Logging Framework for Agent System Observability",
+    "url": "https://arxiv.org/abs/2602.10133",
+    "published": "2026-02-07",
+    "updated": "2026-02-07",
+    "authors": [
+      "Adam AlSayyad",
+      "Kelvin Yuxiang Huang",
+      "Richik Pal"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.10133",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.10133",
+    "pdf_url": "https://arxiv.org/pdf/2602.10133",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2602.05386",
+    "title": "Spider-Sense: Intrinsic Risk Sensing for Efficient Agent Defense with Hierarchical Adaptive Screening",
+    "url": "https://arxiv.org/abs/2602.05386",
+    "published": "2026-02-05",
+    "updated": "2026-02-06",
+    "authors": [
+      "Zhenxiong Yu",
+      "Zhi Yang",
+      "Zhiheng Jin",
+      "Shuhe Wang",
+      "Heng Zhang",
+      "Yanlin Fei",
+      "Lingfeng Zeng",
+      "Fangqi Lou",
+      "Shuo Zhang",
+      "Tu Hu",
+      "Jingping Liu",
+      "Rongze Chen",
+      "Xingyu Zhu",
+      "Kunyi Wang",
+      "Chaofa Yuan",
+      "Xin Guo",
+      "Zhaowei Liu",
+      "Feipeng Zhang",
+      "Jie Huang",
+      "Huacan Wang",
+      "Ronghao Chen",
+      "Liwen Zhang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.05386",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.05386",
+    "pdf_url": "https://arxiv.org/pdf/2602.05386",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2602.03117",
+    "title": "AgentDyn: Are Your Agent Security Defenses Deployable in Real-World Dynamic Environments?",
+    "url": "https://arxiv.org/abs/2602.03117",
+    "published": "2026-02-03",
+    "updated": "2026-05-07",
+    "authors": [
+      "Hao Li",
+      "Ruoyao Wen",
+      "Shanghao Shi",
+      "Ning Zhang",
+      "Yevgeniy Vorobeychik",
+      "Chaowei Xiao"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.03117",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.03117",
+    "pdf_url": "https://arxiv.org/pdf/2602.03117",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2601.12988",
+    "title": "PaperGuide: Making Small Language-Model Paper-Reading Agents More Efficient",
+    "url": "https://arxiv.org/abs/2601.12988",
+    "published": "2026-01-19",
+    "updated": "2026-01-19",
+    "authors": [
+      "Zijian Wang",
+      "Tiancheng Huang",
+      "Hanqi Li",
+      "Da Ma",
+      "Lu Chen",
+      "Kai Yu"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2601.12988",
+    "source": "arxiv",
+    "source_id": "arxiv:2601.12988",
+    "pdf_url": "https://arxiv.org/pdf/2601.12988",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2601.06606",
+    "title": "CEDAR: Context Engineering for Agentic Data Science",
+    "url": "https://arxiv.org/abs/2601.06606",
+    "published": "2026-01-10",
+    "updated": "2026-04-22",
+    "authors": [
+      "Rishiraj Saha Roy",
+      "Chris Hinze",
+      "Luzian Hahn",
+      "Fabian Kuech"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "planning",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2601.06606",
+    "source": "arxiv",
+    "source_id": "arxiv:2601.06606",
+    "pdf_url": "https://arxiv.org/pdf/2601.06606",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2512.23747",
+    "title": "State-of-the-art Small Language Coder Model: Mify-Coder",
+    "url": "https://arxiv.org/abs/2512.23747",
+    "published": "2025-12-26",
+    "updated": "2025-12-26",
+    "authors": [
+      "Abhinav Parmar",
+      "Abhisek Panigrahi",
+      "Abhishek Kumar Dwivedi",
+      "Abhishek Bhattacharya",
+      "Adarsh Ramachandra",
+      "Aditya Choudhary",
+      "Aditya Garg",
+      "Aditya Raj",
+      "Alankrit Bhatt",
+      "Alpesh Yadav",
+      "Anant Vishnu",
+      "Ananthu Pillai",
+      "Ankush Kumar",
+      "Aryan Patnaik",
+      "Aswatha Narayanan S",
+      "Avanish Raj Singh",
+      "Bhavya Shree Gadda",
+      "Brijesh Pankajbhai Kachhadiya",
+      "Buggala Jahnavi",
+      "Chidurala Nithin Krishna",
+      "Chintan Shah",
+      "Chunduru Akshaya",
+      "Debarshi Banerjee",
+      "Debrup Dey",
+      "Deepa R.",
+      "Deepika B G",
+      "Faiz ur Rahman",
+      "Gagan Gayari",
+      "Gudhi Jagadeesh Kumar Naidu",
+      "Gursimar Singh",
+      "Harshal Tyagi",
+      "Harshini K",
+      "James Mani Vathalloor",
+      "Jayarama Nettar",
+      "Jayashree Gajjam",
+      "Joe Walter Sugil George",
+      "Kamalakara Sri Krishna Tadepalli",
+      "Kamalkumar Rathinasamy",
+      "Karan Chaurasia",
+      "Karthikeyan S",
+      "Kashish Arora",
+      "Kaushal Desai",
+      "Khushboo Buwade",
+      "Kiran Manjrekar",
+      "Malikireddy Venkata Sai Likhitha",
+      "Manjunath A",
+      "Mitali Mahavir Bedmutha",
+      "Mohammed Rafee Tarafdar",
+      "Nikhil Tiwari",
+      "Nikitha K Gigi",
+      "Pavan Ravikumar",
+      "Pendyala Swarnanjali",
+      "Piyush Anand",
+      "Prakash Chandrasekar",
+      "Prasanna Bhalchandra Gawade",
+      "Prasanth Sivan",
+      "Preeti Khurana",
+      "Priyanshi Babbar",
+      "Rajab Ali Mondal",
+      "Rajesh Kumar Vissapragada",
+      "Rajeshwari Ganesan",
+      "Rajeswari Koppisetti",
+      "Ramjee R.",
+      "Ramkumar Thiruppathisamy",
+      "Rani G. S.",
+      "S Reka",
+      "Samarth Gupta",
+      "Sandeep Reddy Kothakota",
+      "Sarathy K",
+      "Sathyanarayana Sampath Kumar",
+      "Saurabh Kumar",
+      "Shashank Khasare",
+      "Shenbaga Devi Venkatesh Kumar",
+      "Shiva Rama Krishna Parvatham",
+      "Shoeb Shaikh",
+      "Shrishanmathi A",
+      "Shubham Pathak",
+      "Sree Samhita Koppaka",
+      "Sreenivasa Raghavan K S",
+      "Sreeram Venkatasubramanian",
+      "Suprabha Desai Bojja",
+      "Swetha R",
+      "Syed Ahmed",
+      "Chinmai Harshitha Thota",
+      "Tushar Yadav",
+      "Veeravelly Kusumitha",
+      "V V S S Prasanth Patnaik",
+      "Vidya Sri Sesetti",
+      "Vijayakeerthi K",
+      "Vikram Raj Bakshi",
+      "Vinay K K",
+      "Vinoth Kumar Loganathan",
+      "Vipin Tiwari",
+      "Vivek Kumar Shrivastav",
+      "V Venkata Sri Datta Charan",
+      "Wasim Akhtar Khan"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2512.23747",
+    "source": "arxiv",
+    "source_id": "arxiv:2512.23747",
+    "pdf_url": "https://arxiv.org/pdf/2512.23747",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2510.24645",
+    "title": "FunReason-MT Technical Report: Advanced Data Synthesis Solution for Real-world Multi-Turn Tool-use",
+    "url": "https://arxiv.org/abs/2510.24645",
+    "published": "2025-10-28",
+    "updated": "2025-11-16",
+    "authors": [
+      "Zengzhuang Xu",
+      "Bingguang Hao",
+      "Zechuan Wang",
+      "Yuntao Wen",
+      "Xinyi Xu",
+      "Yang Liu",
+      "Long Chen",
+      "Dong Wang",
+      "Maolin Wang",
+      "Tong Zhao",
+      "Yicheng Chen",
+      "Cunyin Peng",
+      "Jinjie Gu",
+      "Leilei Gan",
+      "Xiangyu Zhao",
+      "Chenyi Zhuang",
+      "Shi Gu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2510.24645",
+    "source": "arxiv",
+    "source_id": "arxiv:2510.24645",
+    "pdf_url": "https://arxiv.org/pdf/2510.24645",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2510.04206",
+    "title": "AgentRL: Scaling Agentic Reinforcement Learning with a Multi-Turn, Multi-Task Framework",
+    "url": "https://arxiv.org/abs/2510.04206",
+    "published": "2025-10-05",
+    "updated": "2025-10-05",
+    "authors": [
+      "Hanchen Zhang",
+      "Xiao Liu",
+      "Bowen Lv",
+      "Xueqiao Sun",
+      "Bohao Jing",
+      "Iat Long Iong",
+      "Zhenyu Hou",
+      "Zehan Qi",
+      "Hanyu Lai",
+      "Yifan Xu",
+      "Rui Lu",
+      "Hongning Wang",
+      "Jie Tang",
+      "Yuxiao Dong"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2510.04206",
+    "source": "arxiv",
+    "source_id": "arxiv:2510.04206",
+    "pdf_url": "https://arxiv.org/pdf/2510.04206",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2509.13311",
+    "title": "Towards General Agentic Intelligence via Environment Scaling",
+    "url": "https://arxiv.org/abs/2509.13311",
+    "published": "2025-09-16",
+    "updated": "2025-09-16",
+    "authors": [
+      "Runnan Fang",
+      "Shihao Cai",
+      "Baixuan Li",
+      "Jialong Wu",
+      "Guangyu Li",
+      "Wenbiao Yin",
+      "Xinyu Wang",
+      "Xiaobin Wang",
+      "Liangcai Su",
+      "Zhen Zhang",
+      "Shibin Wu",
+      "Zhengwei Tao",
+      "Yong Jiang",
+      "Pengjun Xie",
+      "Fei Huang",
+      "Jingren Zhou"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2509.13311",
+    "source": "arxiv",
+    "source_id": "arxiv:2509.13311",
+    "pdf_url": "https://arxiv.org/pdf/2509.13311",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2509.02494",
+    "title": "GridMind: LLMs-Powered Agents for Power System Analysis and Operations",
+    "url": "https://arxiv.org/abs/2509.02494",
+    "published": "2025-09-02",
+    "updated": "2025-09-02",
+    "authors": [
+      "Hongwei Jin",
+      "Kibaek Kim",
+      "Jonghwan Kwon"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2509.02494",
+    "source": "arxiv",
+    "source_id": "arxiv:2509.02494",
+    "pdf_url": "https://arxiv.org/pdf/2509.02494",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2508.17094",
+    "title": "PowerChain: A Verifiable Agentic AI System for Automating Distribution Grid Analyses",
+    "url": "https://arxiv.org/abs/2508.17094",
+    "published": "2025-08-23",
+    "updated": "2025-10-21",
+    "authors": [
+      "Emmanuel O. Badmus",
+      "Peng Sang",
+      "Dimitrios Stamoulis",
+      "Amritanshu Pandey"
+    ],
+    "categories": [
+      "cs.AI",
+      "eess.SY"
+    ],
+    "topics": [
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2508.17094",
+    "source": "arxiv",
+    "source_id": "arxiv:2508.17094",
+    "pdf_url": "https://arxiv.org/pdf/2508.17094",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2508.12685",
+    "title": "ToolACE-MT: Non-Autoregressive Generation for Agentic Multi-Turn Interaction",
+    "url": "https://arxiv.org/abs/2508.12685",
+    "published": "2025-08-18",
+    "updated": "2026-02-13",
+    "authors": [
+      "Xingshan Zeng",
+      "Weiwen Liu",
+      "Lingzhi Wang",
+      "Liangyou Li",
+      "Fei Mi",
+      "Yasheng Wang",
+      "Lifeng Shang",
+      "Xin Jiang",
+      "Qun Liu"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "tool-use",
+      "world-model"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2508.12685",
+    "source": "arxiv",
+    "source_id": "arxiv:2508.12685",
+    "pdf_url": "https://arxiv.org/pdf/2508.12685",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2507.20666",
+    "title": "MIMII-Agent: Leveraging LLMs with Function Calling for Relative Evaluation of Anomalous Sound Detection",
+    "url": "https://arxiv.org/abs/2507.20666",
+    "published": "2025-07-28",
+    "updated": "2025-07-28",
+    "authors": [
+      "Harsh Purohit",
+      "Tomoya Nishida",
+      "Kota Dohi",
+      "Takashi Endo",
+      "Yohei Kawaguchi"
+    ],
+    "categories": [
+      "eess.AS",
+      "cs.AI",
+      "cs.LG",
+      "cs.SD"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2507.20666",
+    "source": "arxiv",
+    "source_id": "arxiv:2507.20666",
+    "pdf_url": "https://arxiv.org/pdf/2507.20666",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2507.20395",
+    "title": "MazeEval: A Benchmark for Testing Sequential Decision-Making in Language Models",
+    "url": "https://arxiv.org/abs/2507.20395",
+    "published": "2025-07-27",
+    "updated": "2025-07-27",
+    "authors": [
+      "Hafsteinn Einarsson"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "embodied-agent",
+      "reasoning"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2507.20395",
+    "source": "arxiv",
+    "source_id": "arxiv:2507.20395",
+    "pdf_url": "https://arxiv.org/pdf/2507.20395",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2607.06503",
+    "title": "Doomed from the Start: Early Abort of LLM Agent Episodes via a Recall-Controlled Probe Cascade",
+    "url": "https://arxiv.org/abs/2607.06503",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Kai Ruan",
+      "Zihe Huang",
+      "Ziqi Zhou",
+      "Qianshan Wei",
+      "Xuan Wang",
+      "Hao Sun"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.06503",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06503",
+    "pdf_url": "https://arxiv.org/pdf/2607.06503",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.04729",
+    "title": "RustMizan: A Compilable, Contamination-Aware Benchmarking Framework for Rust Vulnerabilities",
+    "url": "https://arxiv.org/abs/2607.04729",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Tarek Elsayed",
+      "Shiping Yang",
+      "Eunsong Koh",
+      "Sanika Goyal",
+      "Vincent Huang",
+      "Paul Ngo",
+      "Nathan Young",
+      "Mohammad Omidvar Tehrani",
+      "Alvyn Kang",
+      "Arnell Kang",
+      "Zeyu Chen",
+      "Angélica Moreira",
+      "Xuan Feng",
+      "Angel X. Chang",
+      "Nick Sumner",
+      "Steven Y. Ko"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.04729",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04729",
+    "pdf_url": "https://arxiv.org/pdf/2607.04729",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.05577",
+    "title": "Narrative World Model: Narratology-Grounded Writer Memory for Long-Form Fiction",
+    "url": "https://arxiv.org/abs/2607.05577",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Mohammad Saifullah",
+      "Thomas Kornmaier",
+      "Taaha Kazi",
+      "Vasu Sharma",
+      "Aditya Sanjiv Kanade",
+      "Aanand Kumar Yadav"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2607.05577",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05577",
+    "pdf_url": "https://arxiv.org/pdf/2607.05577",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2607.05477",
+    "title": "Decision Protocols in Multi-Agent Large Language Model Conversations",
+    "url": "https://arxiv.org/abs/2607.05477",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Lars Benedikt Kaesberg"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.05477",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05477",
+    "pdf_url": "https://arxiv.org/pdf/2607.05477",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.04235",
+    "title": "Spinning Straw into Gold: Relabeling LLM Agent Trajectories in Hindsight for Successful Demonstrations",
+    "url": "https://arxiv.org/abs/2607.04235",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Zichao Li",
+      "Gang Wu",
+      "Zichao Wang",
+      "Ruiyi Zhang",
+      "Wanrong Zhu",
+      "Ryan A. Rossi",
+      "Vlad I Morariu",
+      "Jihyung Kil"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "planning",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.04235",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04235",
+    "pdf_url": "https://arxiv.org/pdf/2607.04235",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.01764",
+    "title": "Mastermind: Strategy-grounded Learning for Repository-Scale Vulnerability Reproduction",
+    "url": "https://arxiv.org/abs/2607.01764",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Mingzhe Du",
+      "Luu Anh Tuan",
+      "Tianyi Wu",
+      "Renyang Liu",
+      "Zhijiang Guo",
+      "Dong Huang",
+      "See-Kiong Ng"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.01764",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01764",
+    "pdf_url": "https://arxiv.org/pdf/2607.01764",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.02116",
+    "title": "ContextNest: Verifiable Context Governance for Autonomous AI Agent",
+    "url": "https://arxiv.org/abs/2607.02116",
+    "published": "2026-07-02",
+    "updated": "2026-07-06",
+    "authors": [
+      "Misha Sulpovar",
+      "Benn R. Konsynski",
+      "Qaish Kanchwala",
+      "Gabe Goodhart"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent",
+      "rag-agent"
+    ],
+    "arxiv_id": "2607.02116",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02116",
+    "pdf_url": "https://arxiv.org/pdf/2607.02116",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.02389",
+    "title": "Steerability via constraints: a substrate for scalable oversight of coding agents",
+    "url": "https://arxiv.org/abs/2607.02389",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Thomas Winninger"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CR",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.02389",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02389",
+    "pdf_url": "https://arxiv.org/pdf/2607.02389",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.02802",
+    "title": "Seduced by the Narrative: Assessing Rule Adherence in Semi-Open Textual Sandboxes",
+    "url": "https://arxiv.org/abs/2607.02802",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Weiying Chen",
+      "Junlong Shen",
+      "Zhanyuan Guo",
+      "Xiaoou Zhou"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.02802",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02802",
+    "pdf_url": "https://arxiv.org/pdf/2607.02802",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.01846",
+    "title": "CLAP: Closed-Loop Training, Evaluation, and Release Control for Domain Agent Post-training",
+    "url": "https://arxiv.org/abs/2607.01846",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Fangfei Li",
+      "Chenyang Zhao",
+      "Long Wang",
+      "Feng Tian",
+      "Zhiyue Zheng",
+      "Lv Guo"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2607.01846",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01846",
+    "pdf_url": "https://arxiv.org/pdf/2607.01846",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2607.01136",
+    "title": "Skills Are Not Islands: Measuring Dependency and Risk in Agent Skill Supply Chains",
+    "url": "https://arxiv.org/abs/2607.01136",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Changguo Jia",
+      "Tianqi Zhao",
+      "Runzhi He",
+      "Minghui Zhou"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.01136",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01136",
+    "pdf_url": "https://arxiv.org/pdf/2607.01136",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.00339",
+    "title": "TRACE: State-Aware Query Processing over Temporal Evidence Graphs for Conversational Data",
+    "url": "https://arxiv.org/abs/2607.00339",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Maolin Wang",
+      "Yu Wang",
+      "Zichun Liu",
+      "Baiyuan Qiu",
+      "Chenbin Zhang",
+      "Jiguang Shen",
+      "Haoran Yang",
+      "Hao Miao"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.00339",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00339",
+    "pdf_url": "https://arxiv.org/pdf/2607.00339",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.02605",
+    "title": "A Survey of LLM-Driven Penetration Testing: Taxonomy, Co-Evolution, and Open Challenges",
+    "url": "https://arxiv.org/abs/2607.02605",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Zheyuan He",
+      "Jiaxun Dong",
+      "Zihao Li",
+      "Ting Chen",
+      "Gelei Deng",
+      "Feng Luo",
+      "Jinkun Ji",
+      "Yuanlong Cao",
+      "Xiapu Luo"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2607.02605",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02605",
+    "pdf_url": "https://arxiv.org/pdf/2607.02605",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.31518",
+    "title": "Design and Implementation of Agentic Orchestrations and Orchestration of Agents",
+    "url": "https://arxiv.org/abs/2606.31518",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Stefanie Rinderle-Ma",
+      "Juergen Mangler",
+      "Johannes Loebbecke",
+      "Dominik Voigt",
+      "Nataliia Klievtsova",
+      "Matthias Ehrendorfer"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.31518",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31518",
+    "pdf_url": "https://arxiv.org/pdf/2606.31518",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.31023",
+    "title": "Certified Speculative Execution for Untrusted AI Agents",
+    "url": "https://arxiv.org/abs/2606.31023",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Chenyu Zhou",
+      "Qiliang Jiang",
+      "Shuning Wu",
+      "Xu Zhou"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-safety",
+      "reasoning"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.31023",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31023",
+    "pdf_url": "https://arxiv.org/pdf/2606.31023",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.31613",
+    "title": "Robust Autonomous UAV Landing on Maritime Platforms via Multimodal Agentic AI and Active Wave Compensation",
+    "url": "https://arxiv.org/abs/2606.31613",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Francisco S. Neves",
+      "Pedro N. Pereira",
+      "Raul D. S. G. Campilho",
+      "Andry M. Pinto"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "world-model"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.31613",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31613",
+    "pdf_url": "https://arxiv.org/pdf/2606.31613",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.31392",
+    "title": "ReGRPO: Reflection-Augmented Policy Optimization for Tool-Using Agents",
+    "url": "https://arxiv.org/abs/2606.31392",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Binjie Zhang",
+      "Mike Zheng Shou"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "computer-use",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.31392",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31392",
+    "pdf_url": "https://arxiv.org/pdf/2606.31392",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.31461",
+    "title": "CSTrader: A Testbed for Language-Grounded Trading in a Community-Driven Virtual Asset Market",
+    "url": "https://arxiv.org/abs/2606.31461",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Yao Shi",
+      "Kingfung Luo",
+      "Nan Tang",
+      "Yuyu Luo"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.31461",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31461",
+    "pdf_url": "https://arxiv.org/pdf/2606.31461",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.31039",
+    "title": "Truth or Sophistry? LoFa: A Benchmark for LLM Robustness Against Logical Fallacies",
+    "url": "https://arxiv.org/abs/2606.31039",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Xudong Shen",
+      "Li Yuan",
+      "Ye Chen",
+      "Xin Wu",
+      "Yi Cai",
+      "Zhiyong Wu"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.31039",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31039",
+    "pdf_url": "https://arxiv.org/pdf/2606.31039",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.29722",
+    "title": "Attraction, Not Adaptation: How AI Agent Communities Develop Distinct Linguistic Identities",
+    "url": "https://arxiv.org/abs/2606.29722",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Daming Li",
+      "Simeng Han",
+      "Can Meng",
+      "Wanyu Lei",
+      "Jialu Zhang"
+    ],
+    "categories": [
+      "cs.SI"
+    ],
+    "topics": [
+      "computer-use",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.29722",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29722",
+    "pdf_url": "https://arxiv.org/pdf/2606.29722",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.30531",
+    "title": "Entity Binding Failures in Tool-Augmented Agents",
+    "url": "https://arxiv.org/abs/2606.30531",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Rahul Suresh Babu",
+      "Shashank Indukuri"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.30531",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30531",
+    "pdf_url": "https://arxiv.org/pdf/2606.30531",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.29871",
+    "title": "AI Training Manager: Bounded Closed-Loop Control of Adaptive Training Recipes",
+    "url": "https://arxiv.org/abs/2606.29871",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Anjali Rao",
+      "Nikhil Kamalkumar Advani"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "embodied-agent",
+      "memory",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.29871",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29871",
+    "pdf_url": "https://arxiv.org/pdf/2606.29871",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.30479",
+    "title": "COHORT: Collaborative Orchestration for Hardening via Offensive Replay on Emulated Topologies",
+    "url": "https://arxiv.org/abs/2606.30479",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Chen Frydman",
+      "Aviram Zilberman",
+      "Rubin Krief",
+      "Abed Showgan",
+      "Andres Murillo",
+      "Sekiya Motoyoshi",
+      "Asaf Shabtai",
+      "Yuval Elovici",
+      "Rami Puzis"
+    ],
+    "categories": [
+      "cs.NI",
+      "cs.AI",
+      "cs.CR",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.30479",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30479",
+    "pdf_url": "https://arxiv.org/pdf/2606.30479",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.29033",
+    "title": "Human-in-the-Loop Nugget Annotation for Accountable LLM-as-a-Judge Evaluations",
+    "url": "https://arxiv.org/abs/2606.29033",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Laura Dietz"
+    ],
+    "categories": [
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.29033",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29033",
+    "pdf_url": "https://arxiv.org/pdf/2606.29033",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.27936",
+    "title": "Agentic AI-Powered Re-Identification: An Emerging, Scalable Threat to Mobility Microdata Privacy",
+    "url": "https://arxiv.org/abs/2606.27936",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Oscar Thees",
+      "Roman Müller",
+      "Matthias Templ"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "stat.AP"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.27936",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27936",
+    "pdf_url": "https://arxiv.org/pdf/2606.27936",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.26790",
+    "title": "OPID: On-Policy Skill Distillation for Agentic Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2606.26790",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Shuo Yang",
+      "Jinyang Wu",
+      "Zhengxi Lu",
+      "Yuhao Shen",
+      "Fan Zhang",
+      "Lang Feng",
+      "Shuai Zhang",
+      "Haoran Luo",
+      "Zheng Lian",
+      "Zhengqi Wen",
+      "Jianhua Tao"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "computer-use",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.26790",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26790",
+    "pdf_url": "https://arxiv.org/pdf/2606.26790",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.27443",
+    "title": "When Does Personality Composition Matter for Multi-Agent LLM Teams?",
+    "url": "https://arxiv.org/abs/2606.27443",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Aryan Keluskar",
+      "Amrita Bhattacharjee",
+      "Huan Liu"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "coding-agent",
+      "multi-agent"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.27443",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27443",
+    "pdf_url": "https://arxiv.org/pdf/2606.27443",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.27409",
+    "title": "Delayed Verification Destabilizes Multi-Agent LLM Belief: Instability Thresholds and Optimal Corrector Placement",
+    "url": "https://arxiv.org/abs/2606.27409",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Igor Itkin"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.CL",
+      "cs.LG",
+      "eess.SY"
+    ],
+    "topics": [
+      "multi-agent",
+      "reasoning"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.27409",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27409",
+    "pdf_url": "https://arxiv.org/pdf/2606.27409",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.26289",
+    "title": "Augmentation with Dilution: A Large-Scale Empirical Study of Human Contributor Ecosystems After AI Coding Agent Adoption",
+    "url": "https://arxiv.org/abs/2606.26289",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Weixing Zhang",
+      "Bowen Jiang",
+      "Anne Koziolek"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent",
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.26289",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26289",
+    "pdf_url": "https://arxiv.org/pdf/2606.26289",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.25996",
+    "title": "Autodata: An agentic data scientist to create high quality synthetic data",
+    "url": "https://arxiv.org/abs/2606.25996",
+    "published": "2026-06-24",
+    "updated": "2026-07-04",
+    "authors": [
+      "Ilia Kulikov",
+      "Chenxi Whitehouse",
+      "Tianhao Wu",
+      "Yixin Nie",
+      "Swarnadeep Saha",
+      "Eryk Helenowski",
+      "Weizhe Yuan",
+      "Olga Golovneva",
+      "Jack Lanchantin",
+      "Yoram Bachrach",
+      "Jakob Foerster",
+      "Xian Li",
+      "Han Fang",
+      "Sainbayar Sukhbaatar",
+      "Jason Weston"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.25996",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25996",
+    "pdf_url": "https://arxiv.org/pdf/2606.25996",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.25836",
+    "title": "AI Snitches Get Glitches: Towards Evading Agentic Surveillance",
+    "url": "https://arxiv.org/abs/2606.25836",
+    "published": "2026-06-24",
+    "updated": "2026-06-26",
+    "authors": [
+      "Hyejun Jeong",
+      "Dzung Pham",
+      "Amir Houmansadr",
+      "Eugene Bagdasarian"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.25836",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25836",
+    "pdf_url": "https://arxiv.org/pdf/2606.25836",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.25332",
+    "title": "Decoupling Reconnaissance and Exploitation: Measuring the Capability Boundaries of LLM-Based Web Penetration Testing",
+    "url": "https://arxiv.org/abs/2606.25332",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Liwei Yu",
+      "Shuo Li",
+      "Ming Zhou",
+      "Ge Chu",
+      "Yan Guo"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.25332",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25332",
+    "pdf_url": "https://arxiv.org/pdf/2606.25332",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.24530",
+    "title": "NatureBench: Can Coding Agents Match the Published SOTA of Nature-Family Papers?",
+    "url": "https://arxiv.org/abs/2606.24530",
+    "published": "2026-06-23",
+    "updated": "2026-07-06",
+    "authors": [
+      "Yuru Wang",
+      "Lejun Cheng",
+      "Yuxin Zuo",
+      "Sihang Zeng",
+      "Bingxiang He",
+      "Che Jiang",
+      "Junlin Yang",
+      "Yuchong Wang",
+      "Kaikai Zhao",
+      "Weifeng Huang",
+      "Kai Tian",
+      "Zhenzhao Yuan",
+      "Jincheng Zhong",
+      "Weizhi Wang",
+      "Ning Ding",
+      "Bowen Zhou",
+      "Kaiyan Zhang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "rag"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.24530",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24530",
+    "pdf_url": "https://arxiv.org/pdf/2606.24530",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.24370",
+    "title": "When Helpfulness Overrides Causal Caution: Context-Dependent Suppression and Recovery in LLMs",
+    "url": "https://arxiv.org/abs/2606.24370",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Hiroshi Okumura"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "reasoning"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.24370",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24370",
+    "pdf_url": "https://arxiv.org/pdf/2606.24370",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.22916",
+    "title": "Intent-Governed Tool Authorization for AI Agents",
+    "url": "https://arxiv.org/abs/2606.22916",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Genliang Zhu",
+      "Chu Wang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.22916",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22916",
+    "pdf_url": "https://arxiv.org/pdf/2606.22916",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.23654",
+    "title": "EnterpriseClawBench: Benchmarking Agents from Real Workplace Sessions",
+    "url": "https://arxiv.org/abs/2606.23654",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Jincheng Zhong",
+      "Weizhi Wang",
+      "Che Jiang",
+      "Kai Tian",
+      "Zhenzhao Yuan",
+      "Junlin Yang",
+      "Dianqiao Lei",
+      "Kaiyan Zhang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.23654",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23654",
+    "pdf_url": "https://arxiv.org/pdf/2606.23654",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.23277",
+    "title": "GIF: Locally Sound Geometric Information Flow Control for LLMs",
+    "url": "https://arxiv.org/abs/2606.23277",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Adam Storek",
+      "Nikolaus Holzer",
+      "Zhuo Zhang",
+      "Suman Jana"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.23277",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23277",
+    "pdf_url": "https://arxiv.org/pdf/2606.23277",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.23112",
+    "title": "Self-Evolution for Multi-Turn Tool-Calling Agents via Divergence-Point Preference Learning",
+    "url": "https://arxiv.org/abs/2606.23112",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Jiaqiang Tang"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.23112",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23112",
+    "pdf_url": "https://arxiv.org/pdf/2606.23112",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.23797",
+    "title": "From Task-Guided Conversational Graphs to Goal-Oriented Dialogue Runtimes",
+    "url": "https://arxiv.org/abs/2606.23797",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Mariano Garralda-Barrio"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.CL",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.23797",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23797",
+    "pdf_url": "https://arxiv.org/pdf/2606.23797",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.22329",
+    "title": "BabelJudge: Measuring LLM-as-a-Judge Reliability Across Languages and Agent Trajectories",
+    "url": "https://arxiv.org/abs/2606.22329",
+    "published": "2026-06-21",
+    "updated": "2026-06-21",
+    "authors": [
+      "Shreyas KC"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.22329",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22329",
+    "pdf_url": "https://arxiv.org/pdf/2606.22329",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.22504",
+    "title": "Lingering Authority: Revocable Resource-and-Effect Capabilities for Coding Agents",
+    "url": "https://arxiv.org/abs/2606.22504",
+    "published": "2026-06-21",
+    "updated": "2026-06-21",
+    "authors": [
+      "Igor Santos-Grueiro"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.22504",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22504",
+    "pdf_url": "https://arxiv.org/pdf/2606.22504",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.22337",
+    "title": "Theorist Toolbox: Tools for Agent Based LLM-assisted economic theory Research",
+    "url": "https://arxiv.org/abs/2606.22337",
+    "published": "2026-06-21",
+    "updated": "2026-06-23",
+    "authors": [
+      "Moran Koren"
+    ],
+    "categories": [
+      "econ.TH",
+      "cs.GT",
+      "econ.GN"
+    ],
+    "topics": [
+      "computer-use",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.22337",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22337",
+    "pdf_url": "https://arxiv.org/pdf/2606.22337",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.22692",
+    "title": "VISTA Architect: A graph database-oriented health AI system demonstrated in multidisciplinary tumor boards",
+    "url": "https://arxiv.org/abs/2606.22692",
+    "published": "2026-06-21",
+    "updated": "2026-06-21",
+    "authors": [
+      "Tuomo Kiiskinen",
+      "Jason Fries",
+      "Philip Adamson",
+      "David Wu",
+      "Timothy John Ellis-Caleo",
+      "Aaron Fanous",
+      "Balasubramanian Narasimhan",
+      "Joel Neal",
+      "Sylvia Plevritis",
+      "Manuel A. Rivas"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.DB",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.22692",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22692",
+    "pdf_url": "https://arxiv.org/pdf/2606.22692",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.21955",
+    "title": "From RAN Control to Agentic Intelligence: Architecture and Vision for Energy Efficient AI-RAN",
+    "url": "https://arxiv.org/abs/2606.21955",
+    "published": "2026-06-20",
+    "updated": "2026-06-20",
+    "authors": [
+      "Sabrine Aroua",
+      "Alexis I. Aravanis",
+      "Ilias Chatzistefanidis",
+      "Hamza Abbar",
+      "Anh-Khoa Dang",
+      "Anastasios Giovanidis",
+      "Salah-Eddine El Ayoubi",
+      "Stephane Senecal",
+      "Martha Vlachou Konchylaki",
+      "Navid Nikaein"
+    ],
+    "categories": [
+      "cs.NI",
+      "cs.AI"
+    ],
+    "topics": [
+      "rag",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.21955",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21955",
+    "pdf_url": "https://arxiv.org/pdf/2606.21955",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.21894",
+    "title": "Skills for the future software profession: beyond agentic AI!",
+    "url": "https://arxiv.org/abs/2606.21894",
+    "published": "2026-06-20",
+    "updated": "2026-06-23",
+    "authors": [
+      "Sungmin Kang",
+      "Baishakhi Ray",
+      "Abhik Roychoudhury"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai",
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.21894",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21894",
+    "pdf_url": "https://arxiv.org/pdf/2606.21894",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.20978",
+    "title": "How Should Agents Read Demonstrations? Hierarchical Structure Beats Flat Action Logs",
+    "url": "https://arxiv.org/abs/2606.20978",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Honjar Xing",
+      "Jefferson Lin",
+      "Henry Lieberman"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.20978",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20978",
+    "pdf_url": "https://arxiv.org/pdf/2606.20978",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.20729",
+    "title": "LLM-Guided Test-Time Discovery of Quantum-Chemical Approximation Algorithms",
+    "url": "https://arxiv.org/abs/2606.20729",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Masaya Hagai",
+      "Yuta Suzuki",
+      "Tomoya Murata",
+      "Shuhei Kurita",
+      "Masaki Adachi"
+    ],
+    "categories": [
+      "physics.chem-ph",
+      "cond-mat.mtrl-sci",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.20729",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20729",
+    "pdf_url": "https://arxiv.org/pdf/2606.20729",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.19416",
+    "title": "MortarBench: Evaluating Mortgage Loan Origination Agents",
+    "url": "https://arxiv.org/abs/2606.19416",
+    "published": "2026-06-17",
+    "updated": "2026-06-22",
+    "authors": [
+      "Matthew Toles",
+      "Yunan Lu",
+      "Manav Munjal",
+      "Bojun Liu",
+      "Yuanhao Deng",
+      "Stephanie Selig",
+      "Derek Rindner",
+      "Cheng Li",
+      "Zhou Yu"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.19416",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19416",
+    "pdf_url": "https://arxiv.org/pdf/2606.19416",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.18550",
+    "title": "The Gate Is Only as Honest as Its Contracts: ContractGuard for the Contract Layer of Risk-Aware Causal Gating",
+    "url": "https://arxiv.org/abs/2606.18550",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Laxmipriya Ganesh Iyer",
+      "Rahul Suresh Babu"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.18550",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18550",
+    "pdf_url": "https://arxiv.org/pdf/2606.18550",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.19616",
+    "title": "Before the Pull Request: Mining Multi-Agent Coordination",
+    "url": "https://arxiv.org/abs/2606.19616",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Dipankar Sarkar"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "coding-agent",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.19616",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19616",
+    "pdf_url": "https://arxiv.org/pdf/2606.19616",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.18890",
+    "title": "Skill-Guided Continuation Distillation for GUI Agents",
+    "url": "https://arxiv.org/abs/2606.18890",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Zhimin Fan",
+      "Hongwei Yu",
+      "Yeqing Shen",
+      "Haolong Yan",
+      "Guozhen Peng",
+      "Tianhao Peng",
+      "Yudong Zhang",
+      "Xiaowen Zhang",
+      "Kaijun Tan",
+      "Zheng Ge",
+      "Xiangyu Zhang",
+      "Daxin Jiang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "computer-use",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.18890",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18890",
+    "pdf_url": "https://arxiv.org/pdf/2606.18890",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.19602",
+    "title": "Configurable Clinical Information Extraction with Agentic RAG: What Works, What Breaks, and Why",
+    "url": "https://arxiv.org/abs/2606.19602",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Osman Alperen Çinar-Koraş",
+      "Marie Bauer",
+      "Sameh Khattab",
+      "Merlin Engelke",
+      "Moon Kim",
+      "Stephan Settelmeier",
+      "Shigeyasu Sugawara",
+      "Fabian Freisleben",
+      "Felix Nensa",
+      "Jens Kleesiek"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.19602",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19602",
+    "pdf_url": "https://arxiv.org/pdf/2606.19602",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.30658",
+    "title": "Agentic AI Enhances Physician Trust in Clinical Decision Making",
+    "url": "https://arxiv.org/abs/2606.30658",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Zhiling Yan",
+      "Zhe Fang",
+      "David J King",
+      "Ann Pongsakul",
+      "Eashan Adhikarla",
+      "Hui Ren",
+      "Sunyang Fu",
+      "Quanzheng Li",
+      "Lifang He",
+      "Xiang Li",
+      "Hongfang Liu",
+      "Yonghui Wu",
+      "Lichao Sun"
+    ],
+    "categories": [
+      "cs.CY",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.30658",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30658",
+    "pdf_url": "https://arxiv.org/pdf/2606.30658",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.17628",
+    "title": "OPD-Evolver: Cultivating Holistic Agent Evolver via On-Policy Distillation",
+    "url": "https://arxiv.org/abs/2606.17628",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Guibin Zhang",
+      "Xun Xu",
+      "Yanwei Yue",
+      "Zikun Su",
+      "Wangchunshu Zhou",
+      "Xiaobin Hu",
+      "Shuicheng Yan"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "memory",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.17628",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17628",
+    "pdf_url": "https://arxiv.org/pdf/2606.17628",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.20724",
+    "title": "When Web Agents Finish but Still Fail: Reproducible Triggers and Trace Diagnostics for Parallel Web Exploration",
+    "url": "https://arxiv.org/abs/2606.20724",
+    "published": "2026-06-16",
+    "updated": "2026-06-29",
+    "authors": [
+      "Aagam Sogani",
+      "Botao Rui",
+      "Swetha Vaidyanathan",
+      "Rishi Agarwal",
+      "Minghao Yan",
+      "Shivaram Venkataraman"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.20724",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20724",
+    "pdf_url": "https://arxiv.org/pdf/2606.20724",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.18448",
+    "title": "VISUALSKILL: Multimodal Skills for Computer-Use Agents",
+    "url": "https://arxiv.org/abs/2606.18448",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Ziyan Jiang",
+      "Li An",
+      "Yujian Liu",
+      "Jiabao Ji",
+      "Qiucheng Wu",
+      "Jacob Andreas",
+      "Yang Zhang",
+      "Shiyu Chang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.18448",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18448",
+    "pdf_url": "https://arxiv.org/pdf/2606.18448",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.15139",
+    "title": "Self-Driving Negotiator: An interactive, verifiable benchmark for social negotiation and theory of mind under hidden intent",
+    "url": "https://arxiv.org/abs/2606.15139",
+    "published": "2026-06-13",
+    "updated": "2026-06-13",
+    "authors": [
+      "Ashutosh Kumar"
+    ],
+    "categories": [
+      "cs.GT",
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.15139",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15139",
+    "pdf_url": "https://arxiv.org/pdf/2606.15139",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.15385",
+    "title": "Reward Hacking in Language Model Agents: Revisiting AI Safety Gridworlds",
+    "url": "https://arxiv.org/abs/2606.15385",
+    "published": "2026-06-13",
+    "updated": "2026-06-13",
+    "authors": [
+      "Ömer Veysel Çağatan",
+      "Xuandong Zhao"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2606.15385",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15385",
+    "pdf_url": "https://arxiv.org/pdf/2606.15385",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.14989",
+    "title": "Hierarchical Generative Agents for Simulating Sequential Human Behavior",
+    "url": "https://arxiv.org/abs/2606.14989",
+    "published": "2026-06-12",
+    "updated": "2026-06-12",
+    "authors": [
+      "Maria G. Mendoza",
+      "Lucas Waldburger",
+      "Jin Lee",
+      "Shankar Sastry"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "embodied-agent",
+      "planning",
+      "reasoning",
+      "world-model"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.14989",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.14989",
+    "pdf_url": "https://arxiv.org/pdf/2606.14989",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.13220",
+    "title": "LLM-as-an-Investigator: Evidence-First Reasoning for Robust Interactive Problem Diagnosis",
+    "url": "https://arxiv.org/abs/2606.13220",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Fabrizio Marozzo",
+      "Pietro Liò"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CE",
+      "cs.ET",
+      "cs.LG",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "reasoning"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.13220",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13220",
+    "pdf_url": "https://arxiv.org/pdf/2606.13220",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.12290",
+    "title": "Selection Integrity for LLM Graph Memory: An Accumulability Criterion for Information-Flow-Blind Retrieval",
+    "url": "https://arxiv.org/abs/2606.12290",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Zeming Fei",
+      "Hongming Fei",
+      "Xiaoyang Wang",
+      "Yang yang",
+      "Prosanta Gope",
+      "Biplab Sikdar",
+      "Ying Zhang"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.12290",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12290",
+    "pdf_url": "https://arxiv.org/pdf/2606.12290",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.12587",
+    "title": "Strategic Decision Support for AI Agents",
+    "url": "https://arxiv.org/abs/2606.12587",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Shayan Kiyani",
+      "Sima Noorani",
+      "George Pappas",
+      "Hamed Hassani"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.HC"
+    ],
+    "topics": [
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.12587",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12587",
+    "pdf_url": "https://arxiv.org/pdf/2606.12587",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.10651",
+    "title": "Kwai Keye-VL-2.0 Technical Report",
+    "url": "https://arxiv.org/abs/2606.10651",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Kwai Keye Team",
+      "Bin Wen",
+      "Changyi Liu",
+      "Chengru Song",
+      "Chongling Rao",
+      "Guowang Zhang",
+      "Han Li",
+      "Haonan Fan",
+      "Hengrui Ju",
+      "Jiankang Chen",
+      "Jiapeng Chen",
+      "Jiawei Yuan",
+      "Kaixuan Yang",
+      "Kaiyu Jiang",
+      "Kun Gai",
+      "Lingzhi Zhou",
+      "Na Nie",
+      "Sen Na",
+      "Tianke Zhang",
+      "Tingting Gao",
+      "Xuanyu Zheng",
+      "Yulong Chen",
+      "Fan Yang",
+      "Haixuan Gao",
+      "Lele Yang",
+      "Mingqiao Liu",
+      "Muxi Diao",
+      "Qi Zhang",
+      "Qile Su",
+      "Wei Chen",
+      "Wentao Hong",
+      "Xingyu Lu",
+      "Yancheng Long",
+      "Yankai Yang",
+      "Yingxin Li",
+      "Yiyang Fan",
+      "Yu Xia",
+      "Yuzhe Chen",
+      "Ziliang Lai",
+      "Chuan Yi",
+      "Haonan Jia",
+      "Tianming Liang",
+      "Weixin Xu",
+      "Xiaoxiao Ma",
+      "Yang Tian",
+      "Yufei Han",
+      "Feng Han",
+      "Hang Li",
+      "Jing Wang",
+      "Jinghui Jia",
+      "Junmin Chen",
+      "Junyu Shi",
+      "Ruilin Zhang"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.10651",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10651",
+    "pdf_url": "https://arxiv.org/pdf/2606.10651",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.10956",
+    "title": "Mind the Gap: Can Frontier LLMs Pass a Standardized Office Proficiency Exam?",
+    "url": "https://arxiv.org/abs/2606.10956",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Tengchao Lv",
+      "Dongdong Zhang",
+      "Jiayu Ding",
+      "Yilin Jia",
+      "Yuzhong Zhao",
+      "Yupan Huang",
+      "Wenshan Wu",
+      "Xiangyang Zhou",
+      "Shaohan Huang",
+      "Nan Yang",
+      "Li Dong",
+      "Lei Cui",
+      "Furu Wei"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.10956",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10956",
+    "pdf_url": "https://arxiv.org/pdf/2606.10956",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.08661",
+    "title": "Data Agents Under Attack: Vulnerabilities in LLM-Driven Analytical Systems",
+    "url": "https://arxiv.org/abs/2606.08661",
+    "published": "2026-06-07",
+    "updated": "2026-06-07",
+    "authors": [
+      "Kuncan Wang",
+      "Ziting Wang",
+      "Peizhuo Lv",
+      "Haoyang Li",
+      "Guoliang Li",
+      "Gao Cong",
+      "Wei Dong"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.DB"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2606.08661",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.08661",
+    "pdf_url": "https://arxiv.org/pdf/2606.08661",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.08539",
+    "title": "AgentTrust: A Self-Improving Trust Layer for AI-Agent Actions",
+    "url": "https://arxiv.org/abs/2606.08539",
+    "published": "2026-06-07",
+    "updated": "2026-06-07",
+    "authors": [
+      "Chenglin Yang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.08539",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.08539",
+    "pdf_url": "https://arxiv.org/pdf/2606.08539",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.24896",
+    "title": "Why Memory Components Fail: Eight Years of License and Sustainability Events in Open-Source Data Infrastructure",
+    "url": "https://arxiv.org/abs/2606.24896",
+    "published": "2026-06-05",
+    "updated": "2026-06-05",
+    "authors": [
+      "Dmitrii Dmitrenko"
+    ],
+    "categories": [
+      "cs.DL",
+      "cs.CY"
+    ],
+    "topics": [
+      "coding-agent",
+      "memory",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.24896",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24896",
+    "pdf_url": "https://arxiv.org/pdf/2606.24896",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.07074",
+    "title": "SlimSearcher: Training Efficiency-Aware Web Agents via Adaptive Reward Gating",
+    "url": "https://arxiv.org/abs/2606.07074",
+    "published": "2026-06-05",
+    "updated": "2026-06-05",
+    "authors": [
+      "Zequn Xie",
+      "Junjie Wang",
+      "Dan Yang",
+      "Jie Feng",
+      "Yue Shen",
+      "Jian Wang",
+      "Jinjie Gu"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.07074",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.07074",
+    "pdf_url": "https://arxiv.org/pdf/2606.07074",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.07027",
+    "title": "StainFlow: Entity-Stain Tracking and Evidence Linking for Process Rewards in GUI Agents",
+    "url": "https://arxiv.org/abs/2606.07027",
+    "published": "2026-06-05",
+    "updated": "2026-06-12",
+    "authors": [
+      "Haojie Hao",
+      "Longkun Hao",
+      "Yihang Lou",
+      "Yan Bai",
+      "Zhenyang Li",
+      "Zhichao Yang",
+      "Dongshuo Huang",
+      "Hongyu Lin",
+      "Lanqing Hong",
+      "Jiakai Wang",
+      "Xianglong Liu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.07027",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.07027",
+    "pdf_url": "https://arxiv.org/pdf/2606.07027",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.07486",
+    "title": "OPENPATH: A Supervisor--Specialist Agent System for Personalized, Accessible, and Multi-stop Urban Trip Planning",
+    "url": "https://arxiv.org/abs/2606.07486",
+    "published": "2026-06-05",
+    "updated": "2026-06-05",
+    "authors": [
+      "Ziyang Xiong",
+      "He Zong",
+      "Zhiyuan Xue",
+      "Manxi Wu"
+    ],
+    "categories": [
+      "eess.SY"
+    ],
+    "topics": [
+      "multi-agent",
+      "planning"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.07486",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.07486",
+    "pdf_url": "https://arxiv.org/pdf/2606.07486",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.05553",
+    "title": "ArcANE: Do Role-Playing Language Agents Stay in Character at the Right Time?",
+    "url": "https://arxiv.org/abs/2606.05553",
+    "published": "2026-06-04",
+    "updated": "2026-06-04",
+    "authors": [
+      "Woojung Song",
+      "Nalim Kim",
+      "Sangjun Song",
+      "Chaewon Heo",
+      "Jongwon Lim",
+      "Yohan Jo"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.05553",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.05553",
+    "pdf_url": "https://arxiv.org/pdf/2606.05553",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.05901",
+    "title": "Reducing Hallucinations in Complex Question Answering using Simple Graph-based Retrieval-Augmented Generation (long version)",
+    "url": "https://arxiv.org/abs/2606.05901",
+    "published": "2026-06-04",
+    "updated": "2026-06-04",
+    "authors": [
+      "Christopher J. Wedge",
+      "Joshua Stutter",
+      "Danny Dixon",
+      "Jacek Cała"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.05901",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.05901",
+    "pdf_url": "https://arxiv.org/pdf/2606.05901",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.05233",
+    "title": "Domain-Conditioned Safety in Frontier Computer-Using Agents: A 793-Episode Browser Benchmark, a Coding-Domain Cross-Reference, and a Reproducibility Audit of Recent Red-Teaming",
+    "url": "https://arxiv.org/abs/2606.05233",
+    "published": "2026-06-03",
+    "updated": "2026-06-03",
+    "authors": [
+      "Nicholas Saban"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "computer-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.05233",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.05233",
+    "pdf_url": "https://arxiv.org/pdf/2606.05233",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.09890",
+    "title": "PreAct-Bench: Benchmarking Predictive Monitoring in LLMs",
+    "url": "https://arxiv.org/abs/2606.09890",
+    "published": "2026-06-03",
+    "updated": "2026-06-03",
+    "authors": [
+      "Hainiu Xu",
+      "Italo Luis da Silva",
+      "Jiangnan Ye",
+      "Yuhao Wang",
+      "Wei Liu",
+      "Linyi Yang",
+      "Jonathan Richard Schwarz",
+      "Nicola Paoletti",
+      "Yulan He",
+      "Hanqi Yan"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.09890",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09890",
+    "pdf_url": "https://arxiv.org/pdf/2606.09890",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.03054",
+    "title": "ToolGate: Token-Efficient Pre-Call Control for Tool-Augmented Vision-Language Agents",
+    "url": "https://arxiv.org/abs/2606.03054",
+    "published": "2026-06-02",
+    "updated": "2026-06-02",
+    "authors": [
+      "Anjie Liu",
+      "Yan Song",
+      "Zhixun Chen",
+      "Ziqin Gong",
+      "Zhongwei Yu",
+      "Jun Wang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.03054",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.03054",
+    "pdf_url": "https://arxiv.org/pdf/2606.03054",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.02994",
+    "title": "Inducing Reasoning Primitives from Agent Traces",
+    "url": "https://arxiv.org/abs/2606.02994",
+    "published": "2026-06-02",
+    "updated": "2026-06-02",
+    "authors": [
+      "Zhihan Lei",
+      "Jiarui Yan",
+      "Joshua Momo",
+      "William W. Cohen"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.02994",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.02994",
+    "pdf_url": "https://arxiv.org/pdf/2606.02994",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.02754",
+    "title": "$Ψ$-Bench: Evaluating Persona-Sensitive Influencing in Persuasive Dialogues",
+    "url": "https://arxiv.org/abs/2606.02754",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Peixuan Han",
+      "Hongyi Du",
+      "Jiayu Liu",
+      "Yihang Sun",
+      "Yutong Liu",
+      "Jiaxuan You"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.02754",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.02754",
+    "pdf_url": "https://arxiv.org/pdf/2606.02754",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.02875",
+    "title": "Handoff Debt: The Rediscovery Cost When Coding Agents Take Over Interrupted Tasks",
+    "url": "https://arxiv.org/abs/2606.02875",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Dipesh KC",
+      "Anjila Budathoki"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.02875",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.02875",
+    "pdf_url": "https://arxiv.org/pdf/2606.02875",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2605.29486",
+    "title": "PhoneWorld: Scaling Phone-Use Agent Environments",
+    "url": "https://arxiv.org/abs/2605.29486",
+    "published": "2026-05-28",
+    "updated": "2026-05-28",
+    "authors": [
+      "Zhengyang Tang",
+      "Yuxuan Liu",
+      "Xin Lai",
+      "Junyi Li",
+      "Pengyuan Lyu",
+      "Jason",
+      "Yiduo Guo",
+      "Zhengyao Fang",
+      "Yang Ding",
+      "Yi Zhang",
+      "Weinong Wang",
+      "Huawen Shen",
+      "Xingran Zhou",
+      "Liang Wu",
+      "Fei Tang",
+      "Sunqi Fan",
+      "Shangpin Peng",
+      "Zheng Ruan",
+      "Anran Zhang",
+      "Benyou Wang",
+      "Rui Yan",
+      "Ji-Rong Wen",
+      "Chengquan Zhang",
+      "Han Hu"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2605.29486",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.29486",
+    "pdf_url": "https://arxiv.org/pdf/2605.29486",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2605.28629",
+    "title": "Mobile-Aptus: Confidence-Driven Proactive and Robust Interaction in MLLM-based Mobile-Using Agents",
+    "url": "https://arxiv.org/abs/2605.28629",
+    "published": "2026-05-27",
+    "updated": "2026-05-27",
+    "authors": [
+      "Zheng Wu",
+      "Pengzhou Cheng",
+      "Zongru Wu",
+      "Yuan Guo",
+      "Tianjie Ju",
+      "Aston Zhang",
+      "Gongshen Liu",
+      "Zhuosheng Zhang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2605.28629",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.28629",
+    "pdf_url": "https://arxiv.org/pdf/2605.28629",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2605.27331",
+    "title": "Maat: The Agentic Legal Research Assistant for Competition Protection",
+    "url": "https://arxiv.org/abs/2605.27331",
+    "published": "2026-05-26",
+    "updated": "2026-05-26",
+    "authors": [
+      "Basant Mounir",
+      "Farida Madkour",
+      "Amira Abdelaziz",
+      "Asmaa Sami"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "rag",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.27331",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.27331",
+    "pdf_url": "https://arxiv.org/pdf/2605.27331",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.25641",
+    "title": "Iterate Until Retrieved: Factual Nugget Optimization for Discoverable Continual Corrections in Agentic RAG",
+    "url": "https://arxiv.org/abs/2605.25641",
+    "published": "2026-05-25",
+    "updated": "2026-05-25",
+    "authors": [
+      "Moshe Hazoom",
+      "Gal Patel",
+      "Alon Talmor",
+      "Tom Hope"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.25641",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.25641",
+    "pdf_url": "https://arxiv.org/pdf/2605.25641",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.25480",
+    "title": "Retrieval as Reasoning: Self-Evolving Agent-Native Retrieval via LLM-Wiki",
+    "url": "https://arxiv.org/abs/2605.25480",
+    "published": "2026-05-25",
+    "updated": "2026-05-26",
+    "authors": [
+      "Haoliang Ming",
+      "Feifei Li",
+      "Xiaoqing Wu",
+      "Wenhui Que"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.25480",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.25480",
+    "pdf_url": "https://arxiv.org/pdf/2605.25480",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.25002",
+    "title": "MemMark: State-Evolution Attribution Watermarking for Agent Long-Term Memory Systems",
+    "url": "https://arxiv.org/abs/2605.25002",
+    "published": "2026-05-24",
+    "updated": "2026-05-26",
+    "authors": [
+      "Haobo Zhang",
+      "Xutao Mao",
+      "Guangyuan Dong",
+      "Ziwei Li",
+      "Xuanbo Su",
+      "Kaijie Chen",
+      "Jing Yang",
+      "Zheng Lin"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "computer-use",
+      "memory"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.25002",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.25002",
+    "pdf_url": "https://arxiv.org/pdf/2605.25002",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.20485",
+    "title": "ZEBRA: Zero-shot Budgeted Resource Allocation for LLM Orchestration",
+    "url": "https://arxiv.org/abs/2605.20485",
+    "published": "2026-05-19",
+    "updated": "2026-05-19",
+    "authors": [
+      "May Hamri",
+      "Inbal Talgam-Cohen"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "multi-agent",
+      "reasoning"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.20485",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.20485",
+    "pdf_url": "https://arxiv.org/pdf/2605.20485",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.18597",
+    "title": "Latent Action Reparameterization for Efficient Agent Inference",
+    "url": "https://arxiv.org/abs/2605.18597",
+    "published": "2026-05-18",
+    "updated": "2026-05-19",
+    "authors": [
+      "Wenhao Huang",
+      "Qingwen Zeng",
+      "Qiyue Chen",
+      "Zijie Guo",
+      "Yu Sun",
+      "Cheng Yang",
+      "Siru Ouyang",
+      "Jiri Gesi",
+      "Fang Wu",
+      "Jiayi Zhang",
+      "Huaming Chen",
+      "Bang Liu",
+      "Xiangru Tang",
+      "Chenglin Wu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.18597",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.18597",
+    "pdf_url": "https://arxiv.org/pdf/2605.18597",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.15665",
+    "title": "PRISM: Prompt Reliability via Iterative Simulation and Monitoring for Enterprise Conversational AI",
+    "url": "https://arxiv.org/abs/2605.15665",
+    "published": "2026-05-15",
+    "updated": "2026-05-15",
+    "authors": [
+      "Keshava Chaitanya",
+      "Jahnavi Gundakaram"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.15665",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.15665",
+    "pdf_url": "https://arxiv.org/pdf/2605.15665",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.14051",
+    "title": "SPIN: Structural LLM Planning via Iterative Navigation for Industrial Tasks",
+    "url": "https://arxiv.org/abs/2605.14051",
+    "published": "2026-05-13",
+    "updated": "2026-05-13",
+    "authors": [
+      "Yusuke Ozaki",
+      "Dhaval Patel"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.14051",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.14051",
+    "pdf_url": "https://arxiv.org/pdf/2605.14051",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.13037",
+    "title": "MAP: A Map-then-Act Paradigm for Long-Horizon Interactive Agent Reasoning",
+    "url": "https://arxiv.org/abs/2605.13037",
+    "published": "2026-05-13",
+    "updated": "2026-05-13",
+    "authors": [
+      "Yuxin Liu",
+      "Ziang Ye",
+      "Yueqing Sun",
+      "Mingye Zhu",
+      "Jinwei Xiao",
+      "Zhuowen Han",
+      "Qi GU",
+      "Xunliang Cai",
+      "Lei Zhang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "reasoning"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.13037",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.13037",
+    "pdf_url": "https://arxiv.org/pdf/2605.13037",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.11224",
+    "title": "ABRA: Agent Benchmark for Radiology Applications",
+    "url": "https://arxiv.org/abs/2605.11224",
+    "published": "2026-05-11",
+    "updated": "2026-05-11",
+    "authors": [
+      "Bulat Maksudov",
+      "Vladislav Kurenkov",
+      "Kathleen M. Curran",
+      "Alessandra Mileo"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "planning",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2605.11224",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.11224",
+    "pdf_url": "https://arxiv.org/pdf/2605.11224",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2605.11003",
+    "title": "The Authorization-Execution Gap Is a Major Safety and Security Problem in Open-World Agents",
+    "url": "https://arxiv.org/abs/2605.11003",
+    "published": "2026-05-10",
+    "updated": "2026-05-10",
+    "authors": [
+      "Baoyuan Wu",
+      "Qingshan Liu",
+      "Adel Bibi",
+      "Irwin King",
+      "Siwei Lyu"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.11003",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.11003",
+    "pdf_url": "https://arxiv.org/pdf/2605.11003",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.03855",
+    "title": "Evaluating Generative Models as Interactive Emergent Representations of Human-Like Collaborative Behavior",
+    "url": "https://arxiv.org/abs/2605.03855",
+    "published": "2026-05-05",
+    "updated": "2026-05-06",
+    "authors": [
+      "Shinas Shaji",
+      "Teena Chakkalayil Hassan",
+      "Sebastian Houben",
+      "Alex Mitrevski"
+    ],
+    "categories": [
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "multi-agent",
+      "planning",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.03855",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.03855",
+    "pdf_url": "https://arxiv.org/pdf/2605.03855",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.00380",
+    "title": "ResRL: Boosting LLM Reasoning via Negative Sample Projection Residual Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2605.00380",
+    "published": "2026-05-01",
+    "updated": "2026-05-08",
+    "authors": [
+      "Zihan Lin",
+      "Xiaohan Wang",
+      "Jie Cao",
+      "Jiajun Chai",
+      "Li Wang",
+      "Xiaodong Lu",
+      "Wei Lin",
+      "Ran He",
+      "Guojun Yin"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2605.00380",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.00380",
+    "pdf_url": "https://arxiv.org/pdf/2605.00380",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2605.00060",
+    "title": "TADI: Tool-Augmented Drilling Intelligence via Agentic LLM Orchestration over Heterogeneous Wellsite Data",
+    "url": "https://arxiv.org/abs/2605.00060",
+    "published": "2026-04-30",
+    "updated": "2026-04-30",
+    "authors": [
+      "Rong Lu"
+    ],
+    "categories": [
+      "cs.AI",
+      "eess.SY"
+    ],
+    "topics": [
+      "rag",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2605.00060",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.00060",
+    "pdf_url": "https://arxiv.org/pdf/2605.00060",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2604.27132",
+    "title": "TRUST: A Framework for Decentralized AI Service v.0.1",
+    "url": "https://arxiv.org/abs/2604.27132",
+    "published": "2026-04-29",
+    "updated": "2026-04-29",
+    "authors": [
+      "Yu-Chao Huang",
+      "Zhen Tan",
+      "Mohan Zhang",
+      "Pingzhi Li",
+      "Zhuo Zhang",
+      "Tianlong Chen"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2604.27132",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.27132",
+    "pdf_url": "https://arxiv.org/pdf/2604.27132",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2604.18982",
+    "title": "SAVOIR: Learning Social Savoir-Faire via Shapley-based Reward Attribution",
+    "url": "https://arxiv.org/abs/2604.18982",
+    "published": "2026-04-21",
+    "updated": "2026-04-21",
+    "authors": [
+      "Xiachong Feng",
+      "Yi Jiang",
+      "Xiaocheng Feng",
+      "Deyi Yin",
+      "Libo Qin",
+      "Yangfan Ye",
+      "Lei Huang",
+      "Weitao Ma",
+      "Yuxuan Gu",
+      "Chonghan Qin",
+      "Bing Qin",
+      "Lingpeng Kong"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2604.18982",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.18982",
+    "pdf_url": "https://arxiv.org/pdf/2604.18982",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2604.03070",
+    "title": "How Your Credentials Are Leaked by LLM Agent Skills: An Empirical Study",
+    "url": "https://arxiv.org/abs/2604.03070",
+    "published": "2026-04-03",
+    "updated": "2026-06-19",
+    "authors": [
+      "Zhihao Chen",
+      "Ying Zhang",
+      "Yi Liu",
+      "Gelei Deng",
+      "Yuekang Li",
+      "Yanjun Zhang",
+      "Jianting Ning",
+      "Leo Yu Zhang",
+      "Lei Ma",
+      "Zhiqiang Li"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.03070",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.03070",
+    "pdf_url": "https://arxiv.org/pdf/2604.03070",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.00830",
+    "title": "Learning to Learn-at-Test-Time: Language Agents with Learnable Adaptation Policies",
+    "url": "https://arxiv.org/abs/2604.00830",
+    "published": "2026-04-01",
+    "updated": "2026-04-02",
+    "authors": [
+      "Zhanzhi Lou",
+      "Hui Chen",
+      "Yibo Li",
+      "Qian Wang",
+      "Bryan Hooi"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2604.00830",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.00830",
+    "pdf_url": "https://arxiv.org/pdf/2604.00830",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2604.00992",
+    "title": "Tube-Based Safety for Anticipative Tracking in Multi-Agent Systems",
+    "url": "https://arxiv.org/abs/2604.00992",
+    "published": "2026-04-01",
+    "updated": "2026-04-01",
+    "authors": [
+      "Armel Koulong",
+      "Ali Pakniyat"
+    ],
+    "categories": [
+      "eess.SY"
+    ],
+    "topics": [
+      "agent-safety",
+      "multi-agent",
+      "world-model"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.00992",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.00992",
+    "pdf_url": "https://arxiv.org/pdf/2604.00992",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.29560",
+    "title": "Distributed Predictive Control Barrier Functions: Towards Scalable Safety Certification in Modular Multi-Agent Systems",
+    "url": "https://arxiv.org/abs/2603.29560",
+    "published": "2026-03-31",
+    "updated": "2026-03-31",
+    "authors": [
+      "Jonas Ohnemus",
+      "Alexandre Didier",
+      "Ahmed Aboudonia",
+      "Andrea Carron",
+      "Melanie N. Zeilinger"
+    ],
+    "categories": [
+      "eess.SY",
+      "cs.RO",
+      "math.OC"
+    ],
+    "topics": [
+      "agent-safety",
+      "multi-agent",
+      "world-model"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.29560",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.29560",
+    "pdf_url": "https://arxiv.org/pdf/2603.29560",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.12644",
+    "title": "Uncovering Security Threats and Architecting Defenses in Autonomous Agents: A Case Study of OpenClaw",
+    "url": "https://arxiv.org/abs/2603.12644",
+    "published": "2026-03-13",
+    "updated": "2026-03-13",
+    "authors": [
+      "Zonghao Ying",
+      "Xiao Yang",
+      "Siyang Wu",
+      "Yumeng Song",
+      "Yang Qu",
+      "Hainan Li",
+      "Tianlin Li",
+      "Jiakai Wang",
+      "Aishan Liu",
+      "Xianglong Liu"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-safety",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.12644",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.12644",
+    "pdf_url": "https://arxiv.org/pdf/2603.12644",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.10213",
+    "title": "Sabiá-4 Technical Report",
+    "url": "https://arxiv.org/abs/2603.10213",
+    "published": "2026-03-10",
+    "updated": "2026-03-10",
+    "authors": [
+      "Thiago Laitz",
+      "Thales Sales Almeida",
+      "Hugo Abonizio",
+      "Roseval Malaquias Junior",
+      "Giovana Kerche Bonás",
+      "Marcos Piau",
+      "Celio Larcher",
+      "Ramon Pires",
+      "Rodrigo Nogueira"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "embodied-agent",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2603.10213",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.10213",
+    "pdf_url": "https://arxiv.org/pdf/2603.10213",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2603.08316",
+    "title": "SlowBA: An efficiency backdoor attack towards VLM-based GUI agents",
+    "url": "https://arxiv.org/abs/2603.08316",
+    "published": "2026-03-09",
+    "updated": "2026-07-01",
+    "authors": [
+      "Junxian Li",
+      "Tu Lan",
+      "Haozhen Tan",
+      "Yan Meng",
+      "Haojin Zhu"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.CL",
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.08316",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.08316",
+    "pdf_url": "https://arxiv.org/pdf/2603.08316",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2602.23876",
+    "title": "RF-Agent: Automated Reward Function Design via Language Agent Tree Search",
+    "url": "https://arxiv.org/abs/2602.23876",
+    "published": "2026-02-27",
+    "updated": "2026-02-27",
+    "authors": [
+      "Ning Gao",
+      "Xiuhui Zhang",
+      "Xingyu Jiang",
+      "Mukang You",
+      "Mohan Zhang",
+      "Yue Deng"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "rag",
+      "reasoning"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2602.23876",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.23876",
+    "pdf_url": "https://arxiv.org/pdf/2602.23876",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2602.20156",
+    "title": "Skill-Inject: Measuring Agent Vulnerability to Skill File Attacks",
+    "url": "https://arxiv.org/abs/2602.20156",
+    "published": "2026-02-23",
+    "updated": "2026-02-25",
+    "authors": [
+      "David Schmotz",
+      "Luca Beurer-Kellner",
+      "Sahar Abdelnabi",
+      "Maksym Andriushchenko"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.20156",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.20156",
+    "pdf_url": "https://arxiv.org/pdf/2602.20156",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2602.17875",
+    "title": "MultiVer: Zero-Shot Multi-Agent Vulnerability Detection",
+    "url": "https://arxiv.org/abs/2602.17875",
+    "published": "2026-02-19",
+    "updated": "2026-02-19",
+    "authors": [
+      "Shreshth Rajan"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.17875",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.17875",
+    "pdf_url": "https://arxiv.org/pdf/2602.17875",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2512.00332",
+    "title": "Assertion-Conditioned Compliance: A Provenance-Aware Vulnerability in Multi-Turn Tool-Calling Agents",
+    "url": "https://arxiv.org/abs/2512.00332",
+    "published": "2025-11-29",
+    "updated": "2026-01-21",
+    "authors": [
+      "Daud Waqas",
+      "Aaryamaan Golthi",
+      "Erika Hayashida",
+      "Huanzhi Mao"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2512.00332",
+    "source": "arxiv",
+    "source_id": "arxiv:2512.00332",
+    "pdf_url": "https://arxiv.org/pdf/2512.00332",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2510.24284",
+    "title": "MCP-Flow: Facilitating LLM Agents to Master Real-World, Diverse and Scaling MCP Tools",
+    "url": "https://arxiv.org/abs/2510.24284",
+    "published": "2025-10-28",
+    "updated": "2026-04-16",
+    "authors": [
+      "Wenhao Wang",
+      "Peizhi Niu",
+      "Zhao Xu",
+      "Zhaoyu Chen",
+      "Jian Du",
+      "Yaxin Du",
+      "Xianghe Pang",
+      "Keduan Huang",
+      "Yanfeng Wang",
+      "Qiang Yan",
+      "Siheng Chen"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2510.24284",
+    "source": "arxiv",
+    "source_id": "arxiv:2510.24284",
+    "pdf_url": "https://arxiv.org/pdf/2510.24284",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2509.18420",
+    "title": "Instruction-Following Evaluation in Function Calling for Large Language Models",
+    "url": "https://arxiv.org/abs/2509.18420",
+    "published": "2025-09-22",
+    "updated": "2025-09-22",
+    "authors": [
+      "Nikolai Skripko"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2509.18420",
+    "source": "arxiv",
+    "source_id": "arxiv:2509.18420",
+    "pdf_url": "https://arxiv.org/pdf/2509.18420",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2509.18169",
+    "title": "PiERN: Token-Level Routing for Integrating High-Precision Computation and Reasoning",
+    "url": "https://arxiv.org/abs/2509.18169",
+    "published": "2025-09-17",
+    "updated": "2026-04-20",
+    "authors": [
+      "Hengbo Xiao",
+      "Jingyuan Fan",
+      "Xin Tong",
+      "Jingzhao Zhang",
+      "Chao Lu",
+      "Guannan He"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.CE",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2509.18169",
+    "source": "arxiv",
+    "source_id": "arxiv:2509.18169",
+    "pdf_url": "https://arxiv.org/pdf/2509.18169",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2508.20637",
+    "title": "GDS Agent for Graph Algorithmic Reasoning",
+    "url": "https://arxiv.org/abs/2508.20637",
+    "published": "2025-08-28",
+    "updated": "2025-11-05",
+    "authors": [
+      "Borun Shi",
+      "Ioannis Panagiotas"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 12,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2508.20637",
+    "source": "arxiv",
+    "source_id": "arxiv:2508.20637",
+    "pdf_url": "https://arxiv.org/pdf/2508.20637",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2607.06184",
+    "title": "What Resolve Rate Hides: Trajectory Structure Diagnostics for Coding Agents",
+    "url": "https://arxiv.org/abs/2607.06184",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Rui Shu",
+      "Chun Yong Chong",
+      "Xin Zhou",
+      "Yun Peng",
+      "Zihan Wu",
+      "Xu Han",
+      "Zeyang Zhuang",
+      "Guowen Yuan",
+      "Yuan Wang"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.06184",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06184",
+    "pdf_url": "https://arxiv.org/pdf/2607.06184",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.06065",
+    "title": "SWE-Review: Closing the Loop on Issue Resolution with Agentic Code Review",
+    "url": "https://arxiv.org/abs/2607.06065",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Ruoyu Wang",
+      "Jierun Chen",
+      "Shaowei Wang",
+      "Chaofan Tao",
+      "Sidi Yang",
+      "Yuxin Jiang",
+      "Kim-Hui Yap",
+      "Lifeng Shang",
+      "Xiaohui Li",
+      "Haoli Bai"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.06065",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06065",
+    "pdf_url": "https://arxiv.org/pdf/2607.06065",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.05785",
+    "title": "Can Large Language Models Generate Observability-Aware Code?",
+    "url": "https://arxiv.org/abs/2607.05785",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Yongliang Tao",
+      "Hongyu Zhang",
+      "Pengfei Gao",
+      "Minghua Ma",
+      "Zhiyu Fan",
+      "Yu Kang",
+      "Jue Zhang",
+      "Si Qin",
+      "Liqun Li",
+      "Qingwei Lin",
+      "Saravan Rajmohan"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.05785",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05785",
+    "pdf_url": "https://arxiv.org/pdf/2607.05785",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.05682",
+    "title": "FirstResearch: Auditable Question Formation for LLM Scientific Discovery Agents",
+    "url": "https://arxiv.org/abs/2607.05682",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Yufeng Wang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "planning",
+      "rag"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "llm-agent",
+      "planning-agent"
+    ],
+    "arxiv_id": "2607.05682",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05682",
+    "pdf_url": "https://arxiv.org/pdf/2607.05682",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.04576",
+    "title": "Progressive Disclosure for LLM-Maintained Wiki Knowledge Bases: a Preregistered Ablation",
+    "url": "https://arxiv.org/abs/2607.04576",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Theodore O. Cochran"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.CY",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "llm-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.04576",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04576",
+    "pdf_url": "https://arxiv.org/pdf/2607.04576",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.04558",
+    "title": "EEG-SpikeAgent: Agentic Closed-Loop Program Synthesis for Automated EEG Spike Detection",
+    "url": "https://arxiv.org/abs/2607.04558",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Sonali Santhosh",
+      "Kelly Shuhong Yu",
+      "Eugene Chang",
+      "Jonathan Kim",
+      "Kie Shidara",
+      "Danilo Bernardo"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.04558",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04558",
+    "pdf_url": "https://arxiv.org/pdf/2607.04558",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.05593",
+    "title": "Collective Cognition in Hybrid Groups: A Network Science Synthesis",
+    "url": "https://arxiv.org/abs/2607.05593",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Babak Hemmatian",
+      "Razan Baltaji",
+      "Lav R. Varshney"
+    ],
+    "categories": [
+      "cs.HC"
+    ],
+    "topics": [
+      "memory",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai",
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.05593",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05593",
+    "pdf_url": "https://arxiv.org/pdf/2607.05593",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.04948",
+    "title": "Using Process Mining to Generate AI Agents from Software Engineering Process Records",
+    "url": "https://arxiv.org/abs/2607.04948",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Saimir Bala",
+      "Fabiana Fournier",
+      "Lior Limonad",
+      "Andreas Metzger"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "coding-agent",
+      "multi-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.04948",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04948",
+    "pdf_url": "https://arxiv.org/pdf/2607.04948",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.05462",
+    "title": "Evaluating calibrated refusal and safe usefulness in dual-use biology settings",
+    "url": "https://arxiv.org/abs/2607.05462",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Edwin H. Wintermute",
+      "Harmon Bhasin",
+      "Christina M. Agapakis",
+      "Dianzhuo Wang",
+      "Evan Seeyave",
+      "Arjun Banerjee",
+      "Daniel Fulop",
+      "Matthew C. Watson",
+      "Adam J. Meyer",
+      "Sandrine Boissel",
+      "Jens H. Kuhn",
+      "Rishi Jain",
+      "Noah D. Taylor",
+      "Helena Shomar",
+      "Patrick M. Boyle",
+      "Kenny Workman"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.05462",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05462",
+    "pdf_url": "https://arxiv.org/pdf/2607.05462",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.05483",
+    "title": "PatchOptic for Shared-State LLM Workflows with Projected Views and Verified Structured Updates",
+    "url": "https://arxiv.org/abs/2607.05483",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Zhaoyu Bai",
+      "Jiaqi Cai"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.LO",
+      "cs.MA",
+      "cs.PL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai",
+      "rag-agent"
+    ],
+    "arxiv_id": "2607.05483",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05483",
+    "pdf_url": "https://arxiv.org/pdf/2607.05483",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.05139",
+    "title": "On the risk of coding before testing: An empirical study on LLM-based test generation workflow",
+    "url": "https://arxiv.org/abs/2607.05139",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Michael Konstantinou",
+      "Florian Tambon",
+      "Mike Papadakis"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.05139",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05139",
+    "pdf_url": "https://arxiv.org/pdf/2607.05139",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.04096",
+    "title": "Forethought: Verifiable Reasoning from Neurosymbolic Primitive Programming",
+    "url": "https://arxiv.org/abs/2607.04096",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Vishvesh Bhat",
+      "Jay Vaghasiya",
+      "Emmanuel Anaya Gonzalez"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.04096",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04096",
+    "pdf_url": "https://arxiv.org/pdf/2607.04096",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.04425",
+    "title": "UI-MOPD: Multi-Platform On-Policy Distillation for Continual GUI Agent Learning",
+    "url": "https://arxiv.org/abs/2607.04425",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Niu Lian",
+      "Alan Chen",
+      "Zhehao Yu",
+      "Chengzhen Duan",
+      "Fazhan Liu",
+      "Hui Liu",
+      "Pei Fu",
+      "Jian Luan",
+      "Yaowei Wang",
+      "Shu-Tao Xia",
+      "Jinpeng Wang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.CV",
+      "cs.LG",
+      "cs.MM"
+    ],
+    "topics": [
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2607.04425",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04425",
+    "pdf_url": "https://arxiv.org/pdf/2607.04425",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2607.03651",
+    "title": "LLM-Guided Transportation Hub Capacity Planning with Textual Business Inputs",
+    "url": "https://arxiv.org/abs/2607.03651",
+    "published": "2026-07-04",
+    "updated": "2026-07-04",
+    "authors": [
+      "Xiaoyue Liu",
+      "Zheng Dong"
+    ],
+    "categories": [
+      "cs.LG",
+      "math.OC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "llm-agent",
+      "planning-agent"
+    ],
+    "arxiv_id": "2607.03651",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03651",
+    "pdf_url": "https://arxiv.org/pdf/2607.03651",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.02975",
+    "title": "Evaluating Generative Agents with Actions Grounded in Socially Distributed Task Environments using Incognita",
+    "url": "https://arxiv.org/abs/2607.02975",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Dan C. Hsu",
+      "Luke Lu"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2607.02975",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02975",
+    "pdf_url": "https://arxiv.org/pdf/2607.02975",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2607.03025",
+    "title": "Human-Centric Reflective Architecture for Human-AI Collaborative Decision-Making",
+    "url": "https://arxiv.org/abs/2607.03025",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Andreas Kouridakis",
+      "Dimitrios Patiniotis Spyropoulos",
+      "George Vouros"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "rag"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.03025",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03025",
+    "pdf_url": "https://arxiv.org/pdf/2607.03025",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.03386",
+    "title": "When Aggregate Alignment Misleads: Auditing Policy Repair Without Per-State Expert Actions",
+    "url": "https://arxiv.org/abs/2607.03386",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Peiying Zhu",
+      "Sidi Chang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.03386",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03386",
+    "pdf_url": "https://arxiv.org/pdf/2607.03386",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.03332",
+    "title": "AtomicCommitBench: Can Coding Agents Reconstruct Commit Histories from Squashed Patches?",
+    "url": "https://arxiv.org/abs/2607.03332",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Zhihao Lin",
+      "Mingyi Zhou",
+      "Li Li"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.03332",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03332",
+    "pdf_url": "https://arxiv.org/pdf/2607.03332",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.01810",
+    "title": "Decoupling Code Complexity from Newcomer Participation: A Causal Study of AI Coding Agent Adoption in OSS",
+    "url": "https://arxiv.org/abs/2607.01810",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Weiwei Xu",
+      "Xuanning Cui",
+      "Hengzhi Ye",
+      "Minghui Zhou"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.01810",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01810",
+    "pdf_url": "https://arxiv.org/pdf/2607.01810",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.01760",
+    "title": "Refploit: Facilitating Exploit Construction via Code-Agent Trajectory Repair",
+    "url": "https://arxiv.org/abs/2607.01760",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Zirui Chen",
+      "Zhipeng Xue",
+      "Jiayuan Zhou",
+      "Xing Hu",
+      "Xin Xia",
+      "Xiaohu Yang"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.01760",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01760",
+    "pdf_url": "https://arxiv.org/pdf/2607.01760",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.01418",
+    "title": "Adoption and Impact of Command-Line AI Coding Agents: A Study of Microsoft's Early 2026 Rollout of Claude Code and GitHub Copilot CLI",
+    "url": "https://arxiv.org/abs/2607.01418",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Emerson Murphy-Hill",
+      "Jenna Butler",
+      "Alexandra Savelieva"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.HC"
+    ],
+    "topics": [
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.01418",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01418",
+    "pdf_url": "https://arxiv.org/pdf/2607.01418",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.01087",
+    "title": "Cheap Code, Costly Judgment: A Case Study on Governable Agentic Software Engineering",
+    "url": "https://arxiv.org/abs/2607.01087",
+    "published": "2026-07-01",
+    "updated": "2026-07-04",
+    "authors": [
+      "James C. Davis",
+      "Paschal C. Amusuo",
+      "Tanmay Singla",
+      "Berk Çakar",
+      "Kirsten A. Davis"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.01087",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01087",
+    "pdf_url": "https://arxiv.org/pdf/2607.01087",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.00895",
+    "title": "Beyond Document Grounding: Span-Level Hallucination Detection over Code, Tool Output, and Documents",
+    "url": "https://arxiv.org/abs/2607.00895",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Ádám Kovács",
+      "Bowei He",
+      "Xue Liu",
+      "István Boros",
+      "Szilveszter Tóth",
+      "Gábor Recski"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent",
+      "rag-agent"
+    ],
+    "arxiv_id": "2607.00895",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00895",
+    "pdf_url": "https://arxiv.org/pdf/2607.00895",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.01063",
+    "title": "AutoRestTest at the SBFT 2026 Tool Competition",
+    "url": "https://arxiv.org/abs/2607.01063",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Tyler Stennett",
+      "Myeongsoo Kim",
+      "Saurabh Sinha",
+      "Alessandro Orso"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.01063",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01063",
+    "pdf_url": "https://arxiv.org/pdf/2607.01063",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.31371",
+    "title": "Calibrating the Evaluator: Does Probability Calibration Mitigate Preference Coupling in LLM Agent Feedback Loops?",
+    "url": "https://arxiv.org/abs/2606.31371",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Zewen Liu"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.31371",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31371",
+    "pdf_url": "https://arxiv.org/pdf/2606.31371",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.31182",
+    "title": "AI-Assisted Discovery of Convex Relaxations via Dual Agents",
+    "url": "https://arxiv.org/abs/2606.31182",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Sungyoon Kim",
+      "Mert Pilanci"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "coding-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent",
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.31182",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31182",
+    "pdf_url": "https://arxiv.org/pdf/2606.31182",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.31935",
+    "title": "Delegation Rights: Property, Agency, and Investment Incentives in the Age of AI Agents",
+    "url": "https://arxiv.org/abs/2606.31935",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Yukun Zhang",
+      "Kemu Xu"
+    ],
+    "categories": [
+      "econ.EM"
+    ],
+    "topics": [
+      "agent-safety",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.31935",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31935",
+    "pdf_url": "https://arxiv.org/pdf/2606.31935",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.31041",
+    "title": "A Semantic-Layer-Mediated Agent for Natural Language to SQL over Heterogeneous Enterprise Databases",
+    "url": "https://arxiv.org/abs/2606.31041",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Ha Jeong Kim",
+      "Saksonita Khoeurn",
+      "Ye Ji Yoon"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.31041",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31041",
+    "pdf_url": "https://arxiv.org/pdf/2606.31041",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.31270",
+    "title": "Learning from Failure: Inference-Time Self-Improvement for Computer-Use Agents",
+    "url": "https://arxiv.org/abs/2606.31270",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Xueqiao Sun",
+      "Xiaohan Wang",
+      "Ludwig Schmidt",
+      "Serena Yeung-Levy",
+      "Yuhui Zhang"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI",
+      "cs.CL",
+      "cs.CY",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.31270",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31270",
+    "pdf_url": "https://arxiv.org/pdf/2606.31270",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.31154",
+    "title": "PPT-Eval: A Benchmark for Computer-Use Agents on PowerPoint Tasks",
+    "url": "https://arxiv.org/abs/2606.31154",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Apurva Gandhi",
+      "Vishwas Suryanarayanan",
+      "Raja Hasnain Anwar",
+      "Firoz Shaik",
+      "Shubhang Desai",
+      "Thong Q. Nguyen",
+      "Muhammad Taqi Raza",
+      "Vishal Chowdhary",
+      "Graham Neubig"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.31154",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31154",
+    "pdf_url": "https://arxiv.org/pdf/2606.31154",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.31273",
+    "title": "The Calibration Turn in AI-Assisted Research: A Conceptual and Methodological Framework for Evidence-Licensed Claims",
+    "url": "https://arxiv.org/abs/2606.31273",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Hongmin Li"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent",
+      "workflow-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.31273",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31273",
+    "pdf_url": "https://arxiv.org/pdf/2606.31273",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.29717",
+    "title": "Optimizing Expert-Designed Crystal Graph Networks for Band-Gap Prediction with an Autonomous LLM Research Loop",
+    "url": "https://arxiv.org/abs/2606.29717",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Chenmu Zhang",
+      "Boris I. Yakobson"
+    ],
+    "categories": [
+      "cond-mat.mtrl-sci",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm",
+      "coding-agent",
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.29717",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29717",
+    "pdf_url": "https://arxiv.org/pdf/2606.29717",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.29705",
+    "title": "GUICrafter: Weakly-Supervised GUI Agent Leveraging Massive Unannotated Screenshots",
+    "url": "https://arxiv.org/abs/2606.29705",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Sunqi Fan",
+      "Lingshan Chen",
+      "Runqi Yin",
+      "Qingle Liu",
+      "Yongming Rao",
+      "Meng-Hao Guo",
+      "Shi-Min Hu"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.CV"
+    ],
+    "topics": [
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.29705",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29705",
+    "pdf_url": "https://arxiv.org/pdf/2606.29705",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.30571",
+    "title": "Attractor States Emerge in Multi-Turn LLM Conversations",
+    "url": "https://arxiv.org/abs/2606.30571",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Ting-Wen Ko",
+      "Jonas Geiping"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.30571",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30571",
+    "pdf_url": "https://arxiv.org/pdf/2606.30571",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.29981",
+    "title": "Hephaestus: Toward a Cybersecurity AI Scientist",
+    "url": "https://arxiv.org/abs/2606.29981",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Jiaqi Li",
+      "Yang Zhao",
+      "Wen Lu",
+      "Lvyang Zhang",
+      "Lidong Zhai"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2606.29981",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29981",
+    "pdf_url": "https://arxiv.org/pdf/2606.29981",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.29406",
+    "title": "Adaptive AI Delegation under Uncertainty: A Bayesian Governance Policy for Sequential Decision Authority",
+    "url": "https://arxiv.org/abs/2606.29406",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Matthew Francis Dixon"
+    ],
+    "categories": [
+      "q-fin.RM",
+      "math.OC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.29406",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29406",
+    "pdf_url": "https://arxiv.org/pdf/2606.29406",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.29648",
+    "title": "Hybrid Retriever Evolution for Multimodal Document Reasoning Agents",
+    "url": "https://arxiv.org/abs/2606.29648",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Bohan Yao",
+      "Shruthan Radhakrishna",
+      "Vikas Yadav"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.LG",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.29648",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29648",
+    "pdf_url": "https://arxiv.org/pdf/2606.29648",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.29472",
+    "title": "Agent-Computer Observation Interfaces Enable Dynamic Computer Use",
+    "url": "https://arxiv.org/abs/2606.29472",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Bojie Li",
+      "Noah Shi"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent",
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.29472",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29472",
+    "pdf_url": "https://arxiv.org/pdf/2606.29472",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.29605",
+    "title": "How much of an LLM-generated clinical corpus is actually new? A production-scale measurement of content redundancy for provenance classification",
+    "url": "https://arxiv.org/abs/2606.29605",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Ali H. Lazem",
+      "William J. Teahan"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.29605",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29605",
+    "pdf_url": "https://arxiv.org/pdf/2606.29605",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.27650",
+    "title": "GenWorld: Empirically Grounded Urban Simulation Infrastructure for Scalable LLM-Agent Studies",
+    "url": "https://arxiv.org/abs/2606.27650",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Gen Li",
+      "Jieyuan Lan",
+      "Pengcheng Xu",
+      "Zongyuan Wu",
+      "Masaki Ogura",
+      "Tao Feng"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "computer-use",
+      "planning",
+      "rag",
+      "world-model"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.27650",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27650",
+    "pdf_url": "https://arxiv.org/pdf/2606.27650",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.28235",
+    "title": "Govern the Repository, Not the Agent: Measuring Ecosystem-Level Risk in AI-Native Software",
+    "url": "https://arxiv.org/abs/2606.28235",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Daniel Russo"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai",
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.28235",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28235",
+    "pdf_url": "https://arxiv.org/pdf/2606.28235",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.27909",
+    "title": "Triadic Werewolf: A Jester Role for Multi-Hop Theory of Mind in LLMs",
+    "url": "https://arxiv.org/abs/2606.27909",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Avni Mittal"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.GT",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.27909",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27909",
+    "pdf_url": "https://arxiv.org/pdf/2606.27909",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.28002",
+    "title": "Dialogue to Detection: A Multimodal Hybrid NLP Pipeline for Insurance Fraud Detection",
+    "url": "https://arxiv.org/abs/2606.28002",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Muhammad Shakeel Akram",
+      "Amal Htait",
+      "Abdul Hamid Sadka",
+      "Emma Meisingseth",
+      "Karishma Jaitly"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "eess.AS"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.28002",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28002",
+    "pdf_url": "https://arxiv.org/pdf/2606.28002",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.26722",
+    "title": "Socratic agents for autonomous scientific discovery in high-dimensional physical systems",
+    "url": "https://arxiv.org/abs/2606.26722",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Xianrui Zeng",
+      "Pengfei Liu",
+      "Yirui Zang",
+      "Yang Shen",
+      "Fei Yu",
+      "Chunlei Yu",
+      "Minghao Liu",
+      "Yang Du"
+    ],
+    "categories": [
+      "cs.AI",
+      "physics.optics"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "planning",
+      "workflow-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.26722",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26722",
+    "pdf_url": "https://arxiv.org/pdf/2606.26722",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.27595",
+    "title": "Ko-WideSearch: A Korean Breadth-Search Benchmark for Exhaustive Set Enumeration by Web Agents",
+    "url": "https://arxiv.org/abs/2606.27595",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Minbyul Jeong"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-evaluation",
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.27595",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27595",
+    "pdf_url": "https://arxiv.org/pdf/2606.27595",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.26474",
+    "title": "Localizing RL-Induced Tool Use to a Single Crosscoder Feature",
+    "url": "https://arxiv.org/abs/2606.26474",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Andrii Shportko",
+      "Shubham Bhokare",
+      "Ahmed Zeyad A Alzahrani",
+      "Bowen Cheng",
+      "Gustavo Mercier",
+      "Jessica Hullman"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "rag",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.26474",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26474",
+    "pdf_url": "https://arxiv.org/pdf/2606.26474",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.27122",
+    "title": "Mostly Automatic Translation of Language Interpreters from C to Safe Rust",
+    "url": "https://arxiv.org/abs/2606.27122",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Bo Wang",
+      "Brandon Paulsen",
+      "Joey Dodds",
+      "Daniel Kroening",
+      "Umang Mathur",
+      "Prateek Saxena"
+    ],
+    "categories": [
+      "cs.PL",
+      "cs.MA",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "memory",
+      "multi-agent",
+      "workflow-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.27122",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27122",
+    "pdf_url": "https://arxiv.org/pdf/2606.27122",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.26979",
+    "title": "How Much Static Structure Do Code Agents Need? A Study of Deterministic Anchoring",
+    "url": "https://arxiv.org/abs/2606.26979",
+    "published": "2026-06-25",
+    "updated": "2026-07-02",
+    "authors": [
+      "Zhihao Lin",
+      "Mingyi Zhou",
+      "Yizhuo Yang",
+      "Li Li"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "coding-agent",
+      "computer-use",
+      "embodied-agent",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.26979",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26979",
+    "pdf_url": "https://arxiv.org/pdf/2606.26979",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.26298",
+    "title": "Governing Actions, Not Agents: Institutional Attestation as a Governance Model for Autonomous AI Systems",
+    "url": "https://arxiv.org/abs/2606.26298",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Jakob Salfeld-Nebgen"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.26298",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26298",
+    "pdf_url": "https://arxiv.org/pdf/2606.26298",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.26027",
+    "title": "Why Multi-Step Tool-Use Reinforcement Learning Collapses and How Supervisory Signals Fix It",
+    "url": "https://arxiv.org/abs/2606.26027",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Yupu Hao",
+      "Zhuoran Jin",
+      "Huanxuan Liao",
+      "Kang Liu",
+      "Jun Zhao"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.26027",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26027",
+    "pdf_url": "https://arxiv.org/pdf/2606.26027",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.25257",
+    "title": "How Do Developers Maintain and Evolve Their Agents' Instructions? An Empirical Study",
+    "url": "https://arxiv.org/abs/2606.25257",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Gianmario Voria",
+      "Alfonso Cannavale",
+      "Andrea De Lucia",
+      "Yutaro Kashiwa",
+      "Gemma Catolino",
+      "Fabio Palomba"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "planning",
+      "workflow-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.25257",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25257",
+    "pdf_url": "https://arxiv.org/pdf/2606.25257",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.24429",
+    "title": "Detecting AI Coding Agents in Open Source: A Validated Multi-Method Census of 180 Million Repositories",
+    "url": "https://arxiv.org/abs/2606.24429",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Arsham Khosravani",
+      "Audris Mockus"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.24429",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24429",
+    "pdf_url": "https://arxiv.org/pdf/2606.24429",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.24965",
+    "title": "Project Auto-World: Towards Automated Benchmarking of Neural Relational Reasoners",
+    "url": "https://arxiv.org/abs/2606.24965",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Anirban Das",
+      "Joanne Boisson",
+      "Irtaza Khalid",
+      "Sumita Garai",
+      "Steven Schockaert"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.24965",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24965",
+    "pdf_url": "https://arxiv.org/pdf/2606.24965",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.23449",
+    "title": "AOHP: An Open-Source OS-Level Agent Harness for Personalized, Efficient and Secure Interaction",
+    "url": "https://arxiv.org/abs/2606.23449",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Shanhui Zhao",
+      "Jiacheng Liu",
+      "Guohong Liu",
+      "Jichao Yan",
+      "Jialei Ye",
+      "Yuhao Yang",
+      "Hao Wen",
+      "Shizuo Tian",
+      "Yizhen Yuan",
+      "Yuxuan Chen",
+      "Yunxin Liu",
+      "Ju Ren",
+      "Ya-Qin Zhang",
+      "Chao Huang",
+      "Yao Guo",
+      "Yuanchun Li"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.OS"
+    ],
+    "topics": [
+      "agent-safety",
+      "memory",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.23449",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23449",
+    "pdf_url": "https://arxiv.org/pdf/2606.23449",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.23937",
+    "title": "When Retrieval Metrics Mislead: Measuring Policy Signal in Long-Horizon Tool-Use Agents",
+    "url": "https://arxiv.org/abs/2606.23937",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Tianyu Ding",
+      "Juan Pablo De la Cruz Weinstein"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.23937",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23937",
+    "pdf_url": "https://arxiv.org/pdf/2606.23937",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.23997",
+    "title": "ChartWalker: Benchmarking the Cross-Chart RAG Task with Hierarchical Knowledge Graphs",
+    "url": "https://arxiv.org/abs/2606.23997",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Ning Tang",
+      "Chenghan Xie",
+      "Hanyang Yuan",
+      "Yi Li",
+      "Renhong Huang",
+      "Qian Kou",
+      "Xiaofeng Shi",
+      "Hua Zhou",
+      "Jiarong Xu"
+    ],
+    "categories": [
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.23997",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23997",
+    "pdf_url": "https://arxiv.org/pdf/2606.23997",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.22376",
+    "title": "Vibe Calibration: Autonomous Bring-up of a 112-Qubit Superconducting Quantum Processor by a Skill-Orchestrating Language Agent",
+    "url": "https://arxiv.org/abs/2606.22376",
+    "published": "2026-06-21",
+    "updated": "2026-06-21",
+    "authors": [
+      "Huikai Xu",
+      "Jiaxiu Han",
+      "Shigang Ou",
+      "Cheng Ye",
+      "Zisong Shen",
+      "Jing Gao",
+      "Yijia Wang",
+      "Tianrui Che",
+      "Yu Song",
+      "Weiyang Liu",
+      "Lei Wang",
+      "Lin-Feng Zhang",
+      "Pan Zhang",
+      "Hai-Feng Yu"
+    ],
+    "categories": [
+      "quant-ph"
+    ],
+    "topics": [
+      "workflow-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.22376",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22376",
+    "pdf_url": "https://arxiv.org/pdf/2606.22376",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.22721",
+    "title": "Habituation at the Gate: Rising Approval and Declining Scrutiny in Human Review of AI Agent Code",
+    "url": "https://arxiv.org/abs/2606.22721",
+    "published": "2026-06-21",
+    "updated": "2026-06-21",
+    "authors": [
+      "Haoran Yu",
+      "Lifei Liu",
+      "Xiaochong Jiang",
+      "Yuwen Jia",
+      "Su Wang",
+      "Pin Qian",
+      "Yihang Chen"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "coding-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent",
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.22721",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22721",
+    "pdf_url": "https://arxiv.org/pdf/2606.22721",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.22711",
+    "title": "Beyond Simpson's Paradox: A Cascade of Confounders in AI Agent Pull-Request Co-Authorship",
+    "url": "https://arxiv.org/abs/2606.22711",
+    "published": "2026-06-21",
+    "updated": "2026-06-21",
+    "authors": [
+      "Haoran Yu",
+      "Xiaochong Jiang",
+      "Lifei Liu",
+      "Su Wang",
+      "Pin Qian",
+      "Yihang Chen"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "coding-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent",
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.22711",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22711",
+    "pdf_url": "https://arxiv.org/pdf/2606.22711",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.21843",
+    "title": "Measuring What Persists: Conditioning Mechanisms and a Geometric Framework for AI Agent Identity",
+    "url": "https://arxiv.org/abs/2606.21843",
+    "published": "2026-06-20",
+    "updated": "2026-06-20",
+    "authors": [
+      "Andrew Tanner"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.21843",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21843",
+    "pdf_url": "https://arxiv.org/pdf/2606.21843",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.21562",
+    "title": "Compressing Observation History into Agent Memory: Distilling Transformers into Recurrent Transformers",
+    "url": "https://arxiv.org/abs/2606.21562",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Philippe Weinzaepfel",
+      "Christian Wolf",
+      "Bülent Mert Sariyildiz",
+      "Guillaume Bono",
+      "Gianluca Monaci"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.LG"
+    ],
+    "topics": [
+      "embodied-agent",
+      "memory",
+      "planning"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.21562",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21562",
+    "pdf_url": "https://arxiv.org/pdf/2606.21562",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.19857",
+    "title": "Large Language Models Do Not Always Need Readable Language",
+    "url": "https://arxiv.org/abs/2606.19857",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Jiayi Zhu",
+      "Haoxuan Peng",
+      "Junxi Wang",
+      "Liang Ke",
+      "Chen Zhang",
+      "Linfeng Zhang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.19857",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19857",
+    "pdf_url": "https://arxiv.org/pdf/2606.19857",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.20910",
+    "title": "Whose Agent Are You? Multi-Layer Fingerprinting and Attribution of Autonomous Web Agents",
+    "url": "https://arxiv.org/abs/2606.20910",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Dayeon Kang",
+      "Hyejun Jeong",
+      "Jade Sheffey",
+      "Pubali Datta",
+      "Amir Houmansadr"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "embodied-agent",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.20910",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20910",
+    "pdf_url": "https://arxiv.org/pdf/2606.20910",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.20487",
+    "title": "Beyond Global Replanning: Hierarchical Recovery for Cross-Device Agent Systems",
+    "url": "https://arxiv.org/abs/2606.20487",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Shu Yao",
+      "Yuhua Luo",
+      "Qian Long",
+      "Jingru Fan",
+      "Zhuoyuan Yu",
+      "Yuheng Wang",
+      "Lin Wu",
+      "Yufan Dang",
+      "Huatao Li",
+      "Chen Qian"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.20487",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20487",
+    "pdf_url": "https://arxiv.org/pdf/2606.20487",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.20746",
+    "title": "Amplify, Don't Create: Temporal Accumulation for Slow-Burn Prompt Injection",
+    "url": "https://arxiv.org/abs/2606.20746",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "J Alex Corll"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.LO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.20746",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20746",
+    "pdf_url": "https://arxiv.org/pdf/2606.20746",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.18613",
+    "title": "Are LLMs Ready to Assist Physicians? PhysAssistBench for Interactive Doctor-Patient-EHR Assistance",
+    "url": "https://arxiv.org/abs/2606.18613",
+    "published": "2026-06-17",
+    "updated": "2026-06-18",
+    "authors": [
+      "Tianming Du",
+      "Peijie Yu",
+      "Sihan Shang",
+      "Danli Shi",
+      "My Linh Nguyen",
+      "Shengbo Gao",
+      "Guangyuan Li",
+      "Yinghong Yu",
+      "Yan Jiang",
+      "Qianlong Zhao",
+      "Behzad Bozorgtabar",
+      "Shaoxiong Ji",
+      "Jiazhen Pan",
+      "Daniel Rueckert",
+      "Jiancheng Yang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.18613",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18613",
+    "pdf_url": "https://arxiv.org/pdf/2606.18613",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.19390",
+    "title": "Execution-bound advisory automation for agentic AI: a reproducible AIBOM-driven CSAF-VEX framework",
+    "url": "https://arxiv.org/abs/2606.19390",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Petar Radanliev",
+      "Omar Santos",
+      "Carsten Maple",
+      "Kay Atefi"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "workflow-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.19390",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19390",
+    "pdf_url": "https://arxiv.org/pdf/2606.19390",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.17454",
+    "title": "Dissecting model behavior through agent trajectories",
+    "url": "https://arxiv.org/abs/2606.17454",
+    "published": "2026-06-16",
+    "updated": "2026-06-17",
+    "authors": [
+      "Gaurav Gupta",
+      "Vatshank Chaturvedi",
+      "Jun Huan",
+      "Anoop Deoras"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.17454",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17454",
+    "pdf_url": "https://arxiv.org/pdf/2606.17454",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.17929",
+    "title": "PreAct: Computer-Using Agents that Get Faster on Repeated Tasks",
+    "url": "https://arxiv.org/abs/2606.17929",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Bojie Li"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.17929",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17929",
+    "pdf_url": "https://arxiv.org/pdf/2606.17929",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.16465",
+    "title": "When Agent Automation Becomes Profitable: Quantifying and Insuring Autonomous AI Risk through Trace-Economic Underwriting",
+    "url": "https://arxiv.org/abs/2606.16465",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Binyan Xu",
+      "Xilin Dai",
+      "Fan Yang",
+      "Kehuan Zhang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CE"
+    ],
+    "topics": [
+      "agent-safety",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.16465",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16465",
+    "pdf_url": "https://arxiv.org/pdf/2606.16465",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.28365",
+    "title": "CAMI: Cost-Aware Agent-Guided Multi-Indexing for Semantic Retrieval",
+    "url": "https://arxiv.org/abs/2606.28365",
+    "published": "2026-06-14",
+    "updated": "2026-06-14",
+    "authors": [
+      "Adnan Qidwai",
+      "Anand Eswaran",
+      "Sonam Mishra",
+      "Jaydeep Sen",
+      "Sachindra Joshi"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "rag"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.28365",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28365",
+    "pdf_url": "https://arxiv.org/pdf/2606.28365",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.15485",
+    "title": "The Perils of Agency: How Developers Perceive, Prioritize, and Address Risks in Agentic AI Products",
+    "url": "https://arxiv.org/abs/2606.15485",
+    "published": "2026-06-13",
+    "updated": "2026-06-13",
+    "authors": [
+      "Hao-Ping Lee",
+      "Jessica He",
+      "David Piorkowski",
+      "Thomas Serban von Davier",
+      "Jodi Forlizzi",
+      "Sauvik Das"
+    ],
+    "categories": [
+      "cs.CY",
+      "cs.AI",
+      "cs.HC",
+      "cs.LG",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.15485",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15485",
+    "pdf_url": "https://arxiv.org/pdf/2606.15485",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.15335",
+    "title": "Privacy-Preserving Text Sanitization for Distributed Agents Collaboration via Disentangled Representations",
+    "url": "https://arxiv.org/abs/2606.15335",
+    "published": "2026-06-13",
+    "updated": "2026-06-13",
+    "authors": [
+      "Xuan Liu",
+      "Hefeng Zhou",
+      "Sicheng Chen",
+      "Chao Yang",
+      "Xingcheng Xu",
+      "Jingjing Qu",
+      "Jiong Lou",
+      "Jie LI",
+      "Xia Hu"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "rag"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.15335",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15335",
+    "pdf_url": "https://arxiv.org/pdf/2606.15335",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.15007",
+    "title": "Nemotron 3 Ultra: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning",
+    "url": "https://arxiv.org/abs/2606.15007",
+    "published": "2026-06-12",
+    "updated": "2026-06-12",
+    "authors": [
+      "NVIDIA",
+      ":",
+      "Aaron Blakeman",
+      "Aaron Thomas",
+      "Aastha Jhunjhunwala",
+      "Abhibha Gupta",
+      "Abhinav Khattar",
+      "Adam Rajfer",
+      "Adi Renduchintala",
+      "Adil Asif",
+      "Aditya Vavre",
+      "Adriana Flores Miranda",
+      "Ahmad Bilal",
+      "Aileen Zaman",
+      "Ajay Hotchandani",
+      "Akanksha Shukla",
+      "Akhiad Bercovich",
+      "Aleksander Ficek",
+      "Alex Gronskiy",
+      "Alex Kondratenko",
+      "Alex Steiner",
+      "Alex Ye",
+      "Alexander Bukharin",
+      "Alexandre Milesi",
+      "Ali Taghibakhshi",
+      "Alice Gatti",
+      "Alisa Liu",
+      "Alok Kumar",
+      "Amar Phanishayee",
+      "Ameya Sunil Mahabaleshwarkar",
+      "Amir Klein",
+      "Amit Zuker",
+      "Amnon Geifman",
+      "Anahita Bhiwandiwalla",
+      "Ananth Subramaniam",
+      "Andrea Santilli",
+      "Andrew Fulks",
+      "Andrew McHarg",
+      "Andrew Tao",
+      "Andrii Skliar",
+      "Anjulie Agrusa",
+      "Ankur Srivastava",
+      "Ankur Verma",
+      "Anna Shors",
+      "Anna Warno",
+      "Antoni-Joan Solergibert I Llaquet",
+      "Arham Mehta",
+      "Arkadiusz Nowaczynski",
+      "Arti Jain",
+      "Ashwath Aithal",
+      "Ashwin Poojary",
+      "Asif Ahamed",
+      "Asit Mishra",
+      "Asma Kuriparambil Thekkumpate",
+      "Atefeh Sohrabizadeh",
+      "Avinash Kaur",
+      "Avinash Vem",
+      "Ayush Dattagupta",
+      "Barath Subramaniam Anandan",
+      "Bardiya Sadeghi",
+      "Ben Lanir",
+      "Benedikt Schifferer",
+      "Besmira Nushi",
+      "Bilal Kartal",
+      "Bill Thiede",
+      "Bita Darvish Rouhani",
+      "Bo Deng",
+      "Bob Schatz",
+      "Boris Ginsburg",
+      "Boxin Wang",
+      "Brad Nemire",
+      "Brandon Norick",
+      "Brian Dang",
+      "Brian Westphal",
+      "Brian Yu",
+      "Brucek Khailany",
+      "Bryan Catanzaro",
+      "Carlo del Mundo",
+      "Caryln Aarish",
+      "Chankyu Lee",
+      "Chantal Hwang",
+      "Charbel Sakr",
+      "Charles Wang",
+      "Charlie Truong",
+      "Chen Cui",
+      "Cheng Cheng",
+      "Cheng-Ping Hsieh",
+      "Chenghao Zhang",
+      "Chenhui Deng",
+      "Chintan Patel",
+      "Chris Alexiuk",
+      "Christian Cosgrove",
+      "Christian Munley",
+      "Christine Harvey",
+      "Christopher Parisien",
+      "Chunyang Shen",
+      "Coco Li",
+      "Collin Neale",
+      "Cynthia Gao",
+      "Cyril Meurillon",
+      "Dan Gil",
+      "Dan Su",
+      "Dan Zhao",
+      "Dane Corneil",
+      "Daniel Afrimi",
+      "Daniel Egert",
+      "Daniel Korzekwa",
+      "Daniel Lo",
+      "Daniel Machlab",
+      "Daniel Serebrenik",
+      "Daniil Sorokin",
+      "Daria Gitman",
+      "Daria Levy",
+      "Darko Stosic",
+      "David Mosallanezhad",
+      "David Yu",
+      "Davit Karamyan",
+      "Deena Donia",
+      "Deep Debroy",
+      "Deepak Narayanan",
+      "Devin O'Kelly",
+      "Dheeraj Peri",
+      "Dhruv Nathawani",
+      "Di",
+      "Wu",
+      "Dima Rekesh",
+      "Divyanshu Kakwani",
+      "Donald Plummer",
+      "Dong Anh",
+      "Dongfeng Yu",
+      "Dongfu Jiang",
+      "Donnie Kim",
+      "Dorrin Poorkay",
+      "Duncan Riach",
+      "Dusan Stosic",
+      "Dustin VanStee",
+      "Eavan Meng",
+      "Edgar Minasyan",
+      "Edward Lin",
+      "Eileen Margaret Peters Long",
+      "Elad Sarafin",
+      "Elad Segal",
+      "Elena Lantz",
+      "Ellie Evans",
+      "Elliott Ning",
+      "Eric Chung",
+      "Eric Harper",
+      "Eric Pham-Hung",
+      "Eric Tramel",
+      "Eric Yang",
+      "Erick Galinkin",
+      "Erik Pounds",
+      "Erika Goncalves Goncalves",
+      "Evan Briones",
+      "Evan Wu",
+      "Evelina Bakhturina",
+      "Evgeny Tsykunov",
+      "Ewa Dobrowolska",
+      "Faisal Ladhak",
+      "Farzan Memarian",
+      "Fay Wang",
+      "Fei Jia",
+      "Felipe Soares",
+      "Felipe Vieira Frujeri",
+      "Feng Chen",
+      "Fengguang Lin",
+      "Ferenc Galko",
+      "Frank Sun",
+      "Frankie Siino",
+      "Frida Hou",
+      "Gal Hubara Agam",
+      "Gal Kaplun",
+      "Gantavya Bhatt",
+      "Gargi Prasad",
+      "Garvit Kulshreshtha",
+      "George Armstrong",
+      "Gerald Shen",
+      "Giulio Borghesi",
+      "Gordana Neskovic",
+      "Gorkem Batmaz",
+      "Grace Lam",
+      "Greg Mason",
+      "Greg Pauloski",
+      "Grigor Nalbandyan",
+      "Grzegorz Chlebus",
+      "Grzegorz Karch",
+      "Guan-Ting Liu",
+      "Guoming Zhang",
+      "Guyue Huang",
+      "Haggai Maron",
+      "Haifeng Qian",
+      "Haim Elisha",
+      "Haoxing Ren",
+      "Haran Kumar Shiv Kumar",
+      "Haribhau Hud",
+      "Harris Nover",
+      "Harrison Saturley Hall",
+      "Hayate Iso",
+      "Helen Ngo",
+      "Herbert Hum",
+      "Herman Sahota",
+      "Hexin Wang",
+      "Himanshu Soni",
+      "Hovhannes Tamoyan",
+      "Hua Li",
+      "Huanhuan Chen",
+      "Hui Li",
+      "Hui Wang",
+      "Huy Nguyen",
+      "Ian Chiles",
+      "Ido Galil",
+      "Ido Shahaf",
+      "Igor Gitman",
+      "Igor Shovkun",
+      "Ilya Loshchilov",
+      "Ingo Guehring",
+      "Itamar Schen",
+      "Itay Levy",
+      "Itay Neeman",
+      "Ivan Moshkov",
+      "Izik Golan",
+      "Izzy Putterman",
+      "Jaemin Choi",
+      "Jakub Slowikowski",
+      "Jan Kautz",
+      "Jane Polak Scowcroft",
+      "Jared Casper",
+      "Jatin Mitra",
+      "Jeffrey Glick",
+      "Jenny Chen",
+      "Jesse Oliver",
+      "Jiacheng Xu",
+      "Jiafan Zhu",
+      "Jialin Song",
+      "Jian Zhang",
+      "Jiantao Jiao",
+      "Jiaqi Zeng",
+      "Jie Lou",
+      "Jim King",
+      "Jimmy Zhang",
+      "Jingquan Wang",
+      "Jinhang Choi",
+      "Jinju Chu",
+      "Joey Conway",
+      "Joey Guman",
+      "Johan Jatko",
+      "Johannes Rausch",
+      "John Kamalu",
+      "John Roberts",
+      "Johnny Greco",
+      "Johnny Mensel",
+      "Jonah Alben",
+      "Jonas Yang",
+      "Jonathan Cohen",
+      "Jonathan Raiman",
+      "Joseph Jennings",
+      "Joshua Mabry",
+      "Joshua Pierce",
+      "Joyjit Daw",
+      "Julien Veron Vialard",
+      "Junkeun Yi",
+      "Jupinder Parmar",
+      "Kajal Jain",
+      "Kan Zhu",
+      "Kari Briski",
+      "Katherine Cheung",
+      "Katherine Luna",
+      "Keith Willowhawk",
+      "Keith Wyss",
+      "Keshav Santhanam",
+      "Kevin Shih",
+      "Kezhi Kong",
+      "Khanh Nguyen",
+      "Khushi Bhardwaj",
+      "Kirthi Shankar Sivamani",
+      "Konstantinos Krommydas",
+      "Krishna C. Puvvada",
+      "Krzysztof Pawelec",
+      "Kumar Anik",
+      "Kyle Keprios",
+      "Kylie Day",
+      "Lawrence McAfee",
+      "Leo Du",
+      "Leon Derczynski",
+      "Li Ding",
+      "Linda Liu",
+      "Lingjie Wu",
+      "Lior Kadoch",
+      "Lizzie Wei",
+      "Luis Vega",
+      "Luke Robison",
+      "Lun Su",
+      "Maarten Van Segbroeck",
+      "Maciej Jakub Mikulski",
+      "Maer Rodrigues de Melo",
+      "Magda Sypula",
+      "Mahan Fathi",
+      "Makesh Narsimhan Sreedhar",
+      "Makesh Tarun Chandran",
+      "Manoj Kilaru",
+      "Maor Ashkenazi",
+      "Marc Cuevas",
+      "Marc Romeijn",
+      "Marcin Chochowski",
+      "Mark Cai",
+      "Mark Mozolewski",
+      "Markus Kliegl",
+      "Marta Stepniewska-Dziubinska",
+      "Martyna Patelka",
+      "Mattei Machczynski",
+      "Matvei Novikov",
+      "Mauricio Ferrato",
+      "Maximilian Golub",
+      "Mehrzad Samadi",
+      "Melissa Corpuz",
+      "Mengru Wang",
+      "Mengxi Wu",
+      "Meredith Price",
+      "Meriem Boubdir",
+      "Micah Schaffer",
+      "Michael Andersch",
+      "Michael Boone",
+      "Michael Gschwind",
+      "Michael Lightstone",
+      "Michael Loh",
+      "Michal Bien",
+      "Michal Zawalski",
+      "Michelle Gill",
+      "Miguel Martinez",
+      "Mikail Khona",
+      "Mike Chrzanowski",
+      "Mike Houston",
+      "Mingyuan Ma",
+      "Minseok Lee",
+      "Mohamed Fawzy",
+      "Mohammad Dabbah",
+      "Mohammad Shoeybi",
+      "Mostofa Patwary",
+      "Nabin Mulepati",
+      "Najeeb Nabwani",
+      "Namit Dhameja",
+      "Narimane Hennouni",
+      "Natalie Hereth",
+      "Nathaniel Pinckney",
+      "Nave Algarici",
+      "Nave Assaf",
+      "Netanel Haber",
+      "Nicholas Knight",
+      "Nick Reamaroon",
+      "Nickson Quak",
+      "Nidhi Bhatia",
+      "Nikhil Desai",
+      "Nikolai Ludwig",
+      "Nima Tajbakhsh",
+      "Ning Xu",
+      "Nir Ailon",
+      "Nirmal Juluru",
+      "Nitin Nitin",
+      "Ofri Masad",
+      "Oleg Rybakov",
+      "Oleksii Hrinchuk",
+      "Oleksii Kuchaiev",
+      "Olivia Viessmann",
+      "Olivier Delalleau",
+      "Oluwatobi Olabiyi",
+      "Omer Ullman Argov",
+      "Omri Puny",
+      "Oren Tropp",
+      "Pablo Ribalta",
+      "Pallab Bhattacharya",
+      "Panos Lampropoulos",
+      "Parth Mannan",
+      "Pasha Shamis",
+      "Patrick Legresley",
+      "Paul Gibbons",
+      "Pavlo Molchanov",
+      "Pawel Morkisz",
+      "Peter Dykas",
+      "Peter Jin",
+      "Pierre-Yves Aquilanti",
+      "Pinky Xu",
+      "Piotr Januszewski",
+      "Piotr Laskiewicz",
+      "Pooya Jannaty",
+      "Prakash Gurumurthy",
+      "Pranav Prashant Thombre",
+      "Prasoon Varshney",
+      "Pritam Gundecha",
+      "Przemek Tredak",
+      "Puhui Meng",
+      "Qiyu Wan",
+      "Rabeeh Karimi Mahabadi",
+      "Rachel Oberman",
+      "Rachit Garg",
+      "Radha Sri-Tharan",
+      "Rahul Kandu",
+      "Rakshit Sanadhya",
+      "Ran El-Yaniv",
+      "Ran Zilberstein",
+      "Rasoul Shafipour",
+      "Ray Macalisang",
+      "Rayen Tian",
+      "Reka Kovacs",
+      "Renjie Pi",
+      "Rick Izzo",
+      "Rima Shahbazyan",
+      "Rishabh Garg",
+      "Rishi Puri",
+      "Rita Fernandes Neves",
+      "Ritchie Zhao",
+      "Ritika Borkar",
+      "Ritu Gala",
+      "Riyad Islam",
+      "Robert Clark",
+      "Robert Hesse",
+      "Robert Kirby",
+      "Roger Waleffe",
+      "Rohit Watve",
+      "Roi Koren",
+      "Ron Banner",
+      "Ruoxi Zhang",
+      "Russell J. Hewett",
+      "Ryan Prenger",
+      "Ryan Stewart",
+      "Ryota Egashira",
+      "Sadegh Mahdavi",
+      "Saee Paliwal",
+      "Sagar Singh",
+      "Sahil Modi",
+      "Salika Dave",
+      "Samantha Shinagawa",
+      "Samuel Kriman",
+      "Sandip Bhaskar",
+      "Sangkug Lym",
+      "Sanjay Kariyappa",
+      "Sanjeev Satheesh",
+      "Saran Vikas Murari",
+      "Satish Pasumarthi",
+      "Saurabh Mishra",
+      "Saurav Muralidharan",
+      "Scott Hara",
+      "Sean Narentharen",
+      "Selvaraj Anandaraj",
+      "Seonjin Na",
+      "Seonmeyong Bak",
+      "Seonmyeong Bak",
+      "Sepehr Sameni",
+      "Seph Mard",
+      "Serge Panev",
+      "Seth Henneman",
+      "Seth Poulos",
+      "Shahar Mor",
+      "Shantanu Acharya",
+      "Shaona Ghosh",
+      "Sharath Turuvekere Sreenivas",
+      "Sharon Mendelson",
+      "Shaun Kotek",
+      "Shawn Wang",
+      "Shay Aharon",
+      "Shaya Gharghabi",
+      "Sheng-Chieh Lin",
+      "Shi Chen",
+      "Shiqing Fan",
+      "Shirish Baskaran",
+      "Shreya Gopa",
+      "Shrimai Prabhumoye",
+      "Shubham Pachori",
+      "Shubham Toshniwal",
+      "Shuoyang Ding",
+      "Shwetha Krishnamurthy",
+      "Siddharth Singh",
+      "Simeng Sun",
+      "Sirshak Das",
+      "Sivakumar Arayandi Thottakara",
+      "Smita Ithape",
+      "Somshubra Majumdar",
+      "Soumye Singhal",
+      "Sri Harsha Singudasu",
+      "Sridhar Bhuvanapalli",
+      "Srimukh Veccham",
+      "Stas Sergienko",
+      "Stefania Alborghetti",
+      "Stephen Ge",
+      "Su Rong",
+      "Sugam Dipak Devare",
+      "Sukrit Rao",
+      "Sumeet Kumar Barua",
+      "Sungsoo Ha",
+      "Sunny Gai",
+      "Suriya Gunasekar",
+      "Suseella Panguluri",
+      "Suyog Gupta",
+      "Sviataslau Hinzburh",
+      "Sweta Priyadarshi",
+      "Syeda Nahida Akter",
+      "Talor Abramovich",
+      "Tan Bui",
+      "Tanay Varshney",
+      "Tatevik Ter-Hovhannisyan",
+      "Teodor-Dumitru Ene",
+      "Terry Kong",
+      "Thanh Do",
+      "Tianhe Zhang",
+      "Tiffany Moore",
+      "Tijmen Blankevoort",
+      "Tim Moon",
+      "Tiyasa Mitra",
+      "Tom Balough",
+      "Tomasz Grzegorzek",
+      "Tomasz Hliwiak",
+      "Tomer Asida",
+      "Tomer Bar Natan",
+      "Tomer Keren",
+      "Tomer Ronen",
+      "Tony Salim",
+      "Tony Wang",
+      "Traian Rebedea",
+      "Tugrul Konuk",
+      "Twinkle Vashishth",
+      "Udi Karpas",
+      "Ushnish De",
+      "Vahid Noorozi",
+      "Venkat Srinivasan",
+      "Venmugil Elango",
+      "Vibhor Agrawal",
+      "Victor Cui",
+      "Vijay Korthikanti",
+      "Vikas Mehta",
+      "Vinay Rao",
+      "Virginia Wu",
+      "Vitaly Kurin",
+      "Vitaly Lavrukhin",
+      "Vladimir Anisimov",
+      "Vu Pham",
+      "Wanli Jiang",
+      "Wasi Uddin Ahmad",
+      "Wataru Ishihara",
+      "Wei Du",
+      "Wei Ping",
+      "Weiheng Chai",
+      "Wenliang Dai",
+      "Wesley Helmholz",
+      "Will Jennings",
+      "Will Zhu",
+      "Wojciech Prazuch",
+      "Xiaowei Ren",
+      "Xiwen Yu",
+      "Yan Breek",
+      "Yang Chen",
+      "Yang Yu",
+      "Yangyi Chen",
+      "Yaniv Galron",
+      "Yashaswi Karnati",
+      "Yejin Choi",
+      "Yev Meyer",
+      "Yi-Fu Wu",
+      "Yian Zhang",
+      "Ying Lin",
+      "Yonatan Geifman",
+      "Yonggan Fu",
+      "Youngeun Kwon",
+      "Yu Yao",
+      "Yugi Guvvla",
+      "Yuki Huang",
+      "Yunsheng Liu",
+      "Zach Moshe",
+      "Zachary Newell",
+      "Zhilin Wang",
+      "Zhiyu Li",
+      "Zhongbo Zhu",
+      "Zhuolin Yang",
+      "Zihan Liu",
+      "Zijie Yan",
+      "Zsolt-Alon Wertheimer"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "reasoning"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.15007",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15007",
+    "pdf_url": "https://arxiv.org/pdf/2606.15007",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.14266",
+    "title": "Large Language Model Based Agent for Automated Discovery in Computational Physics",
+    "url": "https://arxiv.org/abs/2606.14266",
+    "published": "2026-06-12",
+    "updated": "2026-06-12",
+    "authors": [
+      "Hang Lin",
+      "Chongwen Liu",
+      "Gang Yan"
+    ],
+    "categories": [
+      "physics.comp-ph"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.14266",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.14266",
+    "pdf_url": "https://arxiv.org/pdf/2606.14266",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.12830",
+    "title": "Perceive, Interact, Reason: Building Tool-Augmented Visual Agents for Spatial Reasoning",
+    "url": "https://arxiv.org/abs/2606.12830",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Changye Li",
+      "Meng Lu",
+      "Yi Wu",
+      "Ligeng Zhu"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.12830",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12830",
+    "pdf_url": "https://arxiv.org/pdf/2606.12830",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.13380",
+    "title": "An LLM System for Autonomous Variational Quantum Circuit Design",
+    "url": "https://arxiv.org/abs/2606.13380",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Kenya Sakka",
+      "Wataru Mizukami",
+      "Kosuke Mitarai"
+    ],
+    "categories": [
+      "quant-ph",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.13380",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13380",
+    "pdf_url": "https://arxiv.org/pdf/2606.13380",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.12666",
+    "title": "CAPED: Context-Aware Privacy Exposure Defense for Mobile GUI Agents",
+    "url": "https://arxiv.org/abs/2606.12666",
+    "published": "2026-06-10",
+    "updated": "2026-06-16",
+    "authors": [
+      "Siyu Shen",
+      "Fenghao Xu",
+      "Wenrui Diao",
+      "Kehuan Zhang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.12666",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12666",
+    "pdf_url": "https://arxiv.org/pdf/2606.12666",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.10522",
+    "title": "GUI-AC: Enhancing Continual Learning in GUI Agents",
+    "url": "https://arxiv.org/abs/2606.10522",
+    "published": "2026-06-09",
+    "updated": "2026-07-06",
+    "authors": [
+      "Can Lin",
+      "Tao Feng",
+      "Hangjie Yuan",
+      "Dan Zhang",
+      "Yifan Zhu",
+      "Zhonghong Ou"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.10522",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10522",
+    "pdf_url": "https://arxiv.org/pdf/2606.10522",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.10062",
+    "title": "Deployment-Time Memorization in Foundation-Model Agents",
+    "url": "https://arxiv.org/abs/2606.10062",
+    "published": "2026-06-08",
+    "updated": "2026-06-08",
+    "authors": [
+      "Lei",
+      "Chen",
+      "Guilin Zhang",
+      "Kai Zhao",
+      "Dalmo Cirne",
+      "Andy Olsen",
+      "Xu Chu",
+      "Zeke Miller",
+      "Alet Blanken",
+      "Amine Anoun",
+      "Jerry Ting"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.10062",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10062",
+    "pdf_url": "https://arxiv.org/pdf/2606.10062",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.06708",
+    "title": "Signal-Driven Observation for Long-Horizon Web Agents",
+    "url": "https://arxiv.org/abs/2606.06708",
+    "published": "2026-06-04",
+    "updated": "2026-06-04",
+    "authors": [
+      "Shubham Gaur",
+      "Ian Lane"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "computer-use",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.06708",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.06708",
+    "pdf_url": "https://arxiv.org/pdf/2606.06708",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.04391",
+    "title": "Online Skill Learning for Web Agents via State-Grounded Dynamic Retrieval",
+    "url": "https://arxiv.org/abs/2606.04391",
+    "published": "2026-06-03",
+    "updated": "2026-06-03",
+    "authors": [
+      "Jiaxi Li",
+      "Ke Deng",
+      "Yun Wang",
+      "Jingyuan Huang",
+      "Yucheng Shi",
+      "Qiaoyu Tan",
+      "Jin Lu",
+      "Ninghao Liu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.04391",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.04391",
+    "pdf_url": "https://arxiv.org/pdf/2606.04391",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.05112",
+    "title": "Evaluating Large Language Models in Dynamic Clinical Decision-Making with Standardized Patient Cases",
+    "url": "https://arxiv.org/abs/2606.05112",
+    "published": "2026-06-03",
+    "updated": "2026-06-03",
+    "authors": [
+      "Cheng Liang",
+      "Pengcheng Qiu",
+      "Ya Zhang",
+      "Yanfeng Wang",
+      "Chaoyi Wu",
+      "Weidi Xie"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "world-model"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.05112",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.05112",
+    "pdf_url": "https://arxiv.org/pdf/2606.05112",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.04435",
+    "title": "Cascading Hallucination in Agentic RAG: The CHARM Framework for Detection and Mitigation",
+    "url": "https://arxiv.org/abs/2606.04435",
+    "published": "2026-06-03",
+    "updated": "2026-06-03",
+    "authors": [
+      "Saroj Mishra"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.CR",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.04435",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.04435",
+    "pdf_url": "https://arxiv.org/pdf/2606.04435",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.28347",
+    "title": "Agentic Safety is an Epistemic Property, Not a Behavioral One",
+    "url": "https://arxiv.org/abs/2606.28347",
+    "published": "2026-06-02",
+    "updated": "2026-06-02",
+    "authors": [
+      "Charles L. Wang",
+      "Keir Dorchen",
+      "Peter Jin"
+    ],
+    "categories": [
+      "cs.CY",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-safety",
+      "embodied-agent",
+      "rag"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2606.28347",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28347",
+    "pdf_url": "https://arxiv.org/pdf/2606.28347",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.28344",
+    "title": "PIXELRAG: Web Screenshots Beat Text for Retrieval-Augmented Generation",
+    "url": "https://arxiv.org/abs/2606.28344",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Yichuan Wang",
+      "Zhifei Li",
+      "Zirui Wang",
+      "Paul Teiletche",
+      "Lesheng Jin",
+      "Matei Zaharia",
+      "Joseph E. Gonzalez",
+      "Sewon Min"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI",
+      "cs.CL",
+      "cs.CV",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-evaluation",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.28344",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28344",
+    "pdf_url": "https://arxiv.org/pdf/2606.28344",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.00183",
+    "title": "Agentic Transformers Provably Learn to Search via Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2606.00183",
+    "published": "2026-05-29",
+    "updated": "2026-05-29",
+    "authors": [
+      "Tong Yang",
+      "Yu Huang",
+      "Yingbin Liang",
+      "Yuejie Chi"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "math.OC",
+      "stat.ML"
+    ],
+    "topics": [
+      "memory",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.00183",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.00183",
+    "pdf_url": "https://arxiv.org/pdf/2606.00183",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.30916",
+    "title": "Welfare, Improvability, and Variance: A Principal-Agent Approach to Optimal Benchmark Item Aggregation",
+    "url": "https://arxiv.org/abs/2605.30916",
+    "published": "2026-05-29",
+    "updated": "2026-05-29",
+    "authors": [
+      "Andreas Haupt",
+      "Justin Hartenstein",
+      "Anka Reuel",
+      "Mykel Kochenderfer",
+      "Sanmi Koyejo"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.GT",
+      "econ.TH"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2605.30916",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.30916",
+    "pdf_url": "https://arxiv.org/pdf/2605.30916",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2605.29786",
+    "title": "Croissant Tasks: A Metadata Format for Reproducible Machine Learning Evaluations",
+    "url": "https://arxiv.org/abs/2605.29786",
+    "published": "2026-05-28",
+    "updated": "2026-05-28",
+    "authors": [
+      "Omar Benjelloun",
+      "Leonardo Martins Bianco",
+      "Isabelle Guyon",
+      "Thanh Gia Hieu Khuong",
+      "Jonathan Lebensold",
+      "Sebastian Lobentanzer",
+      "Luis Oala",
+      "Benedictus Kent Rachmat",
+      "Ihsan Ullah",
+      "Peyman Vahidi",
+      "Joaquin Vanschoren"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.29786",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.29786",
+    "pdf_url": "https://arxiv.org/pdf/2605.29786",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.27995",
+    "title": "AsyncTool: Evaluating the Asynchronous Function Calling Capability under Multi-Task Scenarios",
+    "url": "https://arxiv.org/abs/2605.27995",
+    "published": "2026-05-27",
+    "updated": "2026-05-28",
+    "authors": [
+      "Kou Shi",
+      "Ziao Zhang",
+      "Shiting Huang",
+      "Avery Nie",
+      "Zhen Fang",
+      "Qiuchen Wang",
+      "Lin Chen",
+      "Huaian Chen",
+      "Zehui Chen",
+      "Feng Zhao"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2605.27995",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.27995",
+    "pdf_url": "https://arxiv.org/pdf/2605.27995",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2605.23459",
+    "title": "AI Assurance: A Comprehensive Testing Strategy for Enterprise AI Systems",
+    "url": "https://arxiv.org/abs/2605.23459",
+    "published": "2026-05-22",
+    "updated": "2026-05-22",
+    "authors": [
+      "Chitra Badagi",
+      "Divye Singh",
+      "Animesh Sen",
+      "Adinath Shirsath"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.23459",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.23459",
+    "pdf_url": "https://arxiv.org/pdf/2605.23459",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.20896",
+    "title": "GenAI-Driven Threat Detection with Microsoft Security Copilot",
+    "url": "https://arxiv.org/abs/2605.20896",
+    "published": "2026-05-20",
+    "updated": "2026-05-22",
+    "authors": [
+      "Scott Freitas",
+      "Amir Gharib"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "planning",
+      "rag"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.20896",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.20896",
+    "pdf_url": "https://arxiv.org/pdf/2605.20896",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.20477",
+    "title": "Training Language Agents to Learn from Experience",
+    "url": "https://arxiv.org/abs/2605.20477",
+    "published": "2026-05-19",
+    "updated": "2026-05-19",
+    "authors": [
+      "Yuval Shalev",
+      "Zifeng Ding",
+      "Mateja Jamnik"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.20477",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.20477",
+    "pdf_url": "https://arxiv.org/pdf/2605.20477",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.15077",
+    "title": "Concurrency without Model Changes: Future-based Asynchronous Function Calling for LLMs",
+    "url": "https://arxiv.org/abs/2605.15077",
+    "published": "2026-05-14",
+    "updated": "2026-05-14",
+    "authors": [
+      "Guangyu Feng",
+      "Huanzhi Mao",
+      "Prabal Dutta",
+      "Joseph E. Gonzalez"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2605.15077",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.15077",
+    "pdf_url": "https://arxiv.org/pdf/2605.15077",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2605.12460",
+    "title": "Multi-Stream LLMs: Unblocking Language Models with Parallel Streams of Thoughts, Inputs and Outputs",
+    "url": "https://arxiv.org/abs/2605.12460",
+    "published": "2026-05-12",
+    "updated": "2026-05-12",
+    "authors": [
+      "Guinan Su",
+      "Yanwu Yang",
+      "Xueyan Li",
+      "Jonas Geiping"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.12460",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.12460",
+    "pdf_url": "https://arxiv.org/pdf/2605.12460",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.11442",
+    "title": "Can a Single Message Paralyze the AI Infrastructure? The Rise of AbO-DDoS Attacks through Targeted Mobius Injection",
+    "url": "https://arxiv.org/abs/2605.11442",
+    "published": "2026-05-12",
+    "updated": "2026-05-12",
+    "authors": [
+      "Zi Liang",
+      "Ronghua Li",
+      "Yanyun Wang",
+      "Qingqing Ye",
+      "Haibo Hu"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.11442",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.11442",
+    "pdf_url": "https://arxiv.org/pdf/2605.11442",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.11235",
+    "title": "Internalizing Curriculum Judgment for LLM Reinforcement Fine-Tuning",
+    "url": "https://arxiv.org/abs/2605.11235",
+    "published": "2026-05-11",
+    "updated": "2026-05-11",
+    "authors": [
+      "Han Zheng",
+      "Yining Ma",
+      "Karthick Gunasekaran",
+      "Bharathan Balaji",
+      "Zheng Du",
+      "Shiv Vitaladevuni",
+      "Cathy Wu"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "reasoning"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2605.11235",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.11235",
+    "pdf_url": "https://arxiv.org/pdf/2605.11235",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2605.09990",
+    "title": "Merlin: Deterministic Byte-Exact Deduplication for Lossless Context Optimization in Large Language Model Inference",
+    "url": "https://arxiv.org/abs/2605.09990",
+    "published": "2026-05-11",
+    "updated": "2026-05-11",
+    "authors": [
+      "Sietse Schelpe"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.09990",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.09990",
+    "pdf_url": "https://arxiv.org/pdf/2605.09990",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.06320",
+    "title": "Improving the Efficiency of Language Agent Teams with Adaptive Task Graphs",
+    "url": "https://arxiv.org/abs/2605.06320",
+    "published": "2026-05-07",
+    "updated": "2026-05-07",
+    "authors": [
+      "Elizabeth Mieczkowski",
+      "Alexander Ku",
+      "Tiwalayo Eisape",
+      "Dilip Arumugam",
+      "John Matters",
+      "Katherine M. Collins",
+      "Ilia Sucholutsky",
+      "Thomas L. Griffiths"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "planning"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.06320",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.06320",
+    "pdf_url": "https://arxiv.org/pdf/2605.06320",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.06161",
+    "title": "Beyond Accuracy: Policy Invariance as a Reliability Test for LLM Safety Judges",
+    "url": "https://arxiv.org/abs/2605.06161",
+    "published": "2026-05-07",
+    "updated": "2026-05-07",
+    "authors": [
+      "Shihao Weng",
+      "Yang Feng",
+      "Xiaofei Xie"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.06161",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.06161",
+    "pdf_url": "https://arxiv.org/pdf/2605.06161",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.27264",
+    "title": "Self-Evolving Software Agents",
+    "url": "https://arxiv.org/abs/2604.27264",
+    "published": "2026-04-29",
+    "updated": "2026-04-29",
+    "authors": [
+      "Marco Robol",
+      "Paolo Giorgini"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "reasoning"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2604.27264",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.27264",
+    "pdf_url": "https://arxiv.org/pdf/2604.27264",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2604.10516",
+    "title": "Structure-Grounded Knowledge Retrieval via Code Dependencies for Multi-Step Data Reasoning",
+    "url": "https://arxiv.org/abs/2604.10516",
+    "published": "2026-04-12",
+    "updated": "2026-04-26",
+    "authors": [
+      "Xinyi Huang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "rag",
+      "reasoning"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2604.10516",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.10516",
+    "pdf_url": "https://arxiv.org/pdf/2604.10516",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2604.07960",
+    "title": "TOOLCAD: Exploring Tool-Using Large Language Models in Text-to-CAD Generation with Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2604.07960",
+    "published": "2026-04-09",
+    "updated": "2026-04-20",
+    "authors": [
+      "Yifei Gong",
+      "Xing Wu",
+      "Wenda Liu",
+      "Kang Tu"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2604.07960",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.07960",
+    "pdf_url": "https://arxiv.org/pdf/2604.07960",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.25723",
+    "title": "Natural-Language Agent Harnesses",
+    "url": "https://arxiv.org/abs/2603.25723",
+    "published": "2026-03-26",
+    "updated": "2026-05-18",
+    "authors": [
+      "Linyue Pan",
+      "Lexiao Zou",
+      "Shuo Guo",
+      "Jingchen Ni",
+      "Hai-Tao Zheng"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.25723",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.25723",
+    "pdf_url": "https://arxiv.org/pdf/2603.25723",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.17170",
+    "title": "PAuth - Precise Task-Scoped Authorization For Agents",
+    "url": "https://arxiv.org/abs/2603.17170",
+    "published": "2026-03-17",
+    "updated": "2026-03-17",
+    "authors": [
+      "Reshabh K Sharma",
+      "Linxi Jiang",
+      "Zhiqiang Lin",
+      "Shuo Chen"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.PL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "reasoning"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.17170",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.17170",
+    "pdf_url": "https://arxiv.org/pdf/2603.17170",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2602.22523",
+    "title": "Cognitive Models and AI Algorithms Provide Templates for Designing Language Agents",
+    "url": "https://arxiv.org/abs/2602.22523",
+    "published": "2026-02-26",
+    "updated": "2026-02-26",
+    "authors": [
+      "Ryan Liu",
+      "Dilip Arumugam",
+      "Cedegao E. Zhang",
+      "Sean Escola",
+      "Xaq Pitkow",
+      "Thomas L. Griffiths"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "q-bio.NC"
+    ],
+    "topics": [
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2602.22523",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.22523",
+    "pdf_url": "https://arxiv.org/pdf/2602.22523",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2602.14364",
+    "title": "A Trajectory-Based Safety Audit of Clawdbot (OpenClaw)",
+    "url": "https://arxiv.org/abs/2602.14364",
+    "published": "2026-02-16",
+    "updated": "2026-02-16",
+    "authors": [
+      "Tianyu Chen",
+      "Dongrui Liu",
+      "Xia Hu",
+      "Jingyi Yu",
+      "Wenjie Wang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.14364",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.14364",
+    "pdf_url": "https://arxiv.org/pdf/2602.14364",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2602.10007",
+    "title": "A Collaborative Safety Shield for Safe and Efficient CAV Lane Changes in Congested On-Ramp Merging",
+    "url": "https://arxiv.org/abs/2602.10007",
+    "published": "2026-02-10",
+    "updated": "2026-02-10",
+    "authors": [
+      "Bharathkumar Hegde",
+      "Melanie Bouroche"
+    ],
+    "categories": [
+      "cs.RO",
+      "cs.AI",
+      "cs.MA",
+      "eess.SY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "rag",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.10007",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.10007",
+    "pdf_url": "https://arxiv.org/pdf/2602.10007",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.09554",
+    "title": "LABBench2: An Improved Benchmark for AI Systems Performing Biology Research",
+    "url": "https://arxiv.org/abs/2604.09554",
+    "published": "2026-02-04",
+    "updated": "2026-05-05",
+    "authors": [
+      "Jon M Laurent",
+      "Albert Bou",
+      "Michael Pieler",
+      "Conor Igoe",
+      "Alex Andonian",
+      "Siddharth Narayanan",
+      "James Braza",
+      "Alexandros Sanchez Vassopoulos",
+      "Jacob L Steenwyk",
+      "Blake Lash",
+      "Andrew D White",
+      "Samuel G Rodriques"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2604.09554",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.09554",
+    "pdf_url": "https://arxiv.org/pdf/2604.09554",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.00030",
+    "title": "SimpleTool: Parallel Decoding for Real-Time LLM Function Calling",
+    "url": "https://arxiv.org/abs/2603.00030",
+    "published": "2026-02-04",
+    "updated": "2026-02-04",
+    "authors": [
+      "Xiaoxin Shi",
+      "Jiaxin Wan",
+      "Linkang Dong",
+      "Wei Jiang",
+      "Yue Liu",
+      "Zengfeng Huang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "embodied-agent",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2603.00030",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.00030",
+    "pdf_url": "https://arxiv.org/pdf/2603.00030",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2601.18282",
+    "title": "Think-Augmented Function Calling: Improving LLM Parameter Accuracy Through Embedded Reasoning",
+    "url": "https://arxiv.org/abs/2601.18282",
+    "published": "2026-01-26",
+    "updated": "2026-02-06",
+    "authors": [
+      "Lei Wei",
+      "Xiao Peng",
+      "Jinpeng Ou",
+      "Bin Wang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2601.18282",
+    "source": "arxiv",
+    "source_id": "arxiv:2601.18282",
+    "pdf_url": "https://arxiv.org/pdf/2601.18282",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2603.21013",
+    "title": "A Framework for Low-Latency, LLM-driven Multimodal Interaction on the Pepper Robot",
+    "url": "https://arxiv.org/abs/2603.21013",
+    "published": "2026-01-09",
+    "updated": "2026-01-09",
+    "authors": [
+      "Erich Studerus",
+      "Vivienne Jia Zhong",
+      "Stephan Vonschallen"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG",
+      "cs.RO"
+    ],
+    "topics": [
+      "computer-use",
+      "embodied-agent",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2603.21013",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.21013",
+    "pdf_url": "https://arxiv.org/pdf/2603.21013",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2512.11277",
+    "title": "When Actions Teach You to Think: Reasoning-Action Synergy via Reinforcement Learning in Conversational Agents",
+    "url": "https://arxiv.org/abs/2512.11277",
+    "published": "2025-12-12",
+    "updated": "2025-12-12",
+    "authors": [
+      "Mrinal Rawat",
+      "Arkajyoti Chakraborty",
+      "Neha Gupta",
+      "Roberto Pieraccini"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "computer-use",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2512.11277",
+    "source": "arxiv",
+    "source_id": "arxiv:2512.11277",
+    "pdf_url": "https://arxiv.org/pdf/2512.11277",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2512.01270",
+    "title": "Egent: An Autonomous Agent for Equivalent Width Measurement",
+    "url": "https://arxiv.org/abs/2512.01270",
+    "published": "2025-12-01",
+    "updated": "2026-05-06",
+    "authors": [
+      "Yuan-Sen Ting",
+      "Serat Mahmud Saad",
+      "Fan Liu",
+      "Yuting Shen"
+    ],
+    "categories": [
+      "astro-ph.IM",
+      "astro-ph.GA",
+      "astro-ph.SR"
+    ],
+    "topics": [
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2512.01270",
+    "source": "arxiv",
+    "source_id": "arxiv:2512.01270",
+    "pdf_url": "https://arxiv.org/pdf/2512.01270",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2509.00482",
+    "title": "Talk Less, Call Right: Enhancing Role-Play LLM Agents with Automatic Prompt Optimization and Role Prompting",
+    "url": "https://arxiv.org/abs/2509.00482",
+    "published": "2025-08-30",
+    "updated": "2025-10-12",
+    "authors": [
+      "Saksorn Ruangtanusak",
+      "Pittawat Taveekitworachai",
+      "Kunat Pipatanakul"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.HC"
+    ],
+    "topics": [
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2509.00482",
+    "source": "arxiv",
+    "source_id": "arxiv:2509.00482",
+    "pdf_url": "https://arxiv.org/pdf/2509.00482",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2508.20931",
+    "title": "How Can Input Reformulation Improve Tool Usage Accuracy in a Complex Dynamic Environment? A Study on $τ$-bench",
+    "url": "https://arxiv.org/abs/2508.20931",
+    "published": "2025-08-28",
+    "updated": "2025-09-01",
+    "authors": [
+      "Venkatesh Mishra",
+      "Amir Saeidi",
+      "Satyam Raj",
+      "Mutsumi Nakamura",
+      "Jayanth Srinivasa",
+      "Gaowen Liu",
+      "Ali Payani",
+      "Chitta Baral"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 11,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2508.20931",
+    "source": "arxiv",
+    "source_id": "arxiv:2508.20931",
+    "pdf_url": "https://arxiv.org/pdf/2508.20931",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2607.05999",
+    "title": "AgoraSim: A Hybrid Agent-Based Modeling Framework",
+    "url": "https://arxiv.org/abs/2607.05999",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Chung-Chi Chen"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.05999",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05999",
+    "pdf_url": "https://arxiv.org/pdf/2607.05999",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.05975",
+    "title": "MCP-Enabled Agentic AI for Autonomous IPoDWDM Network Lifecycle Automation",
+    "url": "https://arxiv.org/abs/2607.05975",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Chunmin Xia",
+      "Jakub Harbaczewski",
+      "Nikhil Dsilva",
+      "Julie Raulin",
+      "Dominic Schneider",
+      "Achim Autenrieth"
+    ],
+    "categories": [
+      "cs.NI",
+      "cs.AI",
+      "cs.MA",
+      "eess.SY"
+    ],
+    "topics": [
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.05975",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05975",
+    "pdf_url": "https://arxiv.org/pdf/2607.05975",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.05958",
+    "title": "Agentic AI for IPoDWDM Network Lifecycle Automation: An MCP-Enabled Architecture",
+    "url": "https://arxiv.org/abs/2607.05958",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Chunmin Xia",
+      "Jakub Harbaczewski",
+      "Nikhil Dsilva",
+      "Julie Raulin",
+      "Dominic Schneider",
+      "Achim Autenrieth"
+    ],
+    "categories": [
+      "cs.NI",
+      "cs.AI",
+      "eess.SP",
+      "eess.SY"
+    ],
+    "topics": [
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.05958",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05958",
+    "pdf_url": "https://arxiv.org/pdf/2607.05958",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.06155",
+    "title": "When Does Tool Use Increase the Expressive Power of Finite-Precision Recurrent Models?",
+    "url": "https://arxiv.org/abs/2607.06155",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Nikola Zubić",
+      "Qian Li",
+      "Yuyi Wang",
+      "Davide Scaramuzza"
+    ],
+    "categories": [
+      "cs.FL",
+      "cs.CC",
+      "cs.CL"
+    ],
+    "topics": [
+      "memory",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2607.06155",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06155",
+    "pdf_url": "https://arxiv.org/pdf/2607.06155",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2607.05762",
+    "title": "Articulating Assumptions in AI-Generated Scientific Analyses through Task Decomposition",
+    "url": "https://arxiv.org/abs/2607.05762",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Ahmed Hammad",
+      "Mihoko Nojiri"
+    ],
+    "categories": [
+      "cs.SE",
+      "hep-ex",
+      "hep-ph"
+    ],
+    "topics": [
+      "computer-use",
+      "multi-agent",
+      "planning",
+      "workflow-agent"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.05762",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05762",
+    "pdf_url": "https://arxiv.org/pdf/2607.05762",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.04574",
+    "title": "A Few Teacher Steps Go a Long Way: Cost-Efficient On-Policy Data Augmentation for Agent Post-Training",
+    "url": "https://arxiv.org/abs/2607.04574",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Junze Ye",
+      "Jiayi Cheng",
+      "Miao Lu",
+      "Michal Mankowski",
+      "Jose Blanchet",
+      "Mohsen Bayati"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "rag",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.04574",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04574",
+    "pdf_url": "https://arxiv.org/pdf/2607.04574",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.05277",
+    "title": "Untrusted Content Masking for Web Agents with Security Guarantees",
+    "url": "https://arxiv.org/abs/2607.05277",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Kristina Nikolić",
+      "Egor Zverev",
+      "Javier Rando",
+      "Matthew Jagielski",
+      "Edoardo Debenedetti",
+      "Florian Tramèr"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "tool-use",
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2607.05277",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05277",
+    "pdf_url": "https://arxiv.org/pdf/2607.05277",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2607.04613",
+    "title": "Governed Individuation: Cryptographically Decoupling an Agent's Learning from Its Authority",
+    "url": "https://arxiv.org/abs/2607.04613",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Xue Qin",
+      "Simin Luan",
+      "Cong Yang",
+      "Zhijun Li"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2607.04613",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04613",
+    "pdf_url": "https://arxiv.org/pdf/2607.04613",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2607.04371",
+    "title": "Nemotron-Labs-3-Puzzle-75B-A9B: Compressing Hybrid MoE LLMs",
+    "url": "https://arxiv.org/abs/2607.04371",
+    "published": "2026-07-05",
+    "updated": "2026-07-07",
+    "authors": [
+      "Akhiad Bercovich",
+      "Talor Abramovich",
+      "Daniel Afrimi",
+      "Shay Aharon",
+      "Nir Ailon",
+      "Vladimir Anisimov",
+      "Omer Ullman Argov",
+      "Maor Ashkenazi",
+      "Tomer Asida",
+      "Nave Assaf",
+      "Tomer Bar Natan",
+      "Alexander Bukharin",
+      "Grzegorz Chlebus",
+      "Marcin Chochowski",
+      "Eric Chung",
+      "Mohammad Dabbah",
+      "Carlo del Mundo",
+      "Ewa Dobrowolska",
+      "Ido Galil",
+      "Yaniv Galron",
+      "Amnon Geifman",
+      "Yonatan Geifman",
+      "Izik Golan",
+      "Alex Gronskiy",
+      "Tomasz Grzegorzek",
+      "Netanel Haber",
+      "Lior Kadoch",
+      "Grzegorz Karch",
+      "Tomer Keren",
+      "Abhinav Khattar",
+      "Amir Klein",
+      "Tugrul Konuk",
+      "Roi Koren",
+      "Daniel Korzekwa",
+      "Shaun Kotek",
+      "Konstantinos Krommydas",
+      "Itay Levy",
+      "Ofri Masad",
+      "Yoav Miron",
+      "Pavlo Molchanov",
+      "Shahar Mor",
+      "Zach Moshe",
+      "Saurav Muralidharan",
+      "Najeeb Nabwani",
+      "Besmira Nushi",
+      "Mostofa Patwary",
+      "Omri Puny",
+      "Johannes Rausch",
+      "Tomer Ronen",
+      "Sepehr Sameni",
+      "Itamar Schen",
+      "Elad Segal",
+      "Daniel Serebrenik",
+      "Ido Shahaf",
+      "Soumye Singhal",
+      "Daniil Sorokin",
+      "Sharath Turuvekere Sreenivas",
+      "Marta Stepniewska-Dziubinska",
+      "Ali Taghibakhshi",
+      "Nima Tajbakhsh",
+      "Oren Tropp",
+      "Dor Tzur",
+      "Anna Warno",
+      "Yi-Fu Wu",
+      "Michal Zawalski",
+      "Jiaqi Zeng",
+      "Yian Zhang",
+      "Ran Zilberstein",
+      "Amit Zuker",
+      "Ran El-Yaniv"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "reasoning"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2607.04371",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04371",
+    "pdf_url": "https://arxiv.org/pdf/2607.04371",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.03238",
+    "title": "An Empirical Study of Downstream Adaptation for Agent Skills",
+    "url": "https://arxiv.org/abs/2607.03238",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Xinjian Wu",
+      "Jingzhi Gong",
+      "Gunel Jahangirova",
+      "Zhenpeng Chen",
+      "Jie M. Zhang"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-safety",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.03238",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03238",
+    "pdf_url": "https://arxiv.org/pdf/2607.03238",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.03048",
+    "title": "Compression, structure, and executor capability: a controlled real-cost decomposition of language-model agent skill optimisation",
+    "url": "https://arxiv.org/abs/2607.03048",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Xiaonan Xu",
+      "Wenjing Wu"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "computer-use",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2607.03048",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03048",
+    "pdf_url": "https://arxiv.org/pdf/2607.03048",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2607.02873",
+    "title": "Determinants and Limits of LLM Security-Tool Orchestration: A Study with HexStrike-AI",
+    "url": "https://arxiv.org/abs/2607.02873",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Romain Gerard",
+      "Assmaa Zeghaider",
+      "Yan Guo"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2607.02873",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02873",
+    "pdf_url": "https://arxiv.org/pdf/2607.02873",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2607.03028",
+    "title": "HETERQA: Benchmarking Record Retrieval over Multiple Heterogeneous Sources",
+    "url": "https://arxiv.org/abs/2607.03028",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Yaodong Su",
+      "Hanchang Li",
+      "Quanqing Xu",
+      "Chuanhui Yang",
+      "Yixiang Fang"
+    ],
+    "categories": [
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2607.03028",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03028",
+    "pdf_url": "https://arxiv.org/pdf/2607.03028",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2607.02459",
+    "title": "Language Models as Measurement Apparatus for Culture",
+    "url": "https://arxiv.org/abs/2607.02459",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Kent K. Chang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.02459",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02459",
+    "pdf_url": "https://arxiv.org/pdf/2607.02459",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.02376",
+    "title": "Hardware-Enforced Semantic Coordination for Safety-Critical Real-Time Autonomous Systems",
+    "url": "https://arxiv.org/abs/2607.02376",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Uwe M. Borghoff",
+      "Paolo Bottoni",
+      "Remo Pareschi"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-safety",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.02376",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02376",
+    "pdf_url": "https://arxiv.org/pdf/2607.02376",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.02370",
+    "title": "Understanding Agent-Based Patching of Compiler Missed Optimizations",
+    "url": "https://arxiv.org/abs/2607.02370",
+    "published": "2026-07-02",
+    "updated": "2026-07-03",
+    "authors": [
+      "Batu Guan",
+      "Zirui Wang",
+      "Shaohua Li"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "rag"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.02370",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02370",
+    "pdf_url": "https://arxiv.org/pdf/2607.02370",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.02134",
+    "title": "Coding-agents can replicate scientific machine learning papers",
+    "url": "https://arxiv.org/abs/2607.02134",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Atharva Hans",
+      "Ilias Bilionis"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.02134",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02134",
+    "pdf_url": "https://arxiv.org/pdf/2607.02134",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.01597",
+    "title": "A Single Patch Is Not Enough: Deterministic Fusion of Repair Candidates",
+    "url": "https://arxiv.org/abs/2607.01597",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Boyang Yang",
+      "Xiangliang Hu",
+      "Luyao Ren",
+      "Yanjun Chen",
+      "Bach Le",
+      "Tegawendé F. Bissyandé",
+      "Haoye Tian"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.01597",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01597",
+    "pdf_url": "https://arxiv.org/pdf/2607.01597",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.01557",
+    "title": "DiPS: Dialogue Policy Selection for High-Stakes Persuasion Agents",
+    "url": "https://arxiv.org/abs/2607.01557",
+    "published": "2026-07-02",
+    "updated": "2026-07-03",
+    "authors": [
+      "Tianyi Zhang",
+      "Mousumi Das",
+      "Abrar Anwar",
+      "Jesse Thomason",
+      "David Traum"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2607.01557",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01557",
+    "pdf_url": "https://arxiv.org/pdf/2607.01557",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2607.01456",
+    "title": "From Anatomy to Smells: An Empirical Study of SKILL.md in Agent Skills",
+    "url": "https://arxiv.org/abs/2607.01456",
+    "published": "2026-07-01",
+    "updated": "2026-07-03",
+    "authors": [
+      "David Boram Hong",
+      "Aaron Imani",
+      "Iftekhar Ahmed"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.01456",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01456",
+    "pdf_url": "https://arxiv.org/pdf/2607.01456",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.00394",
+    "title": "When Classic Cache Policies Fail: Learning-Augmented Replacement for Semantic Retrieval Buffers",
+    "url": "https://arxiv.org/abs/2607.00394",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Yushi Sun",
+      "Bowen Cao",
+      "Wai Lam"
+    ],
+    "categories": [
+      "cs.DB",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.00394",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00394",
+    "pdf_url": "https://arxiv.org/pdf/2607.00394",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.01507",
+    "title": "The Agentic Garden of Forking Paths",
+    "url": "https://arxiv.org/abs/2607.01507",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Jiacheng Miao",
+      "Jonathan K Pritchard",
+      "James Zou"
+    ],
+    "categories": [
+      "cs.AI",
+      "stat.ME"
+    ],
+    "topics": [
+      "agent-evaluation"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.01507",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01507",
+    "pdf_url": "https://arxiv.org/pdf/2607.01507",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.00751",
+    "title": "SessionBound: Turning Enterprise Task Approval into Budgeted Database Sessions",
+    "url": "https://arxiv.org/abs/2607.00751",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Minmin Wu"
+    ],
+    "categories": [
+      "cs.DB",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "workflow-agent"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.00751",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00751",
+    "pdf_url": "https://arxiv.org/pdf/2607.00751",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.01465",
+    "title": "Beyond Next-Token Prediction: An RLVR Proof of Concept for Tool-Use Agents on Atlassian Workflows",
+    "url": "https://arxiv.org/abs/2607.01465",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Karthikeya Aditya Vissa",
+      "Sankalp Mane",
+      "Ananya Mantravadi",
+      "Harshit Rajgarhia",
+      "Abhishek Mukherji"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2607.01465",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01465",
+    "pdf_url": "https://arxiv.org/pdf/2607.01465",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2607.00871",
+    "title": "Self-Evolving Agents with Anytime-Valid Certificates",
+    "url": "https://arxiv.org/abs/2607.00871",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Biswa Sengupta"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "reasoning"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.00871",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00871",
+    "pdf_url": "https://arxiv.org/pdf/2607.00871",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.01044",
+    "title": "Robots Ask the Way: Communication-Enabled Social Navigation",
+    "url": "https://arxiv.org/abs/2607.01044",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Valentino Sacco",
+      "Luca Scofano",
+      "Indro Spinelli",
+      "Fabio Galasso"
+    ],
+    "categories": [
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.01044",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01044",
+    "pdf_url": "https://arxiv.org/pdf/2607.01044",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.00245",
+    "title": "Agent-to-Agent Finance: Blockchain Payments and Trust Infrastructure for Autonomous AI Agents",
+    "url": "https://arxiv.org/abs/2607.00245",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Hui Gong"
+    ],
+    "categories": [
+      "q-fin.GN"
+    ],
+    "topics": [
+      "rag",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.00245",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00245",
+    "pdf_url": "https://arxiv.org/pdf/2607.00245",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.31272",
+    "title": "The Decomposition Is the Fingerprint: Per-Component Identity for Agent Skills",
+    "url": "https://arxiv.org/abs/2606.31272",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Hongliang Liu",
+      "Yuhao Wu",
+      "Tung-Ling Li"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.31272",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31272",
+    "pdf_url": "https://arxiv.org/pdf/2606.31272",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.31036",
+    "title": "Teaching LLMs to Recommend and Defer in Underrepresented Epilepsy Care",
+    "url": "https://arxiv.org/abs/2606.31036",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Shreyas Rajesh",
+      "Kartik Sharma",
+      "Tonmoy Monsoor",
+      "Mehmet Yigit Turali",
+      "Richard Idro",
+      "Juliana Kayaga",
+      "Robert Sebunya",
+      "Tracy Tushabe Namata",
+      "Jessica Nichole Pasqua",
+      "Vwani Roychowdhury",
+      "Rajarshi Mazumder"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent",
+      "rag"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.31036",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31036",
+    "pdf_url": "https://arxiv.org/pdf/2606.31036",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.30801",
+    "title": "Using AI Agents to Automate Black-Box Audits of Personalization Algorithms at Scale",
+    "url": "https://arxiv.org/abs/2606.30801",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Alessandro Morosini",
+      "Sarah H. Cen",
+      "Andrew Ilyas",
+      "Hedi Driss",
+      "Aleksander Mądry",
+      "Chara Podimata"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.CY",
+      "cs.LG",
+      "cs.SI"
+    ],
+    "topics": [
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.30801",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30801",
+    "pdf_url": "https://arxiv.org/pdf/2606.30801",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.30182",
+    "title": "MirrorCode: AI can rebuild entire programs from behavior alone",
+    "url": "https://arxiv.org/abs/2606.30182",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Tom Adamczewski",
+      "David Owen",
+      "David Rein",
+      "Florian Brand",
+      "Giles Edkins",
+      "Allen Hart",
+      "Daniel O'Connell"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.30182",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30182",
+    "pdf_url": "https://arxiv.org/pdf/2606.30182",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.30911",
+    "title": "Why Solve It Twice? Hierarchical Accumulation of Skills for Transfer-Efficient ML Engineering",
+    "url": "https://arxiv.org/abs/2606.30911",
+    "published": "2026-06-29",
+    "updated": "2026-07-01",
+    "authors": [
+      "Yongbin Kim",
+      "Yashar Talebirad",
+      "Osmar R. Zaiane"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.30911",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30911",
+    "pdf_url": "https://arxiv.org/pdf/2606.30911",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.29279",
+    "title": "Manufactured Confidence: How Memory Consolidation Turns Hearsay into Confident Facts",
+    "url": "https://arxiv.org/abs/2606.29279",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Alex Kwon"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "memory"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.29279",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29279",
+    "pdf_url": "https://arxiv.org/pdf/2606.29279",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.29194",
+    "title": "AI Trading's Alpha Singularity: Emergent Market Reasoning through Agent-to-Agent Self-Evolution",
+    "url": "https://arxiv.org/abs/2606.29194",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Yuqi Li",
+      "Siyuan Liu",
+      "Bingjun Liu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.29194",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29194",
+    "pdf_url": "https://arxiv.org/pdf/2606.29194",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.29280",
+    "title": "Deterministic Decisions for High-Stakes AI. A Zero-Egress Pipeline with the Deployability of RAG and the Accuracy of Machine Learning",
+    "url": "https://arxiv.org/abs/2606.29280",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Craig Atkinson"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.29280",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29280",
+    "pdf_url": "https://arxiv.org/pdf/2606.29280",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.29113",
+    "title": "LLM Semantic Signaling Game and Mechanism Design: Systematic Blindness, Awareness Shaping, and Mindset Dynamics",
+    "url": "https://arxiv.org/abs/2606.29113",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Quanyan Zhu"
+    ],
+    "categories": [
+      "cs.GT",
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.29113",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29113",
+    "pdf_url": "https://arxiv.org/pdf/2606.29113",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.27845",
+    "title": "LLM Agents as Static Level-k Players in Behavioural Games",
+    "url": "https://arxiv.org/abs/2606.27845",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Po Han Teo"
+    ],
+    "categories": [
+      "econ.GN",
+      "econ.TH"
+    ],
+    "topics": [
+      "agent"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.27845",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27845",
+    "pdf_url": "https://arxiv.org/pdf/2606.27845",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.27974",
+    "title": "ProMSA:Progressive Multimodal Search Agents for Knowledge-Based Visual Question Answering",
+    "url": "https://arxiv.org/abs/2606.27974",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "ZhengXian Wu",
+      "Hangrui Xu",
+      "Kai Shi",
+      "Zhuohong Chen",
+      "Yunyao Yu",
+      "Chuanrui Zhang",
+      "Zirui Liao",
+      "Jun Yang",
+      "Zhenyu Yang",
+      "Haonan Lu",
+      "Haoqian Wang"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.27974",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27974",
+    "pdf_url": "https://arxiv.org/pdf/2606.27974",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.26959",
+    "title": "The Shift to Agentic AI: Evidence from Codex",
+    "url": "https://arxiv.org/abs/2606.26959",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Drew Johnston",
+      "David Holtz",
+      "Alex Martin Richmond",
+      "Christopher Ong",
+      "Prasanna Tambe",
+      "Aaron Chatterji"
+    ],
+    "categories": [
+      "econ.GN"
+    ],
+    "topics": [
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.26959",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26959",
+    "pdf_url": "https://arxiv.org/pdf/2606.26959",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.25605",
+    "title": "Constraint Tax in Open-Weight LLMs: An Empirical Study of Tool Calling Suppression Under Structured Output Constraints",
+    "url": "https://arxiv.org/abs/2606.25605",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Fangzheng Li",
+      "Aimin Zhang",
+      "Chen Lv"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.25605",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25605",
+    "pdf_url": "https://arxiv.org/pdf/2606.25605",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.24424",
+    "title": "Explainable AI for Next-Generation Wireless Physical Layer: Basics, State-of-the-Art, and Open Challenges",
+    "url": "https://arxiv.org/abs/2606.24424",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Bingnan Xiao",
+      "Shuyan Hu",
+      "Xiaojing Chen",
+      "Zhiyuan Zhai",
+      "Bingcong Li",
+      "Wei Ni",
+      "Xin Wang",
+      "Ekram Hossain"
+    ],
+    "categories": [
+      "eess.SP"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.24424",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24424",
+    "pdf_url": "https://arxiv.org/pdf/2606.24424",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.24470",
+    "title": "The Latent Bridge: A Continuous Slow-Fast Channel for Real-Time Game Agents",
+    "url": "https://arxiv.org/abs/2606.24470",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Bojie Li",
+      "Noah Shi"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "computer-use",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.24470",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24470",
+    "pdf_url": "https://arxiv.org/pdf/2606.24470",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.22813",
+    "title": "Active Inference as the Test-Time Scaling Law for Physical AI Agents",
+    "url": "https://arxiv.org/abs/2606.22813",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Omar Hashash",
+      "Christo Kurisummoottil Thomas",
+      "Walid Saad",
+      "Merouane Debbah",
+      "Karl Friston",
+      "Adeel Razi"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "reasoning",
+      "world-model"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.22813",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22813",
+    "pdf_url": "https://arxiv.org/pdf/2606.22813",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.23138",
+    "title": "Rising From the Ashes: How Agentic AI is Unblocking Challenges in Cybersecurity",
+    "url": "https://arxiv.org/abs/2606.23138",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Gabriela F. Ciocarlie",
+      "Kathrin Grosse",
+      "Somesh Jha",
+      "Daryna Oliynyk",
+      "Andrew Paverd",
+      "Christian Wressnegger"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-safety",
+      "reasoning"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.23138",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23138",
+    "pdf_url": "https://arxiv.org/pdf/2606.23138",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.23175",
+    "title": "Position: Correct Answer, Wrong Mechanism -- When AI Scientists Defend General Claims Their Own Data Contradicts",
+    "url": "https://arxiv.org/abs/2606.23175",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Steven Young Eulig"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.23175",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23175",
+    "pdf_url": "https://arxiv.org/pdf/2606.23175",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.22906",
+    "title": "From Fragments to Paths: Task-Level Context Recovery for Large Industrial Codebases",
+    "url": "https://arxiv.org/abs/2606.22906",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Jiawei He",
+      "Weisong Sun",
+      "Mengyu Shi",
+      "Jie Jia",
+      "Tong Bian",
+      "Xikai Yang",
+      "Dong Sun"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "rag"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.22906",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22906",
+    "pdf_url": "https://arxiv.org/pdf/2606.22906",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.23189",
+    "title": "Capable but Careless: Do Computer-Use Agents Follow Contextual Integrity?",
+    "url": "https://arxiv.org/abs/2606.23189",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Anmol Goel",
+      "Iryna Gurevych"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "rag"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.23189",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23189",
+    "pdf_url": "https://arxiv.org/pdf/2606.23189",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.21959",
+    "title": "OpenBioRQ: Unsolved Biomedical Research Questions for Agents",
+    "url": "https://arxiv.org/abs/2606.21959",
+    "published": "2026-06-20",
+    "updated": "2026-06-20",
+    "authors": [
+      "Minbyul Jeong"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.21959",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21959",
+    "pdf_url": "https://arxiv.org/pdf/2606.21959",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.21151",
+    "title": "Context-Aware Generative AI for Automated Telecom Test Script Generation",
+    "url": "https://arxiv.org/abs/2606.21151",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Gautam Prasad",
+      "Chandramohan T. N.",
+      "Joy Bose"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.NI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.21151",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21151",
+    "pdf_url": "https://arxiv.org/pdf/2606.21151",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.21037",
+    "title": "Honeyquest for LLMs: Rethinking Cyber Deception for AI Attackers",
+    "url": "https://arxiv.org/abs/2606.21037",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Kerri Prinos",
+      "Lilianne Brush",
+      "Cameron Denton"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.21037",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21037",
+    "pdf_url": "https://arxiv.org/pdf/2606.21037",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.21804",
+    "title": "Is Agent Code Less Maintainable Than Human Code?",
+    "url": "https://arxiv.org/abs/2606.21804",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Shaswat Patel",
+      "Betty Li Hou",
+      "Arun Purohit",
+      "Kai Xu",
+      "Jane Pan",
+      "He He",
+      "Valerie Chen"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.21804",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21804",
+    "pdf_url": "https://arxiv.org/pdf/2606.21804",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.20002",
+    "title": "Connect the Dots: Training LLMs for Long-Lifecycle Agents with Cross-Domain Generalization Via Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2606.20002",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Yanxi Chen",
+      "Weijie Shi",
+      "Yuexiang Xie",
+      "Boyi Hu",
+      "Yaliang Li",
+      "Bolin Ding",
+      "Jingren Zhou"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.20002",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20002",
+    "pdf_url": "https://arxiv.org/pdf/2606.20002",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.20158",
+    "title": "N-Version Programming with Coding Agents",
+    "url": "https://arxiv.org/abs/2606.20158",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Javier Ron",
+      "Benoit Baudry",
+      "Martin Monperrus"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.20158",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20158",
+    "pdf_url": "https://arxiv.org/pdf/2606.20158",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.19830",
+    "title": "JAMER: Project-Level Code Framework Dataset and Benchmark on Professional Game Engines",
+    "url": "https://arxiv.org/abs/2606.19830",
+    "published": "2026-06-18",
+    "updated": "2026-06-21",
+    "authors": [
+      "Jianwen Sun",
+      "Chuanhao Li",
+      "Zizhen Li",
+      "Yukang Feng",
+      "Fanrui Zhang",
+      "Yifei Huang",
+      "Yu Dai",
+      "Kaipeng Zhang"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.19830",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19830",
+    "pdf_url": "https://arxiv.org/pdf/2606.19830",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.20363",
+    "title": "Automating SKILL.md Generation for Computer-Using Agents via Interaction Trajectory Mining",
+    "url": "https://arxiv.org/abs/2606.20363",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Yuexing Hao",
+      "Xiaomin Li"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.20363",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20363",
+    "pdf_url": "https://arxiv.org/pdf/2606.20363",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.20537",
+    "title": "Execution-State Capsules: Graph-Bound Execution-State Checkpoint and Restore for Low-Latency, Small-Batch, On-Device Physical-AI Serving",
+    "url": "https://arxiv.org/abs/2606.20537",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Liang Su"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.DC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "planning",
+      "rag"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.20537",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20537",
+    "pdf_url": "https://arxiv.org/pdf/2606.20537",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.19116",
+    "title": "Towards an Agent-First Web: Redesigning the Web for AI Agents",
+    "url": "https://arxiv.org/abs/2606.19116",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Eranga Bandara",
+      "Ross Gore",
+      "Ravi Mukkamala",
+      "Asanga Gunaratna",
+      "Safdar H. Bouk",
+      "Xueping Liang",
+      "Peter Foytik",
+      "Abdul Rahman",
+      "Sachini Rajapakse",
+      "Isurunima Kularathna",
+      "Pramoda Karunarathna",
+      "Chalani Rajapakse",
+      "Ng Wee Keong",
+      "Kasun De Zoysa",
+      "Tharaka Hewa",
+      "Amin Hass",
+      "Wathsala Herath",
+      "Aruna Withanage",
+      "Nilaan Loganathan",
+      "Atmaram Yarlagadda",
+      "Sachin Shetty"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CY"
+    ],
+    "topics": [
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.19116",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19116",
+    "pdf_url": "https://arxiv.org/pdf/2606.19116",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.18716",
+    "title": "Human-AI Agent Interaction in a Business Context",
+    "url": "https://arxiv.org/abs/2606.18716",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Kathrin Paimann",
+      "Elizangela Valarini",
+      "Sebastian Juhl"
+    ],
+    "categories": [
+      "cs.HC",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.18716",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18716",
+    "pdf_url": "https://arxiv.org/pdf/2606.18716",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.18831",
+    "title": "Beyond Reward Engineering: A Data Recipe for Long-Context Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2606.18831",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Xiaoyue Xu",
+      "Sikui Zhang",
+      "Xiaorong Wang",
+      "Xu Han",
+      "Chaojun Xiao"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.18831",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18831",
+    "pdf_url": "https://arxiv.org/pdf/2606.18831",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.17645",
+    "title": "Beyond Domains: Reusing Web Skills via Transferable Interaction Patterns",
+    "url": "https://arxiv.org/abs/2606.17645",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Shiqi He",
+      "Yue Cui",
+      "Feijie Wu",
+      "Xinyu Ma",
+      "Jiaheng Lu",
+      "Yaliang Li",
+      "Bolin Ding",
+      "Mosharaf Chowdhury"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.17645",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17645",
+    "pdf_url": "https://arxiv.org/pdf/2606.17645",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.28370",
+    "title": "Conversational Query Engine for Mixed-Modality Heterogeneous Enterprise Data Sources",
+    "url": "https://arxiv.org/abs/2606.28370",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Darshita Rathore",
+      "Vineet Kumar",
+      "Vaibhav Singal",
+      "Ankur Vivek Singh",
+      "Anindya Moitra"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.28370",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28370",
+    "pdf_url": "https://arxiv.org/pdf/2606.28370",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.12908",
+    "title": "SENTINEL: Failure-Driven Reinforcement Learning for Training Tool-Using Language Model Agents",
+    "url": "https://arxiv.org/abs/2606.12908",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Ziyi Wang",
+      "Yuxuan Lu",
+      "Yimeng Zhang",
+      "Qun Liu",
+      "Chen Luo",
+      "Jiri Gesi",
+      "Hanqing Lu",
+      "Yisi Sang",
+      "Manling Li",
+      "Jing Huang",
+      "Dakuo Wang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.12908",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12908",
+    "pdf_url": "https://arxiv.org/pdf/2606.12908",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.13949",
+    "title": "Minim: Privacy-Aware Minimal View for Agents via Trusted Local Sanitization",
+    "url": "https://arxiv.org/abs/2606.13949",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Hexuan Yu",
+      "Chaoyu Zhang",
+      "Heng Jin",
+      "Shanghao Shi",
+      "Ning Zhang",
+      "Y. Thomas Hou",
+      "Wenjing Lou"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.13949",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13949",
+    "pdf_url": "https://arxiv.org/pdf/2606.13949",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.10299",
+    "title": "What Spatial Memory Must Store: Occlusion as the Test for Language-Agent Memory",
+    "url": "https://arxiv.org/abs/2606.10299",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Doeon Kwon",
+      "Junho Bang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CV",
+      "cs.MA"
+    ],
+    "topics": [
+      "memory"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-memory",
+      "language-agent"
+    ],
+    "arxiv_id": "2606.10299",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10299",
+    "pdf_url": "https://arxiv.org/pdf/2606.10299",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.11350",
+    "title": "When More Documents Hurt RAG: Mitigating Vector Search Dilution with Domain-Scoped, Model-Agnostic Retrieval",
+    "url": "https://arxiv.org/abs/2606.11350",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Nabaraj Subedi",
+      "Ahmed Abdelaty",
+      "Shivanand Venkanna Sheshappanavar"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.11350",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.11350",
+    "pdf_url": "https://arxiv.org/pdf/2606.11350",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.09935",
+    "title": "GitInject: Real-World Prompt Injection Attacks in AI-Powered CI/CD Pipelines",
+    "url": "https://arxiv.org/abs/2606.09935",
+    "published": "2026-06-07",
+    "updated": "2026-06-07",
+    "authors": [
+      "Jafar Isbarov",
+      "Umid Suleymanov",
+      "Ilia Shumailov",
+      "Murat Kantarcioglu"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2606.09935",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09935",
+    "pdf_url": "https://arxiv.org/pdf/2606.09935",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.07017",
+    "title": "The Sim-to-Real Gap of Foundation Model Agents: A Unified MDP Perspective",
+    "url": "https://arxiv.org/abs/2606.07017",
+    "published": "2026-06-05",
+    "updated": "2026-06-05",
+    "authors": [
+      "Xiaoou Liu",
+      "Tiejin Chen",
+      "Weibo Li",
+      "Xiyang Hu",
+      "Hua Wei"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.ET"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.07017",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.07017",
+    "pdf_url": "https://arxiv.org/pdf/2606.07017",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.03889",
+    "title": "RealClawBench: Live OpenClaw Benchmarks from Real Developer-Agent Sessions",
+    "url": "https://arxiv.org/abs/2606.03889",
+    "published": "2026-06-02",
+    "updated": "2026-06-05",
+    "authors": [
+      "Zongwei Lv",
+      "Zhewen Tan",
+      "Yaoming Li",
+      "Yilun Yao",
+      "Yuxuan Tian",
+      "Lin Sun",
+      "Xiangzheng Zhang",
+      "Weihong Lin",
+      "Tong Yang",
+      "Guangxiang Zhao"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.03889",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.03889",
+    "pdf_url": "https://arxiv.org/pdf/2606.03889",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.01751",
+    "title": "SparseX: Efficient Segment-Level KV Cache Sharing for Interleaved LLM Serving",
+    "url": "https://arxiv.org/abs/2606.01751",
+    "published": "2026-06-01",
+    "updated": "2026-06-07",
+    "authors": [
+      "Quqing Zhang",
+      "Kai Chen",
+      "Ning Liao",
+      "Zehao Lin",
+      "Bo Tang",
+      "Feiyu Xiong",
+      "Zhiyu Li",
+      "Xiaoxing Wang"
+    ],
+    "categories": [
+      "cs.PF"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.01751",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.01751",
+    "pdf_url": "https://arxiv.org/pdf/2606.01751",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.02643",
+    "title": "Inference Cost Attacks for Retrieval-Augmented Large Language Models",
+    "url": "https://arxiv.org/abs/2606.02643",
+    "published": "2026-05-31",
+    "updated": "2026-05-31",
+    "authors": [
+      "Chengliang Liu",
+      "Liangbo Ning",
+      "Yujuan Ding",
+      "Wenqi Fan"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.DB"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.02643",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.02643",
+    "pdf_url": "https://arxiv.org/pdf/2606.02643",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.00750",
+    "title": "I-WebGenBench : Evaluating Interactivity in LLM-Generated Scientific Web Applications",
+    "url": "https://arxiv.org/abs/2606.00750",
+    "published": "2026-05-30",
+    "updated": "2026-05-30",
+    "authors": [
+      "Dasen Dai",
+      "Biao Wu",
+      "Meng Fang",
+      "Shuoqi Li",
+      "Wenhao Wang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.00750",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.00750",
+    "pdf_url": "https://arxiv.org/pdf/2606.00750",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.29559",
+    "title": "LiteCoder-Terminal: Scaling Long-Horizon Terminal Environments for Learning Language Agents",
+    "url": "https://arxiv.org/abs/2605.29559",
+    "published": "2026-05-28",
+    "updated": "2026-05-28",
+    "authors": [
+      "Xiaoxuan Peng",
+      "Kaiqi Zhang",
+      "Xinyu Lu",
+      "Boxi Cao",
+      "Yaojie Lu",
+      "Hongyu Lin",
+      "Xianpei Han",
+      "Le Sun"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "planning",
+      "workflow-agent"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.29559",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.29559",
+    "pdf_url": "https://arxiv.org/pdf/2605.29559",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.29491",
+    "title": "The Curse of Helpfulness: Inverse Scaling Law in Robustness to Distractor Instructions via DistractionIF",
+    "url": "https://arxiv.org/abs/2605.29491",
+    "published": "2026-05-28",
+    "updated": "2026-05-28",
+    "authors": [
+      "Zeli Su",
+      "Zhankai Xu",
+      "Tianlei Chen",
+      "Longfei Zheng",
+      "Xiaolu Zhang",
+      "Jun Zhou",
+      "Wentao Zhang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.29491",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.29491",
+    "pdf_url": "https://arxiv.org/pdf/2605.29491",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.26352",
+    "title": "RICE-PO: Turning Retrieval Interactions into Credit Signals for Reasoning Agents",
+    "url": "https://arxiv.org/abs/2605.26352",
+    "published": "2026-05-25",
+    "updated": "2026-05-25",
+    "authors": [
+      "Mingchen Li",
+      "Hansi Zeng",
+      "Zhuo Qian",
+      "Jiatan Huang",
+      "Hamed Zamani",
+      "Hong Yu"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.26352",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.26352",
+    "pdf_url": "https://arxiv.org/pdf/2605.26352",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.25988",
+    "title": "What Makes a Medical Checker Trainable? Diagnosing Signal Collapse and Reward Hacking in Checker-Guided RAG for Biomedical QA",
+    "url": "https://arxiv.org/abs/2605.25988",
+    "published": "2026-05-25",
+    "updated": "2026-05-25",
+    "authors": [
+      "Yuelyu Ji",
+      "Min Gu Kwak",
+      "Hang Zhang",
+      "Xizhi Wu",
+      "Chenyu Li",
+      "Yanshan Wan"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "reasoning"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.25988",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.25988",
+    "pdf_url": "https://arxiv.org/pdf/2605.25988",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.23281",
+    "title": "DepthAgent: Towards Better Universal Depth Estimation via Sample-wise Expert Selection",
+    "url": "https://arxiv.org/abs/2605.23281",
+    "published": "2026-05-22",
+    "updated": "2026-05-22",
+    "authors": [
+      "Jie Zhu",
+      "Girish Chandar Ganesan",
+      "Xiaoming Liu"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.23281",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.23281",
+    "pdf_url": "https://arxiv.org/pdf/2605.23281",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.20312",
+    "title": "Pramana: A Protocol-Layer Treatment of Claim Verification in Autonomous Agent Networks",
+    "url": "https://arxiv.org/abs/2605.20312",
+    "published": "2026-05-19",
+    "updated": "2026-05-19",
+    "authors": [
+      "Ravi Kiran Kadaboina"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.LO",
+      "cs.MA"
+    ],
+    "topics": [
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.20312",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.20312",
+    "pdf_url": "https://arxiv.org/pdf/2605.20312",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.13438",
+    "title": "CogniFold: Always-On Proactive Memory via Cognitive Folding",
+    "url": "https://arxiv.org/abs/2605.13438",
+    "published": "2026-05-13",
+    "updated": "2026-06-17",
+    "authors": [
+      "Suli Wang",
+      "Yiqun Duan",
+      "Yu Deng",
+      "Rundong Zhao",
+      "Dai Shi",
+      "Minghua Deng",
+      "Chen Chen",
+      "Xinliang Zhou"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.13438",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.13438",
+    "pdf_url": "https://arxiv.org/pdf/2605.13438",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.11359",
+    "title": "CVEvolve: Autonomous Algorithm Discovery for Unstructured Scientific Data Processing",
+    "url": "https://arxiv.org/abs/2605.11359",
+    "published": "2026-05-12",
+    "updated": "2026-06-01",
+    "authors": [
+      "Ming Du",
+      "Xiangyu Yin",
+      "Yanqi Luo",
+      "Dishant Beniwal",
+      "Songyuan Tang",
+      "Hemant Sharma",
+      "Mathew J. Cherukara"
+    ],
+    "categories": [
+      "cs.AI",
+      "physics.data-an"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.11359",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.11359",
+    "pdf_url": "https://arxiv.org/pdf/2605.11359",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.08721",
+    "title": "Breaking the Impasse: Dual-Scale Evolutionary Policy Training for Social Language Agents",
+    "url": "https://arxiv.org/abs/2605.08721",
+    "published": "2026-05-09",
+    "updated": "2026-05-09",
+    "authors": [
+      "Minzheng Wang",
+      "Run Luo",
+      "Yanbo Wang",
+      "Zichen Liu",
+      "Yuqiao Tan",
+      "Tao Tan",
+      "Xu Nan",
+      "Yinhe Zheng",
+      "Wenji Mao"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.08721",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.08721",
+    "pdf_url": "https://arxiv.org/pdf/2605.08721",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.03159",
+    "title": "Learning Correct Behavior from Examples: Validating Sequential Execution in Autonomous Agents",
+    "url": "https://arxiv.org/abs/2605.03159",
+    "published": "2026-05-04",
+    "updated": "2026-05-04",
+    "authors": [
+      "Reshabh K Sharma",
+      "Gaurav Mittal",
+      "Yu Hu"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "rag"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.03159",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.03159",
+    "pdf_url": "https://arxiv.org/pdf/2605.03159",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2604.11839",
+    "title": "Beyond Static Sandboxing: Learned Capability Governance for Autonomous AI Agents",
+    "url": "https://arxiv.org/abs/2604.11839",
+    "published": "2026-04-12",
+    "updated": "2026-05-03",
+    "authors": [
+      "Bronislav Sidik",
+      "Lior Rokach"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.11839",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.11839",
+    "pdf_url": "https://arxiv.org/pdf/2604.11839",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.27742",
+    "title": "TIR-Agent: Training an Explorative and Efficient Agent for Image Restoration",
+    "url": "https://arxiv.org/abs/2603.27742",
+    "published": "2026-03-29",
+    "updated": "2026-03-29",
+    "authors": [
+      "Yisheng Zhang",
+      "Guoli Jia",
+      "Haote Hu",
+      "Shanxu Zhao",
+      "Kaikai Zhao",
+      "Long Sun",
+      "Xinwei Long",
+      "Kai Tian",
+      "Che Jiang",
+      "Zhaoxiang Liu",
+      "Kai Wang",
+      "Shiguo Lian",
+      "Kaiyan Zhang",
+      "Bowen Zhou"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.27742",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.27742",
+    "pdf_url": "https://arxiv.org/pdf/2603.27742",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.05413",
+    "title": "Building Enterprise Realtime Voice Agents from Scratch: A Technical Tutorial",
+    "url": "https://arxiv.org/abs/2603.05413",
+    "published": "2026-03-05",
+    "updated": "2026-03-17",
+    "authors": [
+      "Jielin Qiu",
+      "Zixiang Chen",
+      "Liangwei Yang",
+      "Ming Zhu",
+      "Zhiwei Liu",
+      "Juntao Tan",
+      "Wenting Zhao",
+      "Rithesh Murthy",
+      "Roshan Ram",
+      "Akshara Prabhakar",
+      "Shelby Heinecke",
+      "Caiming Xiong",
+      "Silvio Savarese",
+      "Huan Wang"
+    ],
+    "categories": [
+      "cs.SD"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2603.05413",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.05413",
+    "pdf_url": "https://arxiv.org/pdf/2603.05413",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2603.00991",
+    "title": "Tracking Capabilities for Safer Agents",
+    "url": "https://arxiv.org/abs/2603.00991",
+    "published": "2026-03-01",
+    "updated": "2026-05-07",
+    "authors": [
+      "Martin Odersky",
+      "Yaoyu Zhao",
+      "Yichen Xu",
+      "Oliver Bračevac",
+      "Cao Nguyen Pham"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.PL"
+    ],
+    "topics": [
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.00991",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.00991",
+    "pdf_url": "https://arxiv.org/pdf/2603.00991",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2602.15197",
+    "title": "OpaqueToolsBench: Learning Nuances of Tool Behavior Through Interaction",
+    "url": "https://arxiv.org/abs/2602.15197",
+    "published": "2026-02-16",
+    "updated": "2026-02-16",
+    "authors": [
+      "Skyler Hallinan",
+      "Thejas Venkatesh",
+      "Xiang Ren",
+      "Sai Praneeth Karimireddy",
+      "Ashwin Paranjape",
+      "Yuhao Zhang",
+      "Jack Hessel"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2602.15197",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.15197",
+    "pdf_url": "https://arxiv.org/pdf/2602.15197",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2602.03025",
+    "title": "RC-GRPO: Reward-Conditioned Group Relative Policy Optimization for Multi-Turn Tool Calling Agents",
+    "url": "https://arxiv.org/abs/2602.03025",
+    "published": "2026-02-03",
+    "updated": "2026-02-03",
+    "authors": [
+      "Haitian Zhong",
+      "Jixiu Zhai",
+      "Lei Song",
+      "Jiang Bian",
+      "Qiang Liu",
+      "Tieniu Tan"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2602.03025",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.03025",
+    "pdf_url": "https://arxiv.org/pdf/2602.03025",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2602.03022",
+    "title": "STAR: Similarity-guided Teacher-Assisted Refinement for Super-Tiny Function Calling Models",
+    "url": "https://arxiv.org/abs/2602.03022",
+    "published": "2026-02-03",
+    "updated": "2026-02-24",
+    "authors": [
+      "Jiliang Ni",
+      "Jiachen Pu",
+      "Zhongyi Yang",
+      "Jingfeng Luo",
+      "Conggang Hu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2602.03022",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.03022",
+    "pdf_url": "https://arxiv.org/pdf/2602.03022",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2601.17829",
+    "title": "Linguistic and Argument Diversity in Synthetic Data for Function-Calling Agents",
+    "url": "https://arxiv.org/abs/2601.17829",
+    "published": "2026-01-25",
+    "updated": "2026-01-25",
+    "authors": [
+      "Dan Greenstein",
+      "Zohar Karnin",
+      "Chen Amiraz",
+      "Oren Somekh"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2601.17829",
+    "source": "arxiv",
+    "source_id": "arxiv:2601.17829",
+    "pdf_url": "https://arxiv.org/pdf/2601.17829",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2512.17052",
+    "title": "Dynamic Tool Dependency Retrieval for Lightweight Function Calling",
+    "url": "https://arxiv.org/abs/2512.17052",
+    "published": "2025-12-18",
+    "updated": "2026-04-17",
+    "authors": [
+      "Bhrij Patel",
+      "Davide Belli",
+      "Amir Jalalirad",
+      "Maximilian Arnold",
+      "Aleksandr Ermolov",
+      "Bence Major"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2512.17052",
+    "source": "arxiv",
+    "source_id": "arxiv:2512.17052",
+    "pdf_url": "https://arxiv.org/pdf/2512.17052",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2510.10197",
+    "title": "Don't Just Fine-tune the Agent, Tune the Environment",
+    "url": "https://arxiv.org/abs/2510.10197",
+    "published": "2025-10-11",
+    "updated": "2026-01-30",
+    "authors": [
+      "Siyuan Lu",
+      "Zechuan Wang",
+      "Hongxuan Zhang",
+      "Qintong Wu",
+      "Leilei Gan",
+      "Chenyi Zhuang",
+      "Jinjie Gu",
+      "Tao Lin"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2510.10197",
+    "source": "arxiv",
+    "source_id": "arxiv:2510.10197",
+    "pdf_url": "https://arxiv.org/pdf/2510.10197",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2510.06727",
+    "title": "Scaling LLM Multi-turn RL with End-to-end Summarization-based Context Management",
+    "url": "https://arxiv.org/abs/2510.06727",
+    "published": "2025-10-08",
+    "updated": "2025-10-08",
+    "authors": [
+      "Miao Lu",
+      "Weiwei Sun",
+      "Weihua Du",
+      "Zhan Ling",
+      "Xuesong Yao",
+      "Kang Liu",
+      "Jiecao Chen"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "tool-use"
+    ],
+    "score": 10,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2510.06727",
+    "source": "arxiv",
+    "source_id": "arxiv:2510.06727",
+    "pdf_url": "https://arxiv.org/pdf/2510.06727",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2607.06126",
+    "title": "Causal Inference with Video Features as Treatments",
+    "url": "https://arxiv.org/abs/2607.06126",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Kentaro Nakamura",
+      "Adam Breuer",
+      "Michael H. Crespin",
+      "Bryce J. Dietrich",
+      "Kosuke Imai"
+    ],
+    "categories": [
+      "stat.AP"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.06126",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06126",
+    "pdf_url": "https://arxiv.org/pdf/2607.06126",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.06214",
+    "title": "A toy framework for single and multi-agent human-AI curiosity ecosystems",
+    "url": "https://arxiv.org/abs/2607.06214",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Ilya E. Monosov"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "multi-agent"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.06214",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06214",
+    "pdf_url": "https://arxiv.org/pdf/2607.06214",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.04763",
+    "title": "Multi-Turn On-Policy Distillation with Prefix Replay",
+    "url": "https://arxiv.org/abs/2607.04763",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Baohao Liao",
+      "Hanze Dong",
+      "Christof Monz",
+      "Xinxing Xu",
+      "Li Dong",
+      "Furu Wei"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL",
+      "stat.ML"
+    ],
+    "topics": [
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.04763",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04763",
+    "pdf_url": "https://arxiv.org/pdf/2607.04763",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.04631",
+    "title": "Formal Disco: Scalable Open-Ended Generation of Formally Verified Programs",
+    "url": "https://arxiv.org/abs/2607.04631",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Gabriel Poesia",
+      "Simon Henniger",
+      "Tzu-Han Hsu",
+      "Yilun Du",
+      "Nada Amin"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.04631",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04631",
+    "pdf_url": "https://arxiv.org/pdf/2607.04631",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.04708",
+    "title": "Strategic Buying Agents",
+    "url": "https://arxiv.org/abs/2607.04708",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Mingyang Fu",
+      "Ming Hu"
+    ],
+    "categories": [
+      "econ.TH",
+      "cs.AI",
+      "cs.CY",
+      "cs.GT",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.04708",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04708",
+    "pdf_url": "https://arxiv.org/pdf/2607.04708",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.05471",
+    "title": "KAT-Coder-V2.5 Technical Report",
+    "url": "https://arxiv.org/abs/2607.05471",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Bo Huang",
+      "Fengxiang Li",
+      "Hao Xu",
+      "Haoyang Huang",
+      "Hongyi Fu",
+      "Jinhua Hao",
+      "Kun Yuan",
+      "Minglei Zhang",
+      "Pengcheng Xu",
+      "Shiyang Liu",
+      "Wenhao Zhuang",
+      "Yuze Shi",
+      "Zongxian Feng",
+      "Chao Wang",
+      "Cheng He",
+      "Chongling Rao",
+      "Deyu Cao",
+      "Fan Yang",
+      "Gang Xiong",
+      "Haochen Liu",
+      "Jiabao Li",
+      "Jian Liang",
+      "Jinghui Jia",
+      "Jingwen Chang",
+      "Jun Du",
+      "Junyu Shi",
+      "Min Li",
+      "Mingqi Wu",
+      "Qiang Gao",
+      "Shangpeng Yan",
+      "Shaotong Qi",
+      "Shu Xu",
+      "Shuo Zhou",
+      "Tiankuo Xu",
+      "Tong Zheng",
+      "Weilun Zhao",
+      "Xiancheng Meng",
+      "Xianda Sun",
+      "Xiaoyu Jiang",
+      "Xunhao Jia",
+      "Yao Xia",
+      "Yimeng Xu",
+      "Yinghan Cui",
+      "Yingpeng Chen",
+      "Yiwen Ning",
+      "Yong Wang",
+      "Yuxuan Sun",
+      "Zhongsheng Liu",
+      "Ming Sun",
+      "Cheng Luo",
+      "Chen Yang",
+      "Han Li",
+      "Kun Gai"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.05471",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05471",
+    "pdf_url": "https://arxiv.org/pdf/2607.05471",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.05465",
+    "title": "CanvasAgent: Enabling Complex Image Creation and Editing via Visual Tool Orchestration",
+    "url": "https://arxiv.org/abs/2607.05465",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Hairui Zhu",
+      "Yiying Yang",
+      "Tengjin Weng",
+      "Ziyu Lu",
+      "Xiao Yao",
+      "Xiaoyang Ye",
+      "Lin Ma",
+      "Wenhao Jiang"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2607.05465",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05465",
+    "pdf_url": "https://arxiv.org/pdf/2607.05465",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2607.04508",
+    "title": "Compressing the Validation Bottleneck: An Agentic Self-Driving Lab for Scientific Discovery",
+    "url": "https://arxiv.org/abs/2607.04508",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Kyunghoon Hur",
+      "Chihun Lee"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.RO"
+    ],
+    "topics": [
+      "planning"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.04508",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04508",
+    "pdf_url": "https://arxiv.org/pdf/2607.04508",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.02514",
+    "title": "Distributed Attacks in Persistent-State AI Control",
+    "url": "https://arxiv.org/abs/2607.02514",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Josh Hills",
+      "Ida Caspary",
+      "Asa Cooper Stickland"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "memory",
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.02514",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02514",
+    "pdf_url": "https://arxiv.org/pdf/2607.02514",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.02357",
+    "title": "Cloak and Detonate: Scanner Evasion and Dynamic Detection of Agent Skill Malware",
+    "url": "https://arxiv.org/abs/2607.02357",
+    "published": "2026-07-02",
+    "updated": "2026-07-03",
+    "authors": [
+      "Zimo Ji",
+      "Congying Xu",
+      "Zongjie Li",
+      "Yudong Gao",
+      "Xin Wei",
+      "Shuai Wang",
+      "Shing-Chi Cheung"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.SE"
+    ],
+    "topics": [
+      "coding-agent"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.02357",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02357",
+    "pdf_url": "https://arxiv.org/pdf/2607.02357",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.02057",
+    "title": "Prompt Coverage Adequacy",
+    "url": "https://arxiv.org/abs/2607.02057",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Florian Tambon",
+      "Michael Konstantinou",
+      "Cedric Richter",
+      "Charles Chenouard",
+      "Mark Harman",
+      "Mike Papadakis"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2607.02057",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02057",
+    "pdf_url": "https://arxiv.org/pdf/2607.02057",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2607.01148",
+    "title": "Emergence of Preferential Attachment and Glass-Ceiling Effects in Autonomous Networks of LLMs",
+    "url": "https://arxiv.org/abs/2607.01148",
+    "published": "2026-07-01",
+    "updated": "2026-07-06",
+    "authors": [
+      "Yiming Zhang",
+      "Vikram Krishnamurthy"
+    ],
+    "categories": [
+      "cs.SI",
+      "eess.SY"
+    ],
+    "topics": [
+      "agent-safety",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.01148",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01148",
+    "pdf_url": "https://arxiv.org/pdf/2607.01148",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.01426",
+    "title": "When Should Service Agents Reconsider? Difficulty-Routed Control in Customer-Service Operations",
+    "url": "https://arxiv.org/abs/2607.01426",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Qian Chen",
+      "Chengyuan Liu",
+      "Xin Yu"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2607.01426",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01426",
+    "pdf_url": "https://arxiv.org/pdf/2607.01426",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.31422",
+    "title": "Ask the World Before Acting: Environment Probing for Calibrated Agent World Models",
+    "url": "https://arxiv.org/abs/2606.31422",
+    "published": "2026-06-30",
+    "updated": "2026-07-05",
+    "authors": [
+      "Xinyuan Song",
+      "Zekun Cai"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.31422",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31422",
+    "pdf_url": "https://arxiv.org/pdf/2606.31422",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.31055",
+    "title": "Reference-Based Prosody and Rhythm Evaluation for Spoken Dialogue Systems",
+    "url": "https://arxiv.org/abs/2606.31055",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Ashish Hallur",
+      "Thomas Thebaud",
+      "Georgi Tinchev",
+      "Venkatesh Ravichandran",
+      "Laureano Moro-Velazquez"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.SD",
+      "eess.AS"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.31055",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31055",
+    "pdf_url": "https://arxiv.org/pdf/2606.31055",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.30775",
+    "title": "A Single Rewrite Suffices: Empirical Lessons from Production Skill Description Optimization",
+    "url": "https://arxiv.org/abs/2606.30775",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Yangqiaoyu Zhou",
+      "Mohammad Alqudah",
+      "Kwei-Herng Lai",
+      "Aaron Halfaker",
+      "Yingqi Xiong",
+      "Yaar Harari"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.30775",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30775",
+    "pdf_url": "https://arxiv.org/pdf/2606.30775",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.30441",
+    "title": "Translating Natural Language to Strategic Temporal Specifications via LLMs",
+    "url": "https://arxiv.org/abs/2606.30441",
+    "published": "2026-06-29",
+    "updated": "2026-07-02",
+    "authors": [
+      "Marco Aruta",
+      "Francesco Improta",
+      "Vadim Malvone",
+      "Aniello Murano",
+      "Vladana Perlić"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.30441",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30441",
+    "pdf_url": "https://arxiv.org/pdf/2606.30441",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.28805",
+    "title": "Physics Models for Sim-to-Real Transfer in Professional-Level Robot Table Tennis",
+    "url": "https://arxiv.org/abs/2606.28805",
+    "published": "2026-06-27",
+    "updated": "2026-07-01",
+    "authors": [
+      "Christian Conti",
+      "Bilan Yang",
+      "Alexander Sigrist",
+      "Lorenzo Miele",
+      "Yamen Saraiji",
+      "Peter Dürr",
+      "Naoya Takahashi"
+    ],
+    "categories": [
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "rag",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.28805",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28805",
+    "pdf_url": "https://arxiv.org/pdf/2606.28805",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.27944",
+    "title": "It Lied to a Doctor to Buy Poison Ingredients: Quantifying Real-World Misuse of Phone-use Agents",
+    "url": "https://arxiv.org/abs/2606.27944",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Yiming Sun",
+      "Chen Chen",
+      "Zifan Zhou",
+      "Mi Zhang"
+    ],
+    "categories": [
+      "cs.MM",
+      "cs.AI",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "rag"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.27944",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27944",
+    "pdf_url": "https://arxiv.org/pdf/2606.27944",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.28277",
+    "title": "Towards Automating Scientific Review with Google's Paper Assistant Tool",
+    "url": "https://arxiv.org/abs/2606.28277",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Rajesh Jayaram",
+      "Drew Tyler",
+      "David Woodruff",
+      "Corinna Cortes",
+      "Yossi Matias",
+      "Vahab Mirrokni",
+      "Vincent Cohen-Addad"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL",
+      "cs.CY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.28277",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28277",
+    "pdf_url": "https://arxiv.org/pdf/2606.28277",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.28125",
+    "title": "How Humans, Bots, and Agents Communicate About Vulnerabilities in Pull Requests",
+    "url": "https://arxiv.org/abs/2606.28125",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Pien Rooijendijk",
+      "Christoph Treude",
+      "Mairieli Wessel"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "planning"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.28125",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28125",
+    "pdf_url": "https://arxiv.org/pdf/2606.28125",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.26978",
+    "title": "To Run or Not to Run: Analyzing the Cost-Effectiveness of Code Execution in LLM-Based Program Repair",
+    "url": "https://arxiv.org/abs/2606.26978",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Zhihao Lin",
+      "Junhua Zhu",
+      "Mingyi Zhou",
+      "Xin Wang",
+      "Zhensu Sun",
+      "Renyu Yang",
+      "David Lo",
+      "Li Li"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "rag"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.26978",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26978",
+    "pdf_url": "https://arxiv.org/pdf/2606.26978",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.26028",
+    "title": "Can Trustless Agents Be Trusted? An Empirical Study of the ERC-8004 Decentralized AI Agent Ecosystem",
+    "url": "https://arxiv.org/abs/2606.26028",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Xihan Xiong",
+      "Zelin Li",
+      "Wei Wei",
+      "Qin Wang",
+      "William Knottenbelt",
+      "Zhipeng Wang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.26028",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26028",
+    "pdf_url": "https://arxiv.org/pdf/2606.26028",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.25342",
+    "title": "Lifelong In-Context Learning with Transformers Requires Parametric Forms of Attention",
+    "url": "https://arxiv.org/abs/2606.25342",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Luke McDermott",
+      "Robert W. Heath",
+      "Rahul Parhi"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.25342",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25342",
+    "pdf_url": "https://arxiv.org/pdf/2606.25342",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.26175",
+    "title": "RMTL: Reinforced Micro-task Learning for Long-Horizon Manipulation with VLM Rewards",
+    "url": "https://arxiv.org/abs/2606.26175",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Anıl Can Ateş",
+      "Orhan Kahraman",
+      "Cihan Topal"
+    ],
+    "categories": [
+      "cs.RO"
+    ],
+    "topics": [
+      "computer-use",
+      "embodied-agent",
+      "planning",
+      "rag"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.26175",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26175",
+    "pdf_url": "https://arxiv.org/pdf/2606.26175",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.22798",
+    "title": "Does the Same Token Mean the Same State? MoE Routing as Signal for Reasoning Control",
+    "url": "https://arxiv.org/abs/2606.22798",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Kang Chen",
+      "Minshen Yu",
+      "Junjie Nian",
+      "Yaoning Wang",
+      "Yixin Cao",
+      "Yugang Jiang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "coding-agent",
+      "reasoning"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.22798",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22798",
+    "pdf_url": "https://arxiv.org/pdf/2606.22798",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.22425",
+    "title": "SVGym (SciVerseGym): An Environment for Reinforcement Learning and Bayesian Optimization in Crystal Discovery",
+    "url": "https://arxiv.org/abs/2606.22425",
+    "published": "2026-06-21",
+    "updated": "2026-06-21",
+    "authors": [
+      "Bin Cao"
+    ],
+    "categories": [
+      "cs.AI",
+      "cond-mat.mtrl-sci"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai",
+      "language-agent"
+    ],
+    "arxiv_id": "2606.22425",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22425",
+    "pdf_url": "https://arxiv.org/pdf/2606.22425",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.21811",
+    "title": "Steer, Don't Solve: Training Small Critic Models for Large Code Agents",
+    "url": "https://arxiv.org/abs/2606.21811",
+    "published": "2026-06-20",
+    "updated": "2026-06-20",
+    "authors": [
+      "Shubham Gandhi",
+      "Yiqing Xie",
+      "Atharva Naik",
+      "Ruichen Zhu",
+      "Carolyn Rose"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "coding-agent",
+      "computer-use",
+      "reasoning"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.21811",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21811",
+    "pdf_url": "https://arxiv.org/pdf/2606.21811",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.21315",
+    "title": "Social World Model for Lifelong Social Intelligence",
+    "url": "https://arxiv.org/abs/2606.21315",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Yu Luo"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.21315",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21315",
+    "pdf_url": "https://arxiv.org/pdf/2606.21315",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.21453",
+    "title": "CORTIS: Text-Only Adaptation of Spoken Language Models for Task-Oriented Voice Agents",
+    "url": "https://arxiv.org/abs/2606.21453",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Youngwon Choi",
+      "Hyeonyu Kim",
+      "Taeyoun Kwon",
+      "Donghyuk Jung",
+      "Myeongkyun Cho"
+    ],
+    "categories": [
+      "cs.HC",
+      "cs.AI",
+      "cs.SD",
+      "eess.AS"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2606.21453",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21453",
+    "pdf_url": "https://arxiv.org/pdf/2606.21453",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2606.21654",
+    "title": "ChainWorld: Composing Long-Horizon Desktop Workloads from Atomic OSWorld Tasks",
+    "url": "https://arxiv.org/abs/2606.21654",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Vincent Siu",
+      "Manasi Sharma",
+      "Dawn Song",
+      "Daniel Yue Zhang",
+      "Chenguang Wang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "rag"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.21654",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21654",
+    "pdf_url": "https://arxiv.org/pdf/2606.21654",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.20753",
+    "title": "Empowering Polymeric Materials Discovery by Artificial Intelligence",
+    "url": "https://arxiv.org/abs/2606.20753",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Chenyao Ma",
+      "Linda Zhang",
+      "Yuheng Chen",
+      "Wei Du",
+      "Shangwen Fang",
+      "Zihao Jiang",
+      "Chuanyu Liu",
+      "Xinyu Ma",
+      "Rui Su",
+      "Gang Wang",
+      "Muyao Yu",
+      "Dong Zhong",
+      "Jie Zhu",
+      "Weibo Gong",
+      "Huan Gu",
+      "Limin Li",
+      "Chen Shen",
+      "Rui Wu",
+      "Zhenghao Wu",
+      "Kan Xu",
+      "Min Zhou",
+      "Donglin He",
+      "Xiayun Huang",
+      "Shan Jiang",
+      "Pengfei Ou",
+      "Jiayu Peng",
+      "Yuwei Zhang",
+      "Jie Zhao",
+      "Di Zhang",
+      "Piao Ma",
+      "Zhenghao Li",
+      "Hao Li"
+    ],
+    "categories": [
+      "physics.chem-ph",
+      "cs.AI"
+    ],
+    "topics": [
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.20753",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20753",
+    "pdf_url": "https://arxiv.org/pdf/2606.20753",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.18733",
+    "title": "SWE-Future: Forecast-Conditioned Data Synthesis for Future-Oriented Software Engineering Agents",
+    "url": "https://arxiv.org/abs/2606.18733",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Qiao Zhao",
+      "JianYing Qu",
+      "Jun Zhang",
+      "Yehua Yang",
+      "Hanwen Du",
+      "Zhongkai Sun"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.18733",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18733",
+    "pdf_url": "https://arxiv.org/pdf/2606.18733",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.18746",
+    "title": "What Must Generalist Agents Remember?",
+    "url": "https://arxiv.org/abs/2606.18746",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Khurram Yamin",
+      "Namrata Deka",
+      "Maitreyi Swaroop",
+      "Albert Ting",
+      "Jeff Schneider",
+      "Bryan Wilder"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "memory",
+      "planning",
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.18746",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18746",
+    "pdf_url": "https://arxiv.org/pdf/2606.18746",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.18996",
+    "title": "TRAP: Benchmark for Task-completion and Resistance to Active Privacy-extraction",
+    "url": "https://arxiv.org/abs/2606.18996",
+    "published": "2026-06-17",
+    "updated": "2026-06-18",
+    "authors": [
+      "Moon Ye-Bin",
+      "Nam Hyeon-Woo",
+      "Baek Seong-Eun",
+      "Yejin Yeo",
+      "Tae-Hyun Oh"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.18996",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18996",
+    "pdf_url": "https://arxiv.org/pdf/2606.18996",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.18385",
+    "title": "CaVe-VLM-CoT: An Interpretable Vision-Language Model Framework",
+    "url": "https://arxiv.org/abs/2606.18385",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Sneha Rao",
+      "Shaina Raza",
+      "Dhanesh Ramachandram"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.18385",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18385",
+    "pdf_url": "https://arxiv.org/pdf/2606.18385",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.18005",
+    "title": "LLM Consumer Behavior Theory: Foundations of a Novel Research Field",
+    "url": "https://arxiv.org/abs/2606.18005",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Manon Reusens",
+      "Sofie Goethals",
+      "David Martens"
+    ],
+    "categories": [
+      "cs.AI",
+      "econ.GN"
+    ],
+    "topics": [
+      "agent-safety",
+      "rag",
+      "world-model"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.18005",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18005",
+    "pdf_url": "https://arxiv.org/pdf/2606.18005",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.16215",
+    "title": "PACT: Privileged Trace Co-Training for Multi-Turn Tool-Use Agents",
+    "url": "https://arxiv.org/abs/2606.16215",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Zhenbang Du",
+      "Jun Luo",
+      "Zhiwei Zheng",
+      "Xiangchi Yuan",
+      "Kejing Xia",
+      "Dachuan Shi",
+      "Qirui Jin",
+      "Qijia He",
+      "Shaofeng Zou",
+      "Yingbin Liang",
+      "Wenke Lee"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.16215",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16215",
+    "pdf_url": "https://arxiv.org/pdf/2606.16215",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.15971",
+    "title": "SAG: SQL-Retrieval Augmented Generation with Query-Time Dynamic Hyperedges",
+    "url": "https://arxiv.org/abs/2606.15971",
+    "published": "2026-06-14",
+    "updated": "2026-06-14",
+    "authors": [
+      "Yuchao Wu",
+      "Junqin Li",
+      "XingCheng Liang",
+      "Yongjie Chen",
+      "Yinghao Liang",
+      "Linyuan Mo",
+      "Guanxian Li"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.15971",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15971",
+    "pdf_url": "https://arxiv.org/pdf/2606.15971",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.13239",
+    "title": "ComAct: Reframing Professional Software Manipulation via COM-as-Action Paradigm",
+    "url": "https://arxiv.org/abs/2606.13239",
+    "published": "2026-06-11",
+    "updated": "2026-06-30",
+    "authors": [
+      "Jiaxin Ai",
+      "Tao Hu",
+      "Xuemeng Yang",
+      "Shu Zou",
+      "Hairong Zhang",
+      "Daocheng Fu",
+      "Yu Yang",
+      "Hongbin Zhou",
+      "Nianchen Deng",
+      "Pinlong Cai",
+      "Zhongyuan Wang",
+      "Botian Shi",
+      "Kaipeng Zhang",
+      "Licheng Wen"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.CL",
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.13239",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13239",
+    "pdf_url": "https://arxiv.org/pdf/2606.13239",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.10478",
+    "title": "3D-CoS: A New 3D Reconstruction Paradigm Based on VLM Code Synthesis",
+    "url": "https://arxiv.org/abs/2606.10478",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Yuhao Wang",
+      "Puyi Wang",
+      "Linjie Li",
+      "Zhengyuan Yang",
+      "Kevin Qinghong Lin",
+      "Yu Cheng"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.10478",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10478",
+    "pdf_url": "https://arxiv.org/pdf/2606.10478",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.10875",
+    "title": "Pushing the Limits of LLM Tool Calling via Experiential Knowledge Integration and Activation",
+    "url": "https://arxiv.org/abs/2606.10875",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Yupu Hao",
+      "Zhuoran Jin",
+      "Huanxuan Liao",
+      "Kang Liu",
+      "Jun Zhao"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.10875",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10875",
+    "pdf_url": "https://arxiv.org/pdf/2606.10875",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.07924",
+    "title": "Decoupling Semantics and Logic: A Training-Free Coarse-to-Fine Pipeline for Video Retrieval-Augmented Generation",
+    "url": "https://arxiv.org/abs/2606.07924",
+    "published": "2026-06-06",
+    "updated": "2026-06-06",
+    "authors": [
+      "Jiaxin Dai",
+      "Zehang Wei",
+      "Jiamin Yan",
+      "Xiang Xiang"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI",
+      "cs.CL",
+      "cs.LG",
+      "cs.MM"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "reasoning"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.07924",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.07924",
+    "pdf_url": "https://arxiv.org/pdf/2606.07924",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.06566",
+    "title": "NTILC: Neural Tool Invocation via Learned Compression",
+    "url": "https://arxiv.org/abs/2606.06566",
+    "published": "2026-06-04",
+    "updated": "2026-06-04",
+    "authors": [
+      "Andrew Krikorian",
+      "Yayuan Li",
+      "Jason J. Corso"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2606.06566",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.06566",
+    "pdf_url": "https://arxiv.org/pdf/2606.06566",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2606.03800",
+    "title": "Trading Human Curation for Synthetic Augmentation in RLVR",
+    "url": "https://arxiv.org/abs/2606.03800",
+    "published": "2026-06-02",
+    "updated": "2026-06-02",
+    "authors": [
+      "Akshansh",
+      "Leonardo Rosa Rodrigues",
+      "Michael Korostelev",
+      "Youssef Hassan",
+      "Mark E. Whiting"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2606.03800",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.03800",
+    "pdf_url": "https://arxiv.org/pdf/2606.03800",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2606.01722",
+    "title": "Post-Deterministic Distributed Systems: A New Foundation for Trustworthy Autonomous Infrastructure",
+    "url": "https://arxiv.org/abs/2606.01722",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Jun He",
+      "Deying Yu"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.DC"
+    ],
+    "topics": [
+      "computer-use",
+      "memory",
+      "planning",
+      "reasoning"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.01722",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.01722",
+    "pdf_url": "https://arxiv.org/pdf/2606.01722",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.02245",
+    "title": "When Knowledge Is Not Free: Cost-Aware Evidence Selection in Retrieval-Augmented Generation",
+    "url": "https://arxiv.org/abs/2606.02245",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Mingyan Wu",
+      "Han Yang",
+      "Omer Ben-Porat",
+      "Yftah Ziser"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.02245",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.02245",
+    "pdf_url": "https://arxiv.org/pdf/2606.02245",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.00734",
+    "title": "EMA: Approximate Nearest Neighbor Search with General Attribute Filtering and Dynamic Updates",
+    "url": "https://arxiv.org/abs/2606.00734",
+    "published": "2026-05-30",
+    "updated": "2026-05-30",
+    "authors": [
+      "Mocheng Li",
+      "Baotong Lu",
+      "James Cheng",
+      "Chenhao Ma"
+    ],
+    "categories": [
+      "cs.DB"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "rag"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-memory",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.00734",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.00734",
+    "pdf_url": "https://arxiv.org/pdf/2606.00734",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.31064",
+    "title": "Fighting Numerical Hallucinations via Data-centric Compilation for Online Financial QA",
+    "url": "https://arxiv.org/abs/2605.31064",
+    "published": "2026-05-29",
+    "updated": "2026-05-29",
+    "authors": [
+      "Hao Chen",
+      "Xing Tang",
+      "Qirui Liu",
+      "Weijie Shi",
+      "Shiwei Li",
+      "Fuyuan Lyu",
+      "Weihong Luo",
+      "Xiku Du",
+      "Xiuqiang He"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.31064",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.31064",
+    "pdf_url": "https://arxiv.org/pdf/2605.31064",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.30628",
+    "title": "The Architecture of Errors: From Universal Impossibility to Patch-Local LLM Reliability",
+    "url": "https://arxiv.org/abs/2605.30628",
+    "published": "2026-05-28",
+    "updated": "2026-05-28",
+    "authors": [
+      "Mikhail L. Arbuzov",
+      "Lee Mosbacker",
+      "Sisong Bei",
+      "Ziwei Dong",
+      "Dmitri Kalaev",
+      "Alexey Shvets"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.30628",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.30628",
+    "pdf_url": "https://arxiv.org/pdf/2605.30628",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.28914",
+    "title": "AIRGuard: Guarding Agent Actions with Runtime Authority Control",
+    "url": "https://arxiv.org/abs/2605.28914",
+    "published": "2026-05-27",
+    "updated": "2026-05-27",
+    "authors": [
+      "Suliu Qin",
+      "Haomin Zhuang",
+      "Yujun Zhou",
+      "Yufei Han",
+      "Xiangliang Zhang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.28914",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.28914",
+    "pdf_url": "https://arxiv.org/pdf/2605.28914",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.25162",
+    "title": "STREAM: A Data-Centric Framework for Mining High-Value Task-Oriented Dialogues from Streaming Media",
+    "url": "https://arxiv.org/abs/2605.25162",
+    "published": "2026-05-24",
+    "updated": "2026-05-24",
+    "authors": [
+      "Liang Xue",
+      "Haoyu Liu",
+      "Cheng Wang",
+      "Pengyu Chen",
+      "Haozhuo Zheng",
+      "Yang Liu"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.25162",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.25162",
+    "pdf_url": "https://arxiv.org/pdf/2605.25162",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.21956",
+    "title": "Detecting Offensive Cyber Agents: A Detection-in-Depth Approach",
+    "url": "https://arxiv.org/abs/2605.21956",
+    "published": "2026-05-21",
+    "updated": "2026-05-21",
+    "authors": [
+      "Matt Mittelsteadt",
+      "Jam Kraprayoon",
+      "Robin Staes-Polet",
+      "Oskar Galeev",
+      "Jan Wehner",
+      "Christopher Covino",
+      "Shaun Ee"
+    ],
+    "categories": [
+      "cs.CY"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.21956",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.21956",
+    "pdf_url": "https://arxiv.org/pdf/2605.21956",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.22177",
+    "title": "Maestro: Reinforcement Learning to Orchestrate Hierarchical Model-Skill Ensembles",
+    "url": "https://arxiv.org/abs/2605.22177",
+    "published": "2026-05-21",
+    "updated": "2026-05-21",
+    "authors": [
+      "Jinyang Wu",
+      "Guocheng Zhai",
+      "Ruihan Jin",
+      "Yuhao Shen",
+      "Zhengxi Lu",
+      "Fan Zhang",
+      "Haoran Luo",
+      "Zheng Lian",
+      "Zhengqi Wen",
+      "Jianhua Tao"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.22177",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.22177",
+    "pdf_url": "https://arxiv.org/pdf/2605.22177",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.18988",
+    "title": "Surviving the Unseen: Predictive Defense for Novel Multi-Turn Multimodal Attacks",
+    "url": "https://arxiv.org/abs/2605.18988",
+    "published": "2026-05-18",
+    "updated": "2026-05-18",
+    "authors": [
+      "Doohee You"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "workflow-agent"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.18988",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.18988",
+    "pdf_url": "https://arxiv.org/pdf/2605.18988",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.12978",
+    "title": "Useful Memories Become Faulty When Continuously Updated by LLMs",
+    "url": "https://arxiv.org/abs/2605.12978",
+    "published": "2026-05-13",
+    "updated": "2026-05-13",
+    "authors": [
+      "Dylan Zhang",
+      "Yanshan Lin",
+      "Zhengkun Wu",
+      "Yihang Sun",
+      "Bingxuan Li",
+      "Dianqi Li",
+      "Hao Peng"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "memory",
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.12978",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.12978",
+    "pdf_url": "https://arxiv.org/pdf/2605.12978",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.13918",
+    "title": "CA2: Code-Aware Agent for Automated Game Testing",
+    "url": "https://arxiv.org/abs/2605.13918",
+    "published": "2026-05-13",
+    "updated": "2026-05-13",
+    "authors": [
+      "Valliappan Chidambaram Adaikkappan",
+      "Vincent Martineau",
+      "Joshua Romoff",
+      "David Meger"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2605.13918",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.13918",
+    "pdf_url": "https://arxiv.org/pdf/2605.13918",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2605.14038",
+    "title": "Model-Adaptive Tool Necessity Reveals the Knowing-Doing Gap in LLM Tool Use",
+    "url": "https://arxiv.org/abs/2605.14038",
+    "published": "2026-05-13",
+    "updated": "2026-05-17",
+    "authors": [
+      "Yize Cheng",
+      "Chenrui Fan",
+      "Mahdi JafariRaviz",
+      "Keivan Rezaei",
+      "Soheil Feizi"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.14038",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.14038",
+    "pdf_url": "https://arxiv.org/pdf/2605.14038",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.13414",
+    "title": "TRIAGE: Evaluating Prospective Metacognitive Control in LLMs under Resource Constraints",
+    "url": "https://arxiv.org/abs/2605.13414",
+    "published": "2026-05-13",
+    "updated": "2026-05-13",
+    "authors": [
+      "Zabir Al Nazi",
+      "Shubhashis Roy Dipta"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "reasoning"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.13414",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.13414",
+    "pdf_url": "https://arxiv.org/pdf/2605.13414",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.13898",
+    "title": "Bidirectional Empowerment of Metamorphic Testing and Large Language Models: A Systematic Survey",
+    "url": "https://arxiv.org/abs/2605.13898",
+    "published": "2026-05-12",
+    "updated": "2026-05-12",
+    "authors": [
+      "Zheng Zheng",
+      "Zenghui Zhou",
+      "Yinwang Xu",
+      "Daixu Ren",
+      "Tsong Yueh Chen"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.13898",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.13898",
+    "pdf_url": "https://arxiv.org/pdf/2605.13898",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.06524",
+    "title": "Process Matters more than Output for Distinguishing Humans from Machines",
+    "url": "https://arxiv.org/abs/2605.06524",
+    "published": "2026-05-07",
+    "updated": "2026-05-09",
+    "authors": [
+      "Milena Rmus",
+      "Mathew D. Hardy",
+      "Thomas L. Griffiths",
+      "Mayank Agrawal"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.06524",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.06524",
+    "pdf_url": "https://arxiv.org/pdf/2605.06524",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2603.09036",
+    "title": "SCALAR: Learning and Composing Skills through LLM Guided Symbolic Planning and Deep RL Grounding",
+    "url": "https://arxiv.org/abs/2603.09036",
+    "published": "2026-03-10",
+    "updated": "2026-03-10",
+    "authors": [
+      "Renos Zabounidis",
+      "Yue Wu",
+      "Simon Stepputtis",
+      "Woojun Kim",
+      "Yuanzhi Li",
+      "Tom Mitchell",
+      "Katia Sycara"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "computer-use",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.09036",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.09036",
+    "pdf_url": "https://arxiv.org/pdf/2603.09036",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.16901",
+    "title": "From Language to Action in Arabic: Reliable Structured Tool Calling via Data-Centric Fine-Tuning",
+    "url": "https://arxiv.org/abs/2603.16901",
+    "published": "2026-03-04",
+    "updated": "2026-03-04",
+    "authors": [
+      "Omer Nacar",
+      "Deema Alquffari",
+      "Saleh Alsharideh",
+      "Adeem AlOtaibi",
+      "Abdulaziz Alabdulkarim",
+      "Leen Alhazmi",
+      "Nada Alomar",
+      "Wareef Alzubaidi",
+      "Nada Alsultan",
+      "Ahmed Alrabghi",
+      "Demah Alhoshan",
+      "Rana Alsayyari",
+      "Hamed Alruwaili",
+      "Albaraa Jaafar",
+      "Khaled Alusmani",
+      "Abdulaziz Alsohimy",
+      "Munirah Alsubaie",
+      "Shahd Aldukhayil",
+      "Arwa Alali",
+      "Yazeed BinShihah",
+      "Razan Alsulaymi",
+      "Nourah Alhumaid",
+      "Razan Abdulsalam",
+      "Reem Alamoudi",
+      "Mohammed Alkhalifa"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2603.16901",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.16901",
+    "pdf_url": "https://arxiv.org/pdf/2603.16901",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2602.09372",
+    "title": "AgentSkiller: Scaling Generalist Agent Intelligence through Semantically Integrated Cross-Domain Data Synthesis",
+    "url": "https://arxiv.org/abs/2602.09372",
+    "published": "2026-02-10",
+    "updated": "2026-02-10",
+    "authors": [
+      "Zexu Sun",
+      "Bokai Ji",
+      "Hengyi Cai",
+      "Shuaiqiang Wang",
+      "Lei Wang",
+      "Guangxia Li",
+      "Xu Chen"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "planning",
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2602.09372",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.09372",
+    "pdf_url": "https://arxiv.org/pdf/2602.09372",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2601.09292",
+    "title": "Blue Teaming Function-Calling Agents",
+    "url": "https://arxiv.org/abs/2601.09292",
+    "published": "2026-01-14",
+    "updated": "2026-01-14",
+    "authors": [
+      "Greta Dolcetti",
+      "Giulio Zizzo",
+      "Sergio Maffeis"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2601.09292",
+    "source": "arxiv",
+    "source_id": "arxiv:2601.09292",
+    "pdf_url": "https://arxiv.org/pdf/2601.09292",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2601.05366",
+    "title": "Lost in Execution: On the Multilingual Robustness of Tool Calling in Large Language Models",
+    "url": "https://arxiv.org/abs/2601.05366",
+    "published": "2026-01-08",
+    "updated": "2026-06-28",
+    "authors": [
+      "Zheng Luo",
+      "T Pranav Kutralingam",
+      "Ogochukwu N Okoani",
+      "Wanpeng Xu",
+      "Hua Wei",
+      "Xiyang Hu"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2601.05366",
+    "source": "arxiv",
+    "source_id": "arxiv:2601.05366",
+    "pdf_url": "https://arxiv.org/pdf/2601.05366",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2509.18076",
+    "title": "Improving Large Language Models Function Calling and Interpretability via Guided-Structured Templates",
+    "url": "https://arxiv.org/abs/2509.18076",
+    "published": "2025-09-22",
+    "updated": "2025-09-22",
+    "authors": [
+      "Hy Dang",
+      "Tianyi Liu",
+      "Zhuofeng Wu",
+      "Jingfeng Yang",
+      "Haoming Jiang",
+      "Tao Yang",
+      "Pei Chen",
+      "Zhengyang Wang",
+      "Helen Wang",
+      "Huasheng Li",
+      "Bing Yin",
+      "Meng Jiang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "computer-use",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2509.18076",
+    "source": "arxiv",
+    "source_id": "arxiv:2509.18076",
+    "pdf_url": "https://arxiv.org/pdf/2509.18076",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2508.09125",
+    "title": "Complex Logical Instruction Generation",
+    "url": "https://arxiv.org/abs/2508.09125",
+    "published": "2025-08-12",
+    "updated": "2026-01-27",
+    "authors": [
+      "Mian Zhang",
+      "Shujian Liu",
+      "Sixun Dong",
+      "Ming Yin",
+      "Yebowen Hu",
+      "Xun Wang",
+      "Steven Ma",
+      "Song Wang",
+      "Sathish Reddy Indurthi",
+      "Haoyun Deng",
+      "Zhiyu Zoey Chen",
+      "Kaiqiang Song"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning"
+    ],
+    "score": 9,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2508.09125",
+    "source": "arxiv",
+    "source_id": "arxiv:2508.09125",
+    "pdf_url": "https://arxiv.org/pdf/2508.09125",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2607.05804",
+    "title": "TurnOPD: Making On-Policy Distillation Turn-Aware for Efficient Long-Horizon Agent Training",
+    "url": "https://arxiv.org/abs/2607.05804",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Yuhang Zhou",
+      "Kai Zheng",
+      "Haoling Li",
+      "Dengyun Peng",
+      "Can Xu",
+      "Jingjing Chen"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "planning"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2607.05804",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05804",
+    "pdf_url": "https://arxiv.org/pdf/2607.05804",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2607.04728",
+    "title": "Turning Off-Policy Tokens On-Policy: A Plug-in Approach for Improving LLM Alignment",
+    "url": "https://arxiv.org/abs/2607.04728",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Yu Li",
+      "Xiuyu Li",
+      "Mingyang Yi",
+      "Jiaxing Wang",
+      "zhangliangxu",
+      "Zhaolong Xing",
+      "Zhen Chen"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2607.04728",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04728",
+    "pdf_url": "https://arxiv.org/pdf/2607.04728",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.04542",
+    "title": "Auto: The AGI Compiler",
+    "url": "https://arxiv.org/abs/2607.04542",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Jaber Jaber",
+      "Osama Jaber"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.04542",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04542",
+    "pdf_url": "https://arxiv.org/pdf/2607.04542",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.03193",
+    "title": "Self-Specializing Vision-Language Transmon Chip Calibration in a Physics-Grounded Environment",
+    "url": "https://arxiv.org/abs/2607.03193",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Animesh Tripathy",
+      "Aswanth Krishnan"
+    ],
+    "categories": [
+      "quant-ph",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "planning",
+      "rag",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2607.03193",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03193",
+    "pdf_url": "https://arxiv.org/pdf/2607.03193",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2607.03451",
+    "title": "SkillOpt-Lite: Better and Faster Agent Self-evolution via One Line of Vibe",
+    "url": "https://arxiv.org/abs/2607.03451",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Yifei Shen",
+      "Bo Li",
+      "Xinjie Zhang"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.03451",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03451",
+    "pdf_url": "https://arxiv.org/pdf/2607.03451",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.02931",
+    "title": "VERITAS: Towards a General-Purpose Replication Tool for Scientific Research",
+    "url": "https://arxiv.org/abs/2607.02931",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Haokun Liu",
+      "Filbert Aurelian Tjiaranata",
+      "Chenhao Tan"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.02931",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02931",
+    "pdf_url": "https://arxiv.org/pdf/2607.02931",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.02217",
+    "title": "Affinage: genome-scale mechanistic gene annotation from the published literature",
+    "url": "https://arxiv.org/abs/2607.02217",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Matteo Di Bernardo",
+      "Iain M. Cheeseman"
+    ],
+    "categories": [
+      "q-bio.GN"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "rag",
+      "reasoning"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.02217",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02217",
+    "pdf_url": "https://arxiv.org/pdf/2607.02217",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.01639",
+    "title": "Autonomous discovery of traffic laws with AI traffic scientists",
+    "url": "https://arxiv.org/abs/2607.01639",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Xingyuan Dai",
+      "Yue Liu",
+      "Xiaoyan Gong",
+      "Qinghai Miao",
+      "Junyou Shang",
+      "Yutong Wang",
+      "Chao Guo",
+      "Yonglin Tian",
+      "Yizhang Chai",
+      "Chao Xiang",
+      "Yisheng Lv",
+      "Fei-Yue Wang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "memory",
+      "planning",
+      "workflow-agent"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.01639",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01639",
+    "pdf_url": "https://arxiv.org/pdf/2607.01639",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.00910",
+    "title": "Calibrating the Instrument: Controllability of an LLM-Driven Synthetic Population",
+    "url": "https://arxiv.org/abs/2607.00910",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Mirko Degli Esposti"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "world-model"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.00910",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00910",
+    "pdf_url": "https://arxiv.org/pdf/2607.00910",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.00941",
+    "title": "From Runtime Records to Legal Findings: An Evidentiary-Adequacy Criterion for Agentic AI Oversight",
+    "url": "https://arxiv.org/abs/2607.00941",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Jeroen Janssen"
+    ],
+    "categories": [
+      "cs.CY"
+    ],
+    "topics": [
+      "agent"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.00941",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00941",
+    "pdf_url": "https://arxiv.org/pdf/2607.00941",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.00316",
+    "title": "Evolving Intelligent Complex Systems via Intellicise Networks: Architecture, Technologies, and Pathways",
+    "url": "https://arxiv.org/abs/2607.00316",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Ping Zhang",
+      "Rui Meng",
+      "Xiaodong Xu",
+      "Song Gao",
+      "Zixuan Huang",
+      "Yaheng Wang",
+      "Yinqiu Liu",
+      "Ruichen Zhang",
+      "Yiming Liu",
+      "Kaiwen Yu",
+      "Yaping Sun",
+      "Han Meng",
+      "Haonan Tong",
+      "Huishi Song",
+      "Qianqian Yang",
+      "Shuoyao Wang",
+      "Lexi Xu",
+      "Qinghe Du",
+      "Geng Sun",
+      "Jiawen Kang",
+      "Gang Wu",
+      "Yiqing Zhou",
+      "Haixia Zhang",
+      "Zesong Fei",
+      "Aimin Hao",
+      "Ming Li"
+    ],
+    "categories": [
+      "eess.SP"
+    ],
+    "topics": [
+      "agent-safety",
+      "embodied-agent",
+      "planning",
+      "tool-use"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.00316",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00316",
+    "pdf_url": "https://arxiv.org/pdf/2607.00316",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.01299",
+    "title": "HYPIC: Accelerating Hybrid-Attention LLM Serving with Position-Independent Caching",
+    "url": "https://arxiv.org/abs/2607.01299",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Yifei Liu",
+      "Juntong Wu",
+      "Yang Liu",
+      "Junhao Hu",
+      "Minghao Li",
+      "Xiaoxu Chen",
+      "Weihang Chen"
+    ],
+    "categories": [
+      "cs.DC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2607.01299",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01299",
+    "pdf_url": "https://arxiv.org/pdf/2607.01299",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.31492",
+    "title": "Higher-order hopping-parameter expansion by human-AI collaboration",
+    "url": "https://arxiv.org/abs/2606.31492",
+    "published": "2026-06-30",
+    "updated": "2026-07-06",
+    "authors": [
+      "Masakiyo Kitazawa",
+      "Tatsuya Wada"
+    ],
+    "categories": [
+      "hep-lat",
+      "hep-ph"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.31492",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31492",
+    "pdf_url": "https://arxiv.org/pdf/2606.31492",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.30774",
+    "title": "What Drives Interactive Improvement from Feedback?",
+    "url": "https://arxiv.org/abs/2606.30774",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Bartłomiej Cupiał",
+      "Jan Łojek",
+      "Mikołaj Garstecki",
+      "Szymon Pobłocki",
+      "Alicja Ziarko",
+      "Piotr Miłoś"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.30774",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30774",
+    "pdf_url": "https://arxiv.org/pdf/2606.30774",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.29916",
+    "title": "EVAF: A Test-Retest Protocol for Selective Parametric Consolidation",
+    "url": "https://arxiv.org/abs/2606.29916",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Haoliang Han"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.29916",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29916",
+    "pdf_url": "https://arxiv.org/pdf/2606.29916",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.30963",
+    "title": "Loc2Repair: A Framework for Evaluating the Impact of File-Level Issue Localization in Repo-Level LLM Repair",
+    "url": "https://arxiv.org/abs/2606.30963",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Mohammad Nour Al Awad",
+      "Sergey Ivanov"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.30963",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30963",
+    "pdf_url": "https://arxiv.org/pdf/2606.30963",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.29556",
+    "title": "Persona-Trained Monte Carlo: Estimating Market-Outcome Distributions via Swarms of Persona-Conditioned Neural Policy Bots in a Limit Order Book",
+    "url": "https://arxiv.org/abs/2606.29556",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Salavat Ishbulatov"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.29556",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29556",
+    "pdf_url": "https://arxiv.org/pdf/2606.29556",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.28690",
+    "title": "Formal Security Analysis of Agent Protocol Composition",
+    "url": "https://arxiv.org/abs/2606.28690",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Shenghan Zheng",
+      "Qifan Zhang",
+      "Zheng Zhang",
+      "Haonan Li",
+      "Christophe Hauser"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.28690",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28690",
+    "pdf_url": "https://arxiv.org/pdf/2606.28690",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.30678",
+    "title": "NanoVer: An open-source framework for interactive molecular dynamics in extended reality (iMD-XR) on commodity hardware",
+    "url": "https://arxiv.org/abs/2606.30678",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Mark D. Wonnacott",
+      "Luis Ernesto Toledo Castro",
+      "Harry J. Stroud",
+      "Ludovica Aisa",
+      "Mohamed Dhouioui",
+      "Rhoslyn Roebuck Williams",
+      "Denis Protopopov",
+      "Sila Sobrado",
+      "David R. Glowacki"
+    ],
+    "categories": [
+      "physics.chem-ph",
+      "physics.bio-ph",
+      "physics.ed-ph"
+    ],
+    "topics": [
+      "computer-use",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.30678",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30678",
+    "pdf_url": "https://arxiv.org/pdf/2606.30678",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.25550",
+    "title": "On the Viability of Requirements Generation From Code: An Experience Report",
+    "url": "https://arxiv.org/abs/2606.25550",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Alexander Korn",
+      "Jone Bartel",
+      "Max Unterbusch",
+      "Andreas Vogelsang"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.25550",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25550",
+    "pdf_url": "https://arxiv.org/pdf/2606.25550",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.23679",
+    "title": "Semantic Browsing: Controllable Diversity for Image Generation",
+    "url": "https://arxiv.org/abs/2606.23679",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Sara Dorfman",
+      "Maya Vishnevsky",
+      "Omer Dahary",
+      "Or Patashnik",
+      "Daniel Cohen-Or"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI",
+      "cs.GR",
+      "cs.LG"
+    ],
+    "topics": [
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.23679",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23679",
+    "pdf_url": "https://arxiv.org/pdf/2606.23679",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.20161",
+    "title": "ARTEMIS: Agent-guided Reliability-aware Temporal Mask Evolution for Imperfectly Supervised Video Polyp Segmentation",
+    "url": "https://arxiv.org/abs/2606.20161",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Tong Wang",
+      "Siwen Wang",
+      "Yaolei Qi",
+      "Jinxing Zhou",
+      "Yuting He",
+      "Guanyu Yang",
+      "Yutong Xie"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "computer-use",
+      "multi-agent"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.20161",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20161",
+    "pdf_url": "https://arxiv.org/pdf/2606.20161",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.20453",
+    "title": "Directors Duties in the Age of Agentic Artificial Intelligence",
+    "url": "https://arxiv.org/abs/2606.20453",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Deirdre Ahern"
+    ],
+    "categories": [
+      "cs.CY",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.20453",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20453",
+    "pdf_url": "https://arxiv.org/pdf/2606.20453",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.19670",
+    "title": "PiMiX 2.0: AI-enhanced Data Fusion for Radiographic Imaging and Tomography",
+    "url": "https://arxiv.org/abs/2606.19670",
+    "published": "2026-06-18",
+    "updated": "2026-06-20",
+    "authors": [
+      "Zhehui Wang",
+      "Shanny Lin",
+      "Nicholas Amano",
+      "Susan S. Glenn",
+      "Ramya Gurunathan",
+      "Katie Liu",
+      "Nathan E. Peterson",
+      "Michelle A. Espy",
+      "Adam Thompson",
+      "Amy J. Clarke",
+      "Ray T. Chen"
+    ],
+    "categories": [
+      "physics.ins-det",
+      "physics.data-an"
+    ],
+    "topics": [
+      "computer-use",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.19670",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19670",
+    "pdf_url": "https://arxiv.org/pdf/2606.19670",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.20708",
+    "title": "Simulated Customers Never Walk Away: Decision Fidelity of LLM User Simulators Measured Against Real Purchase Outcomes",
+    "url": "https://arxiv.org/abs/2606.20708",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Liang Chen"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "world-model"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.20708",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20708",
+    "pdf_url": "https://arxiv.org/pdf/2606.20708",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.16903",
+    "title": "Directory-Aware Query and Maintenance in Vector Databases",
+    "url": "https://arxiv.org/abs/2606.16903",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Mengzhao Wang",
+      "Zheng Gong",
+      "Jingpei Hu",
+      "Jiajie Fu",
+      "Maojia Sheng",
+      "Junwen Chen",
+      "Yifan Zhu"
+    ],
+    "categories": [
+      "cs.DB"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.16903",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16903",
+    "pdf_url": "https://arxiv.org/pdf/2606.16903",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.12485",
+    "title": "Speculative Rollback Correction for Quality-Diverse Web Agent Imitation",
+    "url": "https://arxiv.org/abs/2606.12485",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Longkun Hao",
+      "Hongyu Lin",
+      "Hao Li",
+      "Zhichao Yang",
+      "Haojie Hao",
+      "Dongshuo Huang",
+      "Haitao Yang",
+      "Hongyu Ge",
+      "Ming jie Xie",
+      "Yanjun Wu",
+      "Zi Hao Yin",
+      "Yan Bai",
+      "Yihang Lou"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "computer-use",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.12485",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12485",
+    "pdf_url": "https://arxiv.org/pdf/2606.12485",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.08944",
+    "title": "LongRTL: Graph-Similarity-Guided LLM-driven Long Context RTL Optimization",
+    "url": "https://arxiv.org/abs/2606.08944",
+    "published": "2026-06-08",
+    "updated": "2026-06-08",
+    "authors": [
+      "Yuyang Ye",
+      "Che-Kuan Shen",
+      "Xiangfei Hu",
+      "Yuchen Liu",
+      "Shuo Yin",
+      "Xufeng Yao",
+      "Bei Yu",
+      "Tsung-Yi Ho"
+    ],
+    "categories": [
+      "cs.AR",
+      "cs.PL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.08944",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.08944",
+    "pdf_url": "https://arxiv.org/pdf/2606.08944",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.08755",
+    "title": "Co-Evolving Skill Generation and Policy Optimization",
+    "url": "https://arxiv.org/abs/2606.08755",
+    "published": "2026-06-07",
+    "updated": "2026-06-07",
+    "authors": [
+      "Zhiwei Zhang",
+      "Yudi Lin",
+      "Nikki Lijing Kuang",
+      "Linlin Wu",
+      "Xiaomin Li",
+      "Songtao Liu",
+      "Fenglong Ma"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.08755",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.08755",
+    "pdf_url": "https://arxiv.org/pdf/2606.08755",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.04751",
+    "title": "FALSIFYBENCH: Evaluating Inductive Reasoning in LLMs with Rule Discovery Games",
+    "url": "https://arxiv.org/abs/2606.04751",
+    "published": "2026-06-03",
+    "updated": "2026-06-03",
+    "authors": [
+      "Leonardo Bertolazzi",
+      "Katya Tentori",
+      "Raffaella Bernardi"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.04751",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.04751",
+    "pdf_url": "https://arxiv.org/pdf/2606.04751",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.03777",
+    "title": "From Control Boundary to Insurance Claim: Reconstructing AI-Mediated Losses Through the CER Framework",
+    "url": "https://arxiv.org/abs/2606.03777",
+    "published": "2026-06-02",
+    "updated": "2026-06-02",
+    "authors": [
+      "Alex Leung",
+      "Rex Zhang",
+      "Kentaroh Toyoda",
+      "SiewMei Loh"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CR",
+      "q-fin.RM"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.03777",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.03777",
+    "pdf_url": "https://arxiv.org/pdf/2606.03777",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.02483",
+    "title": "Ghost Tool Calls: Issue-Time Privacy for Speculative Agent Tools",
+    "url": "https://arxiv.org/abs/2606.02483",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Bardia Mohammadi",
+      "Lars Klein",
+      "Akhil Arora",
+      "Laurent Bindschaedler"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.02483",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.02483",
+    "pdf_url": "https://arxiv.org/pdf/2606.02483",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.01212",
+    "title": "DiscourseFlip: An Oblique Discourse-Level Opinion Manipulation Attack against Black-box Retrieval-Augmented Generation",
+    "url": "https://arxiv.org/abs/2606.01212",
+    "published": "2026-05-31",
+    "updated": "2026-06-03",
+    "authors": [
+      "Yuyang Gong",
+      "Miaokun Chen",
+      "Jiawei Liu",
+      "Zhuo Chen",
+      "Guoxiu He",
+      "Wei Lu",
+      "XiaoFeng Wang",
+      "Xiaozhong Liu"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.CR",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "rag"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.01212",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.01212",
+    "pdf_url": "https://arxiv.org/pdf/2606.01212",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.26754",
+    "title": "Cordon-MAS: Defending RAG against Knowledge Poisoning via Information-Flow Control",
+    "url": "https://arxiv.org/abs/2605.26754",
+    "published": "2026-05-26",
+    "updated": "2026-05-26",
+    "authors": [
+      "Zhe Yu",
+      "Wenpeng Xing",
+      "Gaolei Li",
+      "Shuguang Xiong",
+      "Hongzhi Wang",
+      "Xuyang Teng",
+      "Meng Han"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.26754",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.26754",
+    "pdf_url": "https://arxiv.org/pdf/2605.26754",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.25379",
+    "title": "EfficientGraph-RAG: Structured Retrieval-State Management for Cross-Task Retrieval-Augmented Generation",
+    "url": "https://arxiv.org/abs/2605.25379",
+    "published": "2026-05-25",
+    "updated": "2026-05-25",
+    "authors": [
+      "Miaohe Niu",
+      "Lianlei Shan",
+      "Zhengtao Yu",
+      "Jingbo Zhu",
+      "Tong Xiao"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.25379",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.25379",
+    "pdf_url": "https://arxiv.org/pdf/2605.25379",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.21401",
+    "title": "Open-source LLMs administer maximum electric shocks in a Milgram-like obedience experiment",
+    "url": "https://arxiv.org/abs/2605.21401",
+    "published": "2026-05-20",
+    "updated": "2026-06-23",
+    "authors": [
+      "Roland Pihlakas",
+      "Jan Llenzl Dagohoy"
+    ],
+    "categories": [
+      "cs.CY",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.21401",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.21401",
+    "pdf_url": "https://arxiv.org/pdf/2605.21401",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.18414",
+    "title": "Prompts Don't Protect: Architectural Enforcement via MCP Proxy for LLM Tool Access Control",
+    "url": "https://arxiv.org/abs/2605.18414",
+    "published": "2026-05-18",
+    "updated": "2026-05-18",
+    "authors": [
+      "Rohith Uppala"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.18414",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.18414",
+    "pdf_url": "https://arxiv.org/pdf/2605.18414",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.14588",
+    "title": "Silent Collapse in Recursive Learning Systems",
+    "url": "https://arxiv.org/abs/2605.14588",
+    "published": "2026-05-14",
+    "updated": "2026-05-19",
+    "authors": [
+      "Zhipeng Zhang"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.14588",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.14588",
+    "pdf_url": "https://arxiv.org/pdf/2605.14588",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.00651",
+    "title": "EQSANS-CLI: A natural-language, agent-ready command-line tool for small-angle neutron scattering data reduction at EQ-SANS",
+    "url": "https://arxiv.org/abs/2605.00651",
+    "published": "2026-05-01",
+    "updated": "2026-05-01",
+    "authors": [
+      "Changwoo Do"
+    ],
+    "categories": [
+      "physics.ins-det"
+    ],
+    "topics": [
+      "computer-use",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.00651",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.00651",
+    "pdf_url": "https://arxiv.org/pdf/2605.00651",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.01047",
+    "title": "LLM Ghostbusters: Surgical Hallucination Suppression via Adaptive Unlearning",
+    "url": "https://arxiv.org/abs/2605.01047",
+    "published": "2026-05-01",
+    "updated": "2026-05-01",
+    "authors": [
+      "Joseph Spracklen",
+      "Pedram Aghazadeh",
+      "Farinaz Koushanfar",
+      "Murtuza Jadliwala"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.01047",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.01047",
+    "pdf_url": "https://arxiv.org/pdf/2605.01047",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2604.28138",
+    "title": "Crab: A Semantics-Aware Checkpoint/Restore Runtime for Agent Sandboxes",
+    "url": "https://arxiv.org/abs/2604.28138",
+    "published": "2026-04-30",
+    "updated": "2026-04-30",
+    "authors": [
+      "Tianyuan Wu",
+      "Chaokun Chang",
+      "Lunxi Cao",
+      "Wei Gao",
+      "Wei Wang"
+    ],
+    "categories": [
+      "cs.OS",
+      "cs.AI"
+    ],
+    "topics": [
+      "tool-use"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2604.28138",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.28138",
+    "pdf_url": "https://arxiv.org/pdf/2604.28138",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2603.10285",
+    "title": "Conversational AI-Enhanced Exploration System to Query Large-Scale Digitised Collections of Natural History Museums",
+    "url": "https://arxiv.org/abs/2603.10285",
+    "published": "2026-03-11",
+    "updated": "2026-03-11",
+    "authors": [
+      "Yiyuan Wang",
+      "Andrew Johnston",
+      "Zoë Sadokierski",
+      "Rhiannon Stephens",
+      "Shane T. Ahyong"
+    ],
+    "categories": [
+      "cs.HC",
+      "cs.AI",
+      "cs.CY",
+      "cs.DL",
+      "cs.ET"
+    ],
+    "topics": [
+      "rag",
+      "tool-use"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2603.10285",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.10285",
+    "pdf_url": "https://arxiv.org/pdf/2603.10285",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2602.08121",
+    "title": "Initial Risk Probing and Feasibility Testing of Glow: a Generative AI-Powered Dialectical Behavior Therapy Skills Coach for Substance Use Recovery and HIV Prevention",
+    "url": "https://arxiv.org/abs/2602.08121",
+    "published": "2026-02-08",
+    "updated": "2026-02-08",
+    "authors": [
+      "Liying Wang",
+      "Madison Lee",
+      "Yunzhang Jiang",
+      "Steven Chen",
+      "Kewei Sha",
+      "Yunhe Feng",
+      "Frank Wong",
+      "Lisa Hightow-Weidman",
+      "Weichao Yuwen"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.08121",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.08121",
+    "pdf_url": "https://arxiv.org/pdf/2602.08121",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2601.06937",
+    "title": "mind_call: A Dataset for Mental Health Function Calling with Large Language Models",
+    "url": "https://arxiv.org/abs/2601.06937",
+    "published": "2026-01-11",
+    "updated": "2026-01-11",
+    "authors": [
+      "Fozle Rabbi Shafi",
+      "M. Anwar Hossain",
+      "Salimur Choudhury"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2601.06937",
+    "source": "arxiv",
+    "source_id": "arxiv:2601.06937",
+    "pdf_url": "https://arxiv.org/pdf/2601.06937",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2510.13558",
+    "title": "Steer-MoE: Efficient Audio-Language Alignment with a Mixture-of-Experts Steering Module",
+    "url": "https://arxiv.org/abs/2510.13558",
+    "published": "2025-10-15",
+    "updated": "2025-10-15",
+    "authors": [
+      "Ruitao Feng",
+      "Bixi Zhang",
+      "Sheng Liang",
+      "Zheng Yuan"
+    ],
+    "categories": [
+      "cs.SD"
+    ],
+    "topics": [
+      "agent-safety",
+      "reasoning"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2510.13558",
+    "source": "arxiv",
+    "source_id": "arxiv:2510.13558",
+    "pdf_url": "https://arxiv.org/pdf/2510.13558",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2509.26463",
+    "title": "ErrorPrism: Reconstructing Error Propagation Paths in Cloud Service Systems",
+    "url": "https://arxiv.org/abs/2509.26463",
+    "published": "2025-09-30",
+    "updated": "2025-09-30",
+    "authors": [
+      "Junsong Pu",
+      "Yichen Li",
+      "Zhuangbin Chen",
+      "Jinyang Liu",
+      "Zhihan Jiang",
+      "Jianjun Chen",
+      "Rui Shi",
+      "Zibin Zheng",
+      "Tieying Zhang"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2509.26463",
+    "source": "arxiv",
+    "source_id": "arxiv:2509.26463",
+    "pdf_url": "https://arxiv.org/pdf/2509.26463",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2509.24229",
+    "title": "Model Fusion with Multi-LoRA Inference for Tool-Enhanced Game Dialogue Agents",
+    "url": "https://arxiv.org/abs/2509.24229",
+    "published": "2025-09-29",
+    "updated": "2025-09-29",
+    "authors": [
+      "Kangxu Wang",
+      "Ze Chen",
+      "Chengcheng Wei",
+      "Jiewen Zheng",
+      "Jiarong He",
+      "Max Gao"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "tool-use"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2509.24229",
+    "source": "arxiv",
+    "source_id": "arxiv:2509.24229",
+    "pdf_url": "https://arxiv.org/pdf/2509.24229",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2509.04518",
+    "title": "Advancing SLM Tool-Use Capability using Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2509.04518",
+    "published": "2025-09-03",
+    "updated": "2025-09-08",
+    "authors": [
+      "Dhruvi Paprunia",
+      "Vansh Kharidia",
+      "Pankti Doshi"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "tool-use"
+    ],
+    "score": 8,
+    "relevance": "medium",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2509.04518",
+    "source": "arxiv",
+    "source_id": "arxiv:2509.04518",
+    "pdf_url": "https://arxiv.org/pdf/2509.04518",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2607.04103",
+    "title": "Governing Generative AI Across Financial Institutions: An SR 26-2-Compatible Framework for Generative AI Risk Control",
+    "url": "https://arxiv.org/abs/2607.04103",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Yiqing Wang",
+      "Yixin Kang",
+      "Luyun Lin",
+      "Siqi Mao"
+    ],
+    "categories": [
+      "q-fin.RM",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-safety",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 7,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.04103",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04103",
+    "pdf_url": "https://arxiv.org/pdf/2607.04103",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.03181",
+    "title": "Teaming Up with AI: Coordination and Cooperation",
+    "url": "https://arxiv.org/abs/2607.03181",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Nicole Immorlica",
+      "Inbal Talgam-Cohen"
+    ],
+    "categories": [
+      "cs.GT",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 7,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.03181",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03181",
+    "pdf_url": "https://arxiv.org/pdf/2607.03181",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.03100",
+    "title": "Flow-A11y: Flow-Aware Accessibility Testing",
+    "url": "https://arxiv.org/abs/2607.03100",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Nasr Eddine Fliti",
+      "Leisan Kokorina",
+      "Florian Tambon",
+      "Michael Papadakis"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 7,
+    "relevance": "medium",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2607.03100",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03100",
+    "pdf_url": "https://arxiv.org/pdf/2607.03100",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2607.00613",
+    "title": "Ai2-Kit: Streamlining AI-Accelerated Ab Initio Workflows for Complex Chemical Systems",
+    "url": "https://arxiv.org/abs/2607.00613",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Sheng Bi",
+      "Wei-Hong Xu",
+      "Yong-Bin Zhuang",
+      "Jia-Xin Zhu",
+      "Jiang-Peng Qiu",
+      "Yu-Hang Tang",
+      "Xiang-Long Du",
+      "Qi You",
+      "Yun-Pei Liu",
+      "Fu-Qiang Gong",
+      "Yu-Xin Guo",
+      "Yi-Ze Wang",
+      "Cheng-Xuan Wang",
+      "Zi-Heng Gong",
+      "Zi-Qiang Chen",
+      "Chang Liu",
+      "Si-Yuan Han",
+      "Jian Gu",
+      "Jia-Xin Li",
+      "Yi-Ming Chen",
+      "Lin Huang",
+      "Si-Jie Chen",
+      "Bo-Ying Huang",
+      "Jie-Zhen Xia",
+      "Fan-Jie Xu",
+      "Su-Yang Zhong",
+      "Peng-Wei Xu",
+      "Jun-Yi Wang",
+      "Xing-Yun Xie",
+      "Yu-Lei Gong",
+      "Yan-Yi Su",
+      "Yue Liu",
+      "Rui-Hao Bi",
+      "Lang Li",
+      "Fei-Teng Wang",
+      "Jing-Xiang Zou",
+      "Mei Jia",
+      "Jie-Qiong Li",
+      "Min Lin",
+      "Qi-Yuan Fan",
+      "Juan-Juan Sun",
+      "Jia-Bo Le",
+      "Zixuan Wei",
+      "Jin-Yuan Hu",
+      "Meng-Lei Jia",
+      "Yan Sun",
+      "Xiao-Hui Yang",
+      "Fujie Tang",
+      "Feng Wang",
+      "Jun Cheng"
+    ],
+    "categories": [
+      "physics.chem-ph"
+    ],
+    "topics": [
+      "rag",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 7,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.00613",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00613",
+    "pdf_url": "https://arxiv.org/pdf/2607.00613",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.31399",
+    "title": "World-Model Collapse as a Phase Transition",
+    "url": "https://arxiv.org/abs/2606.31399",
+    "published": "2026-06-30",
+    "updated": "2026-07-04",
+    "authors": [
+      "Xinyuan Song",
+      "Zekun Cai"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "planning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 7,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.31399",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31399",
+    "pdf_url": "https://arxiv.org/pdf/2606.31399",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2607.00155",
+    "title": "A Contextual-Bandit Oversight Game with Two-Sided Informational Asymmetry",
+    "url": "https://arxiv.org/abs/2607.00155",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Yunjin Tong"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.GT"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "tool-use"
+    ],
+    "score": 7,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.00155",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00155",
+    "pdf_url": "https://arxiv.org/pdf/2607.00155",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.31214",
+    "title": "EasyScan_HEP 2: Agent-Ready Parameter Scans for High-Energy Physics",
+    "url": "https://arxiv.org/abs/2606.31214",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Yang Xiao",
+      "Yuanfang Yue",
+      "Yang Zhang"
+    ],
+    "categories": [
+      "hep-ph"
+    ],
+    "topics": [
+      "workflow-agent"
+    ],
+    "score": 7,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.31214",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31214",
+    "pdf_url": "https://arxiv.org/pdf/2606.31214",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.29389",
+    "title": "Exploring the Cryptographic Limits of Transformer Networks",
+    "url": "https://arxiv.org/abs/2606.29389",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Stefan Domunco",
+      "Andis Draguns",
+      "Philip Torr",
+      "Isaac Robinson",
+      "Christian Schroeder de Witt"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety"
+    ],
+    "score": 7,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.29389",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29389",
+    "pdf_url": "https://arxiv.org/pdf/2606.29389",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.29100",
+    "title": "Toward Exascale AI for Science: A Scalable AI Skill for Autonomous Microkinetics Discovery",
+    "url": "https://arxiv.org/abs/2606.29100",
+    "published": "2026-06-27",
+    "updated": "2026-07-03",
+    "authors": [
+      "Ken-ichi Nomura",
+      "William Dawson",
+      "Nabankur Dasgupta",
+      "Taufeq Mohammed Razakh",
+      "Thomas Linker",
+      "Kai Ito",
+      "Aiichiro Nakano"
+    ],
+    "categories": [
+      "cs.CE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 7,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.29100",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29100",
+    "pdf_url": "https://arxiv.org/pdf/2606.29100",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.26448",
+    "title": "Closing the Loop to Discover Psychological Theories with an Automated Cognitive Scientist",
+    "url": "https://arxiv.org/abs/2606.26448",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Akshay K. Jagadish",
+      "Younes Strittmatter",
+      "Nori Jacoby",
+      "George Kachergis",
+      "Eric Schulz",
+      "Nathaniel Daw",
+      "Suyog H. Chandramouli",
+      "Thomas L. Griffiths"
+    ],
+    "categories": [
+      "q-bio.NC",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent"
+    ],
+    "score": 7,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai",
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.26448",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26448",
+    "pdf_url": "https://arxiv.org/pdf/2606.26448",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.25525",
+    "title": "The impact of artificial intelligence on enterprise software user roles",
+    "url": "https://arxiv.org/abs/2606.25525",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Isabel Unger",
+      "Elizangela Valarini",
+      "Martin Schrepp",
+      "Nina Hollender",
+      "Gabriela Rocha",
+      "Erik Bertram"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "multi-agent",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 7,
+    "relevance": "medium",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.25525",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25525",
+    "pdf_url": "https://arxiv.org/pdf/2606.25525",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.25496",
+    "title": "Recommendation as Generation: Unifying Personalized Video Generation and Recommendation at Industrial Scale",
+    "url": "https://arxiv.org/abs/2606.25496",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Yanhua Cheng",
+      "Bo Wang",
+      "Haotian Zhang",
+      "Xinyuan Gao",
+      "Zhihui Yin",
+      "Ben Xue",
+      "Yongzhi Li",
+      "Jieting Xue",
+      "Ye Ma",
+      "Minquan Wang",
+      "Jiahui Li",
+      "Tianyu Xu",
+      "Zhiqiang Liu",
+      "Xiao Lin",
+      "Shiyang Wen",
+      "Changcheng Li",
+      "Liu Liu",
+      "Quan Chen",
+      "Peng Jiang",
+      "Kun Gai"
+    ],
+    "categories": [
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag"
+    ],
+    "score": 7,
+    "relevance": "medium",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.25496",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25496",
+    "pdf_url": "https://arxiv.org/pdf/2606.25496",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.23348",
+    "title": "Superhuman AI for Generals.io Using Self-Play Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2606.23348",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Matej Straka",
+      "Viliam Lisý",
+      "Martin Schmid"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag"
+    ],
+    "score": 7,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.23348",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23348",
+    "pdf_url": "https://arxiv.org/pdf/2606.23348",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.21124",
+    "title": "PulseCX: Breaking the Closed-World Assumption in Real-Time CX",
+    "url": "https://arxiv.org/abs/2606.21124",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Rajat Agarwal",
+      "Suvidha Tripathi",
+      "Shubham Sharma"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.IR"
+    ],
+    "topics": [
+      "memory",
+      "tool-use"
+    ],
+    "score": 7,
+    "relevance": "medium",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.21124",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21124",
+    "pdf_url": "https://arxiv.org/pdf/2606.21124",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.16319",
+    "title": "Architectural Wisdom: A Framework for Governing Optimization in AI Systems",
+    "url": "https://arxiv.org/abs/2606.16319",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Edward Y. Chang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "rag",
+      "tool-use"
+    ],
+    "score": 7,
+    "relevance": "medium",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.16319",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16319",
+    "pdf_url": "https://arxiv.org/pdf/2606.16319",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2605.18991",
+    "title": "Agent Security is a Systems Problem",
+    "url": "https://arxiv.org/abs/2605.18991",
+    "published": "2026-05-18",
+    "updated": "2026-05-20",
+    "authors": [
+      "Mihai Christodorescu",
+      "Earlence Fernandes",
+      "Ashish Hooda",
+      "Somesh Jha",
+      "Johann Rehberger",
+      "Kamalika Chaudhuri",
+      "Xiaohan Fu",
+      "Khawaja Shams",
+      "Guy Amir",
+      "Jihye Choi",
+      "Sarthak Choudhary",
+      "Nils Palumbo",
+      "Andrey Labunets",
+      "Nishit V. Pandya"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety"
+    ],
+    "score": 7,
+    "relevance": "medium",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.18991",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.18991",
+    "pdf_url": "https://arxiv.org/pdf/2605.18991",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.02028",
+    "title": "Language models fail at extended rule following",
+    "url": "https://arxiv.org/abs/2605.02028",
+    "published": "2026-05-03",
+    "updated": "2026-05-16",
+    "authors": [
+      "Tianxiang Dai",
+      "Jonathan Fan"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "tool-use"
+    ],
+    "score": 7,
+    "relevance": "medium",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.02028",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.02028",
+    "pdf_url": "https://arxiv.org/pdf/2605.02028",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2604.25610",
+    "title": "Optimizing ground state preparation protocols with autoresearch",
+    "url": "https://arxiv.org/abs/2604.25610",
+    "published": "2026-04-28",
+    "updated": "2026-05-08",
+    "authors": [
+      "Luis Mantilla Calderón",
+      "Jérôme F. Gonthier",
+      "Ignacio Gustin",
+      "Varinia Bernales",
+      "Alán Aspuru-Guzik"
+    ],
+    "categories": [
+      "quant-ph"
+    ],
+    "topics": [
+      "coding-agent",
+      "computer-use",
+      "world-model"
+    ],
+    "score": 7,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2604.25610",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.25610",
+    "pdf_url": "https://arxiv.org/pdf/2604.25610",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.25636",
+    "title": "Designing Any Imaging System from Natural Language: Agent-Constrained Composition over a Finite Primitive Basis",
+    "url": "https://arxiv.org/abs/2603.25636",
+    "published": "2026-03-26",
+    "updated": "2026-03-26",
+    "authors": [
+      "Chengshuai Yang"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "planning",
+      "tool-use"
+    ],
+    "score": 7,
+    "relevance": "medium",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.25636",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.25636",
+    "pdf_url": "https://arxiv.org/pdf/2603.25636",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2607.05835",
+    "title": "Tangent classes of matroids and wonderful compactifications",
+    "url": "https://arxiv.org/abs/2607.05835",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Ronnie Cheng",
+      "Shurui Liu",
+      "Guoxiong Gao"
+    ],
+    "categories": [
+      "math.AG",
+      "cs.AI",
+      "math.CO"
+    ],
+    "topics": [
+      "computer-use",
+      "reasoning"
+    ],
+    "score": 6,
+    "relevance": "low",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.05835",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05835",
+    "pdf_url": "https://arxiv.org/pdf/2607.05835",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.05744",
+    "title": "Unicode TAG-Block Concealment of Tool-Metadata Payloads in the Model Context Protocol: An Approval-View Fidelity Gap Across Three Independent Server Implementations",
+    "url": "https://arxiv.org/abs/2607.05744",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Mohammadreza Rashidi"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.SE"
+    ],
+    "topics": [
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 6,
+    "relevance": "low",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.05744",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05744",
+    "pdf_url": "https://arxiv.org/pdf/2607.05744",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.02905",
+    "title": "Pre-Strings Lectures on Artificial Intelligence",
+    "url": "https://arxiv.org/abs/2607.02905",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "James Halverson"
+    ],
+    "categories": [
+      "hep-th"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "workflow-agent"
+    ],
+    "score": 6,
+    "relevance": "low",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.02905",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02905",
+    "pdf_url": "https://arxiv.org/pdf/2607.02905",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.02609",
+    "title": "Knowledge-Centric Information Systems",
+    "url": "https://arxiv.org/abs/2607.02609",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Mariano Garralda-Barrio"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.DB"
+    ],
+    "topics": [
+      "workflow-agent"
+    ],
+    "score": 6,
+    "relevance": "low",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.02609",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02609",
+    "pdf_url": "https://arxiv.org/pdf/2607.02609",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.01415",
+    "title": "The Rollout Infrastructure Tax in Coding-Agent Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2607.01415",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Daniel Thi Graviet",
+      "Lovre Pesut",
+      "Ivan Dagelic",
+      "Vedran Jukic",
+      "Ivan Burazin"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.DC"
+    ],
+    "topics": [
+      "coding-agent"
+    ],
+    "score": 6,
+    "relevance": "low",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.01415",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01415",
+    "pdf_url": "https://arxiv.org/pdf/2607.01415",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.01380",
+    "title": "Lagrangian evaluation of polymeric stress in viscoelastic fluids",
+    "url": "https://arxiv.org/abs/2607.01380",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Mohammad Majidi",
+      "Rishu Gandhi",
+      "Louison Thorens",
+      "Maliheh Teimouri",
+      "Jeffrey S. Guasto",
+      "Arezoo M. Ardekani"
+    ],
+    "categories": [
+      "physics.flu-dyn"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "world-model"
+    ],
+    "score": 6,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2607.01380",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01380",
+    "pdf_url": "https://arxiv.org/pdf/2607.01380",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.32014",
+    "title": "Scalable Behaviour Cloning on Browser Using via Skill Distillation",
+    "url": "https://arxiv.org/abs/2606.32014",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Kaisen Yang",
+      "Zheng Jiang",
+      "Yuzhao Peng",
+      "Houde Qian",
+      "Boshi Zhang",
+      "Youjie Zheng",
+      "Shijin Hong",
+      "Qingle Liu",
+      "Ruoyu Han",
+      "Bohan Lyu",
+      "Bingxiang He",
+      "Eren Cai",
+      "Calvin Xiao",
+      "Qinhuai Na"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "computer-use",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 6,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.32014",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.32014",
+    "pdf_url": "https://arxiv.org/pdf/2606.32014",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.29532",
+    "title": "SemJoin: Semantic Join Optimization",
+    "url": "https://arxiv.org/abs/2606.29532",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Christopher Gou",
+      "Aditya Banerjee",
+      "Jiaxuan Wang",
+      "Chunwei Liu"
+    ],
+    "categories": [
+      "cs.DB",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation"
+    ],
+    "score": 6,
+    "relevance": "low",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.29532",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29532",
+    "pdf_url": "https://arxiv.org/pdf/2606.29532",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.27951",
+    "title": "AI Persuasive Framing in Collective Dilemmas",
+    "url": "https://arxiv.org/abs/2606.27951",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Anders Giovanni Møller",
+      "Alessia Galdeman",
+      "Arianna Pera",
+      "Luca Maria Aiello"
+    ],
+    "categories": [
+      "cs.CY",
+      "cs.CL",
+      "cs.HC",
+      "physics.soc-ph"
+    ],
+    "topics": [
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 6,
+    "relevance": "low",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.27951",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27951",
+    "pdf_url": "https://arxiv.org/pdf/2606.27951",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.27291",
+    "title": "Designing Reward Signals for Portable Query Generation: A Case Study in Industrial Semantic Job Search",
+    "url": "https://arxiv.org/abs/2606.27291",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Ping Liu",
+      "Qianqi Shen",
+      "Jianqiang Shen",
+      "Wenqiong Liu",
+      "Rajat Arora",
+      "Yunxiang Ren",
+      "Chunnan Yao",
+      "Dan Xu",
+      "Baofen Zheng",
+      "Wanjun Jiang",
+      "Andrii Soviak",
+      "Kevin Kao",
+      "Jingwei Wu",
+      "Wenjing Zhang"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 6,
+    "relevance": "low",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.27291",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27291",
+    "pdf_url": "https://arxiv.org/pdf/2606.27291",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.25244",
+    "title": "Reading AI Model Compilation in MLIR Through the Lens of Formal Theories",
+    "url": "https://arxiv.org/abs/2606.25244",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Javed Absar"
+    ],
+    "categories": [
+      "cs.PL"
+    ],
+    "topics": [
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 6,
+    "relevance": "low",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.25244",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25244",
+    "pdf_url": "https://arxiv.org/pdf/2606.25244",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.23768",
+    "title": "Cryptographic certificates of validity for trustworthy AI",
+    "url": "https://arxiv.org/abs/2606.23768",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Murdoch J. Gabbay"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.LO"
+    ],
+    "topics": [
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 6,
+    "relevance": "low",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.23768",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23768",
+    "pdf_url": "https://arxiv.org/pdf/2606.23768",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.22447",
+    "title": "A Differentiable Atari VCS:A Complex, Fully Known Ground Truth for Explainable AI",
+    "url": "https://arxiv.org/abs/2606.22447",
+    "published": "2026-06-21",
+    "updated": "2026-06-21",
+    "authors": [
+      "Andreas Maier",
+      "Siming Bayer",
+      "Patrick Krauss"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "coding-agent",
+      "planning"
+    ],
+    "score": 6,
+    "relevance": "low",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.22447",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22447",
+    "pdf_url": "https://arxiv.org/pdf/2606.22447",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.19924",
+    "title": "The Tao of Agency: Autotelic AI, Embedded Agency and Dissolution of the Self",
+    "url": "https://arxiv.org/abs/2606.19924",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Aritra Sarkar"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent"
+    ],
+    "score": 6,
+    "relevance": "low",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.19924",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19924",
+    "pdf_url": "https://arxiv.org/pdf/2606.19924",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.19047",
+    "title": "RODS: Reward-Driven Online Data Synthesis for Multi-Turn Tool-Use Agents",
+    "url": "https://arxiv.org/abs/2606.19047",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Ruishan Fang",
+      "Siyuan Lu",
+      "Chenyi Zhuang",
+      "Tao Lin"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "tool-use"
+    ],
+    "score": 6,
+    "relevance": "low",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.19047",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19047",
+    "pdf_url": "https://arxiv.org/pdf/2606.19047",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.19458",
+    "title": "MonaVec: A Training-Free Embedded Vector Search Kernel for Edge and Offline AI Systems",
+    "url": "https://arxiv.org/abs/2606.19458",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Oğuzhan Yenen"
+    ],
+    "categories": [
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag"
+    ],
+    "score": 6,
+    "relevance": "low",
+    "matched_queries": [
+      "function-calling",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.19458",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19458",
+    "pdf_url": "https://arxiv.org/pdf/2606.19458",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2606.17321",
+    "title": "ProCUA-SFT Technical Report",
+    "url": "https://arxiv.org/abs/2606.17321",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Jaehun Jung",
+      "Ximing Lu",
+      "Brandon Cui",
+      "Muhammad Khalifa",
+      "Shaokun Zhang",
+      "Hao Zhang",
+      "Jin Xu",
+      "Amala Sanjay Deshmukh",
+      "Karan Sapra",
+      "Andrew Tao",
+      "Yejin Choi",
+      "Jan Kautz",
+      "Mingjie Liu",
+      "Yi Dong"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.CV"
+    ],
+    "topics": [
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 6,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.17321",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17321",
+    "pdf_url": "https://arxiv.org/pdf/2606.17321",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.15071",
+    "title": "Quantum learning with a single-atom sensor",
+    "url": "https://arxiv.org/abs/2606.15071",
+    "published": "2026-06-13",
+    "updated": "2026-06-13",
+    "authors": [
+      "Yin Mo",
+      "Emilio Bagan",
+      "Giulio Chiribella"
+    ],
+    "categories": [
+      "quant-ph"
+    ],
+    "topics": [
+      "memory",
+      "tool-use"
+    ],
+    "score": 6,
+    "relevance": "low",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.15071",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15071",
+    "pdf_url": "https://arxiv.org/pdf/2606.15071",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.09416",
+    "title": "Harness Engineering for Physical AI: Robot Middleware Is the Harness Layer",
+    "url": "https://arxiv.org/abs/2606.09416",
+    "published": "2026-06-08",
+    "updated": "2026-06-08",
+    "authors": [
+      "Sanghoon Lee",
+      "Jiyeong Chae",
+      "Kyung-Joon Park"
+    ],
+    "categories": [
+      "cs.RO",
+      "cs.AI",
+      "cs.SE"
+    ],
+    "topics": [
+      "embodied-agent",
+      "planning",
+      "tool-use"
+    ],
+    "score": 6,
+    "relevance": "low",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.09416",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09416",
+    "pdf_url": "https://arxiv.org/pdf/2606.09416",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.02840",
+    "title": "Self-Regulation through Communication in Evolved Neural Agents",
+    "url": "https://arxiv.org/abs/2606.02840",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Joshua Nunley"
+    ],
+    "categories": [
+      "q-bio.PE",
+      "cs.MA",
+      "cs.NE",
+      "nlin.AO"
+    ],
+    "topics": [
+      "agent-safety"
+    ],
+    "score": 6,
+    "relevance": "low",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2606.02840",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.02840",
+    "pdf_url": "https://arxiv.org/pdf/2606.02840",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.02528",
+    "title": "Auditing Asset-Specific Preferences in Financial Large Language Models: Evidence from Bitcoin Representations and Portfolio Allocation",
+    "url": "https://arxiv.org/abs/2606.02528",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Wenbin Wu"
+    ],
+    "categories": [
+      "q-fin.GN",
+      "cs.CY",
+      "cs.LG"
+    ],
+    "topics": [
+      "rag"
+    ],
+    "score": 6,
+    "relevance": "low",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.02528",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.02528",
+    "pdf_url": "https://arxiv.org/pdf/2606.02528",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.08378",
+    "title": "Reinforcement Learning for Scalable and Trustworthy Intelligent Systems",
+    "url": "https://arxiv.org/abs/2605.08378",
+    "published": "2026-05-08",
+    "updated": "2026-05-08",
+    "authors": [
+      "Guangchen Lan"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-safety"
+    ],
+    "score": 6,
+    "relevance": "low",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.08378",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.08378",
+    "pdf_url": "https://arxiv.org/pdf/2605.08378",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.05795",
+    "title": "Reward Shaping and Action Masking for Compositional Tasks using Behavior Trees and LLMs",
+    "url": "https://arxiv.org/abs/2605.05795",
+    "published": "2026-05-07",
+    "updated": "2026-05-23",
+    "authors": [
+      "Nicholas Potteiger",
+      "Ankita Samaddar",
+      "Taylor T. Johnson",
+      "Xenofon Koutsoukos"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "tool-use"
+    ],
+    "score": 6,
+    "relevance": "low",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.05795",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.05795",
+    "pdf_url": "https://arxiv.org/pdf/2605.05795",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2509.25378",
+    "title": "Detecting and Fixing API Misuses of Data Science Libraries Using Large Language Models",
+    "url": "https://arxiv.org/abs/2509.25378",
+    "published": "2025-09-29",
+    "updated": "2025-09-29",
+    "authors": [
+      "Akalanka Galappaththi",
+      "Francisco Ribeiro",
+      "Sarah Nadi"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "tool-use"
+    ],
+    "score": 6,
+    "relevance": "low",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2509.25378",
+    "source": "arxiv",
+    "source_id": "arxiv:2509.25378",
+    "pdf_url": "https://arxiv.org/pdf/2509.25378",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2509.06736",
+    "title": "VehicleWorld: A Highly Integrated Multi-Device Environment for Intelligent Vehicle Interaction",
+    "url": "https://arxiv.org/abs/2509.06736",
+    "published": "2025-09-08",
+    "updated": "2025-09-08",
+    "authors": [
+      "Jie Yang",
+      "Jiajun Chen",
+      "Zhangyue Yin",
+      "Shuo Chen",
+      "Yuxin Wang",
+      "Yiran Guo",
+      "Yuan Li",
+      "Yining Zheng",
+      "Xuanjing Huang",
+      "Xipeng Qiu"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 6,
+    "relevance": "low",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2509.06736",
+    "source": "arxiv",
+    "source_id": "arxiv:2509.06736",
+    "pdf_url": "https://arxiv.org/pdf/2509.06736",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2607.01188",
+    "title": "Optimal Resource Utilization for Autonomous Laboratory Orchestrators",
+    "url": "https://arxiv.org/abs/2607.01188",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Austin McDannald",
+      "Julia Tisaranni",
+      "Howie Joress"
+    ],
+    "categories": [
+      "cs.AI",
+      "cond-mat.mtrl-sci"
+    ],
+    "topics": [
+      "planning"
+    ],
+    "score": 5,
+    "relevance": "low",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.01188",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01188",
+    "pdf_url": "https://arxiv.org/pdf/2607.01188",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.31132",
+    "title": "ELASTIC: Efficiently Learning to Adaptively Scale Test-Time Compute for Generative Control Policies",
+    "url": "https://arxiv.org/abs/2606.31132",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Andrew Zou Li",
+      "Gokul Swamy",
+      "Yonatan Bisk",
+      "Andrea Bajcsy"
+    ],
+    "categories": [
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "tool-use"
+    ],
+    "score": 5,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.31132",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31132",
+    "pdf_url": "https://arxiv.org/pdf/2606.31132",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.30543",
+    "title": "TRACE: Temporal Relationship-Aware Conversational Entrainment Detection in Dyadic Speech",
+    "url": "https://arxiv.org/abs/2606.30543",
+    "published": "2026-06-29",
+    "updated": "2026-07-03",
+    "authors": [
+      "Sathvik Manikantan Napa Ugandhar",
+      "Hao Zhang",
+      "Alison Gunzler",
+      "Yuzhe Wang",
+      "Thomas Thebaud",
+      "Georgi Tinchev",
+      "Venkatesh Ravichandran",
+      "Laureano Moro-Velázquez"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "tool-use"
+    ],
+    "score": 5,
+    "relevance": "low",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.30543",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30543",
+    "pdf_url": "https://arxiv.org/pdf/2606.30543",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.27045",
+    "title": "The Spec Growth Engine: Spec-Anchored, Code-Coupled, Drift-Enforced Architecture for AI-Assisted Software Development",
+    "url": "https://arxiv.org/abs/2606.27045",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Hartwig Grabowski"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "coding-agent"
+    ],
+    "score": 5,
+    "relevance": "low",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.27045",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27045",
+    "pdf_url": "https://arxiv.org/pdf/2606.27045",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.27365",
+    "title": "3D Imaging of Complex Skyrmion and Hopf Topologies in an Extended Sample",
+    "url": "https://arxiv.org/abs/2606.27365",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "I. Binnie",
+      "H. Fang",
+      "B. Shearer",
+      "A. Grafov",
+      "N. Jenkins",
+      "Y. Shao",
+      "C. O'Leary",
+      "Y. Liao",
+      "T. Feggeler",
+      "A. Oh",
+      "S. Yazdi",
+      "J. Zou",
+      "B. Wang",
+      "E-E. Cating",
+      "S. A. Montoya",
+      "D. Shapiro",
+      "J. Miao",
+      "H. C. Kapteyn",
+      "M. M. Murnane"
+    ],
+    "categories": [
+      "cond-mat.mtrl-sci",
+      "cond-mat.mes-hall",
+      "physics.app-ph"
+    ],
+    "topics": [
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 5,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.27365",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27365",
+    "pdf_url": "https://arxiv.org/pdf/2606.27365",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.25337",
+    "title": "AI Coaching for Accelerating Human Skill Development with Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2606.25337",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Wei Wang",
+      "Enlin Gu",
+      "Antonio Loquercio",
+      "Haimin Hu",
+      "Rahul Mangharam"
+    ],
+    "categories": [
+      "cs.RO",
+      "cs.AI",
+      "cs.HC"
+    ],
+    "topics": [
+      "embodied-agent"
+    ],
+    "score": 5,
+    "relevance": "low",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.25337",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25337",
+    "pdf_url": "https://arxiv.org/pdf/2606.25337",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.23315",
+    "title": "Test-Driven, AI-Assisted Learning: Replacing Lectures with Weekly Closed-Book Tests",
+    "url": "https://arxiv.org/abs/2606.23315",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Jin-Guo Liu",
+      "Shang-Qi Lu",
+      "Xin-Ran Shi",
+      "Long-Li Zheng",
+      "Wei Wang"
+    ],
+    "categories": [
+      "cs.CY"
+    ],
+    "topics": [
+      "workflow-agent"
+    ],
+    "score": 5,
+    "relevance": "low",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.23315",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23315",
+    "pdf_url": "https://arxiv.org/pdf/2606.23315",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.22568",
+    "title": "SeFi-Image: A Text-to-Image Foundation Model with Semantic-First Diffusion",
+    "url": "https://arxiv.org/abs/2606.22568",
+    "published": "2026-06-21",
+    "updated": "2026-06-26",
+    "authors": [
+      "SeFi-Team"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag"
+    ],
+    "score": 5,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.22568",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22568",
+    "pdf_url": "https://arxiv.org/pdf/2606.22568",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.22226",
+    "title": "Quantifying Theoretical AI Alignment Guarantees: Receiver-Utility Bounds in Bayesian Persuasion",
+    "url": "https://arxiv.org/abs/2606.22226",
+    "published": "2026-06-20",
+    "updated": "2026-06-20",
+    "authors": [
+      "Eric Yachbes",
+      "Eva Tardos"
+    ],
+    "categories": [
+      "cs.GT",
+      "cs.AI",
+      "cs.IT"
+    ],
+    "topics": [
+      "agent-safety"
+    ],
+    "score": 5,
+    "relevance": "low",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.22226",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22226",
+    "pdf_url": "https://arxiv.org/pdf/2606.22226",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.19683",
+    "title": "Exit-and-Join Dynamics for Decentralized Coalition Formation",
+    "url": "https://arxiv.org/abs/2606.19683",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Quanyan Zhu"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA",
+      "eess.SY"
+    ],
+    "topics": [
+      "agent-evaluation"
+    ],
+    "score": 5,
+    "relevance": "low",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.19683",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19683",
+    "pdf_url": "https://arxiv.org/pdf/2606.19683",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.17790",
+    "title": "Distributed Experimental Design: Bayes-optimal Fusion of Local Designs",
+    "url": "https://arxiv.org/abs/2606.17790",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Nagananda K G",
+      "Lav R. Varshney",
+      "Pramod K. Varshney"
+    ],
+    "categories": [
+      "stat.AP",
+      "cs.IT"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning"
+    ],
+    "score": 5,
+    "relevance": "low",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.17790",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17790",
+    "pdf_url": "https://arxiv.org/pdf/2606.17790",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.17388",
+    "title": "Agent Utilities over Generalized Voronoi Regions and their Gradients",
+    "url": "https://arxiv.org/abs/2606.17388",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Andre N. Costa",
+      "Petter Ögren",
+      "Carlos H. C. Ribeiro"
+    ],
+    "categories": [
+      "cs.RO",
+      "cs.CG",
+      "eess.SY"
+    ],
+    "topics": [
+      "agent"
+    ],
+    "score": 5,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.17388",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17388",
+    "pdf_url": "https://arxiv.org/pdf/2606.17388",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.12235",
+    "title": "BenDi: An Energy-Efficient Quasi-Stochastic Systolic Architecture for Edge Bioelectronics",
+    "url": "https://arxiv.org/abs/2606.12235",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Bochen Ye",
+      "Yihan Pan",
+      "Shady Agwa",
+      "Themis Prodromakis"
+    ],
+    "categories": [
+      "cs.AR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag"
+    ],
+    "score": 5,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.12235",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12235",
+    "pdf_url": "https://arxiv.org/pdf/2606.12235",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.04131",
+    "title": "Network node immunization: improving Netshield algorithm through random rooted forests",
+    "url": "https://arxiv.org/abs/2606.04131",
+    "published": "2026-06-02",
+    "updated": "2026-06-02",
+    "authors": [
+      "Luca Avena",
+      "Alexandre Gaudillière",
+      "Irina Gurewitsch",
+      "Adoré Randriamandroso",
+      "Alessio Troiani"
+    ],
+    "categories": [
+      "cs.SI",
+      "math.PR"
+    ],
+    "topics": [
+      "agent-evaluation"
+    ],
+    "score": 5,
+    "relevance": "low",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2606.04131",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.04131",
+    "pdf_url": "https://arxiv.org/pdf/2606.04131",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2605.21792",
+    "title": "Residual Skill Optimization for Text-to-SQL Ensembles",
+    "url": "https://arxiv.org/abs/2605.21792",
+    "published": "2026-05-20",
+    "updated": "2026-05-20",
+    "authors": [
+      "Jiongli Zhu",
+      "Haoquan Guan",
+      "Parjanya Prajakta Prashant",
+      "Nikki Lijing Kuang",
+      "Seyedeh Baharan Khatami",
+      "Canwen Xu",
+      "Xiaodong Yu",
+      "Yingyu Lin",
+      "Zhewei Yao",
+      "Yuxiong He",
+      "Babak Salimi"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.DB",
+      "cs.LG"
+    ],
+    "topics": [
+      "coding-agent"
+    ],
+    "score": 5,
+    "relevance": "low",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2605.21792",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.21792",
+    "pdf_url": "https://arxiv.org/pdf/2605.21792",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2602.23397",
+    "title": "Lifecycle-Integrated Security for AI-Cloud Convergence in Cyber-Physical Infrastructure",
+    "url": "https://arxiv.org/abs/2602.23397",
+    "published": "2026-02-26",
+    "updated": "2026-02-26",
+    "authors": [
+      "S M Zia Ur Rashid",
+      "Deepa Gurung",
+      "Sonam Raj Gupta",
+      "Suman Rath"
+    ],
+    "categories": [
+      "cs.CR",
+      "eess.SY"
+    ],
+    "topics": [
+      "agent-safety"
+    ],
+    "score": 5,
+    "relevance": "low",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.23397",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.23397",
+    "pdf_url": "https://arxiv.org/pdf/2602.23397",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2607.05498",
+    "title": "Non-spherical Cows: Introducing the Asphericity Parameter as a Measure of Accretion Geometry",
+    "url": "https://arxiv.org/abs/2607.05498",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Benjamin A. Seidel",
+      "Rhea-Silvia Remus",
+      "Lucas C. Kimmig",
+      "Lucas M. Valenzuela",
+      "Klaus Dolag"
+    ],
+    "categories": [
+      "astro-ph.GA",
+      "astro-ph.CO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 4,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2607.05498",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05498",
+    "pdf_url": "https://arxiv.org/pdf/2607.05498",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2607.04222",
+    "title": "Unsupervised Features Mining via Activation Geometry",
+    "url": "https://arxiv.org/abs/2607.04222",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Amit LeVi",
+      "Elad David",
+      "Max Fomin"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "reasoning"
+    ],
+    "score": 4,
+    "relevance": "low",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.04222",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04222",
+    "pdf_url": "https://arxiv.org/pdf/2607.04222",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.03955",
+    "title": "Strategy-Proof Probabilistic Social Choice Correspondences under Conditional Expected Utility",
+    "url": "https://arxiv.org/abs/2607.03955",
+    "published": "2026-07-04",
+    "updated": "2026-07-04",
+    "authors": [
+      "Madhuparna Karmokar",
+      "Ujjwal Kumar",
+      "Soumyarup Sadhukhan"
+    ],
+    "categories": [
+      "econ.TH"
+    ],
+    "topics": [
+      "agent-evaluation"
+    ],
+    "score": 4,
+    "relevance": "low",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2607.03955",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03955",
+    "pdf_url": "https://arxiv.org/pdf/2607.03955",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.03949",
+    "title": "TESSERA v2: Scaling Pixel-wise Earth Foundation Models",
+    "url": "https://arxiv.org/abs/2607.03949",
+    "published": "2026-07-04",
+    "updated": "2026-07-04",
+    "authors": [
+      "Zhengpeng Feng",
+      "Sadiq Jaffer",
+      "Ira Shokar",
+      "Jovana Knezevic",
+      "Mark Elvers",
+      "Clement Atzberger",
+      "Robin Young",
+      "Aneesh Naik",
+      "Niall Robinson",
+      "Andrew Blake",
+      "David Coomes",
+      "Anil Madhavapeddy",
+      "Srinivasan Keshav"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag"
+    ],
+    "score": 4,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2607.03949",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03949",
+    "pdf_url": "https://arxiv.org/pdf/2607.03949",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2607.05439",
+    "title": "Design-CP: Context Parallelism for Design of Protein Nanoparticles",
+    "url": "https://arxiv.org/abs/2607.05439",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Lorenzo Tarricone",
+      "Helen E. Eisenach",
+      "Aiko Muraishi",
+      "Charlotte M. Deane"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.DC",
+      "q-bio.QM"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory"
+    ],
+    "score": 4,
+    "relevance": "low",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.05439",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05439",
+    "pdf_url": "https://arxiv.org/pdf/2607.05439",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.28801",
+    "title": "Cross-channel Specific Emitter Identification and Verification via Signal Envelope",
+    "url": "https://arxiv.org/abs/2606.28801",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Yuhao Chen",
+      "Boxiang He",
+      "Shilian Wang",
+      "Jing Lei"
+    ],
+    "categories": [
+      "eess.SP"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "reasoning"
+    ],
+    "score": 4,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.28801",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28801",
+    "pdf_url": "https://arxiv.org/pdf/2606.28801",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.22997",
+    "title": "A Greatest Common Divisor Criterion of Certain Binomial Coefficients",
+    "url": "https://arxiv.org/abs/2606.22997",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Dakai Guo",
+      "Ruichen Qiu",
+      "Yichuan Cao",
+      "Ruyong Feng",
+      "Xiao-Shan Gao"
+    ],
+    "categories": [
+      "math.NT",
+      "cs.LO"
+    ],
+    "topics": [
+      "agent"
+    ],
+    "score": 4,
+    "relevance": "low",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.22997",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22997",
+    "pdf_url": "https://arxiv.org/pdf/2606.22997",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.20208",
+    "title": "Beyond Accuracy: Measuring Logical Compliance of Predictive Models",
+    "url": "https://arxiv.org/abs/2606.20208",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Guillaume Olivier Delplanque",
+      "Pierre Genevès",
+      "Nabil Layaïda",
+      "Zephirin Faure"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.DB",
+      "cs.NE"
+    ],
+    "topics": [
+      "agent-evaluation"
+    ],
+    "score": 4,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.20208",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20208",
+    "pdf_url": "https://arxiv.org/pdf/2606.20208",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.23719",
+    "title": "A Hybrid Quantum-Classical Approach for Melt Pool Prediction in Laser Powder Bed Fusion",
+    "url": "https://arxiv.org/abs/2606.23719",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Matthew M. Sato",
+      "Kincho H. Law"
+    ],
+    "categories": [
+      "quant-ph",
+      "cs.LG"
+    ],
+    "topics": [
+      "coding-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 4,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.23719",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23719",
+    "pdf_url": "https://arxiv.org/pdf/2606.23719",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.17860",
+    "title": "An Epistemic Analysis of Random Coordinated Attack",
+    "url": "https://arxiv.org/abs/2606.17860",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Sophia Knight",
+      "David Lehnherr",
+      "Sergio Rajsbaum"
+    ],
+    "categories": [
+      "cs.DC",
+      "cs.LO"
+    ],
+    "topics": [
+      "computer-use",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 4,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.17860",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17860",
+    "pdf_url": "https://arxiv.org/pdf/2606.17860",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.13225",
+    "title": "The QR Factorization for Banded-Plus-Semiseparable Matrices Is Computable in Linear Complexity",
+    "url": "https://arxiv.org/abs/2606.13225",
+    "published": "2026-06-11",
+    "updated": "2026-06-12",
+    "authors": [
+      "Tao Chen",
+      "Sheehan Olver"
+    ],
+    "categories": [
+      "math.NA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning"
+    ],
+    "score": 4,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.13225",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13225",
+    "pdf_url": "https://arxiv.org/pdf/2606.13225",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.07463",
+    "title": "Amortized Neural Optimization for Pre-Layout Signal Integrity Design Space Exploration using Differentiable Surrogates",
+    "url": "https://arxiv.org/abs/2606.07463",
+    "published": "2026-06-05",
+    "updated": "2026-06-05",
+    "authors": [
+      "Julian Withöft",
+      "Werner John",
+      "Emre Ecik",
+      "Ralf Brüning",
+      "Jürgen Götze"
+    ],
+    "categories": [
+      "eess.SP",
+      "cs.CE",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 4,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.07463",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.07463",
+    "pdf_url": "https://arxiv.org/pdf/2606.07463",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.31582",
+    "title": "Generalized Laura-Andoyer equations and the enumeration of some symmetrical classes of Dziobek configurations",
+    "url": "https://arxiv.org/abs/2606.31582",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Thiago Dias",
+      "Ya-Lun Tsai"
+    ],
+    "categories": [
+      "math.DS"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 3,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.31582",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31582",
+    "pdf_url": "https://arxiv.org/pdf/2606.31582",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.30536",
+    "title": "Evaluating the Fourier Approximation in Pulsar Timing Array Analysis",
+    "url": "https://arxiv.org/abs/2606.30536",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Yongqi Zhang",
+      "Hayden Scholz",
+      "Ken D. Olum",
+      "Lucas Steinberger",
+      "Gabriella Agazie",
+      "Akash Anumarlapudi",
+      "Anne M. Archibald",
+      "Zaven Arzoumanian",
+      "Paul T. Baker",
+      "Paul R. Brook",
+      "H. Thankful Cromartie",
+      "Kathryn Crowter",
+      "Megan E. DeCesar",
+      "Paul B. Demorest",
+      "Timothy Dolch",
+      "Justin A. Ellis",
+      "Elizabeth C. Ferrara",
+      "William Fiore",
+      "Emmanuel Fonseca",
+      "Gabriel E. Freedman",
+      "Nate Garver-Daniels",
+      "Peter A. Gentile",
+      "Joseph Glaser",
+      "Deborah C. Good",
+      "Jeffrey S. Hazboun",
+      "Ross J. Jennings",
+      "Megan L. Jones",
+      "David L. Kaplan",
+      "Matthew Kerr",
+      "Michael T. Lam",
+      "Duncan R. Lorimer",
+      "Jing Luo",
+      "Ryan S. Lynch",
+      "Alexander McEwen",
+      "Maura A. McLaughlin",
+      "Natasha McMann",
+      "Bradley W. Meyers",
+      "Cherry Ng",
+      "David J. Nice",
+      "Timothy T. Pennucci",
+      "Benetge B. P. Perera",
+      "Nihan S. Pol",
+      "Henri A. Radovan",
+      "Scott M. Ransom",
+      "Paul S. Ray",
+      "Ann Schmiedekamp",
+      "Carl Schmiedekamp",
+      "Brent J. Shapiro-Albert",
+      "Ingrid H. Stairs",
+      "Kevin Stovall",
+      "Abhimanyu Susobhanan",
+      "Joseph K. Swiggum",
+      "Stephen R. Taylor",
+      "Michele Vallisneri",
+      "Rutger van Haasteren",
+      "Haley M. Wahl"
+    ],
+    "categories": [
+      "gr-qc",
+      "astro-ph.HE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag"
+    ],
+    "score": 3,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.30536",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30536",
+    "pdf_url": "https://arxiv.org/pdf/2606.30536",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.23478",
+    "title": "ffortissimo: A Freeform Forward-Modeling Pipeline for High-Contrast Images of Circumstellar Disks Based on Automatic Differentiation",
+    "url": "https://arxiv.org/abs/2606.23478",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Jay K. Kueny",
+      "Joseph D. Long",
+      "Jared R. Males",
+      "Alycia J. Weinberger",
+      "Laird M. Close",
+      "Joshua Liberman",
+      "Sebastiaan Haffert",
+      "Eden McEwen",
+      "Maggie Y. Kautz",
+      "Olivier Guyon",
+      "Logan Pearce",
+      "Parker T. Johnson",
+      "Katie Twitchell",
+      "Jialin Li",
+      "Alex Hedglen",
+      "Avalon Gower",
+      "Warren Foster",
+      "Jhen Lumbres",
+      "Lauren Schatz"
+    ],
+    "categories": [
+      "astro-ph.IM",
+      "astro-ph.SR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 3,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.23478",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23478",
+    "pdf_url": "https://arxiv.org/pdf/2606.23478",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.20257",
+    "title": "Measurements of charged-particle pseudorapidity and transverse momentum distributions in O+O and Ne+Ne collisions at $\\sqrt{s_{_\\text{NN}}} = 5.36$ TeV with the ATLAS detector",
+    "url": "https://arxiv.org/abs/2606.20257",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "ATLAS Collaboration"
+    ],
+    "categories": [
+      "nucl-ex",
+      "hep-ex"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 3,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.20257",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20257",
+    "pdf_url": "https://arxiv.org/pdf/2606.20257",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.18975",
+    "title": "On the robustness of the angular homogeneity scale $θ_H$: a comparative analysis of computational approaches",
+    "url": "https://arxiv.org/abs/2606.18975",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Pedro Fanha",
+      "António da Silva",
+      "José Fonseca",
+      "José Pedro Mimoso",
+      "Ziad Sakr"
+    ],
+    "categories": [
+      "astro-ph.CO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "world-model"
+    ],
+    "score": 3,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.18975",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18975",
+    "pdf_url": "https://arxiv.org/pdf/2606.18975",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.15776",
+    "title": "Spin-dependent electron transfer through a ring-wire coupled junction: Role of in-plane electric field",
+    "url": "https://arxiv.org/abs/2606.15776",
+    "published": "2026-06-14",
+    "updated": "2026-06-14",
+    "authors": [
+      "Prabhab Patra",
+      "Santanu K. Maiti"
+    ],
+    "categories": [
+      "cond-mat.mes-hall"
+    ],
+    "topics": [
+      "computer-use",
+      "planning"
+    ],
+    "score": 3,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.15776",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15776",
+    "pdf_url": "https://arxiv.org/pdf/2606.15776",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.15336",
+    "title": "Nonlocal Orbital-Free Kinetic Energy Functional from the Jellium-with-Gap Model for Finite Systems",
+    "url": "https://arxiv.org/abs/2606.15336",
+    "published": "2026-06-13",
+    "updated": "2026-06-13",
+    "authors": [
+      "Abhishek Bhattacharjee",
+      "Subrata Jana",
+      "Szymon Smiga",
+      "Prasanjit Samal"
+    ],
+    "categories": [
+      "cond-mat.mtrl-sci"
+    ],
+    "topics": [
+      "agent-evaluation"
+    ],
+    "score": 3,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.15336",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15336",
+    "pdf_url": "https://arxiv.org/pdf/2606.15336",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.14131",
+    "title": "G-computation for causal effect estimation from observational hierarchical data with unmeasured cluster context",
+    "url": "https://arxiv.org/abs/2606.14131",
+    "published": "2026-06-12",
+    "updated": "2026-06-12",
+    "authors": [
+      "Shafayet Khan Shafee",
+      "Bishal Sarker",
+      "Md. Niamul Islam Sium"
+    ],
+    "categories": [
+      "stat.ME"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "world-model"
+    ],
+    "score": 3,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.14131",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.14131",
+    "pdf_url": "https://arxiv.org/pdf/2606.14131",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.13980",
+    "title": "On Cutting Cakes and Crossing Curves",
+    "url": "https://arxiv.org/abs/2606.13980",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Alexandros Hollender",
+      "Gilbert Maystre",
+      "Kilian Risse"
+    ],
+    "categories": [
+      "cs.GT",
+      "cs.CC"
+    ],
+    "topics": [
+      "agent"
+    ],
+    "score": 3,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.13980",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13980",
+    "pdf_url": "https://arxiv.org/pdf/2606.13980",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.08070",
+    "title": "Earth-Density Stratification and Quantum Gravity Corrections in Long-Baseline Neutrino Oscillation Experiments",
+    "url": "https://arxiv.org/abs/2606.08070",
+    "published": "2026-06-06",
+    "updated": "2026-06-06",
+    "authors": [
+      "Bipin Singh Koranga",
+      "Vivek Kumar Nautiya"
+    ],
+    "categories": [
+      "hep-ph"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning"
+    ],
+    "score": 3,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.08070",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.08070",
+    "pdf_url": "https://arxiv.org/pdf/2606.08070",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2607.01374",
+    "title": "Algebraic conditions for second-moment stability boundaries of linear, time-invariant stochastic delay-differential equations",
+    "url": "https://arxiv.org/abs/2607.01374",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Zsolt Iklodi",
+      "Harry Dankowicz"
+    ],
+    "categories": [
+      "math.DS"
+    ],
+    "topics": [
+      "world-model"
+    ],
+    "score": 2,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2607.01374",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01374",
+    "pdf_url": "https://arxiv.org/pdf/2607.01374",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.30540",
+    "title": "Synthesizability and Mechanical Properties of High-Entropy Borides: First-Principles and Machine Learning Studies",
+    "url": "https://arxiv.org/abs/2606.30540",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Luke Moore",
+      "Ethan Fox",
+      "Bria Storr",
+      "Jayden R. Palomino",
+      "Shane A. Catledge",
+      "Yogesh K. Vohra",
+      "Cheng-Chien Chen"
+    ],
+    "categories": [
+      "cond-mat.mtrl-sci"
+    ],
+    "topics": [
+      "agent-evaluation"
+    ],
+    "score": 2,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.30540",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30540",
+    "pdf_url": "https://arxiv.org/pdf/2606.30540",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.29989",
+    "title": "Rendering Coherent Scattering via Quantum Collision Models",
+    "url": "https://arxiv.org/abs/2606.29989",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "João S. Ferreira",
+      "Spencer S. Topel",
+      "Pierre Fromholz",
+      "James R. Wootton"
+    ],
+    "categories": [
+      "cs.GR",
+      "physics.pop-ph",
+      "quant-ph"
+    ],
+    "topics": [
+      "tool-use"
+    ],
+    "score": 2,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.29989",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29989",
+    "pdf_url": "https://arxiv.org/pdf/2606.29989",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.28146",
+    "title": "A statistically robust framework for detecting and classifying hysteresis patterns in astrophysical spectral evolution",
+    "url": "https://arxiv.org/abs/2606.28146",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Tomislav Terzić"
+    ],
+    "categories": [
+      "astro-ph.HE",
+      "astro-ph.IM"
+    ],
+    "topics": [
+      "planning"
+    ],
+    "score": 2,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.28146",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28146",
+    "pdf_url": "https://arxiv.org/pdf/2606.28146",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.20975",
+    "title": "Solving Einstein Field Equations on a Digital Quantum Computer",
+    "url": "https://arxiv.org/abs/2606.20975",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Clelia Altomonte",
+      "Malcolm Fairbairn"
+    ],
+    "categories": [
+      "gr-qc"
+    ],
+    "topics": [
+      "world-model"
+    ],
+    "score": 2,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.20975",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20975",
+    "pdf_url": "https://arxiv.org/pdf/2606.20975",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.20796",
+    "title": "$N=1$ Supersymmetry, Weil-Petersson Volume Recursion, and a Spectral Curve",
+    "url": "https://arxiv.org/abs/2606.20796",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Clifford V. Johnson"
+    ],
+    "categories": [
+      "hep-th",
+      "math-ph"
+    ],
+    "topics": [
+      "agent-evaluation"
+    ],
+    "score": 2,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.20796",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20796",
+    "pdf_url": "https://arxiv.org/pdf/2606.20796",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.18196",
+    "title": "Receiver-Aware Analysis and Verification of the Spectral Separation Coefficient Under Interference-Induced Degradation",
+    "url": "https://arxiv.org/abs/2606.18196",
+    "published": "2026-06-16",
+    "updated": "2026-06-18",
+    "authors": [
+      "Lucas Heublein",
+      "Fabian Benschuh",
+      "Alexander Rügamer",
+      "Felix Ott"
+    ],
+    "categories": [
+      "eess.SP"
+    ],
+    "topics": [
+      "agent-evaluation"
+    ],
+    "score": 2,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.18196",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18196",
+    "pdf_url": "https://arxiv.org/pdf/2606.18196",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.12698",
+    "title": "Higher Dimensional Loop Quantum Black hole in de Sitter Spacetime: Quasinormal Modes and Shadow Signatures",
+    "url": "https://arxiv.org/abs/2606.12698",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Kourosh Nozari",
+      "Sara Saghafi",
+      "Ali Mohammadpour"
+    ],
+    "categories": [
+      "gr-qc",
+      "hep-ph",
+      "hep-th"
+    ],
+    "topics": [
+      "tool-use"
+    ],
+    "score": 2,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.12698",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12698",
+    "pdf_url": "https://arxiv.org/pdf/2606.12698",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.10341",
+    "title": "Global polarization of $Λ$, $Ξ^{-}$, and $Ω^{-}$ hyperons in Au+Au collisions at RHIC BES-II energies",
+    "url": "https://arxiv.org/abs/2606.10341",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Gen-Hui Li",
+      "Cong Yi",
+      "Xiang-Yu Wu",
+      "Shi Pu",
+      "Guang-You Qin"
+    ],
+    "categories": [
+      "nucl-th"
+    ],
+    "topics": [
+      "tool-use"
+    ],
+    "score": 2,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.10341",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10341",
+    "pdf_url": "https://arxiv.org/pdf/2606.10341",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2607.02160",
+    "title": "Algorithms for hyperelliptic Mumford Curves $p$-adic Uniformization, $p$-adic integrals and $p$-adic heights",
+    "url": "https://arxiv.org/abs/2607.02160",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Enis Kaya",
+      "Marc Masdeu",
+      "J. Steffen Müller",
+      "Marius van der Put"
+    ],
+    "categories": [
+      "math.NT",
+      "math.AG"
+    ],
+    "topics": [],
+    "score": 1,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2607.02160",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02160",
+    "pdf_url": "https://arxiv.org/pdf/2607.02160",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2607.01650",
+    "title": "Computed emissivity of carbon dioxide, water vapor, and their mixtures for a wide range of temperatures and pressure-pathlengths",
+    "url": "https://arxiv.org/abs/2607.01650",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Osama A. Marzouk"
+    ],
+    "categories": [
+      "physics.gen-ph"
+    ],
+    "topics": [],
+    "score": 1,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2607.01650",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01650",
+    "pdf_url": "https://arxiv.org/pdf/2607.01650",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2607.00476",
+    "title": "Complexity of Low-Degree Skew Polynomial Multiplication over Finite Fields",
+    "url": "https://arxiv.org/abs/2607.00476",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Ke Ye",
+      "Yichuan Cao",
+      "Ruichen Qiu"
+    ],
+    "categories": [
+      "cs.SC"
+    ],
+    "topics": [],
+    "score": 1,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2607.00476",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00476",
+    "pdf_url": "https://arxiv.org/pdf/2607.00476",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.29294",
+    "title": "Quantum models of the Riemann zeta function, lattice spin models and algebraic models of entanglement",
+    "url": "https://arxiv.org/abs/2606.29294",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Nikolaj M. Glazunov"
+    ],
+    "categories": [
+      "math.NT"
+    ],
+    "topics": [],
+    "score": 1,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.29294",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29294",
+    "pdf_url": "https://arxiv.org/pdf/2606.29294",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.27415",
+    "title": "Elimination of Flux Trapping in Superconducting Circuits in Ambient Magnetic Fields",
+    "url": "https://arxiv.org/abs/2606.27415",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Rohan T. Kapur",
+      "Alex Wynn",
+      "Sergey K. Tolpygo",
+      "Neel Parmar",
+      "Anil Mankame",
+      "Adam A. Libson",
+      "Rabindra Das",
+      "Michele Kelley",
+      "Pauli Kehayias",
+      "Nathaniel J. O'Connor",
+      "Collin N. Muniz",
+      "Justin L. Mallek",
+      "Jennifer M. Schloss"
+    ],
+    "categories": [
+      "cond-mat.supr-con",
+      "cond-mat.mes-hall",
+      "physics.app-ph",
+      "quant-ph"
+    ],
+    "topics": [],
+    "score": 1,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.27415",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27415",
+    "pdf_url": "https://arxiv.org/pdf/2606.27415",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.26255",
+    "title": "Hochschild (co)homology and cyclic homology via a graded Euler characteristic with applications to higher preprojective algebras",
+    "url": "https://arxiv.org/abs/2606.26255",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Jon Wallem Anundsen",
+      "Mads Hustad Sandøy"
+    ],
+    "categories": [
+      "math.RT"
+    ],
+    "topics": [],
+    "score": 1,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.26255",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26255",
+    "pdf_url": "https://arxiv.org/pdf/2606.26255",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.21829",
+    "title": "Elementary solutions of ordinary tropical differential equations, and vanishing orders of solutions of algebraic differential equations",
+    "url": "https://arxiv.org/abs/2606.21829",
+    "published": "2026-06-20",
+    "updated": "2026-06-20",
+    "authors": [
+      "Cristhian Garay-López",
+      "Johana Luviano-Flores",
+      "Carla Valencia-Negrete"
+    ],
+    "categories": [
+      "math.AG",
+      "math.CO"
+    ],
+    "topics": [],
+    "score": 1,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.21829",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21829",
+    "pdf_url": "https://arxiv.org/pdf/2606.21829",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.19503",
+    "title": "Kernel transformations and bounds for smeared spectral functions",
+    "url": "https://arxiv.org/abs/2606.19503",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "William I. Jay",
+      "Matteo Saccardi"
+    ],
+    "categories": [
+      "hep-lat"
+    ],
+    "topics": [],
+    "score": 1,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.19503",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19503",
+    "pdf_url": "https://arxiv.org/pdf/2606.19503",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.16510",
+    "title": "Petrov-Galerkin Variational Physics-Informed Neural Network Framework for Two-Dimensional Singularly Perturbed Problems",
+    "url": "https://arxiv.org/abs/2606.16510",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Vijay Kumar",
+      "Gautam Singh"
+    ],
+    "categories": [
+      "math.NA",
+      "cs.LG"
+    ],
+    "topics": [],
+    "score": 1,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.16510",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16510",
+    "pdf_url": "https://arxiv.org/pdf/2606.16510",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.12998",
+    "title": "Efficient emulation of nuclear ground states with neural-network variational Monte Carlo and eigenvector continuation",
+    "url": "https://arxiv.org/abs/2606.12998",
+    "published": "2026-06-11",
+    "updated": "2026-06-12",
+    "authors": [
+      "Mao Li",
+      "Yilong Yang",
+      "Pengwei Zhao"
+    ],
+    "categories": [
+      "nucl-th"
+    ],
+    "topics": [],
+    "score": 1,
+    "relevance": "low",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.12998",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12998",
+    "pdf_url": "https://arxiv.org/pdf/2606.12998",
+    "primary_query": "web-gui-agent"
+  }
+]
diff --git a/data/arxiv-agent-papers-2025-07-08-to-2026-07-08.json b/data/arxiv-agent-papers-2025-07-08-to-2026-07-08.json
new file mode 100644
index 0000000..3a51822
--- /dev/null
+++ b/data/arxiv-agent-papers-2025-07-08-to-2026-07-08.json
@@ -0,0 +1,35331 @@
+[
+  {
+    "id": "2606.08340",
+    "title": "Benchmarking Open-Ended Multi-Agent Coordination in Language Agents",
+    "url": "https://arxiv.org/abs/2606.08340",
+    "published": "2026-06-06",
+    "updated": "2026-06-06",
+    "authors": [
+      "Kale-ab Abebe Tessera",
+      "Andras Szecsenyi",
+      "Cameron Barker",
+      "Alexander Rutherford",
+      "Davide Paglieri",
+      "Aidan Scannell",
+      "Henry Gouk",
+      "Elliot J. Crowley",
+      "Tim Rocktäschel",
+      "Amos Storkey"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 28,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm",
+      "language-agent",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.08340",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.08340",
+    "pdf_url": "https://arxiv.org/pdf/2606.08340",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.20833",
+    "title": "MemGym: a Long-Horizon Memory Environment for LLM Agents",
+    "url": "https://arxiv.org/abs/2605.20833",
+    "published": "2026-05-20",
+    "updated": "2026-05-20",
+    "authors": [
+      "Wujiang Xu",
+      "Yu Wang",
+      "Kai Mei",
+      "Kaiqu Liang",
+      "Zhenting Wang",
+      "Mingyu Jin",
+      "Han Zhang",
+      "Shi-Xiong Zhang",
+      "Wenyue Hua",
+      "Sambit Sahu",
+      "Dimitris N. Metaxas"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 26,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.20833",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.20833",
+    "pdf_url": "https://arxiv.org/pdf/2605.20833",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2607.06008",
+    "title": "PolyWorkBench: Benchmarking Multilingual Long-Horizon LLM Agents",
+    "url": "https://arxiv.org/abs/2607.06008",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Hongliang Li",
+      "Yijin Liu",
+      "Zhiwei Zhang",
+      "Zihe Liu",
+      "Xinyue Lou",
+      "Jinan Xu",
+      "Fandong Meng",
+      "Kaiyu Huang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 25,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "llm-agent",
+      "planning-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.06008",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06008",
+    "pdf_url": "https://arxiv.org/pdf/2607.06008",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.28425",
+    "title": "Tool Use Enables Undetectable Steganography in Multi-Agent LLM Systems",
+    "url": "https://arxiv.org/abs/2606.28425",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Jimmy Laurence Rippin",
+      "Simon C. Marshall",
+      "David Demitri Africa",
+      "Christian Schroeder de Witt"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 25,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "ai-agent",
+      "autonomous-agent-llm",
+      "multi-agent-llm",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.28425",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28425",
+    "pdf_url": "https://arxiv.org/pdf/2606.28425",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.24937",
+    "title": "The Hitchhiker's Guide to Agentic AI: From Foundations to Systems",
+    "url": "https://arxiv.org/abs/2606.24937",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Haggai Roitman"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.IR",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "memory",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 25,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "multi-agent-llm",
+      "rag-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.24937",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24937",
+    "pdf_url": "https://arxiv.org/pdf/2606.24937",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.10749",
+    "title": "Toward Secure LLM Agents: Threat Surfaces, Attacks, Defenses, and Evaluation",
+    "url": "https://arxiv.org/abs/2606.10749",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Yuchen Ling",
+      "Shengcheng Yu",
+      "Zhenyu Chen",
+      "Chunrong Fang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 25,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.10749",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10749",
+    "pdf_url": "https://arxiv.org/pdf/2606.10749",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.29824",
+    "title": "Neural Procedural Memory: Empowering LLM Agents with Implicit Activation Steering",
+    "url": "https://arxiv.org/abs/2606.29824",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Chengfeng Zhao",
+      "Yuqiao Tan",
+      "Shizhu He",
+      "Yequan Wang",
+      "Jun Zhao",
+      "Kang Liu"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 24,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "agent-memory",
+      "autonomous-agent-llm",
+      "llm-agent",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.29824",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29824",
+    "pdf_url": "https://arxiv.org/pdf/2606.29824",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.28011",
+    "title": "From Detection to Action: Using LLM Agents for Fault-Tolerant Control",
+    "url": "https://arxiv.org/abs/2606.28011",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Javal Vyas",
+      "Milapji Singh Gill",
+      "Artan Markaj",
+      "Felix Gehlhoff",
+      "Mehmet Mercangöz"
+    ],
+    "categories": [
+      "eess.SY",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 24,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "llm-agent",
+      "multi-agent-llm",
+      "planning-agent",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.28011",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28011",
+    "pdf_url": "https://arxiv.org/pdf/2606.28011",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.20401",
+    "title": "PowerAgentBench-Dyn: A Benchmark for Agentic AI in Power System Dynamic Studies",
+    "url": "https://arxiv.org/abs/2606.20401",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Qian Zhang",
+      "Andrea Pomarico",
+      "Costas Mylonas",
+      "Magda Foti",
+      "Alberto Berizzi",
+      "Le Xie"
+    ],
+    "categories": [
+      "eess.SY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 24,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.20401",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20401",
+    "pdf_url": "https://arxiv.org/pdf/2606.20401",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.18789",
+    "title": "PowerAgentBench-SS: A Benchmark for Agentic AI in Power System Steady-State Studies",
+    "url": "https://arxiv.org/abs/2606.18789",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Costas Mylonas",
+      "Magda Foti",
+      "Andrea Pomarico",
+      "Matheus Duarte",
+      "Qian Zhang",
+      "Emmanouel Varvarigos"
+    ],
+    "categories": [
+      "eess.SY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 24,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "planning-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.18789",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18789",
+    "pdf_url": "https://arxiv.org/pdf/2606.18789",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.08274",
+    "title": "Toward Human-Centered Multi-Agent Systems: Integrating Cognition, Culture, Values, and Cooperation in AI Agents",
+    "url": "https://arxiv.org/abs/2606.08274",
+    "published": "2026-06-06",
+    "updated": "2026-06-06",
+    "authors": [
+      "Safia Baloch",
+      "Rahemeen Khan"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 24,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.08274",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.08274",
+    "pdf_url": "https://arxiv.org/pdf/2606.08274",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2607.05775",
+    "title": "Beyond the Leaderboard: A Synthesis of Tool-Use, Planning, and Reasoning Failures in Large Language Model Agents",
+    "url": "https://arxiv.org/abs/2607.05775",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Wael Albayaydh",
+      "Rui Zhao",
+      "Ivan Flechais"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "embodied-agent",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 23,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm",
+      "planning-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.05775",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05775",
+    "pdf_url": "https://arxiv.org/pdf/2607.05775",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.02255",
+    "title": "AgenticSTS: A Bounded-Memory Testbed for Long-Horizon LLM Agents",
+    "url": "https://arxiv.org/abs/2607.02255",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Xiangchen Cheng",
+      "Yunwei Jiang",
+      "Jianwen Sun",
+      "Zizhen Li",
+      "Chuanhao Li",
+      "Xiangcheng Cao",
+      "Yihao Liu",
+      "Fanrui Zhang",
+      "Li Jin",
+      "Kaipeng Zhang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 23,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.02255",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02255",
+    "pdf_url": "https://arxiv.org/pdf/2607.02255",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.28791",
+    "title": "From Determinism to Delegation: AI-Native Software Engineering and the Evolution of the Agentic Engineer",
+    "url": "https://arxiv.org/abs/2606.28791",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Mamdouh Alenezi"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "memory",
+      "multi-agent",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 23,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "autonomous-agent-llm",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.28791",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28791",
+    "pdf_url": "https://arxiv.org/pdf/2606.28791",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.26614",
+    "title": "HiLSVA: Design and Evaluation of a Human-in-the-Loop Agentic System for Scientific Visualization",
+    "url": "https://arxiv.org/abs/2606.26614",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Kuangshi Ai",
+      "Patrick Phuoc Do",
+      "Chaoli Wang"
+    ],
+    "categories": [
+      "cs.HC",
+      "cs.AI",
+      "cs.GR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 23,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.26614",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26614",
+    "pdf_url": "https://arxiv.org/pdf/2606.26614",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2605.08442",
+    "title": "Defense effectiveness across architectural layers: a mechanistic evaluation of persistent memory attacks on stateful LLM agents",
+    "url": "https://arxiv.org/abs/2605.08442",
+    "published": "2026-05-08",
+    "updated": "2026-07-03",
+    "authors": [
+      "Jun Wen Leong"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 23,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.08442",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.08442",
+    "pdf_url": "https://arxiv.org/pdf/2605.08442",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.06869",
+    "title": "Agentick: A Unified Benchmark for General Sequential Decision-Making Agents",
+    "url": "https://arxiv.org/abs/2605.06869",
+    "published": "2026-05-07",
+    "updated": "2026-05-12",
+    "authors": [
+      "Roger Creus Castanyer",
+      "Pablo Samuel Castro",
+      "Glen Berseth"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 23,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.06869",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.06869",
+    "pdf_url": "https://arxiv.org/pdf/2605.06869",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2607.03233",
+    "title": "Agentic and Generative AI for Open-Source Intelligence and Cyber Investigations: Taxonomy, Evaluation, Challenges, and Future Directions",
+    "url": "https://arxiv.org/abs/2607.03233",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Eduardo Almeida Palmieri",
+      "Mohamed Chahine Ghanem",
+      "Dipo Dunsin",
+      "Zubair Baig",
+      "Ed de Quincey",
+      "Kim-Kwang Raymond Choo"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.IR",
+      "cs.SI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 22,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "rag-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.03233",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03233",
+    "pdf_url": "https://arxiv.org/pdf/2607.03233",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.28061",
+    "title": "ToolPrivacyBench: Benchmarking Purpose-Bound Privacy in Tool-Using LLM Agents",
+    "url": "https://arxiv.org/abs/2606.28061",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Shijing Hu",
+      "Liang Liu",
+      "Zhu Meng",
+      "Zhicheng Zhao"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 22,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "function-calling",
+      "llm-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.28061",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28061",
+    "pdf_url": "https://arxiv.org/pdf/2606.28061",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.17459",
+    "title": "Can LLMs Be CEOs? Benchmarking Strategic Resource Reallocation with Multi-Role Agent Simulation",
+    "url": "https://arxiv.org/abs/2606.17459",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Yuyang Dai",
+      "Xueqing Peng",
+      "Lingfei Qian",
+      "Zhuohan Xie"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 22,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.17459",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17459",
+    "pdf_url": "https://arxiv.org/pdf/2606.17459",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.16613",
+    "title": "CoffeeBench: Benchmarking Long-Horizon LLM Agents in Heterogeneous Multi-Agent Economies",
+    "url": "https://arxiv.org/abs/2606.16613",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Issa Sugiura",
+      "Daichi Hattori",
+      "Kazuo Araragi",
+      "Keita Ogawa",
+      "Shota Onose",
+      "Taro Makino",
+      "Teppei Usuki",
+      "Takashi Ishida"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 22,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.16613",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16613",
+    "pdf_url": "https://arxiv.org/pdf/2606.16613",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.12945",
+    "title": "Learning What to Remember: A Cognitively Grounded Multi-Factor Value Model for Agentic Memory",
+    "url": "https://arxiv.org/abs/2606.12945",
+    "published": "2026-06-11",
+    "updated": "2026-06-20",
+    "authors": [
+      "Zhibao Chen",
+      "Qian Cheng"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 22,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.12945",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12945",
+    "pdf_url": "https://arxiv.org/pdf/2606.12945",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.06399",
+    "title": "CollabSim: A CSCW-Grounded Methodology for Investigating Collaborative Competence of LLM Agents through Controlled Multi-Agent Experiments",
+    "url": "https://arxiv.org/abs/2606.06399",
+    "published": "2026-06-04",
+    "updated": "2026-06-06",
+    "authors": [
+      "Jiaju Chen",
+      "Bo Sun",
+      "Yuxuan Lu",
+      "Yun Wang",
+      "Dakuo Wang",
+      "Bingsheng Yao"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 22,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.06399",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.06399",
+    "pdf_url": "https://arxiv.org/pdf/2606.06399",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.01199",
+    "title": "Can LLM Agents Sustain Long-Horizon Organizational Dynamics?",
+    "url": "https://arxiv.org/abs/2606.01199",
+    "published": "2026-05-31",
+    "updated": "2026-05-31",
+    "authors": [
+      "Xuancheng Zhu",
+      "Yang Yue",
+      "Shuaibing Wan",
+      "Zihan Dou",
+      "Xiaohan Zhang",
+      "Yongrui Liu",
+      "Guoshun Nan"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 22,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.01199",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.01199",
+    "pdf_url": "https://arxiv.org/pdf/2606.01199",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.29861",
+    "title": "Towards Verifiable Multimodal Deep Research: A Multi-Agent Harness for Interleaved Report Generation",
+    "url": "https://arxiv.org/abs/2605.29861",
+    "published": "2026-05-28",
+    "updated": "2026-06-03",
+    "authors": [
+      "Chenghao Zhang",
+      "Guanting Dong",
+      "Yufan Liu",
+      "Tong Zhao",
+      "Xiaoxi Li",
+      "Zhicheng Dou"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 22,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.29861",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.29861",
+    "pdf_url": "https://arxiv.org/pdf/2605.29861",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.18652",
+    "title": "MementoGUI: Learning Agentic Multimodal Memory Control for Long-Horizon GUI Agents",
+    "url": "https://arxiv.org/abs/2605.18652",
+    "published": "2026-05-18",
+    "updated": "2026-05-18",
+    "authors": [
+      "Ziyun Zeng",
+      "Hang Hua",
+      "Bocheng Zou",
+      "Mu Cai",
+      "Rogerio Feris",
+      "Jiebo Luo"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 22,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.18652",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.18652",
+    "pdf_url": "https://arxiv.org/pdf/2605.18652",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.05704",
+    "title": "SafeHarbor: Hierarchical Memory-Augmented Guardrail for LLM Agent Safety",
+    "url": "https://arxiv.org/abs/2605.05704",
+    "published": "2026-05-07",
+    "updated": "2026-05-22",
+    "authors": [
+      "Zhe Liu",
+      "Zonghao Ying",
+      "Wenxin Zhang",
+      "Quanchen Zou",
+      "Deyue Zhang",
+      "Dongdong Yang",
+      "Xiangzheng Zhang",
+      "Hao Peng"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "memory",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 22,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.05704",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.05704",
+    "pdf_url": "https://arxiv.org/pdf/2605.05704",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.03242",
+    "title": "Enhancing Agent Safety Judgment: Controlled Benchmark Rewriting and Analogical Reasoning for Deceptive Out-of-Distribution Scenarios",
+    "url": "https://arxiv.org/abs/2605.03242",
+    "published": "2026-05-05",
+    "updated": "2026-05-05",
+    "authors": [
+      "Zuoyu Zhang",
+      "Yancheng Zhu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 22,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.03242",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.03242",
+    "pdf_url": "https://arxiv.org/pdf/2605.03242",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2607.06118",
+    "title": "WebRetriever: A Large-Scale Comprehensive Benchmark for Efficient Web Agent Evaluation",
+    "url": "https://arxiv.org/abs/2607.06118",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Wei Dong",
+      "Tianyu Fu",
+      "Zhe Yu",
+      "Hanning Wang",
+      "Anyang Su",
+      "Zhizhou Fang",
+      "Yuyang Chen",
+      "Shuo Wang",
+      "Minghui Wu",
+      "Ping Jiang",
+      "Zhen Lei",
+      "Chenxu Zhao"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.MM"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "embodied-agent",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2607.06118",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06118",
+    "pdf_url": "https://arxiv.org/pdf/2607.06118",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.05773",
+    "title": "Beyond Static Evaluation: Building Simulation Environments for Scalable Agentic Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2607.05773",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Akshay Arora",
+      "Ishan Nigam",
+      "Ashutosh Aggarwal",
+      "Shefali Bansal",
+      "Krishna Singh",
+      "Sweta Kumari",
+      "Nikhil Mittal",
+      "Shariq Farhan",
+      "Siddarth Malreddy"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm",
+      "tool-use",
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2607.05773",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05773",
+    "pdf_url": "https://arxiv.org/pdf/2607.05773",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2607.05456",
+    "title": "Prompt-to-Paper: Agentic AI System for Bioinformatics",
+    "url": "https://arxiv.org/abs/2607.05456",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Ramsha Kamran",
+      "Maheera Amjad",
+      "Zartasha Mustansar",
+      "Arsalan Shaukat",
+      "Salma Sherbaz",
+      "Muhammad U. S. Khan"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "q-bio.QM"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "coding-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.05456",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05456",
+    "pdf_url": "https://arxiv.org/pdf/2607.05456",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.04433",
+    "title": "Autonomous Information Seeking: A Roadmap for Agentic Recommender Systems",
+    "url": "https://arxiv.org/abs/2607.04433",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Xinyu Lin",
+      "Yashar Deldjoo",
+      "Sunhao Dai",
+      "Honghui Bao",
+      "Xiaopeng Ye",
+      "Fatemeh Nazary",
+      "Wenjie Wang",
+      "Tommaso Di Noia",
+      "Jun Xu",
+      "Tat-Seng Chua"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2607.04433",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04433",
+    "pdf_url": "https://arxiv.org/pdf/2607.04433",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2607.03601",
+    "title": "ArchEval: Measuring AI Agents as Computer Architects",
+    "url": "https://arxiv.org/abs/2607.03601",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Chenyu Wang",
+      "Zishen Wan",
+      "Jeffrey Ma",
+      "Shvetank Prakash",
+      "Zhenting Qi",
+      "Haebin Do",
+      "Andy Cheng",
+      "Arya Tschand",
+      "Jiahe Shi",
+      "Yilun Du",
+      "Vijay Janapa Reddi"
+    ],
+    "categories": [
+      "cs.AR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "llm-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.03601",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03601",
+    "pdf_url": "https://arxiv.org/pdf/2607.03601",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.02032",
+    "title": "PACE: A Proxy for Agentic Capability Evaluation",
+    "url": "https://arxiv.org/abs/2607.02032",
+    "published": "2026-07-02",
+    "updated": "2026-07-06",
+    "authors": [
+      "Yueqi Song",
+      "Lintang Sutawika",
+      "Jiarui Liu",
+      "Lindia Tjuatja",
+      "Jiayi Geng",
+      "Yunze Xiao",
+      "Daniel Lee",
+      "Aditya Bharat Soni",
+      "Vincent Lo",
+      "Xiang Yue",
+      "Graham Neubig"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "reasoning"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "coding-agent",
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.02032",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02032",
+    "pdf_url": "https://arxiv.org/pdf/2607.02032",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.01641",
+    "title": "When Agents Do Not Stop: Uncovering Infinite Agentic Loops in LLM Agents",
+    "url": "https://arxiv.org/abs/2607.01641",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Xinyi Hou",
+      "Shenao Wang",
+      "Yanjie Zhao",
+      "Haoyu Wang"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "planning-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.01641",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01641",
+    "pdf_url": "https://arxiv.org/pdf/2607.01641",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.30524",
+    "title": "The Illusion of Agentic Complexity in README.md Generation: Evaluating Single-Agent vs. Multi-Agent RAG Systems",
+    "url": "https://arxiv.org/abs/2606.30524",
+    "published": "2026-06-29",
+    "updated": "2026-07-01",
+    "authors": [
+      "Abu Saleh",
+      "Tesfay Welegebreal Tesfay",
+      "Phuong T. Nguyen",
+      "Juri Di Rocco",
+      "Muhammad Umar Zeshan",
+      "Davide Di Ruscio"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "multi-agent",
+      "planning",
+      "rag"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.30524",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30524",
+    "pdf_url": "https://arxiv.org/pdf/2606.30524",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.29537",
+    "title": "OSWorld2.0: Benchmarking Computer Use Agents on Long-Horizon Real-World Tasks",
+    "url": "https://arxiv.org/abs/2606.29537",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Mengqi Yuan",
+      "Zilong Zhou",
+      "Xinzhuang Xiong",
+      "Weiming Wu",
+      "Jiayang Sun",
+      "Jiamin Song",
+      "Kaiqian Cui",
+      "Bowen Wang",
+      "Haoyuan Wu",
+      "Yitong Li",
+      "Dunjie Lu",
+      "Haikong Lu",
+      "Qi Zhen",
+      "Xinyuan Wang",
+      "Jiaqi Deng",
+      "Yuhao Yang",
+      "Cheng Chen",
+      "Boyuan Zheng",
+      "Alex Su",
+      "Xiao Yu",
+      "Hao Zou",
+      "Saaket Agashe",
+      "Xing Han Lu",
+      "Manpreet Kaur",
+      "Zhengyang Qi",
+      "Vincent Sunn Chen",
+      "Frederic Sala",
+      "Dayiheng Liu",
+      "Junyang Lin",
+      "Zhou Yu",
+      "Yu Su",
+      "Siva Reddy",
+      "Xin Eric Wang",
+      "Peng Qi",
+      "Tianbao Xie",
+      "Tao Yu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.29537",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29537",
+    "pdf_url": "https://arxiv.org/pdf/2606.29537",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.28925",
+    "title": "Multi-Agent Routing as Set-Valued Prediction: A WildChat Benchmark and Cost-Aware Evaluation",
+    "url": "https://arxiv.org/abs/2606.28925",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Ananto Nayan Bala",
+      "Faisal Muhammad Shah"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.IR",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.28925",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28925",
+    "pdf_url": "https://arxiv.org/pdf/2606.28925",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.26511",
+    "title": "Temporal Validity in Retrieval Memory: Eliminating Stale-Fact Errors for AI Agents over Evolving Knowledge",
+    "url": "https://arxiv.org/abs/2606.26511",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Neeraj Yadav"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.ET",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.26511",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26511",
+    "pdf_url": "https://arxiv.org/pdf/2606.26511",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.22557",
+    "title": "MacAgentBench: Benchmarking AI Agents on Real-World macOS Desktop",
+    "url": "https://arxiv.org/abs/2606.22557",
+    "published": "2026-06-21",
+    "updated": "2026-06-21",
+    "authors": [
+      "Yikun Fu",
+      "Bowen Fu",
+      "Zhenyu Wu",
+      "Shuang Cheng",
+      "Xiaowei Sun",
+      "Bowen Yang",
+      "Zehao Li",
+      "Yibo Zhao",
+      "Zichen Ding",
+      "Zhoumianze Liu",
+      "Shijie Wang",
+      "Biqing Qi",
+      "Bowen Zhou"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "ai-agent",
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.22557",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22557",
+    "pdf_url": "https://arxiv.org/pdf/2606.22557",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.21877",
+    "title": "AgentRiskBOM: A Risk-Scoping Security Bill of Materials for Agentic AI Systems",
+    "url": "https://arxiv.org/abs/2606.21877",
+    "published": "2026-06-20",
+    "updated": "2026-06-20",
+    "authors": [
+      "Srimonti Dutta",
+      "Akshata Kishore Moharir"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CR",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "memory",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "ai-agent",
+      "rag-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.21877",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21877",
+    "pdf_url": "https://arxiv.org/pdf/2606.21877",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.16802",
+    "title": "LabOSBench: Benchmarking Computer Use Agents for Scientific Instrument Control",
+    "url": "https://arxiv.org/abs/2606.16802",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Anqi Zou",
+      "Han Deng",
+      "Chengyu Zhang",
+      "Junquan Hu",
+      "Yu Wang",
+      "Yuxiang Xing",
+      "Aokai Zhang",
+      "Hanling Zhang",
+      "Zhaoyang Liu",
+      "Ben Fei",
+      "Zhihui Wang",
+      "Wanli Ouyang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "planning",
+      "workflow-agent"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.16802",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16802",
+    "pdf_url": "https://arxiv.org/pdf/2606.16802",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.13994",
+    "title": "Hidden in Plain Sight: Benchmarking Agent Safety Against Decomposition Attacks with DECOMPBENCH",
+    "url": "https://arxiv.org/abs/2606.13994",
+    "published": "2026-06-12",
+    "updated": "2026-06-12",
+    "authors": [
+      "Vikhyath Kothamasu",
+      "Virginia Smith",
+      "Chhavi Yadav"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.13994",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13994",
+    "pdf_url": "https://arxiv.org/pdf/2606.13994",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.04990",
+    "title": "From Agent Traces to Trust: A Survey of Evidence Tracing and Execution Provenance in LLM Agents",
+    "url": "https://arxiv.org/abs/2606.04990",
+    "published": "2026-06-03",
+    "updated": "2026-06-28",
+    "authors": [
+      "Yiqi Wang",
+      "Jiaqi Zhang",
+      "Taotao Cai",
+      "Zirui Liu",
+      "Qingqiang Sun",
+      "Zequn Sun",
+      "Zhangkai Wu",
+      "Manqing Dong",
+      "Mingkai Zheng",
+      "Xuefei Yin",
+      "Yanming Zhu"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.04990",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.04990",
+    "pdf_url": "https://arxiv.org/pdf/2606.04990",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.02461",
+    "title": "AgentCL: Toward Rigorous Evaluation of Continual Learning in Language Agents",
+    "url": "https://arxiv.org/abs/2606.02461",
+    "published": "2026-06-01",
+    "updated": "2026-06-02",
+    "authors": [
+      "Yiheng Shu",
+      "Bernal Jiménez Gutiérrez",
+      "Saisri Padmaja Jonnalagedda",
+      "Yuguang Yao",
+      "Huan Sun",
+      "Yu Su"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.02461",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.02461",
+    "pdf_url": "https://arxiv.org/pdf/2606.02461",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.01385",
+    "title": "Bridging Requirements and Architecture: Multi-Agent Orchestration with External Knowledge and Hierarchical Memory",
+    "url": "https://arxiv.org/abs/2606.01385",
+    "published": "2026-05-31",
+    "updated": "2026-05-31",
+    "authors": [
+      "Ruiyin Li",
+      "Yiran Zhang",
+      "Xiyu Zhou",
+      "Yangxiao Cai",
+      "Peng Liang",
+      "Weisong Sun",
+      "Jifeng Xuan",
+      "Zhi Jin",
+      "Yang Liu"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "memory",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.01385",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.01385",
+    "pdf_url": "https://arxiv.org/pdf/2606.01385",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.00756",
+    "title": "CoMIC: Collaborative Memory and Insights Circulation for Long-Horizon LLM Agents in Cloud-Edge Systems",
+    "url": "https://arxiv.org/abs/2606.00756",
+    "published": "2026-05-30",
+    "updated": "2026-05-30",
+    "authors": [
+      "Yannan Wang",
+      "Longli Yang",
+      "Zhen Liu",
+      "Abhishek Kumar",
+      "Carsten Maple"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.00756",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.00756",
+    "pdf_url": "https://arxiv.org/pdf/2606.00756",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.20315",
+    "title": "Mix-Quant: Quantized Prefilling, Precise Decoding for Agentic LLMs",
+    "url": "https://arxiv.org/abs/2605.20315",
+    "published": "2026-05-19",
+    "updated": "2026-05-19",
+    "authors": [
+      "Haiquan Lu",
+      "Zigeng Chen",
+      "Gongfan Fang",
+      "Xinyin Ma",
+      "Xinchao Wang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.20315",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.20315",
+    "pdf_url": "https://arxiv.org/pdf/2605.20315",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.14498",
+    "title": "GroupMemBench: Benchmarking LLM Agent Memory in Multi-Party Conversations",
+    "url": "https://arxiv.org/abs/2605.14498",
+    "published": "2026-05-14",
+    "updated": "2026-05-16",
+    "authors": [
+      "Jingbo Yang",
+      "Kwei-Herng Lai",
+      "Xiaowen Wang",
+      "Shiyu Chang",
+      "Yaar Harari",
+      "Evgeniy Gabrilovich"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "rag",
+      "reasoning"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.14498",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.14498",
+    "pdf_url": "https://arxiv.org/pdf/2605.14498",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.11633",
+    "title": "Can LLM Agents Respond to Disasters? Benchmarking Heterogeneous Geospatial Reasoning in Emergency Operations",
+    "url": "https://arxiv.org/abs/2605.11633",
+    "published": "2026-05-12",
+    "updated": "2026-05-12",
+    "authors": [
+      "Junjue Wang",
+      "Weihao Xuan",
+      "Heli Qi",
+      "Pengyu Dai",
+      "Kunyi Liu",
+      "Hongruixuan Chen",
+      "Zhuo Zheng",
+      "Junshi Xia",
+      "Stefano Ermon",
+      "Naoto Yokoya"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.11633",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.11633",
+    "pdf_url": "https://arxiv.org/pdf/2605.11633",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.06812",
+    "title": "Towards Security-Auditable LLM Agents: A Unified Graph Representation",
+    "url": "https://arxiv.org/abs/2605.06812",
+    "published": "2026-05-07",
+    "updated": "2026-05-07",
+    "authors": [
+      "Chaofan Li",
+      "Lyuye Zhang",
+      "Jintao Zhai",
+      "Siyue Feng",
+      "Xichun Yang",
+      "Huahao Wang",
+      "Shihan Dou",
+      "Yu Ji",
+      "Yutao Hu",
+      "Yueming Wu",
+      "Yang Liu",
+      "Deqing Zou"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.06812",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.06812",
+    "pdf_url": "https://arxiv.org/pdf/2605.06812",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2602.08412",
+    "title": "From Assistant to Double Agent: Formalizing and Benchmarking Attacks on OpenClaw for Personalized Local AI Agent",
+    "url": "https://arxiv.org/abs/2602.08412",
+    "published": "2026-02-09",
+    "updated": "2026-02-11",
+    "authors": [
+      "Yuhang Wang",
+      "Feiming Xu",
+      "Zheng Lin",
+      "Guangyu He",
+      "Yuzhe Huang",
+      "Haichang Gao",
+      "Zhenxing Niu",
+      "Shiguo Lian",
+      "Zhaoxiang Liu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.08412",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.08412",
+    "pdf_url": "https://arxiv.org/pdf/2602.08412",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2508.07575",
+    "title": "MCPToolBench++: A Large Scale AI Agent Model Context Protocol MCP Tool Use Benchmark",
+    "url": "https://arxiv.org/abs/2508.07575",
+    "published": "2025-08-11",
+    "updated": "2025-08-11",
+    "authors": [
+      "Shiqing Fan",
+      "Xichen Ding",
+      "Liang Zhang",
+      "Linjian Mo"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 21,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2508.07575",
+    "source": "arxiv",
+    "source_id": "arxiv:2508.07575",
+    "pdf_url": "https://arxiv.org/pdf/2508.07575",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2607.05318",
+    "title": "PiSAs: Benchmarking Contextual Integrity in Multi-User Agentic Systems",
+    "url": "https://arxiv.org/abs/2607.05318",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Shubham Gupta",
+      "Nazanin Mohammadi Sepahvand",
+      "Abhinav Kumar",
+      "Cem Subakan",
+      "Spandana Gella",
+      "Pierre-André Noël",
+      "Perouz Taslakian",
+      "Eugene Bagdasarian",
+      "Valentina Zantedeschi"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.05318",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05318",
+    "pdf_url": "https://arxiv.org/pdf/2607.05318",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.04391",
+    "title": "Memory-Orchestrated Semantic System (MOSS): An Auditable Agentic Memory Architecture",
+    "url": "https://arxiv.org/abs/2607.04391",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Serge Lacasse",
+      "Jérémie Hatier",
+      "Alex Baker"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "ai-agent",
+      "rag-agent"
+    ],
+    "arxiv_id": "2607.04391",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04391",
+    "pdf_url": "https://arxiv.org/pdf/2607.04391",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2607.03953",
+    "title": "The Remarkable Effectiveness of Providing AI Agents with Natural Language Tools: A Replication Study Validating NLT Performance Across 14 Models",
+    "url": "https://arxiv.org/abs/2607.03953",
+    "published": "2026-07-04",
+    "updated": "2026-07-04",
+    "authors": [
+      "Alexander Somma",
+      "Isabelle Plante",
+      "Fred Premji"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "ai-agent",
+      "llm-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.03953",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03953",
+    "pdf_url": "https://arxiv.org/pdf/2607.03953",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.03726",
+    "title": "SelfMem: Self-Optimizing Memory for AI Agents",
+    "url": "https://arxiv.org/abs/2607.03726",
+    "published": "2026-07-04",
+    "updated": "2026-07-04",
+    "authors": [
+      "Shu Yang",
+      "Junchao Wu",
+      "Derek F. Wong",
+      "Di Wang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "ai-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.03726",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03726",
+    "pdf_url": "https://arxiv.org/pdf/2607.03726",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2607.01766",
+    "title": "SimWorlds: A Multi-Agent System for Dynamic 3D Scene Creation",
+    "url": "https://arxiv.org/abs/2607.01766",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Chunjiang Liu",
+      "Xiaoyuan Wang",
+      "Haoyu Chen",
+      "Yizhou Zhao",
+      "Ming-Hsuan Yang",
+      "László A. Jeni"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.01766",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01766",
+    "pdf_url": "https://arxiv.org/pdf/2607.01766",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.00454",
+    "title": "Agri-SAGE: Simulation-Grounded Multi-Agent LLM for Context-Aware Agricultural Advisory Generation",
+    "url": "https://arxiv.org/abs/2607.00454",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Vedant Balasubramaniam",
+      "Geetha Charan",
+      "Manojkumar Patil",
+      "Rohit P Suresh",
+      "V Priyanka",
+      "Kodur Sai Vinay Sathvik",
+      "Y. Narahari"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "world-model"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.00454",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00454",
+    "pdf_url": "https://arxiv.org/pdf/2607.00454",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.31073",
+    "title": "MultiUAV-Plat: An LLM-Oriented Platform, Benchmark and Framework for Multi-UAV Collaborative Task Planning",
+    "url": "https://arxiv.org/abs/2606.31073",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Sheng Zhang",
+      "Qinglin Li",
+      "Yuechao Zang",
+      "Xueqin Huang",
+      "Yijia Fu",
+      "Cheng Zhu"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA",
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "llm-agent",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.31073",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31073",
+    "pdf_url": "https://arxiv.org/pdf/2606.31073",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.31179",
+    "title": "HealthAgentBench: A Unified Benchmark Suite of Realistic Agentic Healthcare Environments for Challenging Frontier AI Agents",
+    "url": "https://arxiv.org/abs/2606.31179",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Qianchu Liu",
+      "Sheng Zhang",
+      "Guanghui Qin",
+      "Jeya Maria Jose Valanarasu",
+      "Maximilian Rokuss",
+      "Mingyu Lu",
+      "Timothy Ossowski",
+      "Juan Manuel Zambrano Chaves",
+      "Cliff Wong",
+      "Peniel Argaw",
+      "Yashna Hasija",
+      "Mu Wei",
+      "Wen-wai Yim",
+      "Qin Liu",
+      "Zilin Jing",
+      "Jason Entenmann",
+      "Naoto Usuyama",
+      "Tristan Naumann",
+      "Hoifung Poon"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.31179",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31179",
+    "pdf_url": "https://arxiv.org/pdf/2606.31179",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.31612",
+    "title": "What Memory Do GUI Agents Really Need? From Passive Records to Active Task-Driving States",
+    "url": "https://arxiv.org/abs/2606.31612",
+    "published": "2026-06-30",
+    "updated": "2026-07-02",
+    "authors": [
+      "Chen Liu",
+      "Ling Chen",
+      "Hanzhang Zhou",
+      "Xu Zhang",
+      "Quyu Kong",
+      "Panrong Tong",
+      "Wenhao Wang",
+      "Xin Yu",
+      "Steven Hoi",
+      "Yue Wang"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.31612",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31612",
+    "pdf_url": "https://arxiv.org/pdf/2606.31612",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.30906",
+    "title": "Investigating Multi-Agent Deliberation in Law",
+    "url": "https://arxiv.org/abs/2606.30906",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Cor Steging",
+      "Ludi van Leeuwen",
+      "Tadeusz Zbiegień"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "ai-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.30906",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30906",
+    "pdf_url": "https://arxiv.org/pdf/2606.30906",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.30949",
+    "title": "AgRefactor: Self-Evolving Agentic Workflow for HLS Compatibility and Performance",
+    "url": "https://arxiv.org/abs/2606.30949",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Yang Zou",
+      "Zijian Ding",
+      "Yizhou Sun",
+      "Jason Cong"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.AR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.30949",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30949",
+    "pdf_url": "https://arxiv.org/pdf/2606.30949",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.29116",
+    "title": "Characterizing Large Language Model Agentic Workflows: A Study on N8n Ecosystem",
+    "url": "https://arxiv.org/abs/2606.29116",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Yutian Tang",
+      "Yuming Zhou",
+      "Huaming Chen"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "llm-agent",
+      "planning-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.29116",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29116",
+    "pdf_url": "https://arxiv.org/pdf/2606.29116",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.24535",
+    "title": "Governed Shared Memory for Multi-Agent LLM Systems",
+    "url": "https://arxiv.org/abs/2606.24535",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Yanki Margalit",
+      "Nurit Cohen-Inger",
+      "Erni Avram",
+      "Ran Taig",
+      "Oded Margalit"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.24535",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24535",
+    "pdf_url": "https://arxiv.org/pdf/2606.24535",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.24820",
+    "title": "SHERLOC: Structured Diagnostic Localization for Code Repair Agents",
+    "url": "https://arxiv.org/abs/2606.24820",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Hovhannes Tamoyan",
+      "Sean Narenthiran",
+      "Erik Arakelyan",
+      "Mira Mezini",
+      "Boris Ginsburg"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent",
+      "multi-agent-llm",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.24820",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24820",
+    "pdf_url": "https://arxiv.org/pdf/2606.24820",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.22263",
+    "title": "Revelio: Cost-Efficient Agentic Memory Safety Vulnerability Detection For Repository-Scale Codebases",
+    "url": "https://arxiv.org/abs/2606.22263",
+    "published": "2026-06-20",
+    "updated": "2026-06-20",
+    "authors": [
+      "Yiwei Hou",
+      "Hao Wang",
+      "Muxi Lyu",
+      "Marius Momeu",
+      "Eric Nguyen",
+      "Taige Yang",
+      "Koushik Sen",
+      "Dawn Song",
+      "David Wagner"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.MA",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "memory"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.22263",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22263",
+    "pdf_url": "https://arxiv.org/pdf/2606.22263",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.21627",
+    "title": "Counsel: A Meta-Evaluation Dataset for Agentic Tasks",
+    "url": "https://arxiv.org/abs/2606.21627",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Sashank Pisupati",
+      "Henry Broomfield",
+      "Eujeong Choi",
+      "Antonia Calvi",
+      "Charlie Wang",
+      "Roman Engeler",
+      "Max Bartolo",
+      "Patrick Lewis"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "reasoning"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.21627",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21627",
+    "pdf_url": "https://arxiv.org/pdf/2606.21627",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.16871",
+    "title": "Human-on-the-Bridge: Scalable Evaluation for AI Agents",
+    "url": "https://arxiv.org/abs/2606.16871",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Fouad Bousetouane"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.16871",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16871",
+    "pdf_url": "https://arxiv.org/pdf/2606.16871",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.17246",
+    "title": "GeoDisaster: Benchmarking Orchestrated Agents for Operational Disaster Geo-Intelligence",
+    "url": "https://arxiv.org/abs/2606.17246",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Maram Hasan",
+      "Aman Verma",
+      "Savitra Roy",
+      "Hariseetharam Gunduboina",
+      "Daksh Jain",
+      "Muhammad Haris Khan",
+      "Subhasis Chaudhuri",
+      "Biplab Banerjee"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.17246",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17246",
+    "pdf_url": "https://arxiv.org/pdf/2606.17246",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.17114",
+    "title": "An Evaluation of Data Leakage Risks in Tool-Using LLM Agents in Realistic Scenarios",
+    "url": "https://arxiv.org/abs/2606.17114",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Hankyul Baek",
+      "Jaewon Noh",
+      "Sang Seo",
+      "Yongsu Kim",
+      "Gabriel Waikin Loh Matienzo",
+      "Young Il Kim",
+      "Ee Wei Seah",
+      "Akriti Vij"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.17114",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17114",
+    "pdf_url": "https://arxiv.org/pdf/2606.17114",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.16420",
+    "title": "Transferable Self-Evolving Playbooks for Agentic Security Auditing",
+    "url": "https://arxiv.org/abs/2606.16420",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Ziyue Wang",
+      "Cheuk Wang Maurice Ng",
+      "Chenchen Yu",
+      "Strick Sheng",
+      "Kaihua Qin",
+      "Liyi Zhou"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "embodied-agent",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.16420",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16420",
+    "pdf_url": "https://arxiv.org/pdf/2606.16420",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.14790",
+    "title": "XFlow: An Executable Protocol Programming System for Reliable Multi-Agent Workflows",
+    "url": "https://arxiv.org/abs/2606.14790",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Hanqi Li",
+      "Jing Peng",
+      "Zijian Wang",
+      "Lu Chen",
+      "Kai Yu"
+    ],
+    "categories": [
+      "cs.PL",
+      "cs.AI"
+    ],
+    "topics": [
+      "coding-agent",
+      "memory",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.14790",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.14790",
+    "pdf_url": "https://arxiv.org/pdf/2606.14790",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.08531",
+    "title": "VESTA: A Fully Automated Scenario Generation and Safety Evaluation Framework for LLM Agents",
+    "url": "https://arxiv.org/abs/2606.08531",
+    "published": "2026-06-07",
+    "updated": "2026-06-07",
+    "authors": [
+      "Lu Jia",
+      "Haibo Tong",
+      "Feifei Zhao",
+      "Jindong Li",
+      "Dongqi Liang",
+      "Ping Wu",
+      "Qian Zhang",
+      "Yi Zeng"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2606.08531",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.08531",
+    "pdf_url": "https://arxiv.org/pdf/2606.08531",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.07402",
+    "title": "M$^3$Exam: Benchmarking Multimodal Memory for Realistic User-Agent Interactions",
+    "url": "https://arxiv.org/abs/2606.07402",
+    "published": "2026-06-05",
+    "updated": "2026-06-05",
+    "authors": [
+      "Zhengjun Huang",
+      "Wenxuan Liu",
+      "Zhoujin Tian",
+      "Wei Chen",
+      "Junle Chen",
+      "Yuqian Wu",
+      "Fangyuan Zhang",
+      "Qintian Guo",
+      "Xiaofang Zhou"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.07402",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.07402",
+    "pdf_url": "https://arxiv.org/pdf/2606.07402",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.04780",
+    "title": "PersonaTree: Structured Lifecycle Memory for Person Understanding in LLM Agents",
+    "url": "https://arxiv.org/abs/2606.04780",
+    "published": "2026-06-03",
+    "updated": "2026-06-03",
+    "authors": [
+      "Yubo Hou",
+      "Jingwei Song",
+      "Hongbo Zhang",
+      "Zhisheng Chen",
+      "Bang Xiao",
+      "Tao Wan",
+      "Zengchang Qin"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.04780",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.04780",
+    "pdf_url": "https://arxiv.org/pdf/2606.04780",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.03657",
+    "title": "Diagnosing Knowledge Gaps in LLM Tool Use: An Agentic Benchmark for Novel API Acquisition",
+    "url": "https://arxiv.org/abs/2606.03657",
+    "published": "2026-06-02",
+    "updated": "2026-06-02",
+    "authors": [
+      "Jinnuo Liu",
+      "Yue Peng",
+      "Jinhan Niu",
+      "Hongyi Wen"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.03657",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.03657",
+    "pdf_url": "https://arxiv.org/pdf/2606.03657",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.02109",
+    "title": "BADGER: Bridging Agentic and Deterministic Evaluation for Generative Enterprise Reasoning",
+    "url": "https://arxiv.org/abs/2606.02109",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Shannon Serrao",
+      "Soumitra Chatterjee",
+      "Dorina Strori",
+      "Abhishek Sharma",
+      "Nathan Miller"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.02109",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.02109",
+    "pdf_url": "https://arxiv.org/pdf/2606.02109",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.01416",
+    "title": "Self-Healing Agentic Orchestrators for Reliable Tool-Augmented Large Language Model Systems",
+    "url": "https://arxiv.org/abs/2606.01416",
+    "published": "2026-05-31",
+    "updated": "2026-05-31",
+    "authors": [
+      "Rahul Suresh Babu",
+      "Adarsh Agrawal"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.01416",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.01416",
+    "pdf_url": "https://arxiv.org/pdf/2606.01416",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.00610",
+    "title": "MemGraphRAG: Memory-based Multi-Agent System for Graph Retrieval-Augmented Generation",
+    "url": "https://arxiv.org/abs/2606.00610",
+    "published": "2026-05-30",
+    "updated": "2026-05-30",
+    "authors": [
+      "Chuanjie Wu",
+      "Zhishang Xiang",
+      "Yunbo Tang",
+      "Zerui Chen",
+      "Qinggang Zhang",
+      "Jinsong Su"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.00610",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.00610",
+    "pdf_url": "https://arxiv.org/pdf/2606.00610",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.30883",
+    "title": "TRACE: Task-Aware Adaptive Self-Evolving Agentic Jailbreaking",
+    "url": "https://arxiv.org/abs/2605.30883",
+    "published": "2026-05-29",
+    "updated": "2026-05-29",
+    "authors": [
+      "Churui Zeng",
+      "Weiwei Qi",
+      "Kedong Xiu",
+      "Tianhang Zheng",
+      "Chaochao Lu",
+      "Liang He",
+      "Zhan Qin",
+      "Kui Ren"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.30883",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.30883",
+    "pdf_url": "https://arxiv.org/pdf/2605.30883",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.30090",
+    "title": "DirectorBench: Diagnosing Long-Form Video Generation with Personalized Multi-Agent Evaluation",
+    "url": "https://arxiv.org/abs/2605.30090",
+    "published": "2026-05-28",
+    "updated": "2026-05-28",
+    "authors": [
+      "Jiamin Chen",
+      "Qianben Chen",
+      "Jiawen Zhang",
+      "Yidi Wu",
+      "Yuchen Li",
+      "Xiaokun Zhang",
+      "Wangchunshu Zhou",
+      "Chen Ma"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2605.30090",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.30090",
+    "pdf_url": "https://arxiv.org/pdf/2605.30090",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2605.27134",
+    "title": "Scaling, Benchmarking, and Reasoning of Vision-Language Agents for Mobile GUI Navigation",
+    "url": "https://arxiv.org/abs/2605.27134",
+    "published": "2026-05-26",
+    "updated": "2026-05-26",
+    "authors": [
+      "Heng Qu",
+      "Yike Liu",
+      "Renren Jin",
+      "Wenzong Zhang",
+      "Pengzhi Gao",
+      "Wei Liu",
+      "Jian Luan"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "embodied-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.27134",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.27134",
+    "pdf_url": "https://arxiv.org/pdf/2605.27134",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.22643",
+    "title": "Boiling the Frog: A Multi-Turn Benchmark for Agentic Safety",
+    "url": "https://arxiv.org/abs/2605.22643",
+    "published": "2026-05-21",
+    "updated": "2026-05-22",
+    "authors": [
+      "Piercosma Bisconti",
+      "Matteo Prandi",
+      "Federico Pierucci",
+      "Federico Sartore",
+      "Enrico Panai",
+      "Laura Caroli",
+      "Yue Zhu",
+      "Adam Leon Smith",
+      "Luca Nannini",
+      "Marcello Galisai",
+      "Susanna Cifani",
+      "Francesco Giarrusso",
+      "Marcantonio Bracale Syrnikov",
+      "Daniele Nardi"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.22643",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.22643",
+    "pdf_url": "https://arxiv.org/pdf/2605.22643",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.15040",
+    "title": "Orchard: An Open-Source Agentic Modeling Framework",
+    "url": "https://arxiv.org/abs/2605.15040",
+    "published": "2026-05-14",
+    "updated": "2026-05-21",
+    "authors": [
+      "Baolin Peng",
+      "Wenlin Yao",
+      "Qianhui Wu",
+      "Hao Cheng",
+      "Xiao Yu",
+      "Rui Yang",
+      "Tao Ge",
+      "Alessandro Sordoni",
+      "Xingdi Yuan",
+      "Yelong Shen",
+      "Pengcheng He",
+      "Tong Zhang",
+      "Zhou Yu",
+      "Jianfeng Gao"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.15040",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.15040",
+    "pdf_url": "https://arxiv.org/pdf/2605.15040",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.13542",
+    "title": "RealICU: Do LLM Agents Understand Long-Context ICU Data? A Benchmark Beyond Behavior Imitation",
+    "url": "https://arxiv.org/abs/2605.13542",
+    "published": "2026-05-13",
+    "updated": "2026-05-13",
+    "authors": [
+      "Chengzhi Shen",
+      "Weixiang Shen",
+      "Tobias Susetzky",
+      "Chen",
+      "Chen",
+      "Jun Li",
+      "Yuyuan Liu",
+      "Xuepeng Zhang",
+      "Zhenyu Gong",
+      "Daniel Rueckert",
+      "Jiazhen Pan"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.LG",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.13542",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.13542",
+    "pdf_url": "https://arxiv.org/pdf/2605.13542",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.12015",
+    "title": "SkillSafetyBench: Evaluating Agent Safety under Skill-Facing Attack Surfaces",
+    "url": "https://arxiv.org/abs/2605.12015",
+    "published": "2026-05-12",
+    "updated": "2026-05-27",
+    "authors": [
+      "Chang Jin",
+      "An Wang",
+      "Zeming Wei",
+      "Kai Wang",
+      "Biaojie Zeng",
+      "Qiaosheng Zhang",
+      "Chao Yang",
+      "Jingjing Qu",
+      "Xia Hu",
+      "Xingcheng Xu"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.CL",
+      "cs.LG",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "memory",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.12015",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.12015",
+    "pdf_url": "https://arxiv.org/pdf/2605.12015",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.08374",
+    "title": "MemQ: Integrating Q-Learning into Self-Evolving Memory Agents over Provenance DAGs",
+    "url": "https://arxiv.org/abs/2605.08374",
+    "published": "2026-05-08",
+    "updated": "2026-05-14",
+    "authors": [
+      "Junwei Liao",
+      "Haoting Shi",
+      "Ruiwen Zhou",
+      "Jiaqian Wang",
+      "Shengtao Zhang",
+      "Wei Zhang",
+      "Ying Wen",
+      "Zhiyu Li",
+      "Feiyu Xiong",
+      "Bo Tang",
+      "Weinan Zhang",
+      "Muning Wen"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "embodied-agent",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2605.08374",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.08374",
+    "pdf_url": "https://arxiv.org/pdf/2605.08374",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2605.15206",
+    "title": "AgentStop: Terminating Local AI Agents Early to Save Energy in Consumer Devices",
+    "url": "https://arxiv.org/abs/2605.15206",
+    "published": "2026-05-01",
+    "updated": "2026-05-01",
+    "authors": [
+      "Dzung Pham",
+      "Kleomenis Katevas",
+      "Ali Shahin Shamsabadi",
+      "Hamed Haddadi"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.DC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.15206",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.15206",
+    "pdf_url": "https://arxiv.org/pdf/2605.15206",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.16282",
+    "title": "Taxonomy and Consistency Analysis of Safety Benchmarks for AI Agents",
+    "url": "https://arxiv.org/abs/2605.16282",
+    "published": "2026-04-11",
+    "updated": "2026-04-11",
+    "authors": [
+      "Miles Q. Li",
+      "Benjamin C. M. Fung",
+      "Boyang Li",
+      "Heba Ismail",
+      "Farkhund Iqbal"
+    ],
+    "categories": [
+      "cs.CY",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.16282",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.16282",
+    "pdf_url": "https://arxiv.org/pdf/2605.16282",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.02022",
+    "title": "ATBench: A Diverse and Realistic Agent Trajectory Benchmark for Safety Evaluation and Diagnosis",
+    "url": "https://arxiv.org/abs/2604.02022",
+    "published": "2026-04-02",
+    "updated": "2026-05-13",
+    "authors": [
+      "Yu Li",
+      "Haoyu Luo",
+      "Yuejin Xie",
+      "Yuqian Fu",
+      "Zhonghao Yang",
+      "Shuai Shao",
+      "Qihan Ren",
+      "Wanying Qu",
+      "Yanwei Fu",
+      "Yujiu Yang",
+      "Jing Shao",
+      "Xia Hu",
+      "Dongrui Liu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.02022",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.02022",
+    "pdf_url": "https://arxiv.org/pdf/2604.02022",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.16734",
+    "title": "Differential Harm Propensity in Personalized LLM Agents: The Curious Case of Mental Health Disclosure",
+    "url": "https://arxiv.org/abs/2603.16734",
+    "published": "2026-03-17",
+    "updated": "2026-03-17",
+    "authors": [
+      "Caglar Yildirim"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.16734",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.16734",
+    "pdf_url": "https://arxiv.org/pdf/2603.16734",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2509.20998",
+    "title": "CORE: Full-Path Evaluation of LLM Agents Beyond Final State",
+    "url": "https://arxiv.org/abs/2509.20998",
+    "published": "2025-09-25",
+    "updated": "2025-09-25",
+    "authors": [
+      "Panagiotis Michelakis",
+      "Yiannis Hadjiyiannis",
+      "Dimitrios Stamoulis"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 20,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2509.20998",
+    "source": "arxiv",
+    "source_id": "arxiv:2509.20998",
+    "pdf_url": "https://arxiv.org/pdf/2509.20998",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2607.05174",
+    "title": "AgentGym2: Benchmarking Large Language Model Agents in De-Idealized Real-World Environments",
+    "url": "https://arxiv.org/abs/2607.05174",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Zhiheng Xi",
+      "Dingwen Yang",
+      "Jiaqi Liu",
+      "Jixuan Huang",
+      "Honglin Guo",
+      "Baodai Huang",
+      "Tinggang Chen",
+      "Qi Zhang",
+      "Zhonghang Lu",
+      "Chenyu Liu",
+      "Jiajun Sun",
+      "Jiazheng Zhang",
+      "Dingwei Zhu",
+      "Xin Guo",
+      "Junzhe Wang",
+      "Zhihao Zhang",
+      "Yuming Yang",
+      "Junjie Ye",
+      "Minghe Gao",
+      "Dongrui Liu",
+      "Jiaming Ji",
+      "Guohao Li",
+      "Tao Gui",
+      "Qi Zhang",
+      "Xuanjing Huang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent",
+      "llm-agent",
+      "planning-agent"
+    ],
+    "arxiv_id": "2607.05174",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05174",
+    "pdf_url": "https://arxiv.org/pdf/2607.05174",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2607.05029",
+    "title": "Your Agent's Memories Are Not Its Own: Forged Reasoning Attacks on LLM Agent Memory and Defenses",
+    "url": "https://arxiv.org/abs/2607.05029",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Neeraj Karamchandani",
+      "Piyush Nagasubramaniam",
+      "Sencun Zhu",
+      "Dinghao Wu"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.05029",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05029",
+    "pdf_url": "https://arxiv.org/pdf/2607.05029",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2607.05120",
+    "title": "Agent Data Injection Attacks are Realistic Threats to AI Agents",
+    "url": "https://arxiv.org/abs/2607.05120",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Woohyuk Choi",
+      "Juhee Kim",
+      "Taehyun Kang",
+      "Jihyeon Jeong",
+      "Luyi Xing",
+      "Byoungyoung Lee"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "ai-agent",
+      "coding-agent",
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2607.05120",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05120",
+    "pdf_url": "https://arxiv.org/pdf/2607.05120",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2607.05202",
+    "title": "EvoAgentBench: Benchmarking Agent Self-Evolution via Ability Transfer",
+    "url": "https://arxiv.org/abs/2607.05202",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Xingze Gao",
+      "Chuanrui Hu",
+      "Hongda Chen",
+      "Pengfei Yao",
+      "Zhao Wang",
+      "Yi Bai",
+      "Zhengwei Wu",
+      "Yunyun Han",
+      "Xiaofeng Cong",
+      "Jie Gui",
+      "Yafeng Deng",
+      "Teng Li"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "memory",
+      "planning",
+      "reasoning"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2607.05202",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05202",
+    "pdf_url": "https://arxiv.org/pdf/2607.05202",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.04395",
+    "title": "NKI-Agent: Domain-Specific Fine-Tuning and Agentic Tool Use for Neuron Kernel Generation",
+    "url": "https://arxiv.org/abs/2607.04395",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Junjie Tang",
+      "Jun Huan",
+      "Hao Zhou",
+      "Yuhao Zhang",
+      "Lin Wang"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2607.04395",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04395",
+    "pdf_url": "https://arxiv.org/pdf/2607.04395",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2607.03510",
+    "title": "CAGE-1: Control, Assurance, and Governance Evaluation for Enterprise Agentic AI",
+    "url": "https://arxiv.org/abs/2607.03510",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Roopam W. Sure"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.CY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.03510",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03510",
+    "pdf_url": "https://arxiv.org/pdf/2607.03510",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.02684",
+    "title": "Can Coding Agents Implement Missed Compiler Optimizations? Evaluating LLM Agents on LLVM Peephole Optimizations",
+    "url": "https://arxiv.org/abs/2607.02684",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Hongxu Xu",
+      "Chunhao Liao",
+      "Xintong Zhou",
+      "Chengnian Sun"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "reasoning"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent",
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.02684",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02684",
+    "pdf_url": "https://arxiv.org/pdf/2607.02684",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.02507",
+    "title": "What LLM Agents Say When No One Is Watching: Social Structure and Latent Objective Emergence in Multi-Agent Debates",
+    "url": "https://arxiv.org/abs/2607.02507",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Arman Ghaffarizadeh",
+      "Danyal Mohaddes",
+      "Aliakbar Izadkhah",
+      "Shahriar Noroozizadeh"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.LG",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.02507",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02507",
+    "pdf_url": "https://arxiv.org/pdf/2607.02507",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.30986",
+    "title": "The Organizational Behavior of Agentic AI: Collective Intelligence in Human-Agent Workflows",
+    "url": "https://arxiv.org/abs/2606.30986",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Canhui Liu"
+    ],
+    "categories": [
+      "cs.CY",
+      "cs.HC",
+      "cs.MA",
+      "econ.GN"
+    ],
+    "topics": [
+      "memory",
+      "planning",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.30986",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30986",
+    "pdf_url": "https://arxiv.org/pdf/2606.30986",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.30639",
+    "title": "Self-Evolving World Models for LLM Agent Planning",
+    "url": "https://arxiv.org/abs/2606.30639",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Xuan Zhang",
+      "Wenxuan Zhang",
+      "See-Kiong Ng",
+      "Yang Deng"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.30639",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30639",
+    "pdf_url": "https://arxiv.org/pdf/2606.30639",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.29771",
+    "title": "CLQT: A Closed-Loop, Cost-Aware, Strategy-Consistent Benchmark for Diagnostic Evaluation of LLM Portfolio-Management Agents",
+    "url": "https://arxiv.org/abs/2606.29771",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Bo Qu",
+      "Mingguang Chen"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG",
+      "q-fin.CP",
+      "q-fin.PM"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.29771",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29771",
+    "pdf_url": "https://arxiv.org/pdf/2606.29771",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.00041",
+    "title": "ATM: CID-Brokered Pre-Write Admission for Multi-Agent Code Co-Synthesis",
+    "url": "https://arxiv.org/abs/2607.00041",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Eagl Huang"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "multi-agent",
+      "planning",
+      "rag"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.00041",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00041",
+    "pdf_url": "https://arxiv.org/pdf/2607.00041",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.29774",
+    "title": "Analytic Concept-Centric Memory for Agentic Embodied Manipulation",
+    "url": "https://arxiv.org/abs/2606.29774",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Mingyang Sun",
+      "Xiujian Liang",
+      "Jiude Wei",
+      "Qichen He",
+      "Donglin Wang",
+      "Cewu Lu",
+      "Jianhua Sun"
+    ],
+    "categories": [
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.29774",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29774",
+    "pdf_url": "https://arxiv.org/pdf/2606.29774",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.29193",
+    "title": "A Multi-Dataset Benchmark for Evaluating LLM Agents in Microservice Failure Diagnosis",
+    "url": "https://arxiv.org/abs/2606.29193",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Yuanhong Cai",
+      "Xiaohui Nie",
+      "Kanglin Yin",
+      "Changhua Pei",
+      "Yongqian Sun",
+      "Shenglin Zhang",
+      "Haibin Liu",
+      "Guiyang Liu",
+      "Xidao Wen",
+      "Fang Situ",
+      "Dan Pei"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.29193",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29193",
+    "pdf_url": "https://arxiv.org/pdf/2606.29193",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.29030",
+    "title": "Memory as an Attack Surface in LLM Agents: A Study on Multiple-Choice Question Answering",
+    "url": "https://arxiv.org/abs/2606.29030",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Shahnewaz Karim Sakib",
+      "Anindya Bijoy Das"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.ET"
+    ],
+    "topics": [
+      "memory",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "llm-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.29030",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29030",
+    "pdf_url": "https://arxiv.org/pdf/2606.29030",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.28692",
+    "title": "An AI agent for treatment reasoning over a biomedical tool universe",
+    "url": "https://arxiv.org/abs/2606.28692",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Shanghua Gao",
+      "Ayush Noori",
+      "Richard Zhu",
+      "Curtis Ginder",
+      "Zhenglun Kong",
+      "Xiaorui Su",
+      "Justin Kauffman",
+      "Benjamin S. Glicksberg",
+      "Joshua Lampert",
+      "Ankit Sakhuja",
+      "Ashwin Sawant",
+      "ATHENA-R1 Evaluation Consortium",
+      "David A. Clifton",
+      "Noa Dagan",
+      "Ran Balicer",
+      "Marinka Zitnik"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.28692",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28692",
+    "pdf_url": "https://arxiv.org/pdf/2606.28692",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.27806",
+    "title": "Agent vs. Parametric World Models: Hybrid Planning for Reliable Language Agents",
+    "url": "https://arxiv.org/abs/2606.27806",
+    "published": "2026-06-26",
+    "updated": "2026-07-05",
+    "authors": [
+      "Xinyuan Song",
+      "Zekun Cai"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.27806",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27806",
+    "pdf_url": "https://arxiv.org/pdf/2606.27806",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.28467",
+    "title": "An Agentic AI Pipeline for Appliance-Level Energy Anomaly Detection and LLM-Driven Recommendations",
+    "url": "https://arxiv.org/abs/2606.28467",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Dihia Falouz",
+      "Aida Douaibia",
+      "Amine Bechar",
+      "Youssef Elmir",
+      "Abbes Amira",
+      "Adel Oulefki"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.28467",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28467",
+    "pdf_url": "https://arxiv.org/pdf/2606.28467",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.26960",
+    "title": "Toward Agentic SysAdmin: Rethinking System Administration with AI Agents",
+    "url": "https://arxiv.org/abs/2606.26960",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Gianmaria Frigo",
+      "Davide Saladino",
+      "Alberto Castagnaro",
+      "Francesco Marchiori",
+      "Denis Donadel",
+      "Luca Pajola",
+      "Mauro Conti"
+    ],
+    "categories": [
+      "cs.NI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.26960",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26960",
+    "pdf_url": "https://arxiv.org/pdf/2606.26960",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.26346",
+    "title": "How Do Tool-Augmented LLM Agents Perform on Real-World Energy Analytics Tasks?",
+    "url": "https://arxiv.org/abs/2606.26346",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "David Akinpelu",
+      "Akintonde Abbas",
+      "Rereloluwa Alimi",
+      "Ayodeji Lana"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.26346",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26346",
+    "pdf_url": "https://arxiv.org/pdf/2606.26346",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.26403",
+    "title": "ProfileFoundry: A Synthetic Person-Object Substrate for Privacy, Memory, and Tool-Use Evaluation in LLM Agent",
+    "url": "https://arxiv.org/abs/2606.26403",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Sriram Selvam",
+      "Anneswa Ghosh"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.26403",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26403",
+    "pdf_url": "https://arxiv.org/pdf/2606.26403",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.25161",
+    "title": "TRUSTMEM: Learning Trustworthy Memory Consolidation for LLM Agents with Long-Term Memory",
+    "url": "https://arxiv.org/abs/2606.25161",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Tianyu Yang",
+      "Sudipta Paul",
+      "Vijay Srinivasan",
+      "Vivek Kulkarni",
+      "Srinivas Chappidi"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.25161",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25161",
+    "pdf_url": "https://arxiv.org/pdf/2606.25161",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.24626",
+    "title": "SAFARI: Scaling Long Horizon Agentic Fault Attribution via Active Investigation",
+    "url": "https://arxiv.org/abs/2606.24626",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Chenyang Zhu",
+      "Jiayu Yao",
+      "Kushal Chawla",
+      "Youbing Yin",
+      "Nathan Wolfe",
+      "Pengshan Cai",
+      "Jingyu Wu",
+      "Spencer Hong",
+      "Sangwoo Cho",
+      "Shi-Xiong Zhang",
+      "Daben Liu",
+      "Sambit Sahu",
+      "Erin Babinsky"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.24626",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24626",
+    "pdf_url": "https://arxiv.org/pdf/2606.24626",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.23991",
+    "title": "Critique of Agent Model",
+    "url": "https://arxiv.org/abs/2606.23991",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Eric Xing",
+      "Mingkai Deng",
+      "Jinyu Hou"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG",
+      "cs.MA",
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "ai-agent",
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.23991",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23991",
+    "pdf_url": "https://arxiv.org/pdf/2606.23991",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.22844",
+    "title": "RaMem: Contextual Reinstatement for Long-term Agentic Memory",
+    "url": "https://arxiv.org/abs/2606.22844",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Wei Yang",
+      "Bryce Kan",
+      "Shixuan Li",
+      "Li Li",
+      "Yuehan Qin",
+      "Jiate Li",
+      "Paul Bogdan",
+      "Jesse Thomason"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.22844",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22844",
+    "pdf_url": "https://arxiv.org/pdf/2606.22844",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.23565",
+    "title": "HoloAgent-0: A Unified Embodied Agent Framework with 3D Spatial Memory",
+    "url": "https://arxiv.org/abs/2606.23565",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Xiaolin Zhou",
+      "Liu Liu",
+      "Tingyang Xiao",
+      "Wei Feng",
+      "Fa Fu",
+      "Xinrui Meng",
+      "Xinjie Wang",
+      "Jialiang Han",
+      "Boyang Yu",
+      "Yun Du",
+      "Wei Sui",
+      "Zhizhong Su"
+    ],
+    "categories": [
+      "cs.RO",
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.23565",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23565",
+    "pdf_url": "https://arxiv.org/pdf/2606.23565",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.22678",
+    "title": "RigorBench: Benchmarking Engineering Process Discipline in Autonomous AI Coding Agents",
+    "url": "https://arxiv.org/abs/2606.22678",
+    "published": "2026-06-21",
+    "updated": "2026-06-29",
+    "authors": [
+      "Meher Bhaskar Madiraju",
+      "Meher Sai Preetam Madiraju"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.22678",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22678",
+    "pdf_url": "https://arxiv.org/pdf/2606.22678",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.22417",
+    "title": "Code Isn't Memory: A Structural Codebase Index Inside a Coding Agent",
+    "url": "https://arxiv.org/abs/2606.22417",
+    "published": "2026-06-21",
+    "updated": "2026-06-21",
+    "authors": [
+      "Ishaan Bhola",
+      "Adithyan Krishnan",
+      "Sravanth Kurmala",
+      "Mukunda NS"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "memory",
+      "rag"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.22417",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22417",
+    "pdf_url": "https://arxiv.org/pdf/2606.22417",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.21129",
+    "title": "AgenticOS: An Intent-Oriented Secure Operating System Architecture for Autonomous AI Agents",
+    "url": "https://arxiv.org/abs/2606.21129",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Zhen Zhao",
+      "Yu Zhang",
+      "Yanpeng Zhu",
+      "Jia Wang",
+      "Songqiao Tao",
+      "Xin Cheng",
+      "Jiexin Gao"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.OS"
+    ],
+    "topics": [
+      "agent-safety",
+      "planning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "autonomous-agent-llm",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.21129",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21129",
+    "pdf_url": "https://arxiv.org/pdf/2606.21129",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.21649",
+    "title": "EvoEmbedding: Evolvable Representations for Long-Context Retrieval and Agentic Memory",
+    "url": "https://arxiv.org/abs/2606.21649",
+    "published": "2026-06-19",
+    "updated": "2026-06-25",
+    "authors": [
+      "Chang Nie",
+      "Chaoyou Fu",
+      "Junlan Feng",
+      "Caifeng Shan"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "memory",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "agentic-ai",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.21649",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21649",
+    "pdf_url": "https://arxiv.org/pdf/2606.21649",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.20950",
+    "title": "Power Systems Agent Benchmark: Executable Evaluation of AI Agents in Electric Power Engineering",
+    "url": "https://arxiv.org/abs/2606.20950",
+    "published": "2026-06-18",
+    "updated": "2026-07-02",
+    "authors": [
+      "Sergei Trashchenkov"
+    ],
+    "categories": [
+      "cs.AI",
+      "eess.SY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "ai-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.20950",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20950",
+    "pdf_url": "https://arxiv.org/pdf/2606.20950",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.19704",
+    "title": "Beyond Static Leaderboards: Predictive Validity for the Evaluation of LLM Agents",
+    "url": "https://arxiv.org/abs/2606.19704",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Dhaval C. Patel",
+      "Kaoutar El Maghraoui",
+      "Shuxin Lin",
+      "Yusheng Li",
+      "Tianjun Feng",
+      "Chun-Yi Tsai",
+      "Yihan Sun",
+      "Wei Alexander Xin",
+      "Akshat Bhandari",
+      "Tanisha Rathod",
+      "Aaron Fan",
+      "Sanskruti Vijay Shejwal",
+      "Tomas Pasiecznik",
+      "Sagar Chethan Kumar",
+      "Tanmay Agarwal",
+      "Rohith Kanathur",
+      "Sam Colman",
+      "Amaan Sheikh",
+      "Dev Bahl",
+      "Ann Li",
+      "Krish Veera",
+      "Alimurtaza Mustafa Merchant",
+      "Shambhawi Baswaraj Bhure",
+      "Sajal Kumar Goyla",
+      "Chengrui Li",
+      "Kirthana Natarajan",
+      "Rui Li",
+      "Thomas Ajai",
+      "Rujing Li",
+      "Vivek G. Iyer",
+      "Sanjaii Vijayakumar",
+      "Yitong Bai",
+      "Ayal Yakobe",
+      "Darief Maes",
+      "Yassine Jebbouri",
+      "Tianyang Xu",
+      "Thai Quoc On",
+      "Vera Mazeeva",
+      "Winston Li",
+      "Yuval Shemla",
+      "Yeshitha Bhuvanesh",
+      "Rushin Bhatt",
+      "Siddharth Chethan Gowda",
+      "Alisha Vinod",
+      "Caroline Cahill",
+      "Shriya Aishani Rachakonda",
+      "Yunfeng Chen",
+      "Aryaman Agrawal",
+      "Aman Upganlawar",
+      "Mao Le Jonathan Ang",
+      "Yubin Sally Go",
+      "Madhav Rajkondawar",
+      "Yang-Jung Chen",
+      "Trisha Maturi",
+      "Ananya Kapoor",
+      "Andrew Li",
+      "Shrey Arora",
+      "Mana Abbaszadeh",
+      "Shen Li",
+      "Charles Xu",
+      "Byeolah Kwon"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.19704",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19704",
+    "pdf_url": "https://arxiv.org/pdf/2606.19704",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.20512",
+    "title": "Probe-and-Refine Tuning of Repository Guidance for Coding Agents",
+    "url": "https://arxiv.org/abs/2606.20512",
+    "published": "2026-06-18",
+    "updated": "2026-06-19",
+    "authors": [
+      "Asa Shepard",
+      "Jeannie Albrecht"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.20512",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20512",
+    "pdf_url": "https://arxiv.org/pdf/2606.20512",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.18829",
+    "title": "GateMem: Benchmarking Memory Governance in Multi-Principal Shared-Memory Agents",
+    "url": "https://arxiv.org/abs/2606.18829",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Zhe Ren",
+      "Yibo Yang",
+      "Yimeng Chen",
+      "Zijun Zhao",
+      "Benshuo Fu",
+      "Zhihao Shu",
+      "Bingjie Zhang",
+      "Yangyang Xu",
+      "Dandan Guo",
+      "Shuicheng Yan"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.18829",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18829",
+    "pdf_url": "https://arxiv.org/pdf/2606.18829",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.18356",
+    "title": "SafeClawBench: Separating Semantic, Audit-Evidence, and Sandbox Harm in Tool-Using LLM Agents",
+    "url": "https://arxiv.org/abs/2606.18356",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Yuchuan Tian",
+      "Mengyu Zheng",
+      "Haocheng Mei",
+      "Ye Yuan",
+      "Chao Xu",
+      "Xinghao Chen",
+      "Hanting Chen",
+      "Yu Wang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "memory",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.18356",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18356",
+    "pdf_url": "https://arxiv.org/pdf/2606.18356",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.16774",
+    "title": "OpenClaw-Skill: Collective Skill Tree Search for Agentic Large Language Models",
+    "url": "https://arxiv.org/abs/2606.16774",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Tianyi Lin",
+      "Chuanyu Sun",
+      "Jingyi Zhang",
+      "Changxu Wei",
+      "Huanjin Yao",
+      "Shunyu Liu",
+      "Xikun Zhang",
+      "Liu Liu",
+      "Jiaxing Huang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.16774",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16774",
+    "pdf_url": "https://arxiv.org/pdf/2606.16774",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.15862",
+    "title": "RetailBench: Benchmarking long horizon reasoning and coherent decision making of LLM agents in realistic retail environments",
+    "url": "https://arxiv.org/abs/2606.15862",
+    "published": "2026-06-14",
+    "updated": "2026-06-19",
+    "authors": [
+      "Linghua Zhang",
+      "Jun Wang",
+      "Jingtong Wu",
+      "Zhisong Zhang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.15862",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15862",
+    "pdf_url": "https://arxiv.org/pdf/2606.15862",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.12586",
+    "title": "Beyond Attack Success Rate: Examining Trigger Leakage in Vision-Language Agentic Systems",
+    "url": "https://arxiv.org/abs/2606.12586",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Jiamin Chang",
+      "Salil Kanhere",
+      "Piotr Koniusz",
+      "Jason",
+      "Xue",
+      "Hammond Pearce"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.12586",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12586",
+    "pdf_url": "https://arxiv.org/pdf/2606.12586",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.10507",
+    "title": "HIPIF: Hierarchical Planning and Information Folding for Long-Horizon LLM Agent Learning",
+    "url": "https://arxiv.org/abs/2606.10507",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Juncheng Diao",
+      "Zhicong Lu",
+      "Peiguang Li",
+      "Yongwei Zhou",
+      "Changyuan Tian",
+      "Qingbin Li",
+      "Rongxiang Weng",
+      "Jingang Wang",
+      "Xunliang Cai"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "reasoning"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "autonomous-agent-llm",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.10507",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10507",
+    "pdf_url": "https://arxiv.org/pdf/2606.10507",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.11042",
+    "title": "Workflow-GYM: Towards Long-Horizon Evaluation of Computer-use Agentic tasks in Real-World Professional Fields",
+    "url": "https://arxiv.org/abs/2606.11042",
+    "published": "2026-06-09",
+    "updated": "2026-06-11",
+    "authors": [
+      "Liya Zhu",
+      "Jingzhe Ding",
+      "Jian Zhang",
+      "Jianbo Xue",
+      "Shihao Liang",
+      "Ge Zhang",
+      "Yi Zhu",
+      "Duju Zeng",
+      "Xiang Gao",
+      "Qingshui Gu",
+      "Mailun Gao",
+      "Huimin Che",
+      "Yan Zhao",
+      "Peiheng Zhou",
+      "Haojun Wang",
+      "Chaobo Xian",
+      "Lili Le",
+      "Chi Wu",
+      "Yiwei Liu",
+      "Shengda Long",
+      "Jiale Yang",
+      "Fangzhi Xu",
+      "Sijin Wu",
+      "Haodong Duan",
+      "Chao He",
+      "Zhaojian Li",
+      "Minchao Wang",
+      "Huan Zhou",
+      "Jiani Hou",
+      "Chuqian Yu",
+      "Weiran Shi",
+      "Hongwan Gao",
+      "Jiamin Chen",
+      "Guanhong Chen",
+      "Tingqin Luo",
+      "Kaiyuan Zhang",
+      "Zhixin Yao",
+      "Qing Hua",
+      "Yuhao Jiang",
+      "Jin Chen",
+      "Pu Chen",
+      "Zhenyu Hu",
+      "Xingyu Li",
+      "Zhengxuan Jiang",
+      "Meng Cao",
+      "Tianfeng Long",
+      "Haozhe Wang",
+      "Mingzhang Wang",
+      "Yichen Zhang",
+      "Yiming Dai",
+      "Chenchen Zhang",
+      "Jiaying Wang",
+      "Xinying Liu",
+      "Xingzu Liu",
+      "Lingling Zhang",
+      "Xinjie Chen",
+      "Yujia Qin",
+      "Wangchunshu Zhou",
+      "Zhiyong Wu",
+      "Yang Liu",
+      "Jiaheng Liu",
+      "Lei Zhang",
+      "Shen Yan",
+      "Wenhao Huang",
+      "Zaiyuan Wang",
+      "Xiaolong Chang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.11042",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.11042",
+    "pdf_url": "https://arxiv.org/pdf/2606.11042",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.09483",
+    "title": "Memory Beyond Recall: A Dual-Process Cognitive Memory System for Self-Evolving LLM Agents",
+    "url": "https://arxiv.org/abs/2606.09483",
+    "published": "2026-06-08",
+    "updated": "2026-06-08",
+    "authors": [
+      "Tianxiang Fei",
+      "Mingyang Song",
+      "Mao Zheng",
+      "Xiang Yu"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.09483",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09483",
+    "pdf_url": "https://arxiv.org/pdf/2606.09483",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.07314",
+    "title": "QBugLM: An Agentic Benchmarking Framework for LLM-based Quantum Software Debugging",
+    "url": "https://arxiv.org/abs/2606.07314",
+    "published": "2026-06-05",
+    "updated": "2026-06-05",
+    "authors": [
+      "An B. B. Pham",
+      "Hoa T. Nguyen",
+      "Muhammad Usman"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.ET",
+      "quant-ph"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "reasoning",
+      "world-model"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.07314",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.07314",
+    "pdf_url": "https://arxiv.org/pdf/2606.07314",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.05684",
+    "title": "AdaMEM: Test-Time Adaptive Memory for Language Agents",
+    "url": "https://arxiv.org/abs/2606.05684",
+    "published": "2026-06-04",
+    "updated": "2026-06-04",
+    "authors": [
+      "Yunxiang Zhang",
+      "Yiheng Li",
+      "Ali Payani",
+      "Lu Wang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "language-agent"
+    ],
+    "arxiv_id": "2606.05684",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.05684",
+    "pdf_url": "https://arxiv.org/pdf/2606.05684",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.06448",
+    "title": "Agent Memory: Characterization and System Implications of Stateful Long-Horizon Workloads",
+    "url": "https://arxiv.org/abs/2606.06448",
+    "published": "2026-06-04",
+    "updated": "2026-06-04",
+    "authors": [
+      "Yasmine Omri",
+      "Ziyu Gan",
+      "Zachary Broveak",
+      "Robin Geens",
+      "Zexue He",
+      "Alex Pentland",
+      "Marian Verhelst",
+      "Tsachy Weissman",
+      "Thierry Tambe"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.06448",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.06448",
+    "pdf_url": "https://arxiv.org/pdf/2606.06448",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.04874",
+    "title": "Agent Planning Benchmark: A Diagnostic Framework for Planning Capabilities in LLM Agents",
+    "url": "https://arxiv.org/abs/2606.04874",
+    "published": "2026-06-03",
+    "updated": "2026-06-05",
+    "authors": [
+      "Haoyu Sun",
+      "Wenxuan Wang",
+      "Mingyang Song",
+      "Jujie He",
+      "Weinan Zhang",
+      "Yang Liu",
+      "Yang Yang",
+      "Yu Cheng"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.04874",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.04874",
+    "pdf_url": "https://arxiv.org/pdf/2606.04874",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.28349",
+    "title": "HMARS: A Hierarchical Multi-Agent Memory System for Long-Context Reasoning",
+    "url": "https://arxiv.org/abs/2606.28349",
+    "published": "2026-06-03",
+    "updated": "2026-06-03",
+    "authors": [
+      "Zeju Li",
+      "Ziyang Zheng",
+      "Yizhou Zhou",
+      "Qiang Xu"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.28349",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28349",
+    "pdf_url": "https://arxiv.org/pdf/2606.28349",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.04315",
+    "title": "Exploring Cross-Scenario Generality of Agentic Memory Systems: Diagnostics and a Strong Baseline",
+    "url": "https://arxiv.org/abs/2606.04315",
+    "published": "2026-06-03",
+    "updated": "2026-06-03",
+    "authors": [
+      "Zhikai Chen",
+      "Jialiang Gu",
+      "Junyu Yin",
+      "Xianxuan Long",
+      "Shenglai Zeng",
+      "Xiaoze Liu",
+      "Kai Guo",
+      "Keren Zhou",
+      "Jiliang Tang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.04315",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.04315",
+    "pdf_url": "https://arxiv.org/pdf/2606.04315",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.03374",
+    "title": "eMEM: A Hybrid Spatio-Temporal Memory System For Embodied Agents",
+    "url": "https://arxiv.org/abs/2606.03374",
+    "published": "2026-06-02",
+    "updated": "2026-06-22",
+    "authors": [
+      "A. Haroon Rasheed",
+      "Maria Kabtoul"
+    ],
+    "categories": [
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.03374",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.03374",
+    "pdf_url": "https://arxiv.org/pdf/2606.03374",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.02372",
+    "title": "COMAP: Co-Evolving World Models and Agent Policies for LLM Agents",
+    "url": "https://arxiv.org/abs/2606.02372",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Youwei Liu",
+      "Jian Wang",
+      "Hanlin Wang",
+      "Wenjie Li"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.02372",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.02372",
+    "pdf_url": "https://arxiv.org/pdf/2606.02372",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.01613",
+    "title": "TechRAG: Evidence-Gated Multimodal Agentic RAG for Technical Literature Reasoning",
+    "url": "https://arxiv.org/abs/2606.01613",
+    "published": "2026-06-01",
+    "updated": "2026-06-13",
+    "authors": [
+      "Kanwar Bharat Singh"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.01613",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.01613",
+    "pdf_url": "https://arxiv.org/pdf/2606.01613",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.00939",
+    "title": "FinCom: A Financial Multi-Agent Demo with Disagree-or-Commit Deliberation",
+    "url": "https://arxiv.org/abs/2606.00939",
+    "published": "2026-05-31",
+    "updated": "2026-05-31",
+    "authors": [
+      "Chao Peter Yang",
+      "Zixiao Tan",
+      "Kaisen Yao",
+      "Ziyu Zhou",
+      "Eleanor Jiang",
+      "Michael Wu"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.00939",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.00939",
+    "pdf_url": "https://arxiv.org/pdf/2606.00939",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2605.30690",
+    "title": "ElasticMem: Latent Memory as a Learnable Resource for LLM Agents",
+    "url": "https://arxiv.org/abs/2605.30690",
+    "published": "2026-05-29",
+    "updated": "2026-05-29",
+    "authors": [
+      "Tao Feng",
+      "Chongrui Ye",
+      "Tianyang Luo",
+      "Jingjun Xu",
+      "Xueqiang Xu",
+      "Haozhen Zhang",
+      "Ge Liu",
+      "Jiaxuan You"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.30690",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.30690",
+    "pdf_url": "https://arxiv.org/pdf/2605.30690",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.20629",
+    "title": "Specialize Roles, Mix Deployments: Pushing the Cost-Accuracy Frontier of LLM Agent Teams",
+    "url": "https://arxiv.org/abs/2606.20629",
+    "published": "2026-05-28",
+    "updated": "2026-05-28",
+    "authors": [
+      "Yinsicheng Jiang",
+      "Liang Cheng",
+      "Yeqi Huang",
+      "Yufan Zhao",
+      "Zhan Lu",
+      "Li Dong",
+      "Wenda Li",
+      "Edoardo Ponti",
+      "Luo Mai"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.20629",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20629",
+    "pdf_url": "https://arxiv.org/pdf/2606.20629",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2605.30604",
+    "title": "An Organization-Scoped LLM Agent Runtime Architecture for Regulated Cybersecurity Operations",
+    "url": "https://arxiv.org/abs/2605.30604",
+    "published": "2026-05-28",
+    "updated": "2026-05-28",
+    "authors": [
+      "George Fatouros",
+      "Georgios Makridis",
+      "George Kousiouris",
+      "John Soldatos",
+      "Dimosthenis Kyriazis"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.CL",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.30604",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.30604",
+    "pdf_url": "https://arxiv.org/pdf/2605.30604",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.27240",
+    "title": "ENPMR-Bench: Benchmarking Proactive Memory Retrieval for Emotional Support Agents",
+    "url": "https://arxiv.org/abs/2605.27240",
+    "published": "2026-05-26",
+    "updated": "2026-05-26",
+    "authors": [
+      "Xing Fu",
+      "Yulin Hu",
+      "Mengtong Ji",
+      "Haozhen Li",
+      "Yixin Sun",
+      "Weixiang Zhao",
+      "Yanyan Zhao",
+      "Bing Qin"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.27240",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.27240",
+    "pdf_url": "https://arxiv.org/pdf/2605.27240",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.25200",
+    "title": "GroupTravelBench: Benchmarking LLM Agents on Multi-Person Travel Planning",
+    "url": "https://arxiv.org/abs/2605.25200",
+    "published": "2026-05-24",
+    "updated": "2026-06-03",
+    "authors": [
+      "Xiang Cheng",
+      "Yulan Hu",
+      "Lulu Zheng",
+      "Zheng Pan",
+      "Xin Li",
+      "Yong Liu"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.25200",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.25200",
+    "pdf_url": "https://arxiv.org/pdf/2605.25200",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.24220",
+    "title": "Polar: Agentic RL on Any Harness at Scale",
+    "url": "https://arxiv.org/abs/2605.24220",
+    "published": "2026-05-22",
+    "updated": "2026-05-22",
+    "authors": [
+      "Binfeng Xu",
+      "Hao Zhang",
+      "Shaokun Zhang",
+      "Songyang Han",
+      "Mingjie Liu",
+      "Jian Hu",
+      "Shizhe Diao",
+      "Zhenghui Jin",
+      "Yunheng Zou",
+      "Michael Demoret",
+      "Jan Kautz",
+      "Yi Dong"
+    ],
+    "categories": [
+      "cs.DC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.24220",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.24220",
+    "pdf_url": "https://arxiv.org/pdf/2605.24220",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.23574",
+    "title": "Push Your Agent: Measuring and Enforcing Quantitative Goal Persistence in Long-Horizon LLM Agents",
+    "url": "https://arxiv.org/abs/2605.23574",
+    "published": "2026-05-22",
+    "updated": "2026-05-22",
+    "authors": [
+      "Yuandao Cai",
+      "Yuzhang Zhu",
+      "Liyou Gao",
+      "Wensheng Tang",
+      "Shengchao Qin"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.23574",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.23574",
+    "pdf_url": "https://arxiv.org/pdf/2605.23574",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.24069",
+    "title": "When the Manual Lies: A Realistic Benchmark to Evaluate MCP Poisoning Attacks for LLM Agents",
+    "url": "https://arxiv.org/abs/2605.24069",
+    "published": "2026-05-22",
+    "updated": "2026-05-22",
+    "authors": [
+      "Shi Liu",
+      "Xuehai Tang",
+      "Xikang Yang",
+      "Liang Lin",
+      "Biyu Zhou",
+      "Wenjie Xiao",
+      "Wantao Liu"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.24069",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.24069",
+    "pdf_url": "https://arxiv.org/pdf/2605.24069",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.24216",
+    "title": "Agent-ToM: Learning to Monitor Autonomous LLM Agents via Theory-of-Mind Reasoning",
+    "url": "https://arxiv.org/abs/2605.24216",
+    "published": "2026-05-22",
+    "updated": "2026-05-22",
+    "authors": [
+      "Nesreen K. Ahmed",
+      "Nima Nafisi"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.24216",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.24216",
+    "pdf_url": "https://arxiv.org/pdf/2605.24216",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.18421",
+    "title": "EvoMemBench: Benchmarking Agent Memory from a Self-Evolving Perspective",
+    "url": "https://arxiv.org/abs/2605.18421",
+    "published": "2026-05-18",
+    "updated": "2026-06-15",
+    "authors": [
+      "Yuyao Wang",
+      "Zhongjian Zhang",
+      "Mo Chi",
+      "Kaichi Yu",
+      "Yuhan Li",
+      "Miao Peng",
+      "Bing Tong",
+      "Chen Zhang",
+      "Yan Zhou",
+      "Jia Li"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.18421",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.18421",
+    "pdf_url": "https://arxiv.org/pdf/2605.18421",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.10779",
+    "title": "LITMUS: Benchmarking Behavioral Jailbreaks of LLM Agents in Real OS Environments",
+    "url": "https://arxiv.org/abs/2605.10779",
+    "published": "2026-05-11",
+    "updated": "2026-05-11",
+    "authors": [
+      "Chiyu Zhang",
+      "Huiqin Yang",
+      "Bendong Jiang",
+      "Xiaolei Zhang",
+      "Yiran Zhao",
+      "Ruyi Chen",
+      "Lu Zhou",
+      "Xiaogang Xu",
+      "Jiafei Wu",
+      "Liming Fang",
+      "Zhe Liu"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.10779",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.10779",
+    "pdf_url": "https://arxiv.org/pdf/2605.10779",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.03312",
+    "title": "MemFlow: Intent-Driven Memory Orchestration for Small Language Model Agents",
+    "url": "https://arxiv.org/abs/2605.03312",
+    "published": "2026-05-05",
+    "updated": "2026-05-05",
+    "authors": [
+      "Jiayi Chen",
+      "Yingcong Li",
+      "Guiling Wang"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.03312",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.03312",
+    "pdf_url": "https://arxiv.org/pdf/2605.03312",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.01101",
+    "title": "Virtual Speech Therapist: A Clinician-in-the-Loop AI Speech Therapy Agent for Personalized and Supervised Therapy",
+    "url": "https://arxiv.org/abs/2605.01101",
+    "published": "2026-05-01",
+    "updated": "2026-06-15",
+    "authors": [
+      "Shakeel Sheikh",
+      "Patrick Marmaroli",
+      "MD Sahidullah",
+      "Slim Ouni",
+      "Fabrice Hirsch",
+      "Goncalo Leal",
+      "Bjorn W Schuller"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.SD",
+      "eess.AS"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.01101",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.01101",
+    "pdf_url": "https://arxiv.org/pdf/2605.01101",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2604.19844",
+    "title": "If you're waiting for a sign... that might not be it! Mitigating Trust Boundary Confusion from Visual Injections on Vision-Language Agentic Systems",
+    "url": "https://arxiv.org/abs/2604.19844",
+    "published": "2026-04-21",
+    "updated": "2026-04-21",
+    "authors": [
+      "Jiamin Chang",
+      "Minhui Xue",
+      "Ruoxi Sun",
+      "Shuchao Pang",
+      "Salil S. Kanhere",
+      "Hammond Pearce"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "embodied-agent",
+      "multi-agent"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2604.19844",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.19844",
+    "pdf_url": "https://arxiv.org/pdf/2604.19844",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2604.18658",
+    "title": "Owner-Harm: A Missing Threat Model for AI Agent Safety",
+    "url": "https://arxiv.org/abs/2604.18658",
+    "published": "2026-04-20",
+    "updated": "2026-04-20",
+    "authors": [
+      "Dongcheng Zhang",
+      "Yiqing Jiang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.18658",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.18658",
+    "pdf_url": "https://arxiv.org/pdf/2604.18658",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.17562",
+    "title": "SafeAgent: A Runtime Protection Architecture for Agentic Systems",
+    "url": "https://arxiv.org/abs/2604.17562",
+    "published": "2026-04-19",
+    "updated": "2026-04-19",
+    "authors": [
+      "Hailin Liu",
+      "Eugene Ilyushin",
+      "Jie Ni",
+      "Min Zhu"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "memory",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.17562",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.17562",
+    "pdf_url": "https://arxiv.org/pdf/2604.17562",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.00623",
+    "title": "TraceSIR: A Multi-Agent Framework for Structured Analysis and Reporting of Agentic Execution Traces",
+    "url": "https://arxiv.org/abs/2603.00623",
+    "published": "2026-02-28",
+    "updated": "2026-02-28",
+    "authors": [
+      "Shu-Xun Yang",
+      "Cunxiang Wang",
+      "Haoke Zhang",
+      "Wenbo Yu",
+      "Lindong Wu",
+      "Jiayi Gui",
+      "Dayong Yang",
+      "Yukuo Cen",
+      "Zhuoer Feng",
+      "Bosi Wen",
+      "Yidong Wang",
+      "Lucen Zhong",
+      "Jiamin Ren",
+      "Linfeng Zhang",
+      "Jie Tang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2603.00623",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.00623",
+    "pdf_url": "https://arxiv.org/pdf/2603.00623",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2602.13530",
+    "title": "REMem: Reasoning with Episodic Memory in Language Agent",
+    "url": "https://arxiv.org/abs/2602.13530",
+    "published": "2026-02-13",
+    "updated": "2026-02-28",
+    "authors": [
+      "Yiheng Shu",
+      "Saisri Padmaja Jonnalagedda",
+      "Xiang Gao",
+      "Bernal Jiménez Gutiérrez",
+      "Weijian Qi",
+      "Kamalika Das",
+      "Huan Sun",
+      "Yu Su"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2602.13530",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.13530",
+    "pdf_url": "https://arxiv.org/pdf/2602.13530",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2510.03847",
+    "title": "Small Language Models for Agentic Systems: A Survey of Architectures, Capabilities, and Deployment Trade offs",
+    "url": "https://arxiv.org/abs/2510.03847",
+    "published": "2025-10-04",
+    "updated": "2025-10-04",
+    "authors": [
+      "Raghav Sharma",
+      "Manan Mehta"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 19,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2510.03847",
+    "source": "arxiv",
+    "source_id": "arxiv:2510.03847",
+    "pdf_url": "https://arxiv.org/pdf/2510.03847",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2607.06157",
+    "title": "LLM Agents for Deliberative Collaboration: A Study on Joint Decision Making Under Partial Observability",
+    "url": "https://arxiv.org/abs/2607.06157",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Chenxu Wang",
+      "Yongkun Yang",
+      "Boyuan Du",
+      "Shiwei Lin",
+      "Huaping Liu"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.06157",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06157",
+    "pdf_url": "https://arxiv.org/pdf/2607.06157",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.04686",
+    "title": "ToolFailBench: Diagnosing Tool-Use Failures in LLM Agents",
+    "url": "https://arxiv.org/abs/2607.04686",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Harsh Soni"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "llm-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.04686",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04686",
+    "pdf_url": "https://arxiv.org/pdf/2607.04686",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.04240",
+    "title": "Biological Motifs for Agentic Control",
+    "url": "https://arxiv.org/abs/2607.04240",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Bogdan Banu"
+    ],
+    "categories": [
+      "cs.AI",
+      "q-bio.CB"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "autonomous-agent-llm",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.04240",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04240",
+    "pdf_url": "https://arxiv.org/pdf/2607.04240",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.03441",
+    "title": "No Time Like the Present: Agentic Test-Time Training for LLM Agents",
+    "url": "https://arxiv.org/abs/2607.03441",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Yanbo Wang",
+      "Jinhua Hao",
+      "Yuze Shi",
+      "Kun Yuan",
+      "Ming Sun"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "coding-agent",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent",
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.03441",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03441",
+    "pdf_url": "https://arxiv.org/pdf/2607.03441",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.01874",
+    "title": "SkillCoach: Self-Evolving Rubrics for Evaluating and Enhancing Agentic Skill-Use",
+    "url": "https://arxiv.org/abs/2607.01874",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Jiayin Zhu",
+      "Kelong Mao",
+      "Yudong Guo",
+      "Dengbo He",
+      "Sulong Xu",
+      "Simiu Gu",
+      "Yutao Yue"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.01874",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01874",
+    "pdf_url": "https://arxiv.org/pdf/2607.01874",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.01793",
+    "title": "Safety Testing LLM Agents at Scale: From Risk Discovery to Evidence-Grounded Verification",
+    "url": "https://arxiv.org/abs/2607.01793",
+    "published": "2026-07-02",
+    "updated": "2026-07-04",
+    "authors": [
+      "Yunhao Feng",
+      "Ruixiao Lin",
+      "Ming Wen",
+      "Qinqin He",
+      "Yanming Guo",
+      "Yifan Ding",
+      "Yutao Wu",
+      "Jialuo Chen",
+      "Zhuoer Xu",
+      "Xiaohu Du",
+      "Jianan Ma",
+      "Zixing Chen",
+      "Xingjun Ma",
+      "Yunhao Chen",
+      "Xinhao Deng"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.01793",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01793",
+    "pdf_url": "https://arxiv.org/pdf/2607.01793",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.02703",
+    "title": "LLMoxie: Exploring Agentic AI for Scientific Software Development",
+    "url": "https://arxiv.org/abs/2607.02703",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Landung Setiawan",
+      "Anant Mittal",
+      "Cordero Core",
+      "Anshul Tambay",
+      "Carlos Garcia Jurado Suarez",
+      "David A. C. Beck",
+      "Andrew J. Connolly",
+      "Vani Mandava"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.DC",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.02703",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02703",
+    "pdf_url": "https://arxiv.org/pdf/2607.02703",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.00627",
+    "title": "AGI Maze as a Benchmark Framework for World-Modeling Agents",
+    "url": "https://arxiv.org/abs/2607.00627",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Alexey Potapov"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.00627",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00627",
+    "pdf_url": "https://arxiv.org/pdf/2607.00627",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.02606",
+    "title": "ChainSWE: Benchmarking Coding Agents on Multi-Bug Software Maintenance",
+    "url": "https://arxiv.org/abs/2607.02606",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Qirui Jin",
+      "Lingching Tung",
+      "Kenan Li",
+      "Qiyang Shi",
+      "Yushi She",
+      "Huanzhong Jia",
+      "Harrison Zhao",
+      "Kejing Xia",
+      "Zhenbang Du",
+      "Yikai Zhang",
+      "Jiaxin Pei",
+      "Zhenyu Zhang",
+      "Zhen Qi",
+      "Yuyan Duan",
+      "Wenke Lee",
+      "Zijian Jin"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.02606",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02606",
+    "pdf_url": "https://arxiv.org/pdf/2607.02606",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.32025",
+    "title": "Generative Skill Composition for LLM Agents",
+    "url": "https://arxiv.org/abs/2606.32025",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Xinyu Zhao",
+      "Zhen Tan",
+      "Vaishnav Tadiparthi",
+      "Nakul Agarwal",
+      "Kwonjoon Lee",
+      "Ehsan Moradi Pari",
+      "Hossein Nourkhiz Mahjoub",
+      "Tianlong Chen"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent",
+      "llm-agent",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.32025",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.32025",
+    "pdf_url": "https://arxiv.org/pdf/2606.32025",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.31229",
+    "title": "Agentic-Ideation: Sample Efficient Agentic Trajectories Synthesis for Scientific Ideation Agents",
+    "url": "https://arxiv.org/abs/2606.31229",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Keyu Zhao",
+      "Lingyan Kong",
+      "Fengli Xu",
+      "Yong Li"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "computer-use",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.31229",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31229",
+    "pdf_url": "https://arxiv.org/pdf/2606.31229",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.31410",
+    "title": "Xiaomi-GUI-0 Technical Report",
+    "url": "https://arxiv.org/abs/2606.31410",
+    "published": "2026-06-30",
+    "updated": "2026-07-01",
+    "authors": [
+      "Wanxia Cao",
+      "Chengzhen Duan",
+      "Pei Fu",
+      "Pengzhi Gao",
+      "Niu Lian",
+      "Fazhan Liu",
+      "Hui Liu",
+      "Heng Qu",
+      "Qinzhuo Wu",
+      "Zhehao Yu",
+      "Tongbo Chen",
+      "Shiqi Cui",
+      "Anan Du",
+      "Shukai Jia",
+      "Yuanfa Li",
+      "Wei Liu",
+      "Yike Liu",
+      "Wenchao Lu",
+      "Zhenbo Luo",
+      "Haoyuan Sun",
+      "Jiatong Sun",
+      "Cheng Tan",
+      "Yajie Wang",
+      "Changqiao Wu",
+      "Tao Xiong",
+      "Jiahui Yang",
+      "Yuxuan Yuan",
+      "Ruoceng Zhang",
+      "Shaojie Zhang",
+      "Jian Zhu",
+      "Jian Luan",
+      "Cong Zou"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.31410",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31410",
+    "pdf_url": "https://arxiv.org/pdf/2606.31410",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.29178",
+    "title": "Selective Memory Retention for Long-Horizon LLM Agents",
+    "url": "https://arxiv.org/abs/2606.29178",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Pranath Reddy"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.29178",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29178",
+    "pdf_url": "https://arxiv.org/pdf/2606.29178",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.28456",
+    "title": "Is Lying an Emergent Behaviour in LLMs? Evidence from Gaslighting AI agents in a Sustainability Game",
+    "url": "https://arxiv.org/abs/2606.28456",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Subhendu Bhandary",
+      "Federico Carucci",
+      "Christos Charalambous",
+      "Francesca Dilisante",
+      "Ksenia Dvorkina",
+      "Anna Garbo",
+      "Jiaqi Liang",
+      "Riccardo Vasellini",
+      "Francesco Bertolotti"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "memory",
+      "multi-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.28456",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28456",
+    "pdf_url": "https://arxiv.org/pdf/2606.28456",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.27406",
+    "title": "Towards Evaluation of Implicit Software World Models in Coding LLMs",
+    "url": "https://arxiv.org/abs/2606.27406",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Egor Bogomolov",
+      "Yaroslav Zharov"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "memory",
+      "reasoning",
+      "world-model"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.27406",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27406",
+    "pdf_url": "https://arxiv.org/pdf/2606.27406",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.26627",
+    "title": "Agents That Know Too Much: A Data-Centric Survey of Privacy in LLM Agents",
+    "url": "https://arxiv.org/abs/2606.26627",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Nada Lahjouji",
+      "Ashwin Gerard Colaco"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.26627",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26627",
+    "pdf_url": "https://arxiv.org/pdf/2606.26627",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.26479",
+    "title": "Adaptive Evaluation of Out-of-Band Defenses Against Prompt Injection in LLM Agents",
+    "url": "https://arxiv.org/abs/2606.26479",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Praneeth Narisetty",
+      "Shiva Nagendra Babu Kore",
+      "Uday Kumar Reddy Kattamanchi",
+      "Jayaram Kumarapu"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.26479",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26479",
+    "pdf_url": "https://arxiv.org/pdf/2606.26479",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.26793",
+    "title": "MIRROR: Novelty-Constrained Memory-Guided MCTS Red-Teaming for Agentic RAG",
+    "url": "https://arxiv.org/abs/2606.26793",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Inderjeet Singh",
+      "Andrés Murillo",
+      "Motoyoshi Sekiya",
+      "Yuki Unno",
+      "Junichi Suga"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.26793",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26793",
+    "pdf_url": "https://arxiv.org/pdf/2606.26793",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.27472",
+    "title": "Supersede: Diagnosing and Training the Memory-Update Gap in LLM Agents",
+    "url": "https://arxiv.org/abs/2606.27472",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Vedant Patel"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.27472",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27472",
+    "pdf_url": "https://arxiv.org/pdf/2606.27472",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.25622",
+    "title": "Probabilistic Agents in Deterministic Audits: Evaluating Multi-Agent Systems for Automated Audits Based on the German IT-Grundschutz",
+    "url": "https://arxiv.org/abs/2606.25622",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Lea Roxanne Muth",
+      "Marian Margraf"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.25622",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25622",
+    "pdf_url": "https://arxiv.org/pdf/2606.25622",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.25358",
+    "title": "Agentic Knowledge Tracing: A Multi-Agent LLM Architecture for Stealth Assessment of Financial Literacy in Serious Games",
+    "url": "https://arxiv.org/abs/2606.25358",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Gabriel Santos",
+      "Rita Julia",
+      "Marcelo Nascimento"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.25358",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25358",
+    "pdf_url": "https://arxiv.org/pdf/2606.25358",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.25334",
+    "title": "Bridging the Post-discharge Gap: A Traceable Multi-agent Framework for Safe and Continuous Care",
+    "url": "https://arxiv.org/abs/2606.25334",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Runwei Guan",
+      "Yi Zhou",
+      "Heyi Lin",
+      "Jinjing Zhu",
+      "Mingyuan Hou",
+      "Yang Yang",
+      "Fang Yuan",
+      "Xiaohong Lin",
+      "Shaofeng Liang",
+      "Xuming Hu",
+      "Tao Li",
+      "Tianbin Zhao",
+      "Yutao Yue",
+      "Zhiyuan Wang",
+      "Hui Xiong"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "multi-agent",
+      "rag"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.25334",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25334",
+    "pdf_url": "https://arxiv.org/pdf/2606.25334",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.22673",
+    "title": "AgentLens: Interpretable Safety Steering via Mechanistic Subspaces for Multi-Turn Coding Agent",
+    "url": "https://arxiv.org/abs/2606.22673",
+    "published": "2026-06-21",
+    "updated": "2026-06-21",
+    "authors": [
+      "Weidi Luo",
+      "Qiming Zhang",
+      "Yihao Quan",
+      "Mingyu Jin",
+      "Jie Cai",
+      "Chaowei Xiao",
+      "Jingcheng Niu",
+      "Zhen Xiang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.22673",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22673",
+    "pdf_url": "https://arxiv.org/pdf/2606.22673",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.21710",
+    "title": "PrivacyAlign: Contextual Privacy Alignment for LLM Agents",
+    "url": "https://arxiv.org/abs/2606.21710",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Manveer Singh Tamber",
+      "Abhay Puri",
+      "Marc-Etienne Brunet",
+      "Perouz Taslakian",
+      "Jimmy Lin",
+      "Spandana Gella"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.21710",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21710",
+    "pdf_url": "https://arxiv.org/pdf/2606.21710",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.21013",
+    "title": "Agentic Time Machine as an Infrastructure for Future-Event Forecasting",
+    "url": "https://arxiv.org/abs/2606.21013",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Jingyi Chai",
+      "Bingyang Zheng",
+      "Xiangrui Liu",
+      "Hao Lu",
+      "Zihang Zhou",
+      "Tianchen Wang",
+      "Kemeng Zhang",
+      "Siheng Chen"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.21013",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21013",
+    "pdf_url": "https://arxiv.org/pdf/2606.21013",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.21123",
+    "title": "A Multi-Agent Audit Framework for High-Stakes Reasoning: Evaluation and Interpretability in Clinical Mental Health Screening",
+    "url": "https://arxiv.org/abs/2606.21123",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Jingchen Ye",
+      "Yanpei Yu",
+      "Luyao Zhang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.21123",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21123",
+    "pdf_url": "https://arxiv.org/pdf/2606.21123",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.19899",
+    "title": "Measuring Biological Capabilities and Risks of AI Agents",
+    "url": "https://arxiv.org/abs/2606.19899",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Patricia Paskov",
+      "Jeffrey Lee",
+      "Kyle Brady",
+      "Alyssa Worland"
+    ],
+    "categories": [
+      "cs.CY",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "agentic-ai",
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.19899",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19899",
+    "pdf_url": "https://arxiv.org/pdf/2606.19899",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.20243",
+    "title": "Phoenix: Safe GitHub Issue Resolution via Multi-Agent LLMs",
+    "url": "https://arxiv.org/abs/2606.20243",
+    "published": "2026-06-18",
+    "updated": "2026-06-22",
+    "authors": [
+      "Kipngeno Koech",
+      "Muhammad Adam",
+      "Baimam Boukar Jean Jacques",
+      "Joao Barros"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "multi-agent",
+      "planning",
+      "rag"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.20243",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20243",
+    "pdf_url": "https://arxiv.org/pdf/2606.20243",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.18950",
+    "title": "RTSGameBench: An RTS Benchmark for Strategic Reasoning by Vision-Language Models",
+    "url": "https://arxiv.org/abs/2606.18950",
+    "published": "2026-06-17",
+    "updated": "2026-06-18",
+    "authors": [
+      "San Kim",
+      "Daechul Ahn",
+      "Reokyoung Kim",
+      "Hyeonbeom Choi",
+      "Seungyeon Jwa",
+      "Jonghyun Choi"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.18950",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18950",
+    "pdf_url": "https://arxiv.org/pdf/2606.18950",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.16659",
+    "title": "FraudSMSWalker: Benchmarking Agentic Large Language Models for SMS-to-Webpage Fraud Detection",
+    "url": "https://arxiv.org/abs/2606.16659",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Y. H. Zhou",
+      "Z. M. Ma",
+      "Y. J. Zhou",
+      "Y. T. Li",
+      "H. X. Xiang",
+      "Y. M. Cheng",
+      "T. L. Chen",
+      "K. J. Zhang",
+      "Z. H. Nan",
+      "J. H. Ni",
+      "Z. Wu",
+      "Q. Y. Pan",
+      "S. Zhang",
+      "S. Cheng",
+      "M. Y. Luo"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.16659",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16659",
+    "pdf_url": "https://arxiv.org/pdf/2606.16659",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.17041",
+    "title": "Benchmarking LLM Agents on Meta-Analysis Articles from Nature Portfolio",
+    "url": "https://arxiv.org/abs/2606.17041",
+    "published": "2026-06-15",
+    "updated": "2026-07-01",
+    "authors": [
+      "Anzhe Xie",
+      "Weihang Su",
+      "Yujia Zhou",
+      "Yiqun Liu",
+      "Qingyao Ai"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.17041",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17041",
+    "pdf_url": "https://arxiv.org/pdf/2606.17041",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.16576",
+    "title": "Can LLM Agents Infer World Models? Evidence from Agentic Automata Learning",
+    "url": "https://arxiv.org/abs/2606.16576",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Reef Menaged",
+      "Gili Lior",
+      "Shauli Ravfogel",
+      "Roee Aharoni",
+      "Gabriel Stanovsky"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.16576",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16576",
+    "pdf_url": "https://arxiv.org/pdf/2606.16576",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.12780",
+    "title": "ProPlay: Procedural World Models for Self-Evolving LLM Agents",
+    "url": "https://arxiv.org/abs/2606.12780",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Yijun Ma",
+      "Zehong Wang",
+      "Yiyang Li",
+      "Ziming Li",
+      "Xiaoguang Guo",
+      "Weixiang Sun",
+      "Chuxu Zhang",
+      "Yanfang Ye"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.12780",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12780",
+    "pdf_url": "https://arxiv.org/pdf/2606.12780",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.12320",
+    "title": "A Five-Plane Reference Architecture for Runtime Governance of Production AI Agents",
+    "url": "https://arxiv.org/abs/2606.12320",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Krti Tallam"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CC",
+      "cs.CR",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.12320",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12320",
+    "pdf_url": "https://arxiv.org/pdf/2606.12320",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.11680",
+    "title": "Organize then Retrieve: Hierarchical Memory Navigation for Efficient Agents",
+    "url": "https://arxiv.org/abs/2606.11680",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Hao-Lun Hsu",
+      "Nikki Lijing Kuang",
+      "Boyi Liu",
+      "Zhewei Yao",
+      "Yuxiong He"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.11680",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.11680",
+    "pdf_url": "https://arxiv.org/pdf/2606.11680",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.12657",
+    "title": "TrajGenAgent: A Hierarchical LLM Agent for Human Mobility Trajectory Generation",
+    "url": "https://arxiv.org/abs/2606.12657",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Siyu Li",
+      "Toan Tran",
+      "Lingyi Zhao",
+      "Khurram Shafique",
+      "Li Xiong"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.DB",
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.12657",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12657",
+    "pdf_url": "https://arxiv.org/pdf/2606.12657",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.10677",
+    "title": "Infini Memory: Maintainable Topic Documents for Long-Term LLM Agent Memory",
+    "url": "https://arxiv.org/abs/2606.10677",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Suozhao Ji",
+      "Baodong Wu",
+      "Zehao Wang",
+      "Lei Xia",
+      "Qingping Li",
+      "Ruisong Wang",
+      "Wenbo Ding",
+      "Zhenhua Zhu",
+      "Boxun Li",
+      "Guohao Dai",
+      "Yu Wang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.10677",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10677",
+    "pdf_url": "https://arxiv.org/pdf/2606.10677",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.11354",
+    "title": "A Zero-Shot Multi-Agent Framework for Human-Building Interaction via Programmatic Reasoning",
+    "url": "https://arxiv.org/abs/2606.11354",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Yuqi Wang",
+      "Gulai Shen",
+      "Ali Mehmani"
+    ],
+    "categories": [
+      "cs.ET"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.11354",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.11354",
+    "pdf_url": "https://arxiv.org/pdf/2606.11354",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.05558",
+    "title": "Autoregressive Diffusion World Models for Off-Policy Evaluation of LLM Agents",
+    "url": "https://arxiv.org/abs/2606.05558",
+    "published": "2026-06-04",
+    "updated": "2026-06-04",
+    "authors": [
+      "Kaixuan Liu",
+      "Guojun Xiong",
+      "Weinan Zhang",
+      "Shengpu Tang"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.05558",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.05558",
+    "pdf_url": "https://arxiv.org/pdf/2606.05558",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.04555",
+    "title": "Temporal Order Matters for Agentic Memory: Segment Trees for Long-Horizon Agents",
+    "url": "https://arxiv.org/abs/2606.04555",
+    "published": "2026-06-03",
+    "updated": "2026-06-03",
+    "authors": [
+      "Yifan Simon Liu",
+      "Liam Gallagher",
+      "Faeze Moradi Kalarde",
+      "Jiazhou Liang",
+      "Armin Toroghi",
+      "Scott Sanner"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.04555",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.04555",
+    "pdf_url": "https://arxiv.org/pdf/2606.04555",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.03895",
+    "title": "Agent libOS: A Runtime Substrate for Capability-Controlled Self-Evolving LLM Agents",
+    "url": "https://arxiv.org/abs/2606.03895",
+    "published": "2026-06-02",
+    "updated": "2026-06-29",
+    "authors": [
+      "Yingqi Zhang"
+    ],
+    "categories": [
+      "cs.OS",
+      "cs.AI",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "planning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.03895",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.03895",
+    "pdf_url": "https://arxiv.org/pdf/2606.03895",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.02302",
+    "title": "SeClaw: Spec-Driven Security Task Synthesis for Evaluating Autonomous Agents",
+    "url": "https://arxiv.org/abs/2606.02302",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Hao Cheng",
+      "Changtao Miao",
+      "Tianle Song",
+      "Yin Wu",
+      "He Liu",
+      "Erjia Xiao",
+      "Junchi Chen",
+      "Xiaoyu Shi",
+      "Yichi Wang",
+      "Jing Yang",
+      "Taowen Wang",
+      "Jinhao Duan",
+      "Mengshu Sun",
+      "Peiyan Dong",
+      "Xuan Shen",
+      "Yang Cao",
+      "Renjing Xu",
+      "Kaidi Xu",
+      "Jindong Gu",
+      "Bo Zhang",
+      "Jize Zhang",
+      "Chenhao Lin",
+      "Philip Torr",
+      "Chao Shen"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.02302",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.02302",
+    "pdf_url": "https://arxiv.org/pdf/2606.02302",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.30711",
+    "title": "SAGE: A Novelty Gate for Efficient Memory Evolution in Agentic LLMs",
+    "url": "https://arxiv.org/abs/2605.30711",
+    "published": "2026-05-29",
+    "updated": "2026-06-18",
+    "authors": [
+      "Sijia Wang",
+      "Dhanajit Brahma",
+      "Ricardo Henao"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.LG",
+      "stat.ML"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.30711",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.30711",
+    "pdf_url": "https://arxiv.org/pdf/2605.30711",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.29341",
+    "title": "WorldMemArena: Evaluating Multimodal Agent Memory Through Action-World Interaction",
+    "url": "https://arxiv.org/abs/2605.29341",
+    "published": "2026-05-28",
+    "updated": "2026-06-01",
+    "authors": [
+      "Chengzhi Liu",
+      "Yuzhe Yang",
+      "Sophia Xiao Pu",
+      "Yepeng Liu",
+      "Lin Long",
+      "Yichen Guo",
+      "Nuo Chen",
+      "Zhaotian Weng",
+      "Elena Kochkina",
+      "Simerjot Kaur",
+      "Charese Smiley",
+      "Xiaomo Liu",
+      "James Zou",
+      "Sheng Liu",
+      "Yuheng Bu",
+      "Songyou Peng",
+      "Xin Eric Wang"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.29341",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.29341",
+    "pdf_url": "https://arxiv.org/pdf/2605.29341",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.27690",
+    "title": "TRACES: Proactive Safety Auditing for Multi-Turn LLM Agents via Trajectory-State Modeling",
+    "url": "https://arxiv.org/abs/2605.27690",
+    "published": "2026-05-26",
+    "updated": "2026-05-26",
+    "authors": [
+      "Jiaqian Li",
+      "Yanshu Li",
+      "Boxuan Zhang",
+      "Ruixiang Tang",
+      "Kuan-Hao Huang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.27690",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.27690",
+    "pdf_url": "https://arxiv.org/pdf/2605.27690",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.25141",
+    "title": "LLM Agent Based Renewable Energy Forecasting Using Edge and IoT Data A Review of Solar Wind Weather and Grid Aware Decision Support",
+    "url": "https://arxiv.org/abs/2605.25141",
+    "published": "2026-05-24",
+    "updated": "2026-05-24",
+    "authors": [
+      "Pavan Manjunath",
+      "Thomas Pruefer"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.25141",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.25141",
+    "pdf_url": "https://arxiv.org/pdf/2605.25141",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.19952",
+    "title": "Rethinking How to Remember: Beyond Atomic Facts in Lifelong LLM Agent Memory",
+    "url": "https://arxiv.org/abs/2605.19952",
+    "published": "2026-05-19",
+    "updated": "2026-05-19",
+    "authors": [
+      "Jingwei Sun",
+      "Jianing Zhu",
+      "Jiangchao Yao",
+      "Tongliang Liu",
+      "Bo Han"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.19952",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.19952",
+    "pdf_url": "https://arxiv.org/pdf/2605.19952",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.17625",
+    "title": "Episodic-Semantic Memory Architecture for Long-Horizon Scientific Agents",
+    "url": "https://arxiv.org/abs/2605.17625",
+    "published": "2026-05-17",
+    "updated": "2026-05-17",
+    "authors": [
+      "Nikola Milosevic"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.17625",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.17625",
+    "pdf_url": "https://arxiv.org/pdf/2605.17625",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.16821",
+    "title": "Multi-Paradigm Agent Interaction in Practice:A Systematic Analysis of Generator-Evaluator, ReAct Loop,and Adversarial Evaluation in the buddyMe Framework",
+    "url": "https://arxiv.org/abs/2605.16821",
+    "published": "2026-05-16",
+    "updated": "2026-05-16",
+    "authors": [
+      "Xiaohua Wang",
+      "Chao Han",
+      "Kai Yu",
+      "XiaoLiang Xu",
+      "Liang Wang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "multi-agent",
+      "planning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.16821",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.16821",
+    "pdf_url": "https://arxiv.org/pdf/2605.16821",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.16481",
+    "title": "Visual Agentic Memory: Enabling Online Long Video Understanding via Online Indexing, Hierarchical Memory, and Agentic Retrieval",
+    "url": "https://arxiv.org/abs/2605.16481",
+    "published": "2026-05-15",
+    "updated": "2026-05-15",
+    "authors": [
+      "Aiden Yiliu Li",
+      "Nels Numan",
+      "Anthony Steed"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.16481",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.16481",
+    "pdf_url": "https://arxiv.org/pdf/2605.16481",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.12061",
+    "title": "SAGE: A Self-Evolving Agentic Graph-Memory Engine for Structure-Aware Associative Memory",
+    "url": "https://arxiv.org/abs/2605.12061",
+    "published": "2026-05-12",
+    "updated": "2026-05-12",
+    "authors": [
+      "Juntong Wang",
+      "Haoyue Zhao",
+      "guanghui Pan",
+      "Xiyuan Wang",
+      "Yanbo Wang",
+      "Qiyan Deng",
+      "Muhan Zhang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.12061",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.12061",
+    "pdf_url": "https://arxiv.org/pdf/2605.12061",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.06890",
+    "title": "Beyond the Black Box: Interpretability of Agentic AI Tool Use",
+    "url": "https://arxiv.org/abs/2605.06890",
+    "published": "2026-05-07",
+    "updated": "2026-07-05",
+    "authors": [
+      "Hariom Tatsat",
+      "Ariye Shater"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2605.06890",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.06890",
+    "pdf_url": "https://arxiv.org/pdf/2605.06890",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2605.05716",
+    "title": "More Is Not Always Better: Cross-Component Interference in LLM Agent Scaffolding",
+    "url": "https://arxiv.org/abs/2605.05716",
+    "published": "2026-05-07",
+    "updated": "2026-05-07",
+    "authors": [
+      "Ming Liu"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.05716",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.05716",
+    "pdf_url": "https://arxiv.org/pdf/2605.05716",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.06716",
+    "title": "From Storage to Experience: A Survey on the Evolution of LLM Agent Memory Mechanisms",
+    "url": "https://arxiv.org/abs/2605.06716",
+    "published": "2026-05-07",
+    "updated": "2026-05-07",
+    "authors": [
+      "Jinghao Luo",
+      "Yuchen Tian",
+      "Chuxue Cao",
+      "Ziyang Luo",
+      "Hongzhan Lin",
+      "Kaixin Li",
+      "Chuyi Kong",
+      "Ruichao Yang",
+      "Jing Ma"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.06716",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.06716",
+    "pdf_url": "https://arxiv.org/pdf/2605.06716",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2604.25318",
+    "title": "Cutscene Agent: An LLM Agent Framework for Automated 3D Cutscene Generation",
+    "url": "https://arxiv.org/abs/2604.25318",
+    "published": "2026-04-28",
+    "updated": "2026-04-28",
+    "authors": [
+      "Lanshan He",
+      "Haozhou Pang",
+      "Qi Gan",
+      "Xin Shen",
+      "Ziwei Zhang",
+      "Yibo Liu",
+      "Gang Fang",
+      "Bo Liu",
+      "Kai Sheng",
+      "Shengfeng Zeng",
+      "Chaofan Li",
+      "Zhen Hui",
+      "Keer Zhou",
+      "Lan Zhou",
+      "Shujun Dai"
+    ],
+    "categories": [
+      "cs.GR",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2604.25318",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.25318",
+    "pdf_url": "https://arxiv.org/pdf/2604.25318",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2604.25135",
+    "title": "FAMA: Failure-Aware Meta-Agentic Framework for Open-Source LLMs in Interactive Tool Use Environments",
+    "url": "https://arxiv.org/abs/2604.25135",
+    "published": "2026-04-28",
+    "updated": "2026-04-28",
+    "authors": [
+      "Amir Saeidi",
+      "Venkatesh Mishra",
+      "Souradeep Mukhopadhyay",
+      "Gaowen Liu",
+      "Ali Payani",
+      "Jayanth Srinivasa",
+      "Chitta Baral"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2604.25135",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.25135",
+    "pdf_url": "https://arxiv.org/pdf/2604.25135",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2604.23459",
+    "title": "Architecture Matters for Multi-Agent Security",
+    "url": "https://arxiv.org/abs/2604.23459",
+    "published": "2026-04-25",
+    "updated": "2026-04-25",
+    "authors": [
+      "Ben Hagag",
+      "William L. Anderson",
+      "Christian Schroeder de Witt",
+      "Sarah Scheffler"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.CR",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "memory",
+      "multi-agent",
+      "planning"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.23459",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.23459",
+    "pdf_url": "https://arxiv.org/pdf/2604.23459",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.23374",
+    "title": "Ghost in the Agent: Redefining Information Flow Tracking for LLM Agents",
+    "url": "https://arxiv.org/abs/2604.23374",
+    "published": "2026-04-25",
+    "updated": "2026-04-25",
+    "authors": [
+      "Yuandao Cai",
+      "Wensheng Tang",
+      "Cheng Wen",
+      "Shengchao Qin"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.23374",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.23374",
+    "pdf_url": "https://arxiv.org/pdf/2604.23374",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.22879",
+    "title": "Beyond Single-Agent Alignment: Preventing Context-Fragmented Violations in Multi-Agent Systems",
+    "url": "https://arxiv.org/abs/2604.22879",
+    "published": "2026-04-24",
+    "updated": "2026-04-24",
+    "authors": [
+      "Jie Wu",
+      "Ming Gong"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.AI",
+      "cs.CR",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "rag",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.22879",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.22879",
+    "pdf_url": "https://arxiv.org/pdf/2604.22879",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.18847",
+    "title": "Human-Guided Harm Recovery for Computer Use Agents",
+    "url": "https://arxiv.org/abs/2604.18847",
+    "published": "2026-04-20",
+    "updated": "2026-05-28",
+    "authors": [
+      "Christy Li",
+      "Sky CH-Wang",
+      "Andi Peng",
+      "Andreea Bobu"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.18847",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.18847",
+    "pdf_url": "https://arxiv.org/pdf/2604.18847",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.18245",
+    "title": "Who Tests the Testers? Systematic Enumeration and Coverage Audit of LLM Agent Tool Call Safety",
+    "url": "https://arxiv.org/abs/2603.18245",
+    "published": "2026-03-18",
+    "updated": "2026-03-18",
+    "authors": [
+      "Xuan Chen",
+      "Lu Yan",
+      "Ruqi Zhang",
+      "Xiangyu Zhang"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.18245",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.18245",
+    "pdf_url": "https://arxiv.org/pdf/2603.18245",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.07980",
+    "title": "\\$OneMillion-Bench: How Far are Language Agents from Human Experts?",
+    "url": "https://arxiv.org/abs/2603.07980",
+    "published": "2026-03-09",
+    "updated": "2026-03-09",
+    "authors": [
+      "Qianyu Yang",
+      "Yang Liu",
+      "Jiaqi Li",
+      "Jun Bai",
+      "Hao Chen",
+      "Kaiyuan Chen",
+      "Tiliang Duan",
+      "Jiayun Dong",
+      "Xiaobo Hu",
+      "Zixia Jia",
+      "Yang Liu",
+      "Tao Peng",
+      "Yixin Ren",
+      "Ran Tian",
+      "Zaiyuan Wang",
+      "Yanglihong Xiao",
+      "Gang Yao",
+      "Lingyue Yin",
+      "Ge Zhang",
+      "Chun Zhang",
+      "Jianpeng Jiao",
+      "Zilong Zheng",
+      "Yuan Gong"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.07980",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.07980",
+    "pdf_url": "https://arxiv.org/pdf/2603.07980",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.09002",
+    "title": "Security Considerations for Multi-agent Systems",
+    "url": "https://arxiv.org/abs/2603.09002",
+    "published": "2026-03-09",
+    "updated": "2026-04-26",
+    "authors": [
+      "Tam Nguyen",
+      "Moses Ndebugre",
+      "Dheeraj Arremsetty"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.09002",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.09002",
+    "pdf_url": "https://arxiv.org/pdf/2603.09002",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2602.07962",
+    "title": "LOCA-bench: Benchmarking Language Agents Under Controllable and Extreme Context Growth",
+    "url": "https://arxiv.org/abs/2602.07962",
+    "published": "2026-02-08",
+    "updated": "2026-02-08",
+    "authors": [
+      "Weihao Zeng",
+      "Yuzhen Huang",
+      "Junxian He"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2602.07962",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.07962",
+    "pdf_url": "https://arxiv.org/pdf/2602.07962",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2602.05302",
+    "title": "PieArena: Ranking and Profiling Language Agents in Realistic Negotiation Scenarios",
+    "url": "https://arxiv.org/abs/2602.05302",
+    "published": "2026-02-05",
+    "updated": "2026-06-01",
+    "authors": [
+      "Chris Zhu",
+      "Sasha Cui",
+      "Will Sanok Dufallo",
+      "Runzhi Jin",
+      "Zhen Xu",
+      "Linjun Zhang",
+      "Daylian Cain"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2602.05302",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.05302",
+    "pdf_url": "https://arxiv.org/pdf/2602.05302",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2601.06007",
+    "title": "Don't Break the Cache: An Evaluation of Prompt Caching for Long-Horizon Agentic Tasks",
+    "url": "https://arxiv.org/abs/2601.06007",
+    "published": "2026-01-09",
+    "updated": "2026-01-31",
+    "authors": [
+      "Elias Lumer",
+      "Faheem Nizar",
+      "Akshaya Jangiti",
+      "Kevin Frank",
+      "Anmol Gulati",
+      "Mandar Phadate",
+      "Vamse Kumar Subbiah"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2601.06007",
+    "source": "arxiv",
+    "source_id": "arxiv:2601.06007",
+    "pdf_url": "https://arxiv.org/pdf/2601.06007",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2511.04847",
+    "title": "Test-Time Adaptation for LLM Agents via Environment Interaction",
+    "url": "https://arxiv.org/abs/2511.04847",
+    "published": "2025-11-06",
+    "updated": "2026-02-22",
+    "authors": [
+      "Arthur Chen",
+      "Zuxin Liu",
+      "Jianguo Zhang",
+      "Akshara Prabhakar",
+      "Zhiwei Liu",
+      "Shelby Heinecke",
+      "Silvio Savarese",
+      "Victor Zhong",
+      "Caiming Xiong"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "embodied-agent",
+      "rag",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2511.04847",
+    "source": "arxiv",
+    "source_id": "arxiv:2511.04847",
+    "pdf_url": "https://arxiv.org/pdf/2511.04847",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2509.10769",
+    "title": "AgentArch: A Comprehensive Benchmark to Evaluate Agent Architectures in Enterprise",
+    "url": "https://arxiv.org/abs/2509.10769",
+    "published": "2025-09-13",
+    "updated": "2026-01-06",
+    "authors": [
+      "Tara Bogavelli",
+      "Roshnee Sharma",
+      "Hari Subramani"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 18,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2509.10769",
+    "source": "arxiv",
+    "source_id": "arxiv:2509.10769",
+    "pdf_url": "https://arxiv.org/pdf/2509.10769",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2607.06273",
+    "title": "AgentTether: Graph-Guided Diagnosis and Runtime Intervention for Reliable LLM Agent Operation",
+    "url": "https://arxiv.org/abs/2607.06273",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Chenyu Zhao",
+      "Shenglin Zhang",
+      "Wenwei Gu",
+      "Yongqian Sun",
+      "Dan Pei",
+      "Chetan Bansal",
+      "Saravan Rajmohan",
+      "Minghua Ma"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.06273",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06273",
+    "pdf_url": "https://arxiv.org/pdf/2607.06273",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.06195",
+    "title": "LogicHunter: Testing LLM Agent Frameworks with an Agentic Oracle",
+    "url": "https://arxiv.org/abs/2607.06195",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Minghui Long",
+      "Yanjie Zhao",
+      "Haoyu Wang"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.06195",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06195",
+    "pdf_url": "https://arxiv.org/pdf/2607.06195",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.06080",
+    "title": "From Blueprint to Reality: Modeling and Applying Putnam's Social Capital Theory with LLM-based Multi-agent Simulations",
+    "url": "https://arxiv.org/abs/2607.06080",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Shiyi Ling",
+      "Zhi Zheng",
+      "Hui Zheng",
+      "Wenjun Xue",
+      "Feng Ye",
+      "Tong Xu"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.SI"
+    ],
+    "topics": [
+      "agent-safety",
+      "multi-agent",
+      "rag",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.06080",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06080",
+    "pdf_url": "https://arxiv.org/pdf/2607.06080",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.05297",
+    "title": "MetaSkill-Evolve: Recursive Self-Improvement of LLM Agents via Two-Timescale Meta-Skill Evolution",
+    "url": "https://arxiv.org/abs/2607.05297",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Zefeng Wang",
+      "Minxi Yan",
+      "Jinhe Bi",
+      "Sikuan Yan",
+      "Volker Tresp",
+      "Yunpu Ma"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.05297",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05297",
+    "pdf_url": "https://arxiv.org/pdf/2607.05297",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.04528",
+    "title": "Measuring Harness-Induced Belief Divergence in Multi-Step LLM Agents",
+    "url": "https://arxiv.org/abs/2607.04528",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Haiwen Yi",
+      "Xinyuan Song"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.04528",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04528",
+    "pdf_url": "https://arxiv.org/pdf/2607.04528",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.04293",
+    "title": "CausalGame: Benchmarking Causal Thinking of LLM Agents in Games",
+    "url": "https://arxiv.org/abs/2607.04293",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Zhenhao Chen",
+      "Yongqiang Chen",
+      "Chenxi Liu",
+      "Junchi Yu",
+      "Xiangchen Song",
+      "Zijian Li",
+      "Jialin Li",
+      "Philip Torr",
+      "Bo Han",
+      "Kun Zhang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.LG",
+      "stat.ML"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "reasoning"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.04293",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04293",
+    "pdf_url": "https://arxiv.org/pdf/2607.04293",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.04426",
+    "title": "ACE-Brain-0.5: A Unified Embodied Foundational Model for Physical Agentic AI",
+    "url": "https://arxiv.org/abs/2607.04426",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "ACE-Brain Team",
+      ":",
+      "Ziyang Gong",
+      "Haoming Gu",
+      "Zehang Luo",
+      "Tianyi Zhang",
+      "Tao Tao",
+      "Yixiao Chi",
+      "Zhe Liu",
+      "Lingsi Zhu",
+      "Jingyuan Liu",
+      "Anke Tang",
+      "Songze Li",
+      "Yilun Kong",
+      "Ningjing Liu",
+      "Tianyu Zhu",
+      "Yunpeng Qing",
+      "Shuang Luo",
+      "Xiang Liu",
+      "Shi Fu",
+      "Dawei Nie",
+      "Sixiang Liu",
+      "Zhexi Wen",
+      "Feng Pan",
+      "Xiaofeng Wang",
+      "Zhi Hou",
+      "Chunxiao Liu",
+      "Xue Yang",
+      "Junchi Yan",
+      "Hengshuang Zhao",
+      "Dacheng Tao",
+      "Xiaogang Wang"
+    ],
+    "categories": [
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.04426",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04426",
+    "pdf_url": "https://arxiv.org/pdf/2607.04426",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.04162",
+    "title": "ACE: Agentic Control for Embodied Manipulation via Zero-shot Workflow Reasoning",
+    "url": "https://arxiv.org/abs/2607.04162",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Iok Tong Lei",
+      "QianZhi Li",
+      "Ying Jie Yap",
+      "Yujie Zhang",
+      "Rui Zhong",
+      "Haichao Gui",
+      "Xiaolong Liu",
+      "Zhidong Deng"
+    ],
+    "categories": [
+      "cs.RO",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.04162",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04162",
+    "pdf_url": "https://arxiv.org/pdf/2607.04162",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.03333",
+    "title": "SPORK: Self-Speculative Forking to Accelerate Agentic LLM Inference",
+    "url": "https://arxiv.org/abs/2607.03333",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Huajun Bai",
+      "Weiwei Lv",
+      "Huichuan Zheng",
+      "Youyou Lu",
+      "Jiwu Shu"
+    ],
+    "categories": [
+      "cs.DC",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.03333",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03333",
+    "pdf_url": "https://arxiv.org/pdf/2607.03333",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.02857",
+    "title": "MOSAIC: Knowledge-Guided CLI Command Composition Attack in LLM Coding Agents",
+    "url": "https://arxiv.org/abs/2607.02857",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Jiangrong Wu",
+      "Huaijin Wang",
+      "Yihao Zhang",
+      "Yuhong Nan",
+      "Shuai Wang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.02857",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02857",
+    "pdf_url": "https://arxiv.org/pdf/2607.02857",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.01935",
+    "title": "A-TMA: Decoupling State-Aware Memory Failures in Long-Term Agent Memory",
+    "url": "https://arxiv.org/abs/2607.01935",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Zitong Shi",
+      "Yixuan Tang",
+      "Anthony Kum Hoe Tung"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.01935",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01935",
+    "pdf_url": "https://arxiv.org/pdf/2607.01935",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2607.01640",
+    "title": "AgentFlow: Building Agent Dependency Graphs for Static Analysis of Agent Programs",
+    "url": "https://arxiv.org/abs/2607.01640",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Shenao Wang",
+      "Xinyi Hou",
+      "Yanjie Zhao",
+      "Xiao Cheng",
+      "Haoyu Wang"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.01640",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01640",
+    "pdf_url": "https://arxiv.org/pdf/2607.01640",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.01668",
+    "title": "VeriChat: An Agentic Conversational AI Assistant for Hardware Security Verification",
+    "url": "https://arxiv.org/abs/2607.01668",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Dipayan Saha",
+      "Khan Thamid Hasan",
+      "Shams Tarek",
+      "Sujan Kumar Saha",
+      "Mark Tehranipoor",
+      "Farimah Farahmandi"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "rag",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.01668",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01668",
+    "pdf_url": "https://arxiv.org/pdf/2607.01668",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.02294",
+    "title": "Coding Agents Are Guessing: Measuring Action-Boundary Violations in Underspecified DevOps Instructions",
+    "url": "https://arxiv.org/abs/2607.02294",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Zimo Ji",
+      "Zekai Zhang",
+      "Congying Xu",
+      "Zongjie Li",
+      "Yudong Gao",
+      "Shuai Wang",
+      "Shing-Chi Cheung"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.02294",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02294",
+    "pdf_url": "https://arxiv.org/pdf/2607.02294",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.01916",
+    "title": "ContextSniper: AntTrail's Token-Efficient Code Memory for Repository-Level Program Repair",
+    "url": "https://arxiv.org/abs/2607.01916",
+    "published": "2026-07-02",
+    "updated": "2026-07-06",
+    "authors": [
+      "Chiwang Luk",
+      "Matin Mohammad Najafi",
+      "Zhifeng Jia",
+      "Wei Yang",
+      "Xiuchang Li",
+      "Jinwei Zhu",
+      "Yang Ren",
+      "Lei Chen",
+      "Gao Cong"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "coding-agent",
+      "rag-agent"
+    ],
+    "arxiv_id": "2607.01916",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01916",
+    "pdf_url": "https://arxiv.org/pdf/2607.01916",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2607.01929",
+    "title": "Beyond Textual Repository Exploration: Dual-Modal Structural Reasoning for Agentic Issue Resolution",
+    "url": "https://arxiv.org/abs/2607.01929",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Jiayi Zhang",
+      "Kai Huang",
+      "Yang Liu",
+      "Chunyang Chen"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "embodied-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent",
+      "function-calling"
+    ],
+    "arxiv_id": "2607.01929",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01929",
+    "pdf_url": "https://arxiv.org/pdf/2607.01929",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.01071",
+    "title": "MemSyco-Bench: Benchmarking Sycophancy in Agent Memory",
+    "url": "https://arxiv.org/abs/2607.01071",
+    "published": "2026-07-01",
+    "updated": "2026-07-02",
+    "authors": [
+      "Zhishang Xiang",
+      "Zerui Chen",
+      "Yunbo Tang",
+      "Zhimin Wei",
+      "Ruqin Ning",
+      "Yujie Lin",
+      "Qinggang Zhang",
+      "Jinsong Su"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "reasoning"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2607.01071",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01071",
+    "pdf_url": "https://arxiv.org/pdf/2607.01071",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2607.00939",
+    "title": "Leveraging LLM-Based Agentic Systems to Generate Quantum Applications for Test Optimization",
+    "url": "https://arxiv.org/abs/2607.00939",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Ming Tao",
+      "Yuechen Li",
+      "Tao Yue",
+      "Man Zhang",
+      "Aitor Arrieta Marcos"
+    ],
+    "categories": [
+      "cs.SE",
+      "quant-ph"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.00939",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00939",
+    "pdf_url": "https://arxiv.org/pdf/2607.00939",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.00334",
+    "title": "Managed Autonomy at Runtime: Gear-Based Safety and Governance for Single- and Multi-Agent Cyber-Physical Systems",
+    "url": "https://arxiv.org/abs/2607.00334",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Srini Ramaswamy",
+      "Wang Miaosheng"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "embodied-agent",
+      "multi-agent",
+      "planning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.00334",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00334",
+    "pdf_url": "https://arxiv.org/pdf/2607.00334",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2607.05428",
+    "title": "CHARLIE: An On-Premise Multi-Agent Retrieval-Augmented Generation System for Evidential Reasoning in Forensic Science",
+    "url": "https://arxiv.org/abs/2607.05428",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Leandro D. Carneiro",
+      "Andre L. S. Meirelles",
+      "Juliano de A. Gomes",
+      "Rafael C. A. Cabral"
+    ],
+    "categories": [
+      "cs.DL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2607.05428",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05428",
+    "pdf_url": "https://arxiv.org/pdf/2607.05428",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.31693",
+    "title": "ShopX: A Foundation Model for Intent-to-Item Fulfillment in Agentic Shopping",
+    "url": "https://arxiv.org/abs/2606.31693",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Jiacheng Chen",
+      "Tao Zhang",
+      "Manxi Lin",
+      "Dunxian Huang",
+      "Teng Shi",
+      "Honghao Fu",
+      "Mengyan Li",
+      "Xinming Zhang",
+      "Chenchi Zhang",
+      "Xuan Lu",
+      "Xiaoxiong Du",
+      "Haibin Chen",
+      "Shaolin Ye",
+      "Hao Chang",
+      "Xiaoqi Li",
+      "Shuwen Xiao",
+      "Yujin Yuan",
+      "Jingxuan Feng",
+      "Shaopan Xiong",
+      "Huimin Yi",
+      "Ju Huang",
+      "Qiu Shen",
+      "Ying Chen",
+      "Junjun Zheng",
+      "Xiangheng Kong",
+      "Yuning Jiang"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.31693",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31693",
+    "pdf_url": "https://arxiv.org/pdf/2606.31693",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.31252",
+    "title": "Embodied CAD: Solver-Grounded LLM Agents for Parametric B-Rep Assembly Modeling",
+    "url": "https://arxiv.org/abs/2606.31252",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Fumin Liu",
+      "Haoyu Zhou",
+      "Fei Hao",
+      "Lin Yang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.31252",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31252",
+    "pdf_url": "https://arxiv.org/pdf/2606.31252",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.31174",
+    "title": "ClawArena-Team: Benchmarking Subagent Orchestration and Dynamic Workflows in Language-Model Agents",
+    "url": "https://arxiv.org/abs/2606.31174",
+    "published": "2026-06-30",
+    "updated": "2026-07-02",
+    "authors": [
+      "Kaiwen Xiong",
+      "Haonian Ji",
+      "Shi Qiu",
+      "Zeyu Zheng",
+      "Cihang Xie",
+      "Xinyu Ye",
+      "Huaxiu Yao"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.31174",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31174",
+    "pdf_url": "https://arxiv.org/pdf/2606.31174",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.31046",
+    "title": "OpenLife: Toward Open-World Artificial Life with Autonomous LLM Agents",
+    "url": "https://arxiv.org/abs/2606.31046",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Atsushi Masumori",
+      "Itsuki Doi",
+      "Norihiro Maruyama",
+      "Ryosuke Takata",
+      "Takashi Ikegami"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.31046",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31046",
+    "pdf_url": "https://arxiv.org/pdf/2606.31046",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.31650",
+    "title": "ECHO: Prune to act, trace to learn with selective turn memory in agentic RL",
+    "url": "https://arxiv.org/abs/2606.31650",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Zijun Xie",
+      "Binbin Zheng",
+      "Enlei Gong",
+      "Jihua Liu",
+      "Yuyang You",
+      "Lingfeng Liu",
+      "Jiayao Tang",
+      "Guanqun Zhao",
+      "Aoqi Hu",
+      "Zeyu Chen"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.31650",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31650",
+    "pdf_url": "https://arxiv.org/pdf/2606.31650",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.31639",
+    "title": "A Lifecycle and Application-Stack Survey of Large Language Model Vulnerabilities: Attacks, Risks, Defenses, and Open Problems",
+    "url": "https://arxiv.org/abs/2606.31639",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Seyed Bagher Hashemi Natanzi",
+      "Bo Tang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.GT",
+      "cs.LO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.31639",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31639",
+    "pdf_url": "https://arxiv.org/pdf/2606.31639",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.30566",
+    "title": "Forensic Trajectory Signatures for Agent Memory Poisoning Detection",
+    "url": "https://arxiv.org/abs/2606.30566",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Jun Wen Leong"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.30566",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30566",
+    "pdf_url": "https://arxiv.org/pdf/2606.30566",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.29914",
+    "title": "MemDelta: Controlled Baselines and Hidden Confounds in Agent Memory Evaluation",
+    "url": "https://arxiv.org/abs/2606.29914",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Kuan Wang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.29914",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29914",
+    "pdf_url": "https://arxiv.org/pdf/2606.29914",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.30555",
+    "title": "Linguistic Firewall: Geometry as Defense in Multi-Agent Systems Routing",
+    "url": "https://arxiv.org/abs/2606.30555",
+    "published": "2026-06-29",
+    "updated": "2026-07-05",
+    "authors": [
+      "Dvir Alsheich",
+      "Adar Peleg",
+      "Ben Hagag",
+      "Rom Himelstein",
+      "Amit Levi",
+      "Avi Mendelson"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.30555",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30555",
+    "pdf_url": "https://arxiv.org/pdf/2606.30555",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.30546",
+    "title": "MAS-Lab: A Specification-Driven Validation Framework for Reliable Multi-Agent Systems",
+    "url": "https://arxiv.org/abs/2606.30546",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Jordan Augé",
+      "Giovanna Carofiglio",
+      "Giulio Grassi",
+      "Jacques Samain"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.30546",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30546",
+    "pdf_url": "https://arxiv.org/pdf/2606.30546",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.30259",
+    "title": "Multi-Agentic System Leveraging Open-Source LLMs to Mitigate Disinformation Threats",
+    "url": "https://arxiv.org/abs/2606.30259",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Sebastian Kula",
+      "Martin Tamajka"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.30259",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30259",
+    "pdf_url": "https://arxiv.org/pdf/2606.30259",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.29742",
+    "title": "MicroAgent: Context-Augmented Multi-Agent Framework for Automatic Microservice Decomposition",
+    "url": "https://arxiv.org/abs/2606.29742",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Zishan Su",
+      "Junjie Huang",
+      "Shiwen Shan",
+      "Xingyan Chen",
+      "Hui Zeng",
+      "Yuxin Su",
+      "Yanlin Wang",
+      "Michael R. Lyu"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.29742",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29742",
+    "pdf_url": "https://arxiv.org/pdf/2606.29742",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.29270",
+    "title": "Minority Sentinel: When to Overturn Majority Voting in Multi-Agent LLM Debates",
+    "url": "https://arxiv.org/abs/2606.29270",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Chuan He",
+      "Zebin Chen",
+      "Zhengyi Yang",
+      "Shaobo Qiao",
+      "Mingchen Ju",
+      "Jiate Liu",
+      "Dong Wen",
+      "Guanfeng Liu"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "reasoning"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.29270",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29270",
+    "pdf_url": "https://arxiv.org/pdf/2606.29270",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.29654",
+    "title": "Budgeted Act-or-Defer Multi-Agent LLM Deliberation with Local Reliability Bounds",
+    "url": "https://arxiv.org/abs/2606.29654",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Mengdie Flora Wang",
+      "Haochen Xie",
+      "Guanghui Wang",
+      "Devin Zhang",
+      "Jae Oh Woo"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.29654",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29654",
+    "pdf_url": "https://arxiv.org/pdf/2606.29654",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.28958",
+    "title": "When Latent Agents Lie: KV-Cache Integrity in Multi-Agent LLM Collaboration",
+    "url": "https://arxiv.org/abs/2606.28958",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Luís Brito",
+      "Carlos Baquero"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-safety",
+      "memory",
+      "multi-agent",
+      "reasoning"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.28958",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28958",
+    "pdf_url": "https://arxiv.org/pdf/2606.28958",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.28666",
+    "title": "Why Trust Your Agent? Empirical Security Gains from TRiSM-Guided Agentic Workflows in Healthcare",
+    "url": "https://arxiv.org/abs/2606.28666",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Liam Kearns"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.28666",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28666",
+    "pdf_url": "https://arxiv.org/pdf/2606.28666",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.27990",
+    "title": "AdvancedShelLM: A Stateful Multi-Agent LLM Honeypot for SSH Deception",
+    "url": "https://arxiv.org/abs/2606.27990",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Muris Sladić",
+      "Eman Alibalić",
+      "Veronica Valeros",
+      "Carlos Catania",
+      "Sebastian Garcia"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.27990",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27990",
+    "pdf_url": "https://arxiv.org/pdf/2606.27990",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.27929",
+    "title": "When Multi-Robot Systems Meet Agentic AI:Towards Embodied Collective Intelligence",
+    "url": "https://arxiv.org/abs/2606.27929",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Yuxuan Yan",
+      "Yuanyuan Jia",
+      "Qianqian Yang"
+    ],
+    "categories": [
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.27929",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27929",
+    "pdf_url": "https://arxiv.org/pdf/2606.27929",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.28434",
+    "title": "SWE-MeM: Learning Adaptive Memory Management for Long-Horizon Coding Agents",
+    "url": "https://arxiv.org/abs/2606.28434",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Shuzheng Gao",
+      "Wenhao Zeng",
+      "Zhaojian Yu",
+      "Jianqiao Wangni",
+      "Chaozheng Wang",
+      "Kai Cai",
+      "Shilin He",
+      "Michael R. Lyu"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "coding-agent",
+      "memory",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.28434",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28434",
+    "pdf_url": "https://arxiv.org/pdf/2606.28434",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.28570",
+    "title": "Digitizing Coaching Intelligence: An Agentic Framework for Holistic Athlete Profiling using VLM and RAG",
+    "url": "https://arxiv.org/abs/2606.28570",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Deep Ghosal",
+      "Ishani Sen",
+      "Wazib Ansar",
+      "Amlan Chakrabarti"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.28570",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28570",
+    "pdf_url": "https://arxiv.org/pdf/2606.28570",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.28182",
+    "title": "LLawCo: Learning Laws of Cooperation for Modeling Embodied Multi-Agent Behavior",
+    "url": "https://arxiv.org/abs/2606.28182",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Qinhong Zhou",
+      "Chuang Gan",
+      "Anoop Cherian"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CV",
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.28182",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28182",
+    "pdf_url": "https://arxiv.org/pdf/2606.28182",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.25514",
+    "title": "Unlocking Model Potentials Through Adaptive Multi-Agent Scaffolding for Efficient Issue Resolution",
+    "url": "https://arxiv.org/abs/2606.25514",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Yang Chen",
+      "Aliya Ahmad",
+      "Yiheng Zhou",
+      "Reyhaneh Jabbarvand"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.25514",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25514",
+    "pdf_url": "https://arxiv.org/pdf/2606.25514",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.25588",
+    "title": "IntentTester: Intent-Driven Multi-agent Framework for Cross-Library Test Migration",
+    "url": "https://arxiv.org/abs/2606.25588",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Yi Gao",
+      "Ziyuan Zhang",
+      "Xing Hu",
+      "Xiaohu Yang",
+      "Xin Xia"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.25588",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25588",
+    "pdf_url": "https://arxiv.org/pdf/2606.25588",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.25400",
+    "title": "BrainAgent: A Large Language Model-Driven Multi-Agent Framework for Autonomous Brain Signal Understanding",
+    "url": "https://arxiv.org/abs/2606.25400",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Yangxuan Zhou",
+      "Sha Zhao",
+      "Jiquan Wang",
+      "Shijian Li",
+      "Gang Pan"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.25400",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25400",
+    "pdf_url": "https://arxiv.org/pdf/2606.25400",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.24775",
+    "title": "Are We Ready For An Agent-Native Memory System?",
+    "url": "https://arxiv.org/abs/2606.24775",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Wei Zhou",
+      "Xuanhe Zhou",
+      "Shaokun Han",
+      "Hongming Xu",
+      "Guoliang Li",
+      "Zhiyu Li",
+      "Feiyu Xiong",
+      "Fan Wu"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.DB",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.24775",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24775",
+    "pdf_url": "https://arxiv.org/pdf/2606.24775",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.24649",
+    "title": "Agentic Collaborative Cognition for Zero-Shot 3D Understanding",
+    "url": "https://arxiv.org/abs/2606.24649",
+    "published": "2026-06-23",
+    "updated": "2026-06-25",
+    "authors": [
+      "Wenxin Wang",
+      "Bo Zhang",
+      "Feng Chen",
+      "Zixuan Wang",
+      "Wen Li",
+      "Changsheng Li",
+      "Yinjie Lei"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent",
+      "planning",
+      "rag"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.24649",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24649",
+    "pdf_url": "https://arxiv.org/pdf/2606.24649",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.24437",
+    "title": "ReM-MoA: Reasoning Memory Sustains Mixture-of-Agents Scaling",
+    "url": "https://arxiv.org/abs/2606.24437",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Heng Ping",
+      "Arijit Bhattacharjee",
+      "Peiyu Zhang",
+      "Shixuan Li",
+      "Wei Yang",
+      "Ali Jannesari",
+      "Nesreen Ahmed",
+      "Paul Bogdan"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "reasoning"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.24437",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24437",
+    "pdf_url": "https://arxiv.org/pdf/2606.24437",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.22647",
+    "title": "RAVEN: Agentic RAG for Automated Vulnerability Repair",
+    "url": "https://arxiv.org/abs/2606.22647",
+    "published": "2026-06-21",
+    "updated": "2026-06-21",
+    "authors": [
+      "Varun Gadey",
+      "Zijie Liu",
+      "Alexandra Dmitrienko"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.LG",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "memory",
+      "rag"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.22647",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22647",
+    "pdf_url": "https://arxiv.org/pdf/2606.22647",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.22330",
+    "title": "Hypothesis-Driven Skill Optimization for LLM Agents",
+    "url": "https://arxiv.org/abs/2606.22330",
+    "published": "2026-06-21",
+    "updated": "2026-06-21",
+    "authors": [
+      "Fangxin Shang",
+      "Yehui Yang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.22330",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22330",
+    "pdf_url": "https://arxiv.org/pdf/2606.22330",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.21740",
+    "title": "Training the Orchestrator: A Supervised Approach to End-to-End PDDL Planning with LLM Agents",
+    "url": "https://arxiv.org/abs/2606.21740",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Rajesh Mangannavar",
+      "Zachary Coalson",
+      "Pranay Dugar",
+      "Prasad Tadepalli"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.21740",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21740",
+    "pdf_url": "https://arxiv.org/pdf/2606.21740",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.18467",
+    "title": "ToolChain-CRC: Conformal Risk Control for Agentic AI Under Retrieval and Tool-Use Drift",
+    "url": "https://arxiv.org/abs/2606.18467",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Jeffery Opoku",
+      "David Banahene"
+    ],
+    "categories": [
+      "stat.ML",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "agentic-ai",
+      "ai-agent",
+      "rag-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.18467",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18467",
+    "pdf_url": "https://arxiv.org/pdf/2606.18467",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.18142",
+    "title": "Your AI Travel Agent Would Book You a Bullfight: An Agentic Benchmark for Implicit Animal Welfare in Frontier AI Models",
+    "url": "https://arxiv.org/abs/2606.18142",
+    "published": "2026-06-16",
+    "updated": "2026-07-06",
+    "authors": [
+      "Jasmine Brazilek",
+      "Joel Christoph",
+      "Maheep Chaudhary",
+      "Oliver Tullio",
+      "Carol Kline",
+      "Miles Tidmarsh",
+      "Arturs Kanepajs"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.CY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.18142",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18142",
+    "pdf_url": "https://arxiv.org/pdf/2606.18142",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.16591",
+    "title": "SING: Synthetic Intention Graph for Scalable Active Tool Discovery in LLM Agents",
+    "url": "https://arxiv.org/abs/2606.16591",
+    "published": "2026-06-15",
+    "updated": "2026-06-16",
+    "authors": [
+      "Qiao Xiao",
+      "Haochen Shi",
+      "Yisen Gao",
+      "Wenbin Hu",
+      "Huihao Jing",
+      "Tianshi Zheng",
+      "Baixuan Xu",
+      "Ziheng Zhang",
+      "Weiqi Wang",
+      "Haoran Li",
+      "Jiaxin Bai",
+      "Yangqiu Song"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.16591",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16591",
+    "pdf_url": "https://arxiv.org/pdf/2606.16591",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.15684",
+    "title": "Multi-agent Framework for Time-Sensitive Complementary Collaboration in Minecraft",
+    "url": "https://arxiv.org/abs/2606.15684",
+    "published": "2026-06-14",
+    "updated": "2026-06-14",
+    "authors": [
+      "Juheon Yi",
+      "Jinglu Wang",
+      "Xiaoyi Zhang",
+      "Yan Lu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.15684",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15684",
+    "pdf_url": "https://arxiv.org/pdf/2606.15684",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.15931",
+    "title": "DeepRoot: A KG-Coordinated Multi-Agent System for Therapeutic Reasoning over Historical Medical Texts",
+    "url": "https://arxiv.org/abs/2606.15931",
+    "published": "2026-06-14",
+    "updated": "2026-06-14",
+    "authors": [
+      "Zijian Carl Ma",
+      "Sean J. Wang",
+      "Sijbren Kramer",
+      "Li Erran Li"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.15931",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15931",
+    "pdf_url": "https://arxiv.org/pdf/2606.15931",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.12703",
+    "title": "SMSR: Certified Defence Against Runtime Memory Poisoning in Persistent LLM Agent Systems",
+    "url": "https://arxiv.org/abs/2606.12703",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Tarun Sharma"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.12703",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12703",
+    "pdf_url": "https://arxiv.org/pdf/2606.12703",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.10684",
+    "title": "Divide and Cooperate: Role-Decomposed Multi-Agent LLM Training with Cross-Agent Learning Signals",
+    "url": "https://arxiv.org/abs/2606.10684",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Jaewan Park",
+      "Solbee Cho",
+      "Jay-Yoon Lee"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.10684",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10684",
+    "pdf_url": "https://arxiv.org/pdf/2606.10684",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.10616",
+    "title": "Learning What to Remember: Observability-Safe Memory Retention via Constrained Optimization for Long-Horizon Language Agents",
+    "url": "https://arxiv.org/abs/2606.10616",
+    "published": "2026-06-09",
+    "updated": "2026-06-29",
+    "authors": [
+      "Qingcan Kang",
+      "Liu Mingyang",
+      "Shixiong Kai",
+      "Kaichao Liang",
+      "Tao Zhong",
+      "Mingxuan Yuan"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.10616",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10616",
+    "pdf_url": "https://arxiv.org/pdf/2606.10616",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.10933",
+    "title": "Frontier Coding Agents Use Metaprogramming to Adapt to Unfamiliar Programming Languages",
+    "url": "https://arxiv.org/abs/2606.10933",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Aman Sharma",
+      "Sushrut Thorat",
+      "Paras Chopra"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.10933",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10933",
+    "pdf_url": "https://arxiv.org/pdf/2606.10933",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.10577",
+    "title": "AgenticNav: Zero-Shot Vision-and-Language Navigation as a Tool-Calling Harness",
+    "url": "https://arxiv.org/abs/2606.10577",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Yijian Li",
+      "Changze Li",
+      "Hantian Shi",
+      "Jiaying Luo",
+      "Jiyuan Cai",
+      "Ming Yang",
+      "Tong Qin"
+    ],
+    "categories": [
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.10577",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10577",
+    "pdf_url": "https://arxiv.org/pdf/2606.10577",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.10304",
+    "title": "MIRAGE: A Polarity-Flipping Encoding Subspace in LLM Agents",
+    "url": "https://arxiv.org/abs/2606.10304",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Pratibha Revankar",
+      "Kargi Chauhan",
+      "Jihye Kim",
+      "Sadiba Nusrat Nur",
+      "Vincent Siu",
+      "Chenguang Wang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.10304",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10304",
+    "pdf_url": "https://arxiv.org/pdf/2606.10304",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.07682",
+    "title": "SWE-Marathon: Can Agents Autonomously Complete Ultra-Long-Horizon Software Work?",
+    "url": "https://arxiv.org/abs/2606.07682",
+    "published": "2026-06-05",
+    "updated": "2026-06-05",
+    "authors": [
+      "Rishi Desai",
+      "Jesse Hu",
+      "Joan Cabezas",
+      "Neel Harsola",
+      "Pratyush Shukla",
+      "Roey Ben Chaim",
+      "Adnan El Assadi",
+      "Omkaar Mukund Kamath",
+      "Fenil Faldu",
+      "Prannay Hebbar",
+      "Jiankai Sun",
+      "Yiyuan Li",
+      "Pramod Srinivasan",
+      "Ishan Gupta",
+      "Christopher Settles",
+      "Daniel Wang",
+      "Derek Chen",
+      "Pranav Raja",
+      "Albert Liu",
+      "Marek Šuppa",
+      "Nevasini Sasikumar",
+      "Luyang Kong",
+      "Erik Quintanilla",
+      "Xiangyi Li",
+      "Ivan Bercovich",
+      "Steven Dillmann"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.07682",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.07682",
+    "pdf_url": "https://arxiv.org/pdf/2606.07682",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.07867",
+    "title": "The Cold-Start Safety Gap in LLM Agents",
+    "url": "https://arxiv.org/abs/2606.07867",
+    "published": "2026-06-05",
+    "updated": "2026-06-05",
+    "authors": [
+      "Chung-En Sun",
+      "Linbo Liu",
+      "Tsui-Wei Weng"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2606.07867",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.07867",
+    "pdf_url": "https://arxiv.org/pdf/2606.07867",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.07711",
+    "title": "Rosetta Memory: Adaptive Memory for Cross-LLM Agents",
+    "url": "https://arxiv.org/abs/2606.07711",
+    "published": "2026-06-05",
+    "updated": "2026-06-05",
+    "authors": [
+      "Hao Yang",
+      "Shiqi Shen",
+      "Haoxuan Li",
+      "Zhipeng Wang",
+      "Zhi Gong",
+      "Xu Chen"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "coding-agent",
+      "memory",
+      "planning",
+      "reasoning"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.07711",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.07711",
+    "pdf_url": "https://arxiv.org/pdf/2606.07711",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.06054",
+    "title": "Beyond Similarity: Trustworthy Memory Search for Personal AI Agents",
+    "url": "https://arxiv.org/abs/2606.06054",
+    "published": "2026-06-04",
+    "updated": "2026-06-04",
+    "authors": [
+      "Jiawen Zhang",
+      "Kejia Chen",
+      "Jiachen Ma",
+      "Yangfan Hu",
+      "Lipeng He",
+      "Yechao Zhang",
+      "Jian Liu",
+      "Xiaohu Yang",
+      "Tianwei Zhang",
+      "Ruoxi Jia"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.06054",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.06054",
+    "pdf_url": "https://arxiv.org/pdf/2606.06054",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.05805",
+    "title": "From Risk Classification to Action Plan Remediation: A Guardrail Feedback Driven Framework for LLM Agents",
+    "url": "https://arxiv.org/abs/2606.05805",
+    "published": "2026-06-04",
+    "updated": "2026-06-04",
+    "authors": [
+      "Yuhao Sun",
+      "Jiacheng Zhang",
+      "Shaanan Cohney",
+      "Zhexin Zhang",
+      "Feng Liu",
+      "Xingliang Yuan"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.05805",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.05805",
+    "pdf_url": "https://arxiv.org/pdf/2606.05805",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.04599",
+    "title": "Plan First, Judge Later, Run Better: A DMAIC-Inspired Agentic System for Industrial Anomaly Detection",
+    "url": "https://arxiv.org/abs/2606.04599",
+    "published": "2026-06-03",
+    "updated": "2026-06-03",
+    "authors": [
+      "Yongzi Yu",
+      "Ao Li",
+      "Le Wang",
+      "Ziyue Li",
+      "Fugee Tsung",
+      "Yuxuan Liang",
+      "Man Li"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CE"
+    ],
+    "topics": [
+      "agent-safety",
+      "multi-agent",
+      "planning",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.04599",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.04599",
+    "pdf_url": "https://arxiv.org/pdf/2606.04599",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.04051",
+    "title": "RUBAS: Rubric-Based Reinforcement Learning for Agent Safety",
+    "url": "https://arxiv.org/abs/2606.04051",
+    "published": "2026-06-02",
+    "updated": "2026-06-02",
+    "authors": [
+      "Xian Qi Loye",
+      "Qinglin Su",
+      "Zhexin Zhang",
+      "Shiyao Cui",
+      "Qi Zhu",
+      "Fei Mi",
+      "Hongning Wang",
+      "Minlie Huang"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2606.04051",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.04051",
+    "pdf_url": "https://arxiv.org/pdf/2606.04051",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.02812",
+    "title": "Traj-Evolve: A Self-Evolving Multi-Agent System for Patient Trajectory Modeling in Lung Cancer Early Detection",
+    "url": "https://arxiv.org/abs/2606.02812",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Sihang Zeng",
+      "Matthew Thompson",
+      "Ruth Etzioni",
+      "Meliha Yetisgen"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "multi-agent",
+      "rag",
+      "reasoning"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.02812",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.02812",
+    "pdf_url": "https://arxiv.org/pdf/2606.02812",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.01528",
+    "title": "Joint Agent Memory and Exploration Learning via Novelty Signals",
+    "url": "https://arxiv.org/abs/2606.01528",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Shizuo Tian",
+      "Xiaohong Weng",
+      "Rui Kong",
+      "Yuxuan Chen",
+      "Guohong Liu",
+      "Yuebing Song",
+      "Jiacheng Liu",
+      "Yuchen Li",
+      "Dawei Yin",
+      "Ting Cao",
+      "Yunxin Liu",
+      "Yuanchun Li"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.01528",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.01528",
+    "pdf_url": "https://arxiv.org/pdf/2606.01528",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.01041",
+    "title": "ExpWeaver: LLM Agents Learn from Experience via Latent RAG",
+    "url": "https://arxiv.org/abs/2606.01041",
+    "published": "2026-05-31",
+    "updated": "2026-05-31",
+    "authors": [
+      "Tao Feng",
+      "Tianyang Luo",
+      "Jingjun Xu",
+      "Zhigang Hua",
+      "Yan Xie",
+      "Shuang Yang",
+      "Ge Liu",
+      "Jiaxuan You"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.01041",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.01041",
+    "pdf_url": "https://arxiv.org/pdf/2606.01041",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.30907",
+    "title": "BlueFin: Benchmarking LLM Agents on Financial Spreadsheets",
+    "url": "https://arxiv.org/abs/2605.30907",
+    "published": "2026-05-29",
+    "updated": "2026-05-29",
+    "authors": [
+      "Srivatsa Kundurthy",
+      "Clara Na",
+      "Colton Moraine",
+      "Anoushka Mohta",
+      "Case Winter",
+      "George Fang",
+      "John Ling",
+      "Emma Strubell",
+      "Zach Kirshner"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2605.30907",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.30907",
+    "pdf_url": "https://arxiv.org/pdf/2605.30907",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2605.30858",
+    "title": "ForecastCompass: Guiding Agentic Forecasting with Adaptive Factor Memory",
+    "url": "https://arxiv.org/abs/2605.30858",
+    "published": "2026-05-29",
+    "updated": "2026-05-29",
+    "authors": [
+      "Yurui Chang",
+      "Yongkang Du",
+      "Yuanpu Cao",
+      "Jinghui Chen",
+      "Lu Lin"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.30858",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.30858",
+    "pdf_url": "https://arxiv.org/pdf/2605.30858",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.30058",
+    "title": "HEART-Bench: Do LLM Agents Exhibit Human-like Psychology?",
+    "url": "https://arxiv.org/abs/2605.30058",
+    "published": "2026-05-28",
+    "updated": "2026-05-28",
+    "authors": [
+      "Weihan Peng",
+      "Chenxu Zhang",
+      "Qianao Wang",
+      "Yuling Shi",
+      "Heng Lian",
+      "Qihong Mao",
+      "Jiahao Pang",
+      "Chunliang Feng",
+      "Bowen Li",
+      "Xiaodong Gu"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.30058",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.30058",
+    "pdf_url": "https://arxiv.org/pdf/2605.30058",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.27762",
+    "title": "PEAM: Parametric Embodied Agent Memory through Contrastive Internalization of Experience in Minecraft",
+    "url": "https://arxiv.org/abs/2605.27762",
+    "published": "2026-05-26",
+    "updated": "2026-06-01",
+    "authors": [
+      "Yuchen Guo",
+      "Junli Gong",
+      "Weicheng Wang",
+      "Hongmin Cai",
+      "Yiu-ming Cheung",
+      "Weifeng Su"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.27762",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.27762",
+    "pdf_url": "https://arxiv.org/pdf/2605.27762",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.24659",
+    "title": "IterInject: Indirect Prompt Injection Against LLM Agents via Feedback-Guided Iterative Optimization",
+    "url": "https://arxiv.org/abs/2605.24659",
+    "published": "2026-05-23",
+    "updated": "2026-05-23",
+    "authors": [
+      "Zixuan Chen",
+      "Jiaxiang Chen",
+      "Li Luo",
+      "Ke Xu",
+      "Xiaoxiang Huang",
+      "Tanfeng Sun",
+      "Xinghao Jiang"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.24659",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.24659",
+    "pdf_url": "https://arxiv.org/pdf/2605.24659",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.23723",
+    "title": "MemAudit: Post-hoc Auditing of Poisoned Agent Memory via Causal Attribution and Structural Anomaly Detection",
+    "url": "https://arxiv.org/abs/2605.23723",
+    "published": "2026-05-22",
+    "updated": "2026-05-22",
+    "authors": [
+      "Zhewen Tan",
+      "Yilun Yao",
+      "Huiyan Jin",
+      "Wenhan Yu",
+      "Guoan Wang",
+      "Mengyuan Fan",
+      "liang lu",
+      "Feng Liu",
+      "Xiangzheng Zhang",
+      "Duohe Ma",
+      "Tong Yang",
+      "Lin Sun"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.23723",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.23723",
+    "pdf_url": "https://arxiv.org/pdf/2605.23723",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.24219",
+    "title": "Beyond Final Answers: Auditing Trajectory-Level Hallucinations in Multi-Agent Industrial Workflows",
+    "url": "https://arxiv.org/abs/2605.24219",
+    "published": "2026-05-22",
+    "updated": "2026-05-26",
+    "authors": [
+      "Harshada Badave",
+      "Santosh Borse",
+      "Andrea Gomez",
+      "Harshitha Narahari",
+      "Sara Carter",
+      "Vishwa Bhatt",
+      "Aishani Rachakonda",
+      "Shuxin Lin",
+      "Dhaval Patel"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.24219",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.24219",
+    "pdf_url": "https://arxiv.org/pdf/2605.24219",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.22154",
+    "title": "IdleSpec: Exploiting Idle Time via Speculative Planning for LLM Agents",
+    "url": "https://arxiv.org/abs/2605.22154",
+    "published": "2026-05-21",
+    "updated": "2026-05-21",
+    "authors": [
+      "Daewon Choi",
+      "Kyunghyun Park",
+      "Woomin Song",
+      "Saket Dingliwal",
+      "Sai Muralidhar Jayanthi",
+      "Jinwoo Shin",
+      "Aram Galstyan"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.22154",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.22154",
+    "pdf_url": "https://arxiv.org/pdf/2605.22154",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.21240",
+    "title": "APEX: Autonomous Policy Exploration for Self-Evolving LLM Agents",
+    "url": "https://arxiv.org/abs/2605.21240",
+    "published": "2026-05-20",
+    "updated": "2026-05-20",
+    "authors": [
+      "Yibo Li",
+      "Jiashuo Yang",
+      "Zhi Zheng",
+      "Zhiyuan Hu",
+      "Yuan Sui",
+      "Shizun Wang",
+      "Yufei He",
+      "Bryan Hooi"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.21240",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.21240",
+    "pdf_url": "https://arxiv.org/pdf/2605.21240",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.18930",
+    "title": "OEP: Poisoning Self-Evolving LLM Agents via Locally Correct but Non-Transferable Experiences",
+    "url": "https://arxiv.org/abs/2605.18930",
+    "published": "2026-05-18",
+    "updated": "2026-05-18",
+    "authors": [
+      "Kaixiang Wang",
+      "Jiong Lou",
+      "Zhaojiacheng Zhou",
+      "Jie Li"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "reasoning"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.18930",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.18930",
+    "pdf_url": "https://arxiv.org/pdf/2605.18930",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.18284",
+    "title": "CommitDistill: A Lightweight Knowledge-Centric Memory Layer for Software Repositories",
+    "url": "https://arxiv.org/abs/2605.18284",
+    "published": "2026-05-18",
+    "updated": "2026-05-18",
+    "authors": [
+      "Divya Chukkapalli",
+      "Thejesh Avula",
+      "Aditya Aggarwal",
+      "Harsimran Singh",
+      "Amith Tallanki"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.18284",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.18284",
+    "pdf_url": "https://arxiv.org/pdf/2605.18284",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.14421",
+    "title": "MemLineage: Lineage-Guided Enforcement for LLM Agent Memory",
+    "url": "https://arxiv.org/abs/2605.14421",
+    "published": "2026-05-14",
+    "updated": "2026-05-14",
+    "authors": [
+      "Ciyan Ouyang",
+      "Rui Hou"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.14421",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.14421",
+    "pdf_url": "https://arxiv.org/pdf/2605.14421",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.14892",
+    "title": "Beyond Individual Intelligence: Surveying Collaboration, Failure Attribution, and Self-Evolution in LLM-based Multi-Agent Systems",
+    "url": "https://arxiv.org/abs/2605.14892",
+    "published": "2026-05-14",
+    "updated": "2026-05-15",
+    "authors": [
+      "Shihao Qi",
+      "Jie Ma",
+      "Rui Xing",
+      "Wei Guo",
+      "Xiao Huang",
+      "Zhitao Gao",
+      "Jianhao Deng",
+      "Jun Liu",
+      "Lingling Zhang",
+      "Bifan Wei",
+      "Boqian Yang",
+      "Pinghui Wang",
+      "Jianwen Sun",
+      "Jing Tao",
+      "Yaqiang Wu",
+      "Hui Liu",
+      "Yu Yao",
+      "Tongliang Liu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.14892",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.14892",
+    "pdf_url": "https://arxiv.org/pdf/2605.14892",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.14527",
+    "title": "Lang2MLIP: End-to-End Language-to-Machine Learning Interatomic Potential Development with Autonomous Agentic Workflows",
+    "url": "https://arxiv.org/abs/2605.14527",
+    "published": "2026-05-14",
+    "updated": "2026-05-14",
+    "authors": [
+      "Wenwen Li",
+      "Yuki Orimo",
+      "Nontawat Charoenphakdee"
+    ],
+    "categories": [
+      "cs.LG",
+      "cond-mat.mtrl-sci",
+      "physics.comp-ph"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.14527",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.14527",
+    "pdf_url": "https://arxiv.org/pdf/2605.14527",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.13481",
+    "title": "PersonalAI 2.0: Enhancing knowledge graph traversal/retrieval with planning mechanism for Personalized LLM Agents",
+    "url": "https://arxiv.org/abs/2605.13481",
+    "published": "2026-05-13",
+    "updated": "2026-05-13",
+    "authors": [
+      "Mikhail Menschikov",
+      "Matvey Iskornev",
+      "Alexander Kharitonov",
+      "Alina Bogdanova",
+      "Mikhail Belkin",
+      "Ekaterina Lisitsyna",
+      "Artyom Sosedka",
+      "Victoria Dochkina",
+      "Ruslan Kostoev",
+      "Ilia Perepechkin",
+      "Evgeny Burnaev"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.13481",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.13481",
+    "pdf_url": "https://arxiv.org/pdf/2605.13481",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.12260",
+    "title": "PRISM: Pareto-Efficient Retrieval over Intent-Aware Structured Memory for Long-Horizon Agents",
+    "url": "https://arxiv.org/abs/2605.12260",
+    "published": "2026-05-12",
+    "updated": "2026-05-22",
+    "authors": [
+      "Jingyi Peng",
+      "Zhongwei Wan",
+      "Weiting Liu",
+      "Qiuzhuang Sun"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.12260",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.12260",
+    "pdf_url": "https://arxiv.org/pdf/2605.12260",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.11225",
+    "title": "PIVOT: Bridging Planning and Execution in LLM Agents via Trajectory Refinement",
+    "url": "https://arxiv.org/abs/2605.11225",
+    "published": "2026-05-11",
+    "updated": "2026-05-11",
+    "authors": [
+      "Tuo Zhang",
+      "Alin-Ionut Popa",
+      "Yan Xu",
+      "Rui Song",
+      "Dimitrios Dimitriadis"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "planning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm",
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.11225",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.11225",
+    "pdf_url": "https://arxiv.org/pdf/2605.11225",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.09692",
+    "title": "Causal state binding predicts action control in language agents",
+    "url": "https://arxiv.org/abs/2605.09692",
+    "published": "2026-05-10",
+    "updated": "2026-06-01",
+    "authors": [
+      "Xiao Jia"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "memory",
+      "planning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.09692",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.09692",
+    "pdf_url": "https://arxiv.org/pdf/2605.09692",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.07251",
+    "title": "Can Agents Price a Reaction? Evaluating LLMs on Chemical Cost Reasoning",
+    "url": "https://arxiv.org/abs/2605.07251",
+    "published": "2026-05-08",
+    "updated": "2026-05-08",
+    "authors": [
+      "Yuyang Wu",
+      "Yue Huang",
+      "Shuaike Shen",
+      "Xujian Wang",
+      "Shuhao Zhang",
+      "Qiyao Xue",
+      "Weichen Liu",
+      "Runtian Gao",
+      "Jian Ma",
+      "Xiangliang Zhang",
+      "Olexandr Isayev"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.07251",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.07251",
+    "pdf_url": "https://arxiv.org/pdf/2605.07251",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.06713",
+    "title": "Agentic AI and the Industrialization of Cyber Offense: Forecast, Consequences, and Defensive Priorities for Enterprises and the Mittelstand",
+    "url": "https://arxiv.org/abs/2605.06713",
+    "published": "2026-05-06",
+    "updated": "2026-05-06",
+    "authors": [
+      "Christopher Koch"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.06713",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.06713",
+    "pdf_url": "https://arxiv.org/pdf/2605.06713",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.02240",
+    "title": "PhysicianBench: Evaluating LLM Agents in Real-World EHR Environments",
+    "url": "https://arxiv.org/abs/2605.02240",
+    "published": "2026-05-04",
+    "updated": "2026-05-04",
+    "authors": [
+      "Ruoqi Liu",
+      "Imran Q. Mohiuddin",
+      "Austin J. Schoeffler",
+      "Kavita Renduchintala",
+      "Ashwin Nayak",
+      "Prasantha L. Vemu",
+      "Shivam C. Vedak",
+      "Kameron C. Black",
+      "John L. Havlik",
+      "Isaac Ogunmola",
+      "Stephen P. Ma",
+      "Roopa Dhatt",
+      "Jonathan H. Chen"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.02240",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.02240",
+    "pdf_url": "https://arxiv.org/pdf/2605.02240",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2604.11557",
+    "title": "UniToolCall: Unifying Tool-Use Representation, Data, and Evaluation for LLM Agents",
+    "url": "https://arxiv.org/abs/2604.11557",
+    "published": "2026-04-13",
+    "updated": "2026-05-25",
+    "authors": [
+      "Yijuan Liang",
+      "Xinghao Chen",
+      "Yifan Ge",
+      "Ziyi Wu",
+      "Hao Wu",
+      "Changyu Zeng",
+      "Wei Xing",
+      "Xiaoyu Shen"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2604.11557",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.11557",
+    "pdf_url": "https://arxiv.org/pdf/2604.11557",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2604.10577",
+    "title": "The Blind Spot of Agent Safety: How Benign User Instructions Expose Critical Vulnerabilities in Computer-Use Agents",
+    "url": "https://arxiv.org/abs/2604.10577",
+    "published": "2026-04-12",
+    "updated": "2026-04-17",
+    "authors": [
+      "Xuwei Ding",
+      "Skylar Zhai",
+      "Linxin Song",
+      "Jiate Li",
+      "Taiwei Shi",
+      "Nicholas Meade",
+      "Siva Reddy",
+      "Jian Kang",
+      "Jieyu Zhao"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.10577",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.10577",
+    "pdf_url": "https://arxiv.org/pdf/2604.10577",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.24257",
+    "title": "Memory-Augmented Vision-Language Agents for Persistent and Semantically Consistent Object Captioning",
+    "url": "https://arxiv.org/abs/2603.24257",
+    "published": "2026-03-25",
+    "updated": "2026-03-30",
+    "authors": [
+      "Tommaso Galliena",
+      "Stefano Rosa",
+      "Tommaso Apicella",
+      "Pietro Morerio",
+      "Alessio Del Bue",
+      "Lorenzo Natale"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.24257",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.24257",
+    "pdf_url": "https://arxiv.org/pdf/2603.24257",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.10492",
+    "title": "Human-AI Co-reasoning for Clinical Diagnosis with Evidence-Integrated Language Agent",
+    "url": "https://arxiv.org/abs/2603.10492",
+    "published": "2026-03-11",
+    "updated": "2026-03-18",
+    "authors": [
+      "Zhongzhen Huang",
+      "Yan Ling",
+      "Hong Chen",
+      "Ye Feng",
+      "Li Wu",
+      "Linjie Mu",
+      "Shaoting Zhang",
+      "Xiaofan Zhang",
+      "Kun Qian",
+      "Xiaomu Li"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.10492",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.10492",
+    "pdf_url": "https://arxiv.org/pdf/2603.10492",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.07496",
+    "title": "From Thinker to Society: Security in Hierarchical Autonomy Evolution of AI Agents",
+    "url": "https://arxiv.org/abs/2603.07496",
+    "published": "2026-03-08",
+    "updated": "2026-03-21",
+    "authors": [
+      "Xiaolei Zhang",
+      "Lu Zhou",
+      "Xiaogang Xu",
+      "Jiafei Wu",
+      "Tianyu Du",
+      "Heqing Huang",
+      "Hao Peng",
+      "Zhe Liu"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.07496",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.07496",
+    "pdf_url": "https://arxiv.org/pdf/2603.07496",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.03680",
+    "title": "MAGE: Meta-Reinforcement Learning for Language Agents toward Strategic Exploration and Exploitation",
+    "url": "https://arxiv.org/abs/2603.03680",
+    "published": "2026-03-04",
+    "updated": "2026-03-04",
+    "authors": [
+      "Lu Yang",
+      "Zelai Xu",
+      "Minyang Xie",
+      "Jiaxuan Gao",
+      "Zhao Shok",
+      "Yu Wang",
+      "Yi Wu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "memory",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.03680",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.03680",
+    "pdf_url": "https://arxiv.org/pdf/2603.03680",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.02711",
+    "title": "A Natural Language Agentic Approach to Study Affective Polarization",
+    "url": "https://arxiv.org/abs/2603.02711",
+    "published": "2026-03-03",
+    "updated": "2026-03-03",
+    "authors": [
+      "Stephanie Anneris Malvicini",
+      "Ewelina Gajewska",
+      "Arda Derbent",
+      "Katarzyna Budzynska",
+      "Jarosław A. Chudziak",
+      "Maria Vanina Martinez"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.02711",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.02711",
+    "pdf_url": "https://arxiv.org/pdf/2603.02711",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.03515",
+    "title": "The Controllability Trap: A Governance Framework for Military AI Agents",
+    "url": "https://arxiv.org/abs/2603.03515",
+    "published": "2026-03-03",
+    "updated": "2026-03-03",
+    "authors": [
+      "Subramanyam Sahoo"
+    ],
+    "categories": [
+      "cs.CY",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.03515",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.03515",
+    "pdf_url": "https://arxiv.org/pdf/2603.03515",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.03242",
+    "title": "DRAFT: Task Decoupled Latent Reasoning for Agent Safety",
+    "url": "https://arxiv.org/abs/2604.03242",
+    "published": "2026-02-11",
+    "updated": "2026-02-11",
+    "authors": [
+      "Lin Wang",
+      "Junfeng Fang",
+      "Dan Zhang",
+      "Fei Shen",
+      "Xiang Wang",
+      "Tat-Seng Chua"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.03242",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.03242",
+    "pdf_url": "https://arxiv.org/pdf/2604.03242",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.08721",
+    "title": "KernelCraft: Benchmarking for Agentic Close-to-Metal Kernel Generation on Emerging Hardware",
+    "url": "https://arxiv.org/abs/2603.08721",
+    "published": "2026-02-10",
+    "updated": "2026-05-29",
+    "authors": [
+      "Jiayi Nie",
+      "Haoran Wu",
+      "Yao Lai",
+      "Zeyu Cao",
+      "Cheng Zhang",
+      "Binglei Lou",
+      "Erwei Wang",
+      "Jianyi Cheng",
+      "Timothy M. Jones",
+      "Robert Mullins",
+      "Rika Antonova",
+      "Yiren Zhao"
+    ],
+    "categories": [
+      "cs.AR",
+      "cs.LG",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2603.08721",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.08721",
+    "pdf_url": "https://arxiv.org/pdf/2603.08721",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2602.03224",
+    "title": "TAME: A Trustworthy Test-Time Evolution of Agent Memory with Systematic Benchmarking",
+    "url": "https://arxiv.org/abs/2602.03224",
+    "published": "2026-02-03",
+    "updated": "2026-06-06",
+    "authors": [
+      "Yu Cheng",
+      "Yongkang Hu",
+      "Jiuan Zhou",
+      "Yushuo Zhang",
+      "Yihang Chen",
+      "Huichi Zhou",
+      "Mingang Chen",
+      "Zhizhong Zhang",
+      "Kun Shao",
+      "Yuan Xie",
+      "Zhaoxia Yin"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "memory",
+      "reasoning"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.03224",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.03224",
+    "pdf_url": "https://arxiv.org/pdf/2602.03224",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2512.11682",
+    "title": "MedAI: Evaluating TxAgent's Therapeutic Agentic Reasoning in the NeurIPS CURE-Bench Competition",
+    "url": "https://arxiv.org/abs/2512.11682",
+    "published": "2025-12-12",
+    "updated": "2026-06-15",
+    "authors": [
+      "Tim Cofala",
+      "Christian Kalfar",
+      "Jingge Xiao",
+      "Johanna Schrader",
+      "Michelle Tang",
+      "Wolfgang Nejdl"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2512.11682",
+    "source": "arxiv",
+    "source_id": "arxiv:2512.11682",
+    "pdf_url": "https://arxiv.org/pdf/2512.11682",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2511.15203",
+    "title": "Taxonomy, Evaluation and Exploitation of IPI-Centric LLM Agent Defense Frameworks",
+    "url": "https://arxiv.org/abs/2511.15203",
+    "published": "2025-11-19",
+    "updated": "2025-11-19",
+    "authors": [
+      "Zimo Ji",
+      "Xunguang Wang",
+      "Zongjie Li",
+      "Pingchuan Ma",
+      "Yudong Gao",
+      "Daoyuan Wu",
+      "Xincheng Yan",
+      "Tian Tian",
+      "Shuai Wang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2511.15203",
+    "source": "arxiv",
+    "source_id": "arxiv:2511.15203",
+    "pdf_url": "https://arxiv.org/pdf/2511.15203",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2510.18586",
+    "title": "TokenCake: A KV-Cache-centric Serving Framework for LLM-based Multi-Agent Applications",
+    "url": "https://arxiv.org/abs/2510.18586",
+    "published": "2025-10-21",
+    "updated": "2026-05-20",
+    "authors": [
+      "Zhuohang Bian",
+      "Feiyang Wu",
+      "Zhuoran Li",
+      "Teng Ma",
+      "Youwei Zhuo"
+    ],
+    "categories": [
+      "cs.DC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "multi-agent"
+    ],
+    "score": 17,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2510.18586",
+    "source": "arxiv",
+    "source_id": "arxiv:2510.18586",
+    "pdf_url": "https://arxiv.org/pdf/2510.18586",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2607.06001",
+    "title": "Information Limits and Attractor Dynamics in Economies of Frontier LLM Agents: A Pre-Registered Test",
+    "url": "https://arxiv.org/abs/2607.06001",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Cheng Qian"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-safety",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.06001",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06001",
+    "pdf_url": "https://arxiv.org/pdf/2607.06001",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.06000",
+    "title": "Context-to-Execution Integrity for LLM Agents",
+    "url": "https://arxiv.org/abs/2607.06000",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Igor Santos-Grueiro"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "coding-agent",
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.06000",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06000",
+    "pdf_url": "https://arxiv.org/pdf/2607.06000",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.06413",
+    "title": "An Experimental Design Approach to Evaluating Agentic AI's Autonomous Model Discovery",
+    "url": "https://arxiv.org/abs/2607.06413",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Hao He",
+      "Xueying Liu",
+      "Chris J. Kuhlman",
+      "Xinwei Deng"
+    ],
+    "categories": [
+      "stat.ME",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.06413",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06413",
+    "pdf_url": "https://arxiv.org/pdf/2607.06413",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.06411",
+    "title": "RuBench: A Repository-Level Agentic Coding Benchmark with Natively Authored Russian Task Specifications",
+    "url": "https://arxiv.org/abs/2607.06411",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Evgeny Shilov"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.06411",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06411",
+    "pdf_url": "https://arxiv.org/pdf/2607.06411",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.06101",
+    "title": "Agents That Teach: Towards Designing Incidental Learning Back into AI-Assisted Software Development",
+    "url": "https://arxiv.org/abs/2607.06101",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Rohit Mehra",
+      "Samdyuti Suri",
+      "Prithviraj K Tagadinamani",
+      "Kapil Singi",
+      "Vikrant Kaulgud",
+      "Adam P. Burden"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.CY",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.06101",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06101",
+    "pdf_url": "https://arxiv.org/pdf/2607.06101",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.05659",
+    "title": "Agents with Feelings? Personality and Emotion in Multi-Agent Software Teams",
+    "url": "https://arxiv.org/abs/2607.05659",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Yunyan Ding",
+      "Thomas Zimmermann",
+      "Iftekhar Ahmed"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.05659",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05659",
+    "pdf_url": "https://arxiv.org/pdf/2607.05659",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.04713",
+    "title": "RSPO: Reward-Swap Policy Optimization for Multi-Turn LLM Agents",
+    "url": "https://arxiv.org/abs/2607.04713",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Qiang Liu",
+      "Taian Guo",
+      "Ruizhi Qiao",
+      "Xing Sun"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.04713",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04713",
+    "pdf_url": "https://arxiv.org/pdf/2607.04713",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.04569",
+    "title": "LLMs for Agentic Home Energy Management",
+    "url": "https://arxiv.org/abs/2607.04569",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Sokipriala Jonah"
+    ],
+    "categories": [
+      "eess.SY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling",
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.04569",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04569",
+    "pdf_url": "https://arxiv.org/pdf/2607.04569",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2607.04617",
+    "title": "MRMS: A Multi-Resolution Memory Substrate for Long-Lived AI Agents",
+    "url": "https://arxiv.org/abs/2607.04617",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Jizhizi Li",
+      "Amy Shi-Nash"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.04617",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04617",
+    "pdf_url": "https://arxiv.org/pdf/2607.04617",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.05363",
+    "title": "SovereignPA-Bench: Evaluating User-Owned Personal Agents under Evolving Intent, Platform Mediation, and Consent Constraints",
+    "url": "https://arxiv.org/abs/2607.05363",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Dylan Zongmin Liu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "embodied-agent",
+      "memory",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.05363",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05363",
+    "pdf_url": "https://arxiv.org/pdf/2607.05363",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.05458",
+    "title": "Learning to Control LLM Agent Harnesses with Offline Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2607.05458",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Haiwen Yi",
+      "Xinyuan Song"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.05458",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05458",
+    "pdf_url": "https://arxiv.org/pdf/2607.05458",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.04219",
+    "title": "Agentic IoT: Architectures, Applications, and Challenges Toward the Internet of Agents",
+    "url": "https://arxiv.org/abs/2607.04219",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Rümeysa Hilal Sevinç",
+      "Bahaeddin Türkoğlu",
+      "İbrahim Kök"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA",
+      "cs.NI"
+    ],
+    "topics": [
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.04219",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04219",
+    "pdf_url": "https://arxiv.org/pdf/2607.04219",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.04212",
+    "title": "An Evaluation of Role-Based Multi-Agent Code Generation on Repository-Scale Problems",
+    "url": "https://arxiv.org/abs/2607.04212",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Benedetta Donato",
+      "Noah Hagar-Dent",
+      "Aaron Worsnop",
+      "Leonardo Mariani",
+      "Valerio Terragni"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.04212",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04212",
+    "pdf_url": "https://arxiv.org/pdf/2607.04212",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.04149",
+    "title": "Beyond Scene Priors: Fine-Grained Traffic Scene Reasoning with Benchmarking and Query-Guided Small-Object Focus",
+    "url": "https://arxiv.org/abs/2607.04149",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Waikit Xiu",
+      "Qiang Lu",
+      "Zian Wang",
+      "Xinjie Yang",
+      "Zhiwei Chen",
+      "Chen Sun",
+      "Xiying Li"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "multi-agent",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.04149",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04149",
+    "pdf_url": "https://arxiv.org/pdf/2607.04149",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.04034",
+    "title": "The \"I Don't Know\" Filter: Enhancing Agentic Reliability in Function Calling",
+    "url": "https://arxiv.org/abs/2607.04034",
+    "published": "2026-07-04",
+    "updated": "2026-07-04",
+    "authors": [
+      "Stefan Broecker",
+      "Mason del Rosario",
+      "Boris Selitser",
+      "Thomas Strohmer"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "function-calling"
+    ],
+    "arxiv_id": "2607.04034",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04034",
+    "pdf_url": "https://arxiv.org/pdf/2607.04034",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2607.03691",
+    "title": "Don't Blame the Large Language Model: How Scaffolding Evolution Shapes Coding Agent Quality",
+    "url": "https://arxiv.org/abs/2607.03691",
+    "published": "2026-07-04",
+    "updated": "2026-07-04",
+    "authors": [
+      "Oussama Ben Sghaier",
+      "Hao Li",
+      "Bram Adams",
+      "Ahmed E. Hassan"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.03691",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03691",
+    "pdf_url": "https://arxiv.org/pdf/2607.03691",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.02927",
+    "title": "VideoSearcher: Empowering Video Deep Research with Multi-Tool Agentic Reasoning via Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2607.02927",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Zhenkun Gao",
+      "Yicheng Bao",
+      "Jinlong Peng",
+      "Xueheng Li",
+      "Theo Huang",
+      "Bangwei Liu",
+      "Kunquan Li",
+      "Zhenye Gan",
+      "Tao Hu",
+      "Chengjun Xie",
+      "Mingqian Yang",
+      "Xuanhua He",
+      "Zhizhong Zhang",
+      "Xin Tan",
+      "Chengjie Wang",
+      "Yuan Xie"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2607.02927",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02927",
+    "pdf_url": "https://arxiv.org/pdf/2607.02927",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2607.03525",
+    "title": "GameEngineBench: Evaluating Coding Agents on Real C++ Runtime Environments",
+    "url": "https://arxiv.org/abs/2607.03525",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Brian La",
+      "Sejoon Chang",
+      "Ben Kim",
+      "Junyoung Bae",
+      "Aamish Ahmad Beg",
+      "Sei Chang",
+      "Gonzalo Gonzalez-Pumariega"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "embodied-agent",
+      "memory",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.03525",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03525",
+    "pdf_url": "https://arxiv.org/pdf/2607.03525",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.02689",
+    "title": "S-EMBER: A Large-Scale Benchmark for Streaming Egocentric Memory Retrieval",
+    "url": "https://arxiv.org/abs/2607.02689",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Xiaodong Wang",
+      "Xuanyi Zhao",
+      "Pedro Rodriguez",
+      "Devendra Singh Sachan",
+      "Barlas Oguz",
+      "Seungwhan Moon",
+      "Shang-Wen Li",
+      "Gargi Ghosh",
+      "Xin Dong",
+      "Wen-Tau Yih"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.02689",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02689",
+    "pdf_url": "https://arxiv.org/pdf/2607.02689",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.01788",
+    "title": "KRCA: An Efficient Root Cause Analysis System in Hyper-Scale Microservice Systems via Agentic AI",
+    "url": "https://arxiv.org/abs/2607.01788",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Jiamin Jiang",
+      "Jingfei Feng",
+      "Yu Luo",
+      "Qingliang Zhang",
+      "Yongqian Su",
+      "Wenwei Gu",
+      "Shenglin Zhang",
+      "Tianyu Cui",
+      "Yao Wu",
+      "Jielong Huang",
+      "Nan Qi",
+      "Dan Pei"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.01788",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01788",
+    "pdf_url": "https://arxiv.org/pdf/2607.01788",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.02716",
+    "title": "Evaluating Large Language Models for Decision-Making in Agent-Based Urban Mobility Simulations",
+    "url": "https://arxiv.org/abs/2607.02716",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Bruno Cascaes Alves",
+      "Míriam Blank Born",
+      "Ulisses Gilioli Francescatto Júnior",
+      "Felipe Moura Goulart",
+      "Letícia Brandão Caldas",
+      "Marilton Sanchotene de Aguiar"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "multi-agent",
+      "planning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.02716",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02716",
+    "pdf_url": "https://arxiv.org/pdf/2607.02716",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.02186",
+    "title": "UA-ChatDev: Uncertainty-Aware Multi-Agent Collaboration for Reliable Software Development",
+    "url": "https://arxiv.org/abs/2607.02186",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Temitayo Olamilekan Ogunsusi",
+      "Lijun Qian",
+      "Xishuang Dong"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.02186",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02186",
+    "pdf_url": "https://arxiv.org/pdf/2607.02186",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.01084",
+    "title": "Can Agents Generalize to the Open World? Unveiling the Fragility of Static Training in Tool Use",
+    "url": "https://arxiv.org/abs/2607.01084",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Song-Lin Lv",
+      "Weiming Wu",
+      "Rui Zhu",
+      "Zi-Jian Cheng",
+      "Lan-Zhe Guo"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.01084",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01084",
+    "pdf_url": "https://arxiv.org/pdf/2607.01084",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.02599",
+    "title": "AgentLTL: A Trace-Verification Framework for Measuring, Enforcing, and Training Procedural Compliance in Tool-Using LLM Agents",
+    "url": "https://arxiv.org/abs/2607.02599",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Laïla Elkoussy",
+      "Julien Perez"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.LO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.02599",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02599",
+    "pdf_url": "https://arxiv.org/pdf/2607.02599",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.00555",
+    "title": "Rise From The Ashes: LLM-based Static Analysis for Deep Learning Framework Bugs",
+    "url": "https://arxiv.org/abs/2607.00555",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Shaoyu Yang",
+      "Haifeng Lin",
+      "Chunrong Fang",
+      "Xiang Chen",
+      "Wei Cheng",
+      "Jiawei Liu",
+      "Yiyu Zhang",
+      "Hongyu Liu",
+      "Zhenyu Chen"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "multi-agent",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.00555",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00555",
+    "pdf_url": "https://arxiv.org/pdf/2607.00555",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.00436",
+    "title": "PHREEQC-MCQ-200: A Diagnostic Benchmark for Tool-Augmented Scientific Simulator Agents",
+    "url": "https://arxiv.org/abs/2607.00436",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Ke Zhang",
+      "Sahchit Chundur",
+      "Mohammad Javad Qomi",
+      "Maziar Raissi"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2607.00436",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00436",
+    "pdf_url": "https://arxiv.org/pdf/2607.00436",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2607.00502",
+    "title": "A Task-State Representation for Long-Horizon Mobile GUI Agents",
+    "url": "https://arxiv.org/abs/2607.00502",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Yujie Zheng",
+      "Zikang Liu",
+      "Xin Zhao",
+      "Ji-Rong Wen"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2607.00502",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00502",
+    "pdf_url": "https://arxiv.org/pdf/2607.00502",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2607.00440",
+    "title": "Minos: A Multi-Agent Collaborative Framework for Provenance-Based Backward Tracking",
+    "url": "https://arxiv.org/abs/2607.00440",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Jiahui Wang",
+      "Zhenyuan Li",
+      "Zhengkai Wang",
+      "Xiangmin Shen",
+      "Fan Zhang"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "rag",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.00440",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00440",
+    "pdf_url": "https://arxiv.org/pdf/2607.00440",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.00233",
+    "title": "From Signals to Structure: How Memory Architecture Drives Language Emergence in LLM Agents",
+    "url": "https://arxiv.org/abs/2607.00233",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Yashar Talebirad",
+      "Eden Redman",
+      "Ali Parsaee",
+      "Osmar R. Zaiane"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.IT",
+      "cs.MA"
+    ],
+    "topics": [
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.00233",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00233",
+    "pdf_url": "https://arxiv.org/pdf/2607.00233",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.32034",
+    "title": "QVal: Cheaply Evaluating Dense Supervision Signals for Long-Horizon LLM Agents",
+    "url": "https://arxiv.org/abs/2606.32034",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Sergio Hernández-Gutiérrez",
+      "Matteo Merler",
+      "Ilze Amanda Auzina",
+      "Joschka Strüber",
+      "Ameya Prabhu",
+      "Matthias Bethge"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.32034",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.32034",
+    "pdf_url": "https://arxiv.org/pdf/2606.32034",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.02579",
+    "title": "When Not to Write Memory: Governing False Promotion from Correlated Agent Traces",
+    "url": "https://arxiv.org/abs/2607.02579",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Yijiashun Qi",
+      "Xiang Xu",
+      "Yuxuan Li"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "language-agent"
+    ],
+    "arxiv_id": "2607.02579",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02579",
+    "pdf_url": "https://arxiv.org/pdf/2607.02579",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.31339",
+    "title": "Verification-Gated Agentic Mission-State Governance for Intelligent Industrial Multi-Robot Systems",
+    "url": "https://arxiv.org/abs/2606.31339",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Guoqin Tang",
+      "Qingxuan Jia",
+      "Yichen Tan",
+      "Zeyuan Huang",
+      "Ning Ji",
+      "Gang Chen"
+    ],
+    "categories": [
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "embodied-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.31339",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31339",
+    "pdf_url": "https://arxiv.org/pdf/2606.31339",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.31980",
+    "title": "DigitalCoach: Communication and Grounding Gaps in Human and Agentic Computer Use Coaching",
+    "url": "https://arxiv.org/abs/2606.31980",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Meng Chen",
+      "Anya Ji",
+      "Tsung-Han Wu",
+      "Tobias Maringgele",
+      "David M. Chan",
+      "Alane Suhr",
+      "Amy Pavel"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "rag"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.31980",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31980",
+    "pdf_url": "https://arxiv.org/pdf/2606.31980",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.31134",
+    "title": "Beyond the Library: An Agentic Framework for Autoformalizing Research Mathematics",
+    "url": "https://arxiv.org/abs/2606.31134",
+    "published": "2026-06-30",
+    "updated": "2026-07-01",
+    "authors": [
+      "Arshia Soltani Moakhar",
+      "Iman Gholami",
+      "Max Springer",
+      "Mahdi JafariRaviz",
+      "MohammadTaghi Hajiaghayi"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "rag",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.31134",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31134",
+    "pdf_url": "https://arxiv.org/pdf/2606.31134",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.31200",
+    "title": "Agentic RAG-VLM: Affordance-Aware Retrieval-Augmented Generation with Self-Reflective Planning for Robotic Grasping",
+    "url": "https://arxiv.org/abs/2606.31200",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Tao Chen",
+      "Lizheng Liu",
+      "Jiaxu Wang",
+      "Ziyue Jiang",
+      "Ruiqi Tian",
+      "JiGuang Huo",
+      "Zhongxue Gan"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.31200",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31200",
+    "pdf_url": "https://arxiv.org/pdf/2606.31200",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.30251",
+    "title": "TACO: Tool-Augmented Credit Optimization for Agentic Tool Use",
+    "url": "https://arxiv.org/abs/2606.30251",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Mingkuan Feng",
+      "Jinyang Wu",
+      "Hao Gu",
+      "Fangrui Lv",
+      "Ruihan Jin",
+      "Chuyuan Zhang",
+      "Zhengqi Wen",
+      "Jianhua Tao"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.30251",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30251",
+    "pdf_url": "https://arxiv.org/pdf/2606.30251",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.30294",
+    "title": "Rehearsed Multi-Agent Live Product Demonstrations with Real-Time Voice Question Answering",
+    "url": "https://arxiv.org/abs/2606.30294",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Rahul Khedar",
+      "Mayank Malhotra",
+      "Avinash Karn",
+      "Mouli V",
+      "Prakhar Mehrotra"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.HC",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "multi-agent",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.30294",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30294",
+    "pdf_url": "https://arxiv.org/pdf/2606.30294",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.29354",
+    "title": "When LLMs Develop Languages: Symbolic Communication for Efficient Multi-Agent Reasoning",
+    "url": "https://arxiv.org/abs/2606.29354",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Zhengqi Pei",
+      "Qingming Huang",
+      "Shuhui Wang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.NE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.29354",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29354",
+    "pdf_url": "https://arxiv.org/pdf/2606.29354",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.29315",
+    "title": "Hierarchical Experimentalist Agents",
+    "url": "https://arxiv.org/abs/2606.29315",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Abhranil Chandra",
+      "Sankaran Vaidyanathan",
+      "Utsav Dhanuka",
+      "Varun Gandhi",
+      "Scott Niekum"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.29315",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29315",
+    "pdf_url": "https://arxiv.org/pdf/2606.29315",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.29445",
+    "title": "Bridging VideoQA and Video-Guided Agentic Tasks via Generalized Keyframe Extraction",
+    "url": "https://arxiv.org/abs/2606.29445",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Sunqi Fan",
+      "Qingle Liu",
+      "Runqi Yin",
+      "Meng-Hao Guo",
+      "Shuojin Yang"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.29445",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29445",
+    "pdf_url": "https://arxiv.org/pdf/2606.29445",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.28841",
+    "title": "LAMP: Lean-based Agentic framework with MCP and Proof Repair",
+    "url": "https://arxiv.org/abs/2606.28841",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Santhana Srinivasan R",
+      "Maithilee Patawar"
+    ],
+    "categories": [
+      "cs.LO",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "coding-agent",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.28841",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28841",
+    "pdf_url": "https://arxiv.org/pdf/2606.28841",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.28450",
+    "title": "LLM agents security duality: a comprehensive survey of self-security and empowered cybersecurity",
+    "url": "https://arxiv.org/abs/2606.28450",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Yiwei Xu",
+      "Yong Zhuang",
+      "Xuanming Liu",
+      "Tian Zhang",
+      "Bowen Xiao",
+      "Xiaoyang Xu",
+      "Delong Jiang",
+      "Juan Wang",
+      "Hongxin Hu"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "llm-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.28450",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28450",
+    "pdf_url": "https://arxiv.org/pdf/2606.28450",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.28480",
+    "title": "TUA-Bench: A Benchmark for General-Purpose Terminal-Use Agents",
+    "url": "https://arxiv.org/abs/2606.28480",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Shoufa Chen",
+      "Luyuan Wang",
+      "Xuan Yang",
+      "Zhiheng Liu",
+      "Yuren Cong",
+      "Yuanfeng Ji",
+      "Feiyan Zhou",
+      "Xiaohui Zhang",
+      "Fanny Yang",
+      "Belinda Zeng"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.28480",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28480",
+    "pdf_url": "https://arxiv.org/pdf/2606.28480",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.27350",
+    "title": "CHIA: An open-source framework for principled, agentic AI-driven hardware/software co-design research",
+    "url": "https://arxiv.org/abs/2606.27350",
+    "published": "2026-06-25",
+    "updated": "2026-06-27",
+    "authors": [
+      "Angela Cui",
+      "Ferran Hermida-Rivera",
+      "Jack Toubes",
+      "Raghav Gupta",
+      "Jim Fang",
+      "Chengyi Lux Zhang",
+      "Ella Schwarz",
+      "Junha Kim",
+      "Yakun Sophia Shao",
+      "Borivoje Nikolic",
+      "Christopher W. Fletcher",
+      "Sagar Karandikar"
+    ],
+    "categories": [
+      "cs.AR"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.27350",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27350",
+    "pdf_url": "https://arxiv.org/pdf/2606.27350",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.26721",
+    "title": "Knowledge-Based Pull Requests: A Trusted Workflow for Agent-Mediated Knowledge Collaboration",
+    "url": "https://arxiv.org/abs/2606.26721",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Xinyu Zhang",
+      "Weiwei Sun"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "memory",
+      "multi-agent",
+      "planning",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.26721",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26721",
+    "pdf_url": "https://arxiv.org/pdf/2606.26721",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.27330",
+    "title": "Empowering GUI Agents via Autonomous Experience Exploration and Hindsight Experience Utilization for Task Planning",
+    "url": "https://arxiv.org/abs/2606.27330",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Tianyi Men",
+      "Zhuoran Jin",
+      "Pengfei Cao",
+      "Yubo Chen",
+      "Kang Liu",
+      "Jun Zhao"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.CV",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.27330",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27330",
+    "pdf_url": "https://arxiv.org/pdf/2606.27330",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.27009",
+    "title": "Semantic Early-Stopping for Iterative LLM Agent Loops",
+    "url": "https://arxiv.org/abs/2606.27009",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Sahil Shrivastava"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.27009",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27009",
+    "pdf_url": "https://arxiv.org/pdf/2606.27009",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.27483",
+    "title": "Internalizing the Future: A Unified Agentic Training Paradigm for World Model Planning",
+    "url": "https://arxiv.org/abs/2606.27483",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Xuan Zhang",
+      "Zhijian Zhou",
+      "Lingfeng Qiao",
+      "Yulei Qin",
+      "Ke Li",
+      "Xing Sun",
+      "Xiaoyu Tan",
+      "Chao Qu",
+      "Yuan Qi"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "world-model"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.27483",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27483",
+    "pdf_url": "https://arxiv.org/pdf/2606.27483",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.26216",
+    "title": "CyberChainBench: Can AI Agents Secure Smart Contracts Against Real-World On-Chain Vulnerabilities?",
+    "url": "https://arxiv.org/abs/2606.26216",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Jintao Huang",
+      "Fengqing Jiang",
+      "Radha Poovendran",
+      "Zhiqiang Lin"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.26216",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26216",
+    "pdf_url": "https://arxiv.org/pdf/2606.26216",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.26203",
+    "title": "Agentic Analysis for Agentic Infrastructure: An LLM-Powered Pipeline for Comparative Governance of DAO and Corporate AI Protocols",
+    "url": "https://arxiv.org/abs/2606.26203",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Yutian Wang",
+      "Luyao Zhang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CY",
+      "cs.MA",
+      "cs.SI"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.26203",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26203",
+    "pdf_url": "https://arxiv.org/pdf/2606.26203",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.25760",
+    "title": "Uncertainty Quantification for Computer-Use Agents: A Benchmark across Vision-Language Models and GUI Grounding Datasets",
+    "url": "https://arxiv.org/abs/2606.25760",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Divake Kumar",
+      "Sina Tayebati",
+      "Devashri Naik",
+      "Amanda Sofie Rios",
+      "Nilesh Ahuja",
+      "Omesh Tickoo",
+      "Ranganath Krishnan",
+      "Amit Ranjan Trivedi"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL",
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "rag"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.25760",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25760",
+    "pdf_url": "https://arxiv.org/pdf/2606.25760",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.25651",
+    "title": "MedGuards: Multi-Agent System for Reliable Medical Error Detection and Correction",
+    "url": "https://arxiv.org/abs/2606.25651",
+    "published": "2026-06-24",
+    "updated": "2026-06-26",
+    "authors": [
+      "Congbo Ma",
+      "Hu Wang",
+      "Yichun Zhang",
+      "Farah E. Shamout"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.25651",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25651",
+    "pdf_url": "https://arxiv.org/pdf/2606.25651",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.24595",
+    "title": "MEMPROBE: Probing Long-Term Agent Memory via Hidden User-State Recovery",
+    "url": "https://arxiv.org/abs/2606.24595",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Enze Ma",
+      "Yufan Zhou",
+      "Wei-Chieh Huang",
+      "Jie Yang",
+      "Huanhuan Ma",
+      "Zixuan Wang",
+      "Chengze Li",
+      "Chunyu Miao",
+      "Philip S. Yu",
+      "Zhen Wang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.24595",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24595",
+    "pdf_url": "https://arxiv.org/pdf/2606.24595",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.24453",
+    "title": "Bayesian control for coding agents",
+    "url": "https://arxiv.org/abs/2606.24453",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Theodore Papamarkou",
+      "Vladislav Smirnov",
+      "Viktor Mazanov",
+      "Artem Vazhentsev",
+      "Preslav Nakov",
+      "Timothy Baldwin",
+      "Artem Shelmanov"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.24453",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24453",
+    "pdf_url": "https://arxiv.org/pdf/2606.24453",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.24525",
+    "title": "VisCritic: Visual State Comparison as Process Reward for GUI Agents",
+    "url": "https://arxiv.org/abs/2606.24525",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Jiachen Qian"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.24525",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24525",
+    "pdf_url": "https://arxiv.org/pdf/2606.24525",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.24689",
+    "title": "Automated Summarization of Software Documents: An LLM-based Multi-Agent Approach",
+    "url": "https://arxiv.org/abs/2606.24689",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Duc S. H. Nguyen",
+      "Minh T. Nguyen",
+      "Phuong T. Nguyen",
+      "Juri Di Rocco",
+      "Davide Di Ruscio"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.24689",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24689",
+    "pdf_url": "https://arxiv.org/pdf/2606.24689",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.24623",
+    "title": "Privacy-Preserving RAG via Multi-Agent Semantic Rewriting: Achieving Confidentiality Without Compromising Contextual Fidelity",
+    "url": "https://arxiv.org/abs/2606.24623",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Yuanhe Zhao",
+      "Tianyu Zhang",
+      "Huafei Xing",
+      "Derek F. Wong",
+      "Jianbin Li",
+      "Tao Fang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.24623",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24623",
+    "pdf_url": "https://arxiv.org/pdf/2606.24623",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.23664",
+    "title": "MAS-PromptBench: When Does Prompt Optimization Improve Multi-Agent LLM Systems?",
+    "url": "https://arxiv.org/abs/2606.23664",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Juyang Bai",
+      "Laixi Shi"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.23664",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23664",
+    "pdf_url": "https://arxiv.org/pdf/2606.23664",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.23032",
+    "title": "IPO Finance Agent: Benchmark of LLM Financial Analysts Beyond Finance Agent v2, with Automated Rubric Generation, on the SpaceX (SPCX) IPO",
+    "url": "https://arxiv.org/abs/2606.23032",
+    "published": "2026-06-22",
+    "updated": "2026-06-30",
+    "authors": [
+      "Mostapha Benhenda"
+    ],
+    "categories": [
+      "cs.AI",
+      "q-fin.GN"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.23032",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23032",
+    "pdf_url": "https://arxiv.org/pdf/2606.23032",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.22741",
+    "title": "GRADE: Graph Representation of LLM Agent Dependency and Execution",
+    "url": "https://arxiv.org/abs/2606.22741",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Yue Zhao"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "coding-agent",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.22741",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22741",
+    "pdf_url": "https://arxiv.org/pdf/2606.22741",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.23752",
+    "title": "ESAA-Conversational: An Event-Sourced Memory Layer for Continuity, Handoff, and Curation Across Heterogeneous LLM Coding Agents",
+    "url": "https://arxiv.org/abs/2606.23752",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Elzo Brito dos Santos Filho"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "coding-agent",
+      "memory",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.23752",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23752",
+    "pdf_url": "https://arxiv.org/pdf/2606.23752",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.23327",
+    "title": "VideoAgent: All-in-One Framework for Video Understanding and Editing",
+    "url": "https://arxiv.org/abs/2606.23327",
+    "published": "2026-06-22",
+    "updated": "2026-07-03",
+    "authors": [
+      "Hengji Zhou",
+      "Lingxuan Huang",
+      "Jian Wang",
+      "Bing Zhou",
+      "Si Wu",
+      "Lianghao Xia",
+      "Chao Huang"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.23327",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23327",
+    "pdf_url": "https://arxiv.org/pdf/2606.23327",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.22110",
+    "title": "TraceView: Interactive Visualization of Agentic Program Repair Trajectories",
+    "url": "https://arxiv.org/abs/2606.22110",
+    "published": "2026-06-20",
+    "updated": "2026-06-20",
+    "authors": [
+      "Amirali Sajadi",
+      "Tu Nguyen",
+      "Kimmie Huynh",
+      "Esteban Parra",
+      "Preetha Chatterjee"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.22110",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22110",
+    "pdf_url": "https://arxiv.org/pdf/2606.22110",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.22082",
+    "title": "CodeTeam: An LLM-Powered Multi-Agent Framework for Repository-Level Code Generation",
+    "url": "https://arxiv.org/abs/2606.22082",
+    "published": "2026-06-20",
+    "updated": "2026-06-20",
+    "authors": [
+      "Yifei Wang",
+      "Ruiyin Li",
+      "Peng Liang",
+      "Qiong Feng",
+      "Zengyang Li",
+      "Mojtaba Shahin",
+      "Arif Ali Khan"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "planning",
+      "rag"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.22082",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22082",
+    "pdf_url": "https://arxiv.org/pdf/2606.22082",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.21963",
+    "title": "Holmes: Multimodal Agentic Diagnosis for Mixed-Language Mobile Crashes at Industrial Scale",
+    "url": "https://arxiv.org/abs/2606.21963",
+    "published": "2026-06-20",
+    "updated": "2026-06-20",
+    "authors": [
+      "Jia Li",
+      "Wenyuan Ma",
+      "Ting Peng",
+      "Haibin Zheng",
+      "Yuetang Deng"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.21963",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21963",
+    "pdf_url": "https://arxiv.org/pdf/2606.21963",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.21445",
+    "title": "AutoRAS: Learning Robust Agentic Systems with Primitive Representations",
+    "url": "https://arxiv.org/abs/2606.21445",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Yang Yue",
+      "Xuancheng Zhu",
+      "Yuyang Ma",
+      "Guoshun Nan",
+      "Zihan Dou",
+      "Jingru Shan",
+      "Congyu Guo",
+      "Ji Zhang",
+      "Hua Wang",
+      "Jingfeng Zhang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "multi-agent",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.21445",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21445",
+    "pdf_url": "https://arxiv.org/pdf/2606.21445",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.20954",
+    "title": "Learning What Not to Forget: Long-Horizon Agent Memory from a Few Kilobytes of Learning",
+    "url": "https://arxiv.org/abs/2606.20954",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Nusrat Jahan Lia",
+      "Aritra Mazumder"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.20954",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20954",
+    "pdf_url": "https://arxiv.org/pdf/2606.20954",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.19980",
+    "title": "ENPIRE: Agentic Robot Policy Self-Improvement in the Real World",
+    "url": "https://arxiv.org/abs/2606.19980",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Wenli Xiao",
+      "Jia Xie",
+      "Tonghe Zhang",
+      "Haotian Lin",
+      "Letian \"Max\" Fu",
+      "Haoru Xue",
+      "Jalen Lu",
+      "Yi Yang",
+      "Cunxi Dai",
+      "Zi Wang",
+      "Jimmy Wu",
+      "Guanzhi Wang",
+      "S. Shankar Sastry",
+      "Ken Goldberg",
+      "Linxi \"Jim\" Fan",
+      "Yuke Zhu",
+      "Guanya Shi"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "embodied-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.19980",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19980",
+    "pdf_url": "https://arxiv.org/pdf/2606.19980",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.19245",
+    "title": "TxBench-PP: Analyzing AI Agent Performance on Small-Molecule Preclinical Pharmacology",
+    "url": "https://arxiv.org/abs/2606.19245",
+    "published": "2026-06-17",
+    "updated": "2026-06-18",
+    "authors": [
+      "Hannah Le",
+      "Ramesh Ramasamy",
+      "Alex Urrutia",
+      "Mahsa Yazdani",
+      "Tim Proctor",
+      "Kenny Workman"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.19245",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19245",
+    "pdf_url": "https://arxiv.org/pdf/2606.19245",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.18619",
+    "title": "Code-Augur: Agentic Vulnerability Detection via Specification Inference",
+    "url": "https://arxiv.org/abs/2606.18619",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Zhengxiong Luo",
+      "Mehtab Zafar",
+      "Dylan Wolff",
+      "Abhik Roychoudhury"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "rag",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.18619",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18619",
+    "pdf_url": "https://arxiv.org/pdf/2606.18619",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.19464",
+    "title": "Deontic Policies for Runtime Governance of Agentic AI Systems",
+    "url": "https://arxiv.org/abs/2606.19464",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Anupam Joshi",
+      "Tim Finin",
+      "Karuna Pande Joshi",
+      "Lalana Kagal"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.19464",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19464",
+    "pdf_url": "https://arxiv.org/pdf/2606.19464",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.18502",
+    "title": "Towards Scalable Customization and Deployment of Multi-Agent Systems for Enterprise Applications",
+    "url": "https://arxiv.org/abs/2606.18502",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Paresh Dashore",
+      "Shreyas Kulkarni",
+      "Uttam Gurram",
+      "Nadia Bathaee",
+      "Kartik Balasubramaniam",
+      "Genta Indra Winata",
+      "Sambit Sahu",
+      "Shi-Xiong Zhang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "coding-agent",
+      "multi-agent",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.18502",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18502",
+    "pdf_url": "https://arxiv.org/pdf/2606.18502",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.18037",
+    "title": "ProvenanceGuard: Source-Aware Factuality Verification for MCP-Based LLM Agents",
+    "url": "https://arxiv.org/abs/2606.18037",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Ander Alvarez",
+      "Santhiya Rajan",
+      "Samuel Mugel",
+      "Román Orús"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.18037",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18037",
+    "pdf_url": "https://arxiv.org/pdf/2606.18037",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.17573",
+    "title": "Cordon: Semantic Transactions for Tool-Using LLM Agents",
+    "url": "https://arxiv.org/abs/2606.17573",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Zheng Chen",
+      "Hanqing Liu",
+      "Duling Xu",
+      "Dong Dong",
+      "Jialin Li",
+      "Bangzheng Pu",
+      "Jidong Zhai"
+    ],
+    "categories": [
+      "cs.OS",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.17573",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17573",
+    "pdf_url": "https://arxiv.org/pdf/2606.17573",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.18051",
+    "title": "Compositional Skill Routing for LLM Agents: Decompose, Retrieve, and Compose",
+    "url": "https://arxiv.org/abs/2606.18051",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Xueping Gao"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.18051",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18051",
+    "pdf_url": "https://arxiv.org/pdf/2606.18051",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.16295",
+    "title": "VisualClaw: A Real-Time, Personalized Agent for the Physical World",
+    "url": "https://arxiv.org/abs/2606.16295",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Haoqin Tu",
+      "Jianwen Chen",
+      "Zijun Wang",
+      "Siwei Han",
+      "Juncheng Wu",
+      "Hardy Chen",
+      "Haonian Ji",
+      "Kaiwen Xiong",
+      "Jiaqi Liu",
+      "Peng Xia",
+      "Jieru Mei",
+      "Hongliang Fei",
+      "Jason Eshraghian",
+      "Zeyu Zheng",
+      "Yuyin Zhou",
+      "Huaxiu Yao",
+      "Cihang Xie"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "tool-use",
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.16295",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16295",
+    "pdf_url": "https://arxiv.org/pdf/2606.16295",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.16534",
+    "title": "Generated, Parallel, Scalable? A Study of Agentic AI-Generated Julia Code on Supercomputers",
+    "url": "https://arxiv.org/abs/2606.16534",
+    "published": "2026-06-15",
+    "updated": "2026-06-16",
+    "authors": [
+      "Linus Bantel",
+      "Anna-Lena Roth",
+      "Jonas Posner",
+      "Dirk Pflüger"
+    ],
+    "categories": [
+      "cs.DC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.16534",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16534",
+    "pdf_url": "https://arxiv.org/pdf/2606.16534",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.15376",
+    "title": "CoAgent: Concurrency Control for Multi-Agent Systems",
+    "url": "https://arxiv.org/abs/2606.15376",
+    "published": "2026-06-13",
+    "updated": "2026-06-13",
+    "authors": [
+      "Hongtao Lyu",
+      "Dingyan Zhang",
+      "Mingyu Wu",
+      "Xingda Wei",
+      "Haibo Chen"
+    ],
+    "categories": [
+      "cs.DC",
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "coding-agent",
+      "multi-agent",
+      "planning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.15376",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15376",
+    "pdf_url": "https://arxiv.org/pdf/2606.15376",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.14517",
+    "title": "From Shield to Target: Denial-of-Service Attacks on LLM-Based Agent Guardrails",
+    "url": "https://arxiv.org/abs/2606.14517",
+    "published": "2026-06-12",
+    "updated": "2026-06-16",
+    "authors": [
+      "Yuguang Zhou",
+      "Xunguang Wang",
+      "Pingchuan Ma",
+      "Zhantong Xue",
+      "Zhaoyu Wang",
+      "Shuai Wang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.14517",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.14517",
+    "pdf_url": "https://arxiv.org/pdf/2606.14517",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.14470",
+    "title": "GitOfThoughts: Version-Controlled Reasoning and Agent Memory You Can Replay, Diff, and Merge",
+    "url": "https://arxiv.org/abs/2606.14470",
+    "published": "2026-06-12",
+    "updated": "2026-06-22",
+    "authors": [
+      "Pavan C Shekar",
+      "Abhishek H S",
+      "Aswanth Krishnan"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "memory",
+      "rag",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.14470",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.14470",
+    "pdf_url": "https://arxiv.org/pdf/2606.14470",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.14106",
+    "title": "Naive Visual Memory is Not Enough: A Failure-Mode Study of GUI Agents",
+    "url": "https://arxiv.org/abs/2606.14106",
+    "published": "2026-06-12",
+    "updated": "2026-06-12",
+    "authors": [
+      "Seoyoung Choi",
+      "Minseok Ko",
+      "Hyunseok Lee",
+      "Kunwoong Kim",
+      "Woomin Song",
+      "Chanseok Jeon",
+      "Jinwoo Shin"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.CV"
+    ],
+    "topics": [
+      "computer-use",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.14106",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.14106",
+    "pdf_url": "https://arxiv.org/pdf/2606.14106",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.13608",
+    "title": "AgentBeats: Agentifying Agent Assessment for Openness, Standardization, and Reproducibility",
+    "url": "https://arxiv.org/abs/2606.13608",
+    "published": "2026-06-11",
+    "updated": "2026-06-14",
+    "authors": [
+      "Xiaoyuan Liu",
+      "Jianhong Tu",
+      "Yuqi Chen",
+      "Siyuan Xie",
+      "Sihan Ren",
+      "Tianneng Shi",
+      "Gal Gantar",
+      "Evan Sandoval",
+      "Donghyun Lee",
+      "Daniel Miao",
+      "Peter J. Gilbert",
+      "Nick Hynes",
+      "Mauro Staver",
+      "Warren He",
+      "David Marn",
+      "Andrew Low",
+      "Xi Zhang",
+      "Elron Bandel",
+      "Michal Shmueli-Scheuer",
+      "Siva Reddy",
+      "Alexandre Drouin",
+      "Alexandre Lacoste",
+      "Ramayya Krishnan",
+      "Elham Tabassi",
+      "Yu Su",
+      "Victor Barres",
+      "Chenguang Wang",
+      "Wenbo Guo",
+      "Dawn Song"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.13608",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13608",
+    "pdf_url": "https://arxiv.org/pdf/2606.13608",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.13177",
+    "title": "MemRefine: LLM-Guided Compression for Long-Term Agent Memory",
+    "url": "https://arxiv.org/abs/2606.13177",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Minjae Kim",
+      "Jinheon Baek",
+      "Soyeong Jeong",
+      "Sung Ju Hwang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.13177",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13177",
+    "pdf_url": "https://arxiv.org/pdf/2606.13177",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.13148",
+    "title": "TerraBench: Can Agents Reason Over Heterogeneous Earth-System Data?",
+    "url": "https://arxiv.org/abs/2606.13148",
+    "published": "2026-06-11",
+    "updated": "2026-07-01",
+    "authors": [
+      "Dat Tien Nguyen",
+      "Thao Nguyen",
+      "Fadillah Adamsyah Maani",
+      "Huy M. Le",
+      "Muhammad Umer Sheikh",
+      "Numan Saeed",
+      "Muhammad Haris Khan",
+      "Salman Khan"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.13148",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13148",
+    "pdf_url": "https://arxiv.org/pdf/2606.13148",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.28360",
+    "title": "Carolina Guide: A Multi-Agent RAG System with Institutional Guardrails for Academic Policy Assistance",
+    "url": "https://arxiv.org/abs/2606.28360",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Ben Torsion",
+      "Jun Zhou"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "rag"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.28360",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28360",
+    "pdf_url": "https://arxiv.org/pdf/2606.28360",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.12344",
+    "title": "Claw-SWE-Bench: A Benchmark for Evaluating OpenClaw-style Agent Harnesses on Coding Tasks",
+    "url": "https://arxiv.org/abs/2606.12344",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Mengyu Zheng",
+      "Kai Han",
+      "Boxun Li",
+      "Haiyang Xu",
+      "Yuchuan Tian",
+      "Wei He",
+      "Hang Zhou",
+      "Jianyuan Guo",
+      "Hailin Hu",
+      "Lin Ma",
+      "Chao Xu",
+      "Guohao Dai",
+      "Lixue Xia",
+      "Yunchao Wei",
+      "Yunhe Wang",
+      "Yu Wang"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.12344",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12344",
+    "pdf_url": "https://arxiv.org/pdf/2606.12344",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.11702",
+    "title": "MedCTA: A Benchmark for Clinical Tool Agents",
+    "url": "https://arxiv.org/abs/2606.11702",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Tajamul Ashraf",
+      "Hyewon Jeong",
+      "Fida Mohammad Thoker",
+      "Bernard Ghanem"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.11702",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.11702",
+    "pdf_url": "https://arxiv.org/pdf/2606.11702",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.11119",
+    "title": "TRACE: A Unified Rollout Budget Allocation Framework for Efficient Agentic Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2606.11119",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Heming Zou",
+      "Qi Wang",
+      "Yun Qu",
+      "Yuhang Jiang",
+      "Lizhou Cai",
+      "Yixiu Mao",
+      "Ru Peng",
+      "Xin Xu",
+      "Weijie Liu",
+      "Kai Yang",
+      "Saiyong Yang",
+      "Xiangyang Ji"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.11119",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.11119",
+    "pdf_url": "https://arxiv.org/pdf/2606.11119",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.10921",
+    "title": "Trace Only What You Need: Structure-Aware On-Demand Hypergraph Memory for Long-Document Question Answering",
+    "url": "https://arxiv.org/abs/2606.10921",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Xiangjun Zai",
+      "Xingyu Tan",
+      "Chen Chen",
+      "Xiaoyang Wang",
+      "Wenjie Zhang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.10921",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10921",
+    "pdf_url": "https://arxiv.org/pdf/2606.10921",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.10316",
+    "title": "TabClaw: An Interactive and Self-Evolving Agent for Spreadsheet Manipulation and Table Reasoning",
+    "url": "https://arxiv.org/abs/2606.10316",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Mingyue Cheng",
+      "Shuo Yu",
+      "Daoyu Wang",
+      "Qingchuan Li",
+      "Xiaoyu Tao",
+      "Qingyang Mao",
+      "Yitong Zhou",
+      "Qi Liu"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.10316",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10316",
+    "pdf_url": "https://arxiv.org/pdf/2606.10316",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.09037",
+    "title": "A Multi-Agent System for IPMSM Design Optimization via an FEA-AI Hybrid Approach",
+    "url": "https://arxiv.org/abs/2606.09037",
+    "published": "2026-06-08",
+    "updated": "2026-06-08",
+    "authors": [
+      "Jinseong Han",
+      "Sunwoong Yang",
+      "Namwoo Kang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.09037",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09037",
+    "pdf_url": "https://arxiv.org/pdf/2606.09037",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.09738",
+    "title": "HDSL: A Hierarchical Domain-Specific Language for Structured 3D Indoor Scene Generation and Localized Editing with LLM Agents",
+    "url": "https://arxiv.org/abs/2606.09738",
+    "published": "2026-06-08",
+    "updated": "2026-06-08",
+    "authors": [
+      "Letian Li",
+      "Chao Shen",
+      "Shuzhao Xie",
+      "Chenghao Gu",
+      "ZhengXiao He",
+      "Yu Meng",
+      "Xin Yang",
+      "Wenyuan Jiang",
+      "Zhi Wang"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "planning",
+      "rag"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.09738",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09738",
+    "pdf_url": "https://arxiv.org/pdf/2606.09738",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.10209",
+    "title": "Less Context, Better Agents: Efficient Context Engineering for Long-Horizon Tool-Using LLM Agents",
+    "url": "https://arxiv.org/abs/2606.10209",
+    "published": "2026-06-08",
+    "updated": "2026-06-08",
+    "authors": [
+      "Abhilasha Lodha",
+      "Mahsa Pahlavikhah Varnosfaderani",
+      "Abir Chakraborty",
+      "Abhinav Mithal"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.10209",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10209",
+    "pdf_url": "https://arxiv.org/pdf/2606.10209",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.05711",
+    "title": "Beyond tokens: a unified framework for latent communication in LLM-based multi-agent systems",
+    "url": "https://arxiv.org/abs/2606.05711",
+    "published": "2026-06-04",
+    "updated": "2026-06-05",
+    "authors": [
+      "Yingzhuo Liu"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.05711",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.05711",
+    "pdf_url": "https://arxiv.org/pdf/2606.05711",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.06090",
+    "title": "Beyond Semantic Organization: Memory as Execution State Management for Long-Horizon Agents",
+    "url": "https://arxiv.org/abs/2606.06090",
+    "published": "2026-06-04",
+    "updated": "2026-06-04",
+    "authors": [
+      "Yaoqi Chen",
+      "Haibin Lai",
+      "Yuru Feng",
+      "Chuyu Han",
+      "Qianxi Zhang",
+      "Baotong Lu",
+      "Menghao Li",
+      "Xinjiang Wang",
+      "Zhirui Wang",
+      "Shusen Xu",
+      "Zengzhong Li",
+      "Zewen Jin",
+      "Hao Wu",
+      "Cheng Li",
+      "Qi Chen"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "computer-use",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.06090",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.06090",
+    "pdf_url": "https://arxiv.org/pdf/2606.06090",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.06388",
+    "title": "Humans' ALMANAC: A Human Collaboration Dataset of Action-Level Mental Model Annotations for Agent Collaboration",
+    "url": "https://arxiv.org/abs/2606.06388",
+    "published": "2026-06-04",
+    "updated": "2026-06-06",
+    "authors": [
+      "Jiaju Chen",
+      "Yuxuan Lu",
+      "Jiayi Su",
+      "Chaoran Chen",
+      "Songlin Xiao",
+      "Zheng Zhang",
+      "Yun Wang",
+      "Yunyao Li",
+      "Jian Zhao",
+      "Tongshuang Wu",
+      "Toby Jia-Jun Li",
+      "Dakuo Wang",
+      "Bingsheng Yao"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.06388",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.06388",
+    "pdf_url": "https://arxiv.org/pdf/2606.06388",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.05241",
+    "title": "Search-Time Contamination in Deep Research Agents: Measuring Performance Inflation in Public Benchmark Evaluation",
+    "url": "https://arxiv.org/abs/2606.05241",
+    "published": "2026-06-03",
+    "updated": "2026-06-03",
+    "authors": [
+      "Yongjie Wang",
+      "Xinyue Zhang",
+      "Kunhong Yao",
+      "Zhiwei Zeng",
+      "Kaisong Song",
+      "Jun Lin",
+      "Zhiqi Shen"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.05241",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.05241",
+    "pdf_url": "https://arxiv.org/pdf/2606.05241",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.03197",
+    "title": "MemTrain: Self-Supervised Context Memory Training",
+    "url": "https://arxiv.org/abs/2606.03197",
+    "published": "2026-06-02",
+    "updated": "2026-06-02",
+    "authors": [
+      "Ziheng Li",
+      "Xingrun Xing",
+      "Haoqing Wang",
+      "Zhi-Hong Deng",
+      "Yehui Tang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.03197",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.03197",
+    "pdf_url": "https://arxiv.org/pdf/2606.03197",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.02497",
+    "title": "Bridging the Last Mile of Time Series Forecasting with LLM Agents",
+    "url": "https://arxiv.org/abs/2606.02497",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Yuhua Liao",
+      "Zetian Wang",
+      "Qiangqiang Nie",
+      "Zhenhua Zhang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.02497",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.02497",
+    "pdf_url": "https://arxiv.org/pdf/2606.02497",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.01222",
+    "title": "RAG-driven Multi-Agent LLM Framework with Task Decomposition for Beyond 5G Auto-Configuration",
+    "url": "https://arxiv.org/abs/2606.01222",
+    "published": "2026-05-31",
+    "updated": "2026-05-31",
+    "authors": [
+      "İrşat Emin Sarıdaş",
+      "Onur Salan",
+      "Ali Görçin",
+      "Ibrahim Hokelek",
+      "Hakan Ali Çırpan"
+    ],
+    "categories": [
+      "eess.SP"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.01222",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.01222",
+    "pdf_url": "https://arxiv.org/pdf/2606.01222",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.00619",
+    "title": "MemPro: Agentic Memory Systems as Evolvable Programs",
+    "url": "https://arxiv.org/abs/2606.00619",
+    "published": "2026-05-30",
+    "updated": "2026-05-30",
+    "authors": [
+      "Qingshan Liu",
+      "Guoqing Wang",
+      "Wen Wu",
+      "Jingqi Huang",
+      "Xinqi Tao",
+      "Dejia Song",
+      "Jie Zhou",
+      "Liang He"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "rag"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.00619",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.00619",
+    "pdf_url": "https://arxiv.org/pdf/2606.00619",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.00922",
+    "title": "A Machine-to-Machine Knowledge-Guided LLM Agent for Generalizable Radiotherapy Treatment Planning",
+    "url": "https://arxiv.org/abs/2606.00922",
+    "published": "2026-05-30",
+    "updated": "2026-05-30",
+    "authors": [
+      "Md Mainul Abrar",
+      "Xun Jia",
+      "Yujie Chi"
+    ],
+    "categories": [
+      "physics.med-ph",
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "planning",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.00922",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.00922",
+    "pdf_url": "https://arxiv.org/pdf/2606.00922",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.07595",
+    "title": "VisualLeakBench: Reproducible Action-Boundary Propagation Failures in Vision-Language Agents",
+    "url": "https://arxiv.org/abs/2606.07595",
+    "published": "2026-05-29",
+    "updated": "2026-05-29",
+    "authors": [
+      "Youting Wang",
+      "Yuan Tang",
+      "Yitian Qian",
+      "Chen Zhao"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.07595",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.07595",
+    "pdf_url": "https://arxiv.org/pdf/2606.07595",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.00341",
+    "title": "ROGUE: Misaligned Agent Behavior Arising from Ordinary Computer Use",
+    "url": "https://arxiv.org/abs/2606.00341",
+    "published": "2026-05-29",
+    "updated": "2026-05-29",
+    "authors": [
+      "Jeremy Tien",
+      "Abishek Anand",
+      "Yu-Rou Tuan",
+      "Yuchen Shen",
+      "J. Zico Kolter",
+      "Aran Nayebi"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2606.00341",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.00341",
+    "pdf_url": "https://arxiv.org/pdf/2606.00341",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.31377",
+    "title": "DynaTree: Dynamic Agentic Retrieval Tree for Time-Sensitive News Retrieval",
+    "url": "https://arxiv.org/abs/2605.31377",
+    "published": "2026-05-29",
+    "updated": "2026-05-29",
+    "authors": [
+      "Siyuan Qi",
+      "Xinyuan Wang",
+      "Yingxuan Yang",
+      "Haochuan Guo",
+      "Jianghao Lin",
+      "Weiwen Liu",
+      "Yong Yu",
+      "Weinan Zhang"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.31377",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.31377",
+    "pdf_url": "https://arxiv.org/pdf/2605.31377",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.30947",
+    "title": "Extending AI for Research to the Humanities: A Multi-Agent Framework for Evidence-Grounded Scholarship",
+    "url": "https://arxiv.org/abs/2605.30947",
+    "published": "2026-05-29",
+    "updated": "2026-06-03",
+    "authors": [
+      "Yating Pan",
+      "Jiajun Zhang",
+      "Jun Wang",
+      "Qi Su"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.30947",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.30947",
+    "pdf_url": "https://arxiv.org/pdf/2605.30947",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.29960",
+    "title": "Hijacking Agent Memory: Stealthy Trojan Attacks Through Conversational Interaction",
+    "url": "https://arxiv.org/abs/2605.29960",
+    "published": "2026-05-28",
+    "updated": "2026-05-28",
+    "authors": [
+      "Hongtao Wang",
+      "Se Yang",
+      "Yu Chen",
+      "Puzhuo Liu"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.29960",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.29960",
+    "pdf_url": "https://arxiv.org/pdf/2605.29960",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.25435",
+    "title": "Security of OpenClaw Agents: Fundamentals, Attacks, and Countermeasures",
+    "url": "https://arxiv.org/abs/2605.25435",
+    "published": "2026-05-25",
+    "updated": "2026-05-25",
+    "authors": [
+      "Yuntao Wang",
+      "Jianle Ba",
+      "Han Liu",
+      "Yanghe Pan",
+      "Jintao Wei",
+      "Zhou Su",
+      "Tom H. Luan",
+      "Linkang Du"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "memory",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.25435",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.25435",
+    "pdf_url": "https://arxiv.org/pdf/2605.25435",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.24812",
+    "title": "CoRe-Code: Collaborative Reinforcement Learning for Code Generation",
+    "url": "https://arxiv.org/abs/2605.24812",
+    "published": "2026-05-24",
+    "updated": "2026-05-24",
+    "authors": [
+      "Zhihao Dou",
+      "Qinjian Zhao",
+      "Zhongwei Wan",
+      "Xiaoyu Xia",
+      "Sumon Biswas"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.24812",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.24812",
+    "pdf_url": "https://arxiv.org/pdf/2605.24812",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.23067",
+    "title": "What Training Data Teaches RL Memory Agents: An Empirical Study of Curriculum Effects in Memory-Augmented QA",
+    "url": "https://arxiv.org/abs/2605.23067",
+    "published": "2026-05-21",
+    "updated": "2026-05-21",
+    "authors": [
+      "Xinjie He",
+      "Zhiyuan Lin",
+      "Su Liu",
+      "Jialun Wu",
+      "Qiyang Xie",
+      "Weikai Zhou",
+      "Shuai Xiao"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.23067",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.23067",
+    "pdf_url": "https://arxiv.org/pdf/2605.23067",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.20616",
+    "title": "Auto-Dreamer: Learning Offline Memory Consolidation for Language Agents",
+    "url": "https://arxiv.org/abs/2605.20616",
+    "published": "2026-05-20",
+    "updated": "2026-05-20",
+    "authors": [
+      "Chongrui Ye",
+      "Yuxiang Liu",
+      "Yu Wang",
+      "Haofei Yu",
+      "Yining Zhao",
+      "Ge Liu",
+      "Julian McAuley",
+      "Jiaxuan You"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "language-agent"
+    ],
+    "arxiv_id": "2605.20616",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.20616",
+    "pdf_url": "https://arxiv.org/pdf/2605.20616",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.16233",
+    "title": "FORGE: Self-Evolving Agent Memory With No Weight Updates via Population Broadcast",
+    "url": "https://arxiv.org/abs/2605.16233",
+    "published": "2026-05-15",
+    "updated": "2026-05-15",
+    "authors": [
+      "Igor Bogdanov",
+      "Chung-Horng Lung",
+      "Thomas Kunz",
+      "Jie Gao",
+      "Adrian Taylor",
+      "Marzia Zaman"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.LG",
+      "cs.MA",
+      "eess.SY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "reasoning"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.16233",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.16233",
+    "pdf_url": "https://arxiv.org/pdf/2605.16233",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.15759",
+    "title": "DimMem: Dimensional Structuring for Efficient Long-Term Agent Memory",
+    "url": "https://arxiv.org/abs/2605.15759",
+    "published": "2026-05-15",
+    "updated": "2026-05-24",
+    "authors": [
+      "Wentao Qiu",
+      "Haotian Hu",
+      "Fanyi Wang",
+      "Jinwei Kong",
+      "Yu Zhang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.15759",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.15759",
+    "pdf_url": "https://arxiv.org/pdf/2605.15759",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.15710",
+    "title": "SMMBench: A Benchmark for Source-Distributed Multimodal Agent Memory",
+    "url": "https://arxiv.org/abs/2605.15710",
+    "published": "2026-05-15",
+    "updated": "2026-05-15",
+    "authors": [
+      "Huacan Chai",
+      "Yukai Wang",
+      "Yingxuan Yang",
+      "Dan Peng",
+      "Yuanyi Song",
+      "Zhihui Fu",
+      "Weiwen Liu",
+      "Jianghao Lin",
+      "Jun Wang",
+      "Weinan Zhang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.15710",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.15710",
+    "pdf_url": "https://arxiv.org/pdf/2605.15710",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.15128",
+    "title": "MemEye: A Visual-Centric Evaluation Framework for Multimodal Agent Memory",
+    "url": "https://arxiv.org/abs/2605.15128",
+    "published": "2026-05-14",
+    "updated": "2026-05-14",
+    "authors": [
+      "Minghao Guo",
+      "Qingyue Jiao",
+      "Zeru Shi",
+      "Yihao Quan",
+      "Boxuan Zhang",
+      "Danrui Li",
+      "Liwei Che",
+      "Wujiang Xu",
+      "Shilong Liu",
+      "Zirui Liu",
+      "Mubbasir Kapadia",
+      "Vladimir Pavlovic",
+      "Jiang Liu",
+      "Mengdi Wang",
+      "Yiyu Shi",
+      "Dimitris N. Metaxas",
+      "Ruixiang Tang"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.CL",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.15128",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.15128",
+    "pdf_url": "https://arxiv.org/pdf/2605.15128",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.14906",
+    "title": "MemLens: Benchmarking Multimodal Long-Term Memory in Large Vision-Language Models",
+    "url": "https://arxiv.org/abs/2605.14906",
+    "published": "2026-05-14",
+    "updated": "2026-05-14",
+    "authors": [
+      "Xiyu Ren",
+      "Zhaowei Wang",
+      "Yiming Du",
+      "Zhongwei Xie",
+      "Chi Liu",
+      "Xinlin Yang",
+      "Haoyue Feng",
+      "Wenjun Pan",
+      "Tianshi Zheng",
+      "Baixuan Xu",
+      "Zhengnan Li",
+      "Yangqiu Song",
+      "Ginny Wong",
+      "Simon See"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.14906",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.14906",
+    "pdf_url": "https://arxiv.org/pdf/2605.14906",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.14932",
+    "title": "Toward Securing AI Agents Like Operating Systems",
+    "url": "https://arxiv.org/abs/2605.14932",
+    "published": "2026-05-14",
+    "updated": "2026-05-14",
+    "authors": [
+      "Lukas Pirch",
+      "Micha Horlboge",
+      "Patrick Großmann",
+      "Syeda Mahnur Asif",
+      "Klim Kireev",
+      "Thorsten Holz",
+      "Konrad Rieck"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.14932",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.14932",
+    "pdf_url": "https://arxiv.org/pdf/2605.14932",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.11946",
+    "title": "Counterfactual Trace Auditing of LLM Agent Skills",
+    "url": "https://arxiv.org/abs/2605.11946",
+    "published": "2026-05-12",
+    "updated": "2026-05-28",
+    "authors": [
+      "Xiaolin Zhou",
+      "Jinbo Liu",
+      "Li Li",
+      "Ryan A. Rossi",
+      "Xiyang Hu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "rag"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.11946",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.11946",
+    "pdf_url": "https://arxiv.org/pdf/2605.11946",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.08964",
+    "title": "Trustworthy AI: Ensuring Reliability and Accountability from Models to Agents",
+    "url": "https://arxiv.org/abs/2605.08964",
+    "published": "2026-05-09",
+    "updated": "2026-05-09",
+    "authors": [
+      "Carol Xuan Long"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.08964",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.08964",
+    "pdf_url": "https://arxiv.org/pdf/2605.08964",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.07830",
+    "title": "CyBiasBench: Benchmarking Bias in LLM Agents for Cyber-Attack Scenarios",
+    "url": "https://arxiv.org/abs/2605.07830",
+    "published": "2026-05-08",
+    "updated": "2026-05-08",
+    "authors": [
+      "Taein Lim",
+      "Seongyong Ju",
+      "Munhyeok Kim",
+      "Hyunjun Kim",
+      "Hoki Kim"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.07830",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.07830",
+    "pdf_url": "https://arxiv.org/pdf/2605.07830",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.04808",
+    "title": "DecodingTrust-Agent Platform (DTap): A Controllable and Interactive Red-Teaming Platform for AI Agents",
+    "url": "https://arxiv.org/abs/2605.04808",
+    "published": "2026-05-06",
+    "updated": "2026-05-06",
+    "authors": [
+      "Zhaorun Chen",
+      "Xun Liu",
+      "Haibo Tong",
+      "Chengquan Guo",
+      "Yuzhou Nie",
+      "Jiawei Zhang",
+      "Mintong Kang",
+      "Chejian Xu",
+      "Qichang Liu",
+      "Xiaogeng Liu",
+      "Tianneng Shi",
+      "Chaowei Xiao",
+      "Sanmi Koyejo",
+      "Percy Liang",
+      "Wenbo Guo",
+      "Dawn Song",
+      "Bo Li"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "planning",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.04808",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.04808",
+    "pdf_url": "https://arxiv.org/pdf/2605.04808",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.01644",
+    "title": "Toward a Principled Framework for Agent Safety Measurement",
+    "url": "https://arxiv.org/abs/2605.01644",
+    "published": "2026-05-02",
+    "updated": "2026-05-02",
+    "authors": [
+      "Shuyi Lin",
+      "Anshuman Suri",
+      "Alina Oprea",
+      "Cheng Tan"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.01644",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.01644",
+    "pdf_url": "https://arxiv.org/pdf/2605.01644",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.00741",
+    "title": "Self-Adaptive Multi-Agent LLM-Based Security Pattern Selection for IoT Systems",
+    "url": "https://arxiv.org/abs/2605.00741",
+    "published": "2026-05-01",
+    "updated": "2026-05-01",
+    "authors": [
+      "Saeid Jamshidi",
+      "Foutse Khomh",
+      "Carol Fung",
+      "Kawser Wazed Nafi"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.00741",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.00741",
+    "pdf_url": "https://arxiv.org/pdf/2605.00741",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.24212",
+    "title": "Empowering Autonomous Debugging Agents with Efficient Dynamic Analysis",
+    "url": "https://arxiv.org/abs/2604.24212",
+    "published": "2026-04-27",
+    "updated": "2026-04-27",
+    "authors": [
+      "Jiahong Xiang",
+      "Xiaoyang Xu",
+      "Xiaopan Chu",
+      "Hongliang Tian",
+      "Yuqun Zhang"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "embodied-agent",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2604.24212",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.24212",
+    "pdf_url": "https://arxiv.org/pdf/2604.24212",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2604.23210",
+    "title": "Discovering Agentic Safety Specifications from 1-Bit Danger Signals",
+    "url": "https://arxiv.org/abs/2604.23210",
+    "published": "2026-04-25",
+    "updated": "2026-04-25",
+    "authors": [
+      "Víctor Gallego"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.23210",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.23210",
+    "pdf_url": "https://arxiv.org/pdf/2604.23210",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.13954",
+    "title": "HINTBench: Horizon-agent Intrinsic Non-attack Trajectory Benchmark",
+    "url": "https://arxiv.org/abs/2604.13954",
+    "published": "2026-04-15",
+    "updated": "2026-04-15",
+    "authors": [
+      "Jiacheng Wang",
+      "Jinchang Hou",
+      "Fabian Wang",
+      "Ping Jian",
+      "Chenfu Bao",
+      "Zhonghou Lv"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.13954",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.13954",
+    "pdf_url": "https://arxiv.org/pdf/2604.13954",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.13536",
+    "title": "Don't Let AI Agents YOLO Your Files: Shifting Information and Control to Filesystems for Agent Safety and Autonomy",
+    "url": "https://arxiv.org/abs/2604.13536",
+    "published": "2026-04-15",
+    "updated": "2026-04-16",
+    "authors": [
+      "Shawn Wanxiang Zhong",
+      "Junxuan Liao",
+      "Jing Liu",
+      "Mai Zheng",
+      "Andrea C. Arpaci-Dusseau",
+      "Remzi H. Arpaci-Dusseau"
+    ],
+    "categories": [
+      "cs.OS"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.13536",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.13536",
+    "pdf_url": "https://arxiv.org/pdf/2604.13536",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.13298",
+    "title": "Can Agents Secure Hardware? Evaluating Agentic LLM-Driven Obfuscation for IP Protection",
+    "url": "https://arxiv.org/abs/2604.13298",
+    "published": "2026-04-14",
+    "updated": "2026-04-14",
+    "authors": [
+      "Sujan Ghimire",
+      "Parsa Mirfasihi",
+      "Muhtasim Alam Chowdhury",
+      "Veeramani Pugazhenthi",
+      "Harish Kumar Dharavath",
+      "Farshad Firouzi",
+      "Rozhin Yasaei",
+      "Pratik Satam",
+      "Soheil Salehi"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.13298",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.13298",
+    "pdf_url": "https://arxiv.org/pdf/2604.13298",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.28835",
+    "title": "GenesisFunc: Multi-Agent Data Generation for Accurate and Generalizable Function-Calling",
+    "url": "https://arxiv.org/abs/2605.28835",
+    "published": "2026-04-10",
+    "updated": "2026-04-10",
+    "authors": [
+      "Hao-Xiang Xu",
+      "Chong Deng",
+      "Jiaqing Liu",
+      "Wen Wang",
+      "Qian Chen",
+      "Lujia Bao",
+      "Xiangang Li",
+      "Zhen-Hua Ling"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2605.28835",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.28835",
+    "pdf_url": "https://arxiv.org/pdf/2605.28835",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2605.00845",
+    "title": "Graph Query Generation with Constraint-guided Large Language Agents",
+    "url": "https://arxiv.org/abs/2605.00845",
+    "published": "2026-04-09",
+    "updated": "2026-04-09",
+    "authors": [
+      "Mengying Wang",
+      "Nicolaas Jedema",
+      "Rahul Pandey",
+      "RaviKiran Krishnan",
+      "Jens Lehmann",
+      "Yinghui Wu"
+    ],
+    "categories": [
+      "cs.DB",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.00845",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.00845",
+    "pdf_url": "https://arxiv.org/pdf/2605.00845",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2604.26959",
+    "title": "CareGuardAI: Context-Aware Multi-Agent Guardrails for Clinical Safety & Hallucination Mitigation in Patient-Facing LLMs",
+    "url": "https://arxiv.org/abs/2604.26959",
+    "published": "2026-04-07",
+    "updated": "2026-04-07",
+    "authors": [
+      "Elham Nasarian",
+      "Abhilash Neog",
+      "Kwok-Leung Tsui",
+      "Niyousha HosseiniChimeh"
+    ],
+    "categories": [
+      "cs.CY",
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.26959",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.26959",
+    "pdf_url": "https://arxiv.org/pdf/2604.26959",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.04131",
+    "title": "Profile-Then-Reason: Bounded Semantic Complexity for Tool-Augmented Language Agents",
+    "url": "https://arxiv.org/abs/2604.04131",
+    "published": "2026-04-05",
+    "updated": "2026-04-05",
+    "authors": [
+      "Paulo Akira F. Enabe"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2604.04131",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.04131",
+    "pdf_url": "https://arxiv.org/pdf/2604.04131",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.28428",
+    "title": "Synergy: A Next-Generation General-Purpose Agent for Open Agentic Web",
+    "url": "https://arxiv.org/abs/2603.28428",
+    "published": "2026-03-30",
+    "updated": "2026-03-30",
+    "authors": [
+      "Xiaohang Nie",
+      "Zihan Guo",
+      "Kezhuo Yang",
+      "Zhichong Zheng",
+      "Bochen Ge",
+      "Shuai Pan",
+      "Zeyi Chen",
+      "Youling Xiang",
+      "Yu Zhang",
+      "Weiwen Liu",
+      "Yuanjian Zhou",
+      "Weinan Zhang"
+    ],
+    "categories": [
+      "cs.CY",
+      "cs.MA"
+    ],
+    "topics": [
+      "coding-agent",
+      "embodied-agent",
+      "memory",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2603.28428",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.28428",
+    "pdf_url": "https://arxiv.org/pdf/2603.28428",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2603.28166",
+    "title": "Evaluating Privilege Usage of Agents with Real-World Tools",
+    "url": "https://arxiv.org/abs/2603.28166",
+    "published": "2026-03-30",
+    "updated": "2026-04-20",
+    "authors": [
+      "Quan Zhang",
+      "Lianhang Fu",
+      "Lvsi Lian",
+      "Gwihwan Go",
+      "Yujue Wang",
+      "Chijin Zhou",
+      "Yu Jiang",
+      "Geguang Pu"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.28166",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.28166",
+    "pdf_url": "https://arxiv.org/pdf/2603.28166",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.21564",
+    "title": "Toward a Theory of Hierarchical Memory for Language Agents",
+    "url": "https://arxiv.org/abs/2603.21564",
+    "published": "2026-03-23",
+    "updated": "2026-03-23",
+    "authors": [
+      "Yashar Talebirad",
+      "Ali Parsaee",
+      "Csongor Y. Szepesvari",
+      "Amirhossein Nadiri",
+      "Osmar Zaiane"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI",
+      "cs.IT",
+      "cs.SI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.21564",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.21564",
+    "pdf_url": "https://arxiv.org/pdf/2603.21564",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.21357",
+    "title": "AgentHER: Hindsight Experience Replay for LLM Agent Trajectory Relabeling",
+    "url": "https://arxiv.org/abs/2603.21357",
+    "published": "2026-03-22",
+    "updated": "2026-05-10",
+    "authors": [
+      "Liang Ding"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "computer-use",
+      "memory",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.21357",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.21357",
+    "pdf_url": "https://arxiv.org/pdf/2603.21357",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.05553",
+    "title": "EigenData: A Self-Evolving Multi-Agent Platform for Function-Calling Data Synthesis, Auditing, and Repair",
+    "url": "https://arxiv.org/abs/2603.05553",
+    "published": "2026-03-05",
+    "updated": "2026-03-05",
+    "authors": [
+      "Jiaao Chen",
+      "Jingyuan Qi",
+      "Mingye Gao",
+      "Wei-Chen Wang",
+      "Hanrui Wang",
+      "Di Jin"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2603.05553",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.05553",
+    "pdf_url": "https://arxiv.org/pdf/2603.05553",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2602.07652",
+    "title": "Agent-Fence: Mapping Security Vulnerabilities Across Deep Research Agents",
+    "url": "https://arxiv.org/abs/2602.07652",
+    "published": "2026-02-07",
+    "updated": "2026-02-07",
+    "authors": [
+      "Sai Puppala",
+      "Ismail Hossain",
+      "Md Jahangir Alam",
+      "Yoonpyo Lee",
+      "Jay Yoo",
+      "Tanzim Ahad",
+      "Syed Bahauddin Alam",
+      "Sajedul Talukder"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.07652",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.07652",
+    "pdf_url": "https://arxiv.org/pdf/2602.07652",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2602.05115",
+    "title": "SocialVeil: Probing Social Intelligence of Language Agents under Communication Barriers",
+    "url": "https://arxiv.org/abs/2602.05115",
+    "published": "2026-02-04",
+    "updated": "2026-02-04",
+    "authors": [
+      "Keyang Xuan",
+      "Pengda Wang",
+      "Chongrui Ye",
+      "Haofei Yu",
+      "Tal August",
+      "Jiaxuan You"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2602.05115",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.05115",
+    "pdf_url": "https://arxiv.org/pdf/2602.05115",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2602.03786",
+    "title": "AOrchestra: Automating Sub-Agent Creation for Agentic Orchestration",
+    "url": "https://arxiv.org/abs/2602.03786",
+    "published": "2026-02-03",
+    "updated": "2026-02-07",
+    "authors": [
+      "Jianhao Ruan",
+      "Zhihao Xu",
+      "Yiran Peng",
+      "Fashen Ren",
+      "Zhaoyang Yu",
+      "Xinbing Liang",
+      "Jinyu Xiang",
+      "Yongru Chen",
+      "Bang Liu",
+      "Chenglin Wu",
+      "Yuyu Luo",
+      "Jiayi Zhang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2602.03786",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.03786",
+    "pdf_url": "https://arxiv.org/pdf/2602.03786",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2601.00268",
+    "title": "Beyond Perfect APIs: A Comprehensive Evaluation of LLM Agents Under Real-World API Complexity",
+    "url": "https://arxiv.org/abs/2601.00268",
+    "published": "2026-01-01",
+    "updated": "2026-01-01",
+    "authors": [
+      "Doyoung Kim",
+      "Zhiwei Ren",
+      "Jie Hao",
+      "Zhongkai Sun",
+      "Lichao Wang",
+      "Xiyao Ma",
+      "Zack Ye",
+      "Xu Han",
+      "Jun Yin",
+      "Heng Ji",
+      "Wei Shen",
+      "Xing Fan",
+      "Benjamin Yao",
+      "Chenlei Guo"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2601.00268",
+    "source": "arxiv",
+    "source_id": "arxiv:2601.00268",
+    "pdf_url": "https://arxiv.org/pdf/2601.00268",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2511.22138",
+    "title": "TinyLLM: Evaluation and Optimization of Small Language Models for Agentic Tasks on Edge Devices",
+    "url": "https://arxiv.org/abs/2511.22138",
+    "published": "2025-11-27",
+    "updated": "2025-11-27",
+    "authors": [
+      "Mohd Ariful Haque",
+      "Fahad Rahman",
+      "Kishor Datta Gupta",
+      "Khalil Shujaee",
+      "Roy George"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2511.22138",
+    "source": "arxiv",
+    "source_id": "arxiv:2511.22138",
+    "pdf_url": "https://arxiv.org/pdf/2511.22138",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2511.11169",
+    "title": "Refine and Align: Confidence Calibration through Multi-Agent Interaction in VQA",
+    "url": "https://arxiv.org/abs/2511.11169",
+    "published": "2025-11-14",
+    "updated": "2025-11-14",
+    "authors": [
+      "Ayush Pandey",
+      "Jai Bardhan",
+      "Ishita Jain",
+      "Ramya S Hebbalaguppe",
+      "Rohan Raju Dhanakshirur",
+      "Lovekesh Vig"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2511.11169",
+    "source": "arxiv",
+    "source_id": "arxiv:2511.11169",
+    "pdf_url": "https://arxiv.org/pdf/2511.11169",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2510.21524",
+    "title": "EU-Agent-Bench: Measuring Illegal Behavior of LLM Agents Under EU Law",
+    "url": "https://arxiv.org/abs/2510.21524",
+    "published": "2025-10-24",
+    "updated": "2025-10-24",
+    "authors": [
+      "Ilija Lichkovski",
+      "Alexander Müller",
+      "Mariam Ibrahim",
+      "Tiwai Mhundwa"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2510.21524",
+    "source": "arxiv",
+    "source_id": "arxiv:2510.21524",
+    "pdf_url": "https://arxiv.org/pdf/2510.21524",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2509.08863",
+    "title": "GeoJSON Agents:A Multi-Agent LLM Architecture for Geospatial Analysis-Function Calling vs Code Generation",
+    "url": "https://arxiv.org/abs/2509.08863",
+    "published": "2025-09-10",
+    "updated": "2025-12-03",
+    "authors": [
+      "Qianqian Luo",
+      "Qingming Lin",
+      "Liuchang Xu",
+      "Sensen Wu",
+      "Ruichen Mao",
+      "Chao Wang",
+      "Hailin Feng",
+      "Bo Huang",
+      "Zhenhong Du"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2509.08863",
+    "source": "arxiv",
+    "source_id": "arxiv:2509.08863",
+    "pdf_url": "https://arxiv.org/pdf/2509.08863",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2508.11027",
+    "title": "Hell or High Water: Evaluating Agentic Recovery from External Failures",
+    "url": "https://arxiv.org/abs/2508.11027",
+    "published": "2025-08-14",
+    "updated": "2025-08-14",
+    "authors": [
+      "Andrew Wang",
+      "Sophia Hager",
+      "Adi Asija",
+      "Daniel Khashabi",
+      "Nicholas Andrews"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 16,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2508.11027",
+    "source": "arxiv",
+    "source_id": "arxiv:2508.11027",
+    "pdf_url": "https://arxiv.org/pdf/2508.11027",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2607.06223",
+    "title": "Information Gain-based Rollout Policy Optimization: An Adaptive Tree-Structured Rollout Approach for Multi-Turn LLM Agents",
+    "url": "https://arxiv.org/abs/2607.06223",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Yijun Zhang",
+      "Fan Xu",
+      "Jiaxin Ding",
+      "Yule Xie",
+      "Shiqing Gao",
+      "Xin Ding",
+      "Haoxiang Zhang",
+      "Luoyi Fu",
+      "Xinbing Wang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "rag"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.06223",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06223",
+    "pdf_url": "https://arxiv.org/pdf/2607.06223",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.05915",
+    "title": "PCBWorld: A Benchmark Environment for Engine-Grounded PCB Design Automation",
+    "url": "https://arxiv.org/abs/2607.05915",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Hyungseok Song",
+      "Junseok Park",
+      "Won-Seok Choi",
+      "Seohui Bae",
+      "Han-Seul Jeong",
+      "Youngjoon Park",
+      "Soonyoung Lee"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "llm-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2607.05915",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05915",
+    "pdf_url": "https://arxiv.org/pdf/2607.05915",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.05805",
+    "title": "Onnes: A Physics-Grounded Multi-Agent LLM Simulator for Cryogenic Fault Diagnosis in Quantum Computing Infrastructure",
+    "url": "https://arxiv.org/abs/2607.05805",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Praneeth Narisetty",
+      "Uday Kumar Reddy Kattamanchi",
+      "Shiva Nagendra Babu Kore"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG",
+      "quant-ph"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.05805",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05805",
+    "pdf_url": "https://arxiv.org/pdf/2607.05805",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.05743",
+    "title": "The Balkanization of Execution-Security Research for AI Coding Agents: Isolation, Access Control, and Time-of-Check-to-Time-of-Use Vulnerabilities",
+    "url": "https://arxiv.org/abs/2607.05743",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Mohammadreza Rashidi"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.05743",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05743",
+    "pdf_url": "https://arxiv.org/pdf/2607.05743",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.05690",
+    "title": "Memory in the Loop: In-Process Retrieval as ExtendedWorking Memory for Language Agents",
+    "url": "https://arxiv.org/abs/2607.05690",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Yusuf Khan",
+      "Carlo Lipizzi"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2607.05690",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05690",
+    "pdf_url": "https://arxiv.org/pdf/2607.05690",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2607.05055",
+    "title": "Toward Trustworthy Large Language Model Agents in Healthcare",
+    "url": "https://arxiv.org/abs/2607.05055",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Hadi Hasan",
+      "Safaa Salman",
+      "Adam Tai Abou Dargham",
+      "Ammar Mohanna",
+      "Ali Chehab"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling",
+      "rag-agent"
+    ],
+    "arxiv_id": "2607.05055",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05055",
+    "pdf_url": "https://arxiv.org/pdf/2607.05055",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2607.04089",
+    "title": "PLACEMEM: Toward a Compute-Aware Memory Plane for Lifelong Agents",
+    "url": "https://arxiv.org/abs/2607.04089",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Sukanta Ganguly"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2607.04089",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04089",
+    "pdf_url": "https://arxiv.org/pdf/2607.04089",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2607.03702",
+    "title": "Agent Reinforcement Learning via Pivotal-Aware Self-Feedback Retry",
+    "url": "https://arxiv.org/abs/2607.03702",
+    "published": "2026-07-04",
+    "updated": "2026-07-04",
+    "authors": [
+      "Weiyang Guo",
+      "Zesheng Shi",
+      "Longhui Zhang",
+      "Zeen Zhu",
+      "Min Zhang",
+      "Jing Li"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.03702",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03702",
+    "pdf_url": "https://arxiv.org/pdf/2607.03702",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.03695",
+    "title": "Social Networks of LLM Agents",
+    "url": "https://arxiv.org/abs/2607.03695",
+    "published": "2026-07-04",
+    "updated": "2026-07-04",
+    "authors": [
+      "Kaixuan Liu",
+      "Guojun Xiong",
+      "Weinan Zhang",
+      "Shengpu Tang"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.03695",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03695",
+    "pdf_url": "https://arxiv.org/pdf/2607.03695",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.03821",
+    "title": "DualView: Preventing Indirect Prompt Injection in Personal AI Agents",
+    "url": "https://arxiv.org/abs/2607.03821",
+    "published": "2026-07-04",
+    "updated": "2026-07-04",
+    "authors": [
+      "Juhee Kim",
+      "Woohyuk Choi",
+      "Taehyun Kang",
+      "Youngmin Kim",
+      "Byoungyoung Lee"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.03821",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03821",
+    "pdf_url": "https://arxiv.org/pdf/2607.03821",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.04009",
+    "title": "PhysMiner: An Agentic AI Framework for Discovering Turbulence Physics",
+    "url": "https://arxiv.org/abs/2607.04009",
+    "published": "2026-07-04",
+    "updated": "2026-07-04",
+    "authors": [
+      "Jiawei Chen",
+      "Han Gao",
+      "Ping He"
+    ],
+    "categories": [
+      "physics.flu-dyn"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.04009",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04009",
+    "pdf_url": "https://arxiv.org/pdf/2607.04009",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.03968",
+    "title": "Refused in Chat, Written in Code: Workflow-Level Jailbreak Construction in IDE Coding Agents",
+    "url": "https://arxiv.org/abs/2607.03968",
+    "published": "2026-07-04",
+    "updated": "2026-07-04",
+    "authors": [
+      "Abhishek Kumar",
+      "Carsten Maple"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.03968",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03968",
+    "pdf_url": "https://arxiv.org/pdf/2607.03968",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.02846",
+    "title": "Object-Centric Environment Modeling for Agentic Tasks",
+    "url": "https://arxiv.org/abs/2607.02846",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Yiyang Li",
+      "Tianyi Ma",
+      "Zehong Wang",
+      "Yijun Ma",
+      "Yanfang Ye"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.02846",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02846",
+    "pdf_url": "https://arxiv.org/pdf/2607.02846",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.03269",
+    "title": "Agentic-SecPBFT: Agentic AI-Driven Proactive Security Framework for Wireless PBFT Consensus in Mobile Ad-Hoc Networks",
+    "url": "https://arxiv.org/abs/2607.03269",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Haoxiang Luo",
+      "Yinqiu Liu",
+      "Ruichen Zhang",
+      "Guangyuan Liu",
+      "Gang Sun",
+      "Hongfang Yu",
+      "Zhu Han",
+      "Dong In Kim"
+    ],
+    "categories": [
+      "cs.NI"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "rag",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.03269",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03269",
+    "pdf_url": "https://arxiv.org/pdf/2607.03269",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.02911",
+    "title": "CoACT: Action-Preserving Observation Compression for Coding Agents",
+    "url": "https://arxiv.org/abs/2607.02911",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Haorui Chen",
+      "Yuancheng Zhu",
+      "Yitong Zhang",
+      "Jia Li"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "coding-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.02911",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02911",
+    "pdf_url": "https://arxiv.org/pdf/2607.02911",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.01767",
+    "title": "Repair the Amplifier, Not the Symptom: Stable World-Model Correction for Agent Rollouts",
+    "url": "https://arxiv.org/abs/2607.01767",
+    "published": "2026-07-02",
+    "updated": "2026-07-05",
+    "authors": [
+      "Xinyuan Song",
+      "Zekun Cai"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2607.01767",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01767",
+    "pdf_url": "https://arxiv.org/pdf/2607.01767",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2607.01709",
+    "title": "COMFYCLAW: Self-Evolving Skill Harnesses for Image Generation Workflows",
+    "url": "https://arxiv.org/abs/2607.01709",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Zongxia Li",
+      "Dawei Liu",
+      "Fuxiao Liu",
+      "Yuhang Zhou",
+      "Xiyang Wu",
+      "Jingxi Chen",
+      "Jing Xie",
+      "Xiaomin Wu",
+      "Lichao Sun"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2607.01709",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01709",
+    "pdf_url": "https://arxiv.org/pdf/2607.01709",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2607.02807",
+    "title": "SwarmResearch: Orchestrating Coding Agents for Open-Ended Discovery",
+    "url": "https://arxiv.org/abs/2607.02807",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Yuvraj Virk",
+      "Zack Edds",
+      "Chunqiu Steven Xia",
+      "Lingming Zhang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "coding-agent",
+      "computer-use",
+      "multi-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.02807",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02807",
+    "pdf_url": "https://arxiv.org/pdf/2607.02807",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.02448",
+    "title": "AgentsCAD: Automated Design for Manufacturing of FDM Parts via Multi-Agent LLM Reasoning and Geometric Feature Recognition",
+    "url": "https://arxiv.org/abs/2607.02448",
+    "published": "2026-07-02",
+    "updated": "2026-07-07",
+    "authors": [
+      "Emmanuel George",
+      "Christopher Keefe",
+      "Peter Pak",
+      "Amir Barati Farimani"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.02448",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02448",
+    "pdf_url": "https://arxiv.org/pdf/2607.02448",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.01213",
+    "title": "RepoRescue: An Empirical Study of LLM Agents on Whole-Repository Compatibility Rescue",
+    "url": "https://arxiv.org/abs/2607.01213",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Zhihao Lin",
+      "Mingyi Zhou",
+      "Zhensu Sun",
+      "Yizhuo Yang",
+      "Renyu Yang",
+      "David Lo",
+      "Li Li"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.01213",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01213",
+    "pdf_url": "https://arxiv.org/pdf/2607.01213",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.01120",
+    "title": "Next-Generation Agentic Reinforcement Learning Systems Enable Self-Evolving Agents",
+    "url": "https://arxiv.org/abs/2607.01120",
+    "published": "2026-07-01",
+    "updated": "2026-07-02",
+    "authors": [
+      "Ran Yan",
+      "Wei Fu",
+      "Jiale Li",
+      "Shusheng Xu",
+      "Zhiyu Mei",
+      "Jiaxuan Gao",
+      "Jiarui Zhang",
+      "Wentai Zhang",
+      "Hao Dai",
+      "Xujie Shen",
+      "Chuyi He",
+      "Zhen Pu",
+      "Jun Mei",
+      "Zhiyao Lin",
+      "Haitao Wang",
+      "Zhiqiang Ding",
+      "Jiawei Zhang",
+      "Huaijie Wang",
+      "Ruida Xu",
+      "Honghua Dong",
+      "Youhe Jiang",
+      "Yi Wu",
+      "Tongkai Yang",
+      "Binhang Yuan"
+    ],
+    "categories": [
+      "cs.DC"
+    ],
+    "topics": [
+      "coding-agent",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.01120",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01120",
+    "pdf_url": "https://arxiv.org/pdf/2607.01120",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.00345",
+    "title": "Registry-Governed Agent Lifecycle:Completing EDDOps with Evaluation-DrivenRegistration, Promotion, and Retirement on AWS AgentCore",
+    "url": "https://arxiv.org/abs/2607.00345",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Richard Kang",
+      "Vincent Wang"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.00345",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00345",
+    "pdf_url": "https://arxiv.org/pdf/2607.00345",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.00911",
+    "title": "From Registry to Repository: How AI Agent Skills Are Written, Adapted, and Maintained",
+    "url": "https://arxiv.org/abs/2607.00911",
+    "published": "2026-07-01",
+    "updated": "2026-07-06",
+    "authors": [
+      "Haoyu Gao",
+      "Jai Lal Lulla",
+      "Hong Yi Lin",
+      "Sebastian Baltes",
+      "Christoph Treude",
+      "Mansooreh Zahedi"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "coding-agent",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.00911",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00911",
+    "pdf_url": "https://arxiv.org/pdf/2607.00911",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.01421",
+    "title": "Risk Architecture for AI-Native Engineering Teams: An Organizational Framework for Agentic System Governance",
+    "url": "https://arxiv.org/abs/2607.01421",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Laxmipriya Ganesh Iyer"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.01421",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01421",
+    "pdf_url": "https://arxiv.org/pdf/2607.01421",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.01061",
+    "title": "Agentic generation of verifiable rules for deterministic, self-expanding reaction classification",
+    "url": "https://arxiv.org/abs/2607.01061",
+    "published": "2026-07-01",
+    "updated": "2026-07-05",
+    "authors": [
+      "Daniel Armstrong",
+      "Maarten Dobbelaere",
+      "Valentas Olikauskas",
+      "Helena Avila",
+      "Octavian Susanu",
+      "Jérôme Waser",
+      "Philippe Schwaller"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "coding-agent",
+      "multi-agent",
+      "planning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.01061",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01061",
+    "pdf_url": "https://arxiv.org/pdf/2607.01061",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.31744",
+    "title": "A Conversational Agentic Interface to Physics-Based Household Digital Twins for Residential Energy Decision Support",
+    "url": "https://arxiv.org/abs/2606.31744",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Costas Mylonas",
+      "Titos Georgoulakis",
+      "Magda Foti"
+    ],
+    "categories": [
+      "eess.SY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.31744",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31744",
+    "pdf_url": "https://arxiv.org/pdf/2606.31744",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.31209",
+    "title": "Long-term Traffic Simulation via Structured Autoregressive Modeling",
+    "url": "https://arxiv.org/abs/2606.31209",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Lingyu Xiao",
+      "Zexin Feng",
+      "Xintao Yan"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.31209",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31209",
+    "pdf_url": "https://arxiv.org/pdf/2606.31209",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.31085",
+    "title": "DDIAgents: Mechanism-Conditioned Context Flow for Drug-Drug Interaction Prediction",
+    "url": "https://arxiv.org/abs/2606.31085",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Zhenqian Shen",
+      "Yu Liu",
+      "Xiaoyi Fu",
+      "Quanming Yao"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.31085",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31085",
+    "pdf_url": "https://arxiv.org/pdf/2606.31085",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.30383",
+    "title": "Whose Side Is Your Agent On? Multi-Party Principal Loyalty in LLM Agents",
+    "url": "https://arxiv.org/abs/2606.30383",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Bojie Li",
+      "Noah Shi"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.30383",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30383",
+    "pdf_url": "https://arxiv.org/pdf/2606.30383",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.30697",
+    "title": "LUMOS: A Semantic Operating-System Layer for Accessibility-Grounded AI Agents",
+    "url": "https://arxiv.org/abs/2606.30697",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Yogeswar Reddy Thota"
+    ],
+    "categories": [
+      "cs.OS",
+      "cs.AI",
+      "cs.CV"
+    ],
+    "topics": [
+      "computer-use",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.30697",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30697",
+    "pdf_url": "https://arxiv.org/pdf/2606.30697",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.30877",
+    "title": "A Systematic Approach to Multi-Agent AI from Advanced Regulatory Control Theory: Safe and Auditable LLM Operator Agents for Process Control",
+    "url": "https://arxiv.org/abs/2606.30877",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Idelfonso B. R. Nogueira",
+      "Sigurd Skogestad"
+    ],
+    "categories": [
+      "eess.SY",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.30877",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30877",
+    "pdf_url": "https://arxiv.org/pdf/2606.30877",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.29957",
+    "title": "SWE-Together: Evaluating Coding Agents in Interactive User Sessions",
+    "url": "https://arxiv.org/abs/2606.29957",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Yifan Wu",
+      "Zhuokai Zhao",
+      "Songlin Li",
+      "Ho Hin Lee",
+      "Jiacheng Zhu",
+      "Shirley Wu",
+      "Tianhe Yu",
+      "Serena Li",
+      "Lizhu Zhang",
+      "Xiangjun Fan",
+      "Shengzhi Li"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.29957",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29957",
+    "pdf_url": "https://arxiv.org/pdf/2606.29957",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.29778",
+    "title": "Mandol: An Agglomerative Agent Memory System for Long-Term Conversations",
+    "url": "https://arxiv.org/abs/2606.29778",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Yuhan Zhang",
+      "Zhiyuan Guo",
+      "Ziheng Zeng",
+      "Wei Wang",
+      "Wentao Wu",
+      "Lijie Xu"
+    ],
+    "categories": [
+      "cs.DB",
+      "cs.AI",
+      "cs.CL",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.29778",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29778",
+    "pdf_url": "https://arxiv.org/pdf/2606.29778",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.30573",
+    "title": "SWE-INTERACT: Reimagining SWE Benchmarks as User-Driven Long-Horizon Coding Sessions",
+    "url": "https://arxiv.org/abs/2606.30573",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Mohit Raghavendra",
+      "Anisha Gunjal",
+      "Aakash Sabharwal",
+      "Yunzhong He"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.30573",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30573",
+    "pdf_url": "https://arxiv.org/pdf/2606.30573",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.30560",
+    "title": "TraceLab: Characterizing Coding Agent Workloads for LLM Serving",
+    "url": "https://arxiv.org/abs/2606.30560",
+    "published": "2026-06-29",
+    "updated": "2026-06-30",
+    "authors": [
+      "Kan Zhu",
+      "Mathew Jacob",
+      "Chenxi Ma",
+      "Yi Pan",
+      "Stephanie Wang",
+      "Arvind Krishnamurthy",
+      "Baris Kasikci"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.PF"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.30560",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30560",
+    "pdf_url": "https://arxiv.org/pdf/2606.30560",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.30887",
+    "title": "Training Therapeutic Judges and Multi-Agent Systems for Human-Aligned Mental Health Support",
+    "url": "https://arxiv.org/abs/2606.30887",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Mizanur Rahman",
+      "Abeer Badawi",
+      "Elahe Rahimi",
+      "Laleh Seyyed-Kalantari",
+      "Frank Rudzicz",
+      "Enamul Hoque",
+      "Elham Dolatabadi"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.30887",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30887",
+    "pdf_url": "https://arxiv.org/pdf/2606.30887",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.00038",
+    "title": "Stop Hand-Holding Your Coding Agent: Engineering the Loops that Replace Step-by-Step Prompting",
+    "url": "https://arxiv.org/abs/2607.00038",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Sandeco Macedo"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.00038",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00038",
+    "pdf_url": "https://arxiv.org/pdf/2607.00038",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.29014",
+    "title": "Customized Generative AI Agent for Transportation Engineering Practice: A Development and Continued Pre-training Guideline",
+    "url": "https://arxiv.org/abs/2606.29014",
+    "published": "2026-06-27",
+    "updated": "2026-07-03",
+    "authors": [
+      "Dianwei Chen",
+      "Yuan-Zheng Lei",
+      "Zifan Zhang",
+      "Yuchen Liu",
+      "Xianfeng Yang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.DL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "planning",
+      "reasoning"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.29014",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29014",
+    "pdf_url": "https://arxiv.org/pdf/2606.29014",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.28781",
+    "title": "HyphaeDB: A Living Knowledge Topology for Agent-First Memory",
+    "url": "https://arxiv.org/abs/2606.28781",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Krishna Halaharvi"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "coding-agent",
+      "memory",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.28781",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28781",
+    "pdf_url": "https://arxiv.org/pdf/2606.28781",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.28839",
+    "title": "The Contagion Tensor: A Framework for Measuring Output-Distribution Coupling in Multi-Agent LLM Systems -- and Auditing the Claims It Enables",
+    "url": "https://arxiv.org/abs/2606.28839",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Zewen Liu"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.28839",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28839",
+    "pdf_url": "https://arxiv.org/pdf/2606.28839",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.28270",
+    "title": "Agent-Native Immune System: Architecture, Taxonomy, and Engineering",
+    "url": "https://arxiv.org/abs/2606.28270",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Bo Shen",
+      "Lifeng Chang",
+      "Tianyuan Wei",
+      "Yunpeng Li",
+      "Feng Shi",
+      "Yichen Han",
+      "Peijie Gao",
+      "Shiyi Kuang",
+      "Xin Chang",
+      "Dehui Li"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "multi-agent",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.28270",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28270",
+    "pdf_url": "https://arxiv.org/pdf/2606.28270",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.28436",
+    "title": "Dockerless: Environment-Free Program Verifier for Coding Agents",
+    "url": "https://arxiv.org/abs/2606.28436",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Wenhao Zeng",
+      "Yuling Shi",
+      "Xiaodong Gu",
+      "Chao Hu",
+      "Chaofan Wang",
+      "Yuhao Cui",
+      "Hongting Zhou",
+      "Mengnan Qi",
+      "Jianqiao Wangni",
+      "Zhaojian Yu",
+      "Shuzheng Gao",
+      "Kai Cai",
+      "Shilin He"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "reasoning"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.28436",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28436",
+    "pdf_url": "https://arxiv.org/pdf/2606.28436",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.26918",
+    "title": "Diagnosing Task Insensitivity in Language Agents",
+    "url": "https://arxiv.org/abs/2606.26918",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Jingyu Liu",
+      "Xiaopeng Wu",
+      "Kehan Chen",
+      "Chuan Yu",
+      "Yong Liu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.26918",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26918",
+    "pdf_url": "https://arxiv.org/pdf/2606.26918",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.26524",
+    "title": "VIGIL: Runtime Enforcement of Behavioral Specifications in AI Agent Skills",
+    "url": "https://arxiv.org/abs/2606.26524",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Ying Li",
+      "Yanju Chen",
+      "Hongbo Wen",
+      "Bosi Zhang",
+      "Hanzhi Liu",
+      "Peiran Wang",
+      "Yu Feng",
+      "Yuan Tian"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.26524",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26524",
+    "pdf_url": "https://arxiv.org/pdf/2606.26524",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.27499",
+    "title": "DMV-Bench: Diagnosing Long-Horizon Multimodal Agents' Visual Memory with Incidental Cue Injection",
+    "url": "https://arxiv.org/abs/2606.27499",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Yujin Tang",
+      "Chenming Shang",
+      "Ruize Xu",
+      "Nikhil Singh"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.27499",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27499",
+    "pdf_url": "https://arxiv.org/pdf/2606.27499",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.27243",
+    "title": "NOVA: A Verification-Aware Agent Harness for Architecture Evolution in Industrial Recommender Systems",
+    "url": "https://arxiv.org/abs/2606.27243",
+    "published": "2026-06-25",
+    "updated": "2026-06-26",
+    "authors": [
+      "Shaohua Liu",
+      "Liang Fang",
+      "Yilong Sun",
+      "Shudong Huang",
+      "Qingsong Luo",
+      "Shaoxin Liu",
+      "Xiaoyang Chen",
+      "Dongqiang Liu",
+      "Chuangang Ma",
+      "Zhenzhen Chai",
+      "Henghuan Wang",
+      "Shijie Quan",
+      "Changyuan Cui",
+      "Zhangbin Zhu",
+      "Peng Chen",
+      "Wei Xu",
+      "Lei Xiao",
+      "Haijie Gu",
+      "Jie Jiang"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "memory",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.27243",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27243",
+    "pdf_url": "https://arxiv.org/pdf/2606.27243",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.26924",
+    "title": "A Deterministic Control Plane for LLM Coding Agents",
+    "url": "https://arxiv.org/abs/2606.26924",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Padmaraj Madatha"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.26924",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26924",
+    "pdf_url": "https://arxiv.org/pdf/2606.26924",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.26883",
+    "title": "EconSimulacra: A Digital Twin Platform of Socio-Economic Systems Powered by LLM Agents",
+    "url": "https://arxiv.org/abs/2606.26883",
+    "published": "2026-06-25",
+    "updated": "2026-06-30",
+    "authors": [
+      "Ryuji Hashimoto",
+      "Masahiro Kaneko",
+      "Kentaro Ueda",
+      "Takehiro Takayanagi",
+      "Kiyoshi Izumi"
+    ],
+    "categories": [
+      "cs.DL"
+    ],
+    "topics": [
+      "memory",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.26883",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26883",
+    "pdf_url": "https://arxiv.org/pdf/2606.26883",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.28409",
+    "title": "Evidence-Driven LLM Agent for C-to-Synthesizable-C Conversion and Verification",
+    "url": "https://arxiv.org/abs/2606.28409",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Zhe Zhao",
+      "Hongbing Lang",
+      "Zhihan Xiao",
+      "Luke Ztz Hu",
+      "John Imoleayo Adebisi",
+      "Songping Mai"
+    ],
+    "categories": [
+      "cs.AR",
+      "cs.AI"
+    ],
+    "topics": [
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.28409",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28409",
+    "pdf_url": "https://arxiv.org/pdf/2606.28409",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.25819",
+    "title": "Beyond Function Calling: Benchmarking Tool-Using Agents under Tool-Environment Unreliability",
+    "url": "https://arxiv.org/abs/2606.25819",
+    "published": "2026-06-24",
+    "updated": "2026-06-27",
+    "authors": [
+      "Yang Tian",
+      "Zhengpeng Shi",
+      "Yu Zhou",
+      "Bo Zhao"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.25819",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25819",
+    "pdf_url": "https://arxiv.org/pdf/2606.25819",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2606.26453",
+    "title": "Optimizing CUDA like a Human: Micro-Profiling Tools as Expert Surrogates for LLM-Based GPU Kernel Optimization",
+    "url": "https://arxiv.org/abs/2606.26453",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Jiading Gai",
+      "Shuai Zhang",
+      "Kaj Bostrom",
+      "Jin Huang",
+      "Vihang Patil",
+      "Haoyang Fang",
+      "Bernie Wang",
+      "Huzefa Rangwala",
+      "George Karypis"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "memory",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.26453",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26453",
+    "pdf_url": "https://arxiv.org/pdf/2606.26453",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.25361",
+    "title": "Memory Makes the Difference: Evaluating How Different Memory Roles Shape Conversational Agents",
+    "url": "https://arxiv.org/abs/2606.25361",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Yuxin Wang",
+      "Paul Thomas",
+      "Zhiwei Yu",
+      "Yuan Gao",
+      "Saeed Hassanpour",
+      "Soroush Vosoughi",
+      "Robert Sim",
+      "Nick Craswell"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.25361",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25361",
+    "pdf_url": "https://arxiv.org/pdf/2606.25361",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.27397",
+    "title": "SidConArena: An Environment Evaluating Agents in Open-Ended,Positive-Sum Bargaining Game",
+    "url": "https://arxiv.org/abs/2606.27397",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Yeqi Feng",
+      "Yuxin Chen",
+      "Tianxing He"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.AI",
+      "cs.GT"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.27397",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27397",
+    "pdf_url": "https://arxiv.org/pdf/2606.27397",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.25139",
+    "title": "Buildrix: An Open Platform for Sharing and Benchmarking Agentic AI Skills in Building Engineering",
+    "url": "https://arxiv.org/abs/2606.25139",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Zixin Jiang",
+      "Bing Dong"
+    ],
+    "categories": [
+      "eess.SY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.25139",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25139",
+    "pdf_url": "https://arxiv.org/pdf/2606.25139",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.24779",
+    "title": "DeepBD: A Grounded Agentic Workflow for Variant Prioritization and Diagnosis of Genetic Birth Defects",
+    "url": "https://arxiv.org/abs/2606.24779",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Shiyu Li",
+      "Ziqi Yan",
+      "Zhihao Wu",
+      "Jielong Lu",
+      "Weiran Liao",
+      "Jiajun Yu",
+      "Genjie Li",
+      "Zeyu Chu",
+      "Jiajun Bu",
+      "Haishuai Wang"
+    ],
+    "categories": [
+      "q-bio.GN",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.24779",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24779",
+    "pdf_url": "https://arxiv.org/pdf/2606.24779",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.24193",
+    "title": "SkyChain Intelligence: A Blockchain-Secured Multi-Agent DRL Framework for Low-Altitude Embodied Artificial Intelligence",
+    "url": "https://arxiv.org/abs/2606.24193",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Haoxiang Luo",
+      "Tianqi Jiang",
+      "Ruichen Zhang",
+      "Yinqiu Liu",
+      "Gang Sun",
+      "Hongfang Yu",
+      "Abbas Jamalipour",
+      "Dong In Kim"
+    ],
+    "categories": [
+      "cs.NI",
+      "cs.DC"
+    ],
+    "topics": [
+      "agent-safety",
+      "embodied-agent",
+      "multi-agent",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.24193",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24193",
+    "pdf_url": "https://arxiv.org/pdf/2606.24193",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.24597",
+    "title": "Qwen-AgentWorld: Language World Models for General Agents",
+    "url": "https://arxiv.org/abs/2606.24597",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Yuxin Zuo",
+      "Zikai Xiao",
+      "Li Sheng",
+      "Fei Huang",
+      "Jianhong Tu",
+      "Yuxuan Liu",
+      "Tianyi Tang",
+      "Xiaomeng Hu",
+      "Yang Su",
+      "Qingfeng Lan",
+      "Yantao Liu",
+      "Qin Zhu",
+      "Yinger Zhang",
+      "Bowen Yu",
+      "Haiquan Zhao",
+      "Haiyang Xu",
+      "Jianxin Yang",
+      "Jiayang Cheng",
+      "Junyang Wang",
+      "Lianghao Deng",
+      "Mingfeng Xue",
+      "Tianyi Bai",
+      "Yang Fan",
+      "Yubo Ma",
+      "Yucheng Li",
+      "Zeyu Cui",
+      "Zhihai Wang",
+      "Zhihui Xie",
+      "Zhuorui Ye",
+      "An Yang",
+      "Dayiheng Liu",
+      "Jingren Zhou",
+      "Ning Ding"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.24597",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24597",
+    "pdf_url": "https://arxiv.org/pdf/2606.24597",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.25115",
+    "title": "Forget to Improve: On-Device LLM-Agent Continual Learning via Budget-Curated Memory",
+    "url": "https://arxiv.org/abs/2606.25115",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Beining Wu",
+      "Zihao Ding",
+      "Jun Huang",
+      "Yanxiao Zhao"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.NI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.25115",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25115",
+    "pdf_url": "https://arxiv.org/pdf/2606.25115",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.24322",
+    "title": "Securing LLM-Agent Long-Term Memory Against Poisoning: Non-Malleable, Origin-Bound Authority with Machine-Checked Guarantees",
+    "url": "https://arxiv.org/abs/2606.24322",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Yedidel Louck"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.24322",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24322",
+    "pdf_url": "https://arxiv.org/pdf/2606.24322",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.24839",
+    "title": "Grading the Grader: Lessons from Evaluating an Agentic Data Analysis System",
+    "url": "https://arxiv.org/abs/2606.24839",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Tian Zheng",
+      "Kai-Tai Hsu"
+    ],
+    "categories": [
+      "cs.AI",
+      "stat.AP"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.24839",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24839",
+    "pdf_url": "https://arxiv.org/pdf/2606.24839",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.23927",
+    "title": "RIFT-Bench: Dynamic Red-teaming For Agentic AI Systems",
+    "url": "https://arxiv.org/abs/2606.23927",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Yarin Yerushalmi Levi",
+      "Roy Betser",
+      "Amit Giloni",
+      "Lidor Erez",
+      "Itay Gershon",
+      "Oren Rachmil",
+      "Sindhu Padakandla",
+      "Roman Vainshtein"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.23927",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23927",
+    "pdf_url": "https://arxiv.org/pdf/2606.23927",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.24551",
+    "title": "GUI vs. CLI: Execution Bottlenecks in Screen-Only and Skill-Mediated Computer-Use Agents",
+    "url": "https://arxiv.org/abs/2606.24551",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Xiao Zhou",
+      "Siyue Zhang",
+      "Yilun Zhao",
+      "Jinbiao Wei",
+      "Tingyu Song",
+      "Arman Cohan",
+      "Chen Zhao"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.24551",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24551",
+    "pdf_url": "https://arxiv.org/pdf/2606.24551",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.22948",
+    "title": "ENVS: Environment-Native Verified Search for Long-Horizon GUI Agents",
+    "url": "https://arxiv.org/abs/2606.22948",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Yincheng Zhou",
+      "Athena Zhuoming Zhong",
+      "Shijie Zhang",
+      "Kevin Zhang",
+      "Teresa Xiaotao Shang",
+      "Shanghang Zhang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.22948",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22948",
+    "pdf_url": "https://arxiv.org/pdf/2606.22948",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.23764",
+    "title": "Emergent Relational Order in LLM Agent Societies: From Collective Affect to Authority Stratification",
+    "url": "https://arxiv.org/abs/2606.23764",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Zhiyuan Ji",
+      "Xinyu Chen",
+      "Ziqi Dai",
+      "Shiyun Tang",
+      "Chunyu Wei",
+      "Yueguo Chen"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.AI"
+    ],
+    "topics": [
+      "multi-agent",
+      "planning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.23764",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23764",
+    "pdf_url": "https://arxiv.org/pdf/2606.23764",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.23343",
+    "title": "Group Selection Promotes Prosocial Prompts in Populations of LLM Agents",
+    "url": "https://arxiv.org/abs/2606.23343",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Luis Celiktemel",
+      "Edward Eichhorn",
+      "Levin Brinkmann",
+      "Robin Schimmelpfennig",
+      "Aron Vallinder",
+      "Yaomin Jiang",
+      "Edward Hughes",
+      "Iyad Rahwan"
+    ],
+    "categories": [
+      "cs.CY"
+    ],
+    "topics": [
+      "coding-agent",
+      "computer-use",
+      "multi-agent",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.23343",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23343",
+    "pdf_url": "https://arxiv.org/pdf/2606.23343",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.22388",
+    "title": "PlanBench-XL: Evaluating Long-Horizon Planning of LLM Tool-Use Agents in Large-Scale Tool Ecosystems",
+    "url": "https://arxiv.org/abs/2606.22388",
+    "published": "2026-06-21",
+    "updated": "2026-06-21",
+    "authors": [
+      "Jiayu Liu",
+      "Qihan Lin",
+      "Cheng Qian",
+      "Rui Wang",
+      "Emre Can Acikgoz",
+      "Xiaocheng Yang",
+      "Jiateng Liu",
+      "Zhenhailong Wang",
+      "Xiusi Chen",
+      "Heng Ji",
+      "Dilek Hakkani-Tür"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.22388",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22388",
+    "pdf_url": "https://arxiv.org/pdf/2606.22388",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.22610",
+    "title": "PaperClaw: Harnessing Agents for Autonomous Research and Human-in-the-Loop Refinement",
+    "url": "https://arxiv.org/abs/2606.22610",
+    "published": "2026-06-21",
+    "updated": "2026-06-21",
+    "authors": [
+      "Weiwei Ye",
+      "Hangchen Liu",
+      "Dongyuan Li",
+      "Renhe Jiang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.22610",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22610",
+    "pdf_url": "https://arxiv.org/pdf/2606.22610",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.21565",
+    "title": "Composing Verifiable Conceptual Models via Building Blocks: Towards Design-Time Verification of Agentic AI Workflows",
+    "url": "https://arxiv.org/abs/2606.21565",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Noe Y. Flandre",
+      "Alexander C. Nwala",
+      "Philippe J. Giabbanelli"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.21565",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21565",
+    "pdf_url": "https://arxiv.org/pdf/2606.21565",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.21732",
+    "title": "Safe to Check, Unsafe to Use: Relinking at the Compression Boundary of LLM Agents",
+    "url": "https://arxiv.org/abs/2606.21732",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Zesen Liu",
+      "Zihan Zhang",
+      "Dongdong She"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.21732",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21732",
+    "pdf_url": "https://arxiv.org/pdf/2606.21732",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.20479",
+    "title": "GroundControl: Anticipating Navigation Failures in Vision-Language Agents via Trajectory-Consistent Uncertainty Estimates",
+    "url": "https://arxiv.org/abs/2606.20479",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Nastaran Darabi",
+      "Divake Kumar",
+      "Sina Tayebati",
+      "Devashri Naik",
+      "Amit Ranjan Trivedi"
+    ],
+    "categories": [
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "embodied-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.20479",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20479",
+    "pdf_url": "https://arxiv.org/pdf/2606.20479",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.20041",
+    "title": "AI Economist Agent: An Agentic Framework for Model-Grounded Economic Analysis with RAG, Knowledge Graphs, and Large Language Models",
+    "url": "https://arxiv.org/abs/2606.20041",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Masahiro Kato"
+    ],
+    "categories": [
+      "econ.GN",
+      "cs.AI",
+      "cs.LG",
+      "q-fin.GN"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.20041",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20041",
+    "pdf_url": "https://arxiv.org/pdf/2606.20041",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.19852",
+    "title": "Prompt, Plan, Extract: Zero-Shot Agentic LLMs Workflows for Lung Pathology Extraction from Clinical Narratives",
+    "url": "https://arxiv.org/abs/2606.19852",
+    "published": "2026-06-18",
+    "updated": "2026-06-25",
+    "authors": [
+      "Aman Pathak",
+      "Cheng Peng",
+      "Mengxian Lyu",
+      "Ziyi Chen",
+      "Reema Solan",
+      "Sankalp Talankar",
+      "Yasir Khan",
+      "Hiren Mehta",
+      "Aokun Chen",
+      "Yi Guo",
+      "Yonghui Wu"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.19852",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19852",
+    "pdf_url": "https://arxiv.org/pdf/2606.19852",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.20047",
+    "title": "PACMS: Submodular Context Selection as a Pluggable Engine for LLM Agents",
+    "url": "https://arxiv.org/abs/2606.20047",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Manu Ghulyani",
+      "Arunabh Singh",
+      "Karan Bharadwaj",
+      "Ankit Nath",
+      "Suranjan Goswami"
+    ],
+    "categories": [
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.20047",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20047",
+    "pdf_url": "https://arxiv.org/pdf/2606.20047",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.19926",
+    "title": "MemGUI-Agent: An End-to-End Long-Horizon Mobile GUI Agent with Proactive Context Management",
+    "url": "https://arxiv.org/abs/2606.19926",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Guangyi Liu",
+      "Gao Wu",
+      "Congxiao Liu",
+      "Pengxiang Zhao",
+      "Liang Liu",
+      "Mading Li",
+      "Qi Zhang",
+      "Mengyan Wang",
+      "Liang Guo",
+      "Yong Liu"
+    ],
+    "categories": [
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.19926",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19926",
+    "pdf_url": "https://arxiv.org/pdf/2606.19926",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.20922",
+    "title": "Think Twice Before You Act: Protecting LLM Agents Against Tool Description Poisoning via Isolated Planning",
+    "url": "https://arxiv.org/abs/2606.20922",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Shanghao Shi",
+      "Xiao Wang",
+      "Chaoyu Zhang",
+      "Hao Li",
+      "Wenjing Lou",
+      "Thomas Hou",
+      "Yevgeniy Vorobeychik",
+      "Chongjie Zhang",
+      "Ning Zhang"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.20922",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20922",
+    "pdf_url": "https://arxiv.org/pdf/2606.20922",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.19787",
+    "title": "ORAgentBench: Can LLM Agents Solve Challenging Operations Research Tasks End to End?",
+    "url": "https://arxiv.org/abs/2606.19787",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Jiajun Li",
+      "Mingshu Cai",
+      "Yixuan Li",
+      "Yu Ding",
+      "Ran Hou",
+      "Guanyu Nie",
+      "Xiongwei Han",
+      "Wanyuan Wang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.19787",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19787",
+    "pdf_url": "https://arxiv.org/pdf/2606.19787",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.18068",
+    "title": "Agentic AI-based Framework for Mitigating Premature Diagnostic Handoff and Silent Hallucination in Healthcare Applications",
+    "url": "https://arxiv.org/abs/2606.18068",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Divyansh Srivastava",
+      "Shreya Ghosh",
+      "Anshul Verma",
+      "Rajkumar Buyya"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "reasoning"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.18068",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18068",
+    "pdf_url": "https://arxiv.org/pdf/2606.18068",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.17680",
+    "title": "EnvRL: Learn from Environment Dynamics in Agentic Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2606.17680",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Zhitong Wang",
+      "Songze Li",
+      "Hao Peng",
+      "Shuzheng Si",
+      "Yi Wang",
+      "Maosong Sun",
+      "Juanzi Li"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.17680",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17680",
+    "pdf_url": "https://arxiv.org/pdf/2606.17680",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.18406",
+    "title": "CoreMem: Riemannian Retrieval and Fisher-Guided Distillation for Long-Term Memory in Dialogue Agents",
+    "url": "https://arxiv.org/abs/2606.18406",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Jiaqi Chen",
+      "Yongqin Zeng",
+      "Shaoshen Chen",
+      "Yijian Zhang",
+      "Hai-Tao Zheng",
+      "Chunxia Ma",
+      "XiuTeng Zhou"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.18406",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18406",
+    "pdf_url": "https://arxiv.org/pdf/2606.18406",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.17449",
+    "title": "MODE-RAG: Manifold Outlier Diagnosis and Energy-based Retrieval-Augmented Generation Evaluation",
+    "url": "https://arxiv.org/abs/2606.17449",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Zehang Wei",
+      "Jiaxin Dai",
+      "Jiamin Yan",
+      "Xiang Xiang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.CV",
+      "cs.LG",
+      "cs.MM"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "rag",
+      "reasoning"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.17449",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17449",
+    "pdf_url": "https://arxiv.org/pdf/2606.17449",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.16432",
+    "title": "ACCORD: Action-Conditioned Contextual Grounding for Language Agents",
+    "url": "https://arxiv.org/abs/2606.16432",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Lai Jiang",
+      "Cheng Qian",
+      "Zhenhailong Wang",
+      "Pan Lu",
+      "Heng Ji",
+      "Hao Peng"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.16432",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16432",
+    "pdf_url": "https://arxiv.org/pdf/2606.16432",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.15903",
+    "title": "Control-Plane Placement Shapes Forgetting: An Architectural Study of Agent Memory Across Thirteen System Configurations",
+    "url": "https://arxiv.org/abs/2606.15903",
+    "published": "2026-06-14",
+    "updated": "2026-06-16",
+    "authors": [
+      "Dongxu Yang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.15903",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15903",
+    "pdf_url": "https://arxiv.org/pdf/2606.15903",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.15609",
+    "title": "FragFuse: Bypassing Access Control of Large Language Model Agents via Memory-Based Query Fragmentation and Fusion",
+    "url": "https://arxiv.org/abs/2606.15609",
+    "published": "2026-06-14",
+    "updated": "2026-06-14",
+    "authors": [
+      "Zixin Rao",
+      "Wentian Zhu",
+      "Chan Aristella Lu",
+      "Zhaorun Chen",
+      "Wei Niu",
+      "Le Guan",
+      "Bo Li",
+      "Zhen Xiang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.15609",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15609",
+    "pdf_url": "https://arxiv.org/pdf/2606.15609",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.15079",
+    "title": "Ling and Ring 2.6 Technical Report: Efficient and Instant Agentic Intelligence at Trillion-Parameter Scale",
+    "url": "https://arxiv.org/abs/2606.15079",
+    "published": "2026-06-13",
+    "updated": "2026-06-13",
+    "authors": [
+      "Ang Li",
+      "Ben Liu",
+      "Bin Han",
+      "Bin Hu",
+      "Bin Jing",
+      "Binbin Hu",
+      "Bing Li",
+      "Cai Chen",
+      "Caizhi Tang",
+      "Changxin Tian",
+      "Chao Huang",
+      "Chao Zhang",
+      "Chen Liang",
+      "Chen Qian",
+      "Chengfu Tang",
+      "Chengyao Wen",
+      "Chilin Fu",
+      "Chunwei Wu",
+      "Cong Zhang",
+      "Cunyin Peng",
+      "Daixin Wang",
+      "Dalong Zhang",
+      "Deng Zhao",
+      "Dingnan Jin",
+      "Dingyuan Zhu",
+      "Donghao Zhang",
+      "Fan Yuan",
+      "Fangzheng Zhao",
+      "Fanzhuang Meng",
+      "Feifan Wu",
+      "Feng Xu",
+      "Fengbin Fang",
+      "Gangshan Wang",
+      "Guodong Yang",
+      "Hailin Zhao",
+      "Haitao Wang",
+      "Haitao Zhang",
+      "Hanxiao Zhang",
+      "Hanzi Wang",
+      "Hao Dai",
+      "Hao Liu",
+      "Hao Qian",
+      "Hao Wu",
+      "Haoxiong Liu",
+      "Haoyu Xu",
+      "Heng Zhang",
+      "Hong Liu",
+      "Hongliang Zhang",
+      "Hongrui Liu",
+      "Hongxun Li",
+      "Hongzhi Ruan",
+      "Huaidong Xiong",
+      "Huihuang Zheng",
+      "Huikang Tang",
+      "Jia Guo",
+      "Jia Li",
+      "Jia Liu",
+      "Jiameng Wang",
+      "Jiaming Liu",
+      "Jiannan Shi",
+      "Jianping Wei",
+      "Jiaolong Yang",
+      "Jiapeng Wang",
+      "Jie Gao",
+      "Jie Wang",
+      "Jiewei Wu",
+      "Jin Yang",
+      "Jinjin Li",
+      "Jinjing Huang",
+      "Jinquan Sun",
+      "Jinyao Chen",
+      "Juanhui Tu",
+      "Jun Liu",
+      "Jun Mei",
+      "Jun Xu",
+      "Jun Zhou",
+      "Junjie Ou",
+      "Junnan Sipan",
+      "Junpeng Fang",
+      "Kaihong Zhang",
+      "Kaiqin Hu",
+      "Ke Shi",
+      "Kuan Xu",
+      "Kun Tang",
+      "Kunlong Chen",
+      "Lanyin Mei",
+      "Lei Chen",
+      "Lei Liang",
+      "Lei Xu",
+      "Li Tang",
+      "Liang Jiang",
+      "Liangcheng Fu",
+      "Lihui Zhang",
+      "Linfeng Shi",
+      "Lintao Ma",
+      "Liyuan Liu",
+      "Longfei Li",
+      "Longfei Zheng",
+      "Lu Liu",
+      "Lu Yu",
+      "Man Li",
+      "Meiqi Zhu",
+      "Meng Li",
+      "Mengjie Gao",
+      "Mengshu Sun",
+      "Mingming Yin",
+      "Mingyang Zhang",
+      "Mingyuan Fan",
+      "Nuo Xu",
+      "Pan Tang",
+      "Peijie Jiang",
+      "Peilong Zhao",
+      "Peng Lin",
+      "Pingping Liu",
+      "Qi Zuo",
+      "Qian Zhao",
+      "Qiang Cheng",
+      "Qianggang Cao",
+      "Qiaoben Bao",
+      "Qing Cui",
+      "Qingyuan Yang",
+      "Qitao Shi",
+      "Qiyin Huang",
+      "Qizheng Zhou",
+      "Quan Wan",
+      "Runyuan Zhao",
+      "Shaomian Zheng",
+      "Shaowei Wei",
+      "Shengnan Zhang",
+      "Shuaicheng Li",
+      "Shujie Li",
+      "Shuo Zhang",
+      "Sikang Bian",
+      "Tianchu Yao",
+      "Tiange Xu",
+      "Tianshu Wang",
+      "Ting Guo",
+      "Tinghao Wang",
+      "Tingwei Huang",
+      "Tong Zhao",
+      "Tongkai Yang",
+      "Wang Hong",
+      "Wanli Gu",
+      "Wei Lu",
+      "Weichang Wu",
+      "Weiguang Han",
+      "Weiquan Li",
+      "Wenbo Shen",
+      "Wenjing Fang",
+      "Wenzhi Tang",
+      "Xiang Shu",
+      "Xiao Shi",
+      "Xiaodong Yan",
+      "Xiaolu Zhang",
+      "Xiaopei Wan",
+      "Xiaqing Sun",
+      "Xin Zhao",
+      "Xingyu Lu",
+      "Xinxing Yang",
+      "Xinyao Tang",
+      "Xinyu Kong",
+      "Xinyu Liu",
+      "Xiong Xu",
+      "Xuan Sun",
+      "Xudong Han",
+      "Xudong Wang",
+      "Xujie Shen",
+      "Yalin Zhang",
+      "Yangyang Hou",
+      "Yankun Ren",
+      "Yao Zhao",
+      "Ye Chen",
+      "Yeyang Chen",
+      "Yibo Cao",
+      "Yifan Zuo",
+      "Yijie Chen",
+      "Ying Li",
+      "Yingjie Song",
+      "Yingxue Li",
+      "Yiqi Wang",
+      "Yixuan Sun",
+      "Yizhu Xiao",
+      "Yongfei Xu",
+      "Yu Liu",
+      "Yuchen Fang",
+      "Yue Gao",
+      "Yue Yu",
+      "Yue Zhang",
+      "Yuqi Zhang",
+      "Yuxiao He",
+      "Yuxiao Lu",
+      "Yuxin Tian",
+      "Yuxuan Li",
+      "Yuzhuo Fu",
+      "Zhankai Xu",
+      "Zhaoxin Huan",
+      "Zhenduo Zhang",
+      "Zhengke Gui",
+      "Zhengyu Huang",
+      "Zhenjun Ma",
+      "Zhenxuan Pan",
+      "Zheping Qu",
+      "Zhibo Zhu",
+      "Zhidong Fan",
+      "Zhigang Huangfu",
+      "Zhihao Wang",
+      "Zhiqiang Zhang",
+      "Zhizhen Liu",
+      "Zhuyan Zhou",
+      "Zibin Lin",
+      "Zihang Zeng",
+      "Zihao Wang",
+      "Zilong Wang",
+      "Ziqi Liu",
+      "Zitao Xuan",
+      "Zixuan Cheng",
+      "Zujie Wen",
+      "Zuoli Tang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.15079",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15079",
+    "pdf_url": "https://arxiv.org/pdf/2606.15079",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.14571",
+    "title": "StreamMemBench: Streaming Evaluation of Agent Memory for Future-Oriented Assistance",
+    "url": "https://arxiv.org/abs/2606.14571",
+    "published": "2026-06-12",
+    "updated": "2026-06-12",
+    "authors": [
+      "Guanming Liu",
+      "Yuqi Ren",
+      "Hansu Gu",
+      "Peng Zhang",
+      "Weihang Wang",
+      "Jiahao Liu",
+      "Ning Gu",
+      "Tun Lu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.14571",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.14571",
+    "pdf_url": "https://arxiv.org/pdf/2606.14571",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.13643",
+    "title": "Recursive Agent Harnesses",
+    "url": "https://arxiv.org/abs/2606.13643",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Elias Lumer",
+      "Sahil Sen",
+      "Kevin Paul",
+      "Vamse Kumar Subbiah"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2606.13643",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13643",
+    "pdf_url": "https://arxiv.org/pdf/2606.13643",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2606.13385",
+    "title": "Who Pays the Price? Stakeholder-Centric Prompt Injection Benchmarking for Real-world Web Agents",
+    "url": "https://arxiv.org/abs/2606.13385",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Zihao Wang",
+      "Yiming Li",
+      "Yutong Wu",
+      "Zheyu Liu",
+      "Kangjie Chen",
+      "Fok Kar Wai",
+      "Pin-Yu Chen",
+      "Vrizlynn L. L. Thing",
+      "Bo Li",
+      "Dacheng Tao",
+      "Tianwei Zhang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.CY",
+      "cs.HC",
+      "cs.MM"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.13385",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13385",
+    "pdf_url": "https://arxiv.org/pdf/2606.13385",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.13904",
+    "title": "SANA: What Matters for QA Agents over Massive Data Lakes?",
+    "url": "https://arxiv.org/abs/2606.13904",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Austin Senna Wijaya",
+      "Jiaxiang Liu",
+      "Haonan Wang",
+      "Eugene Wu"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.DB"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "embodied-agent",
+      "planning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.13904",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13904",
+    "pdf_url": "https://arxiv.org/pdf/2606.13904",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.12341",
+    "title": "OCELOT: Inference-Leakage Budgets for Privacy-Preserving LLM Agents",
+    "url": "https://arxiv.org/abs/2606.12341",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Jin Xie",
+      "Songze Li"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.12341",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12341",
+    "pdf_url": "https://arxiv.org/pdf/2606.12341",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.12195",
+    "title": "InternVideo3: Agentify Foundation Models with Multimodal Contextual Reasoning",
+    "url": "https://arxiv.org/abs/2606.12195",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Ziang Yan",
+      "Sheng Xia",
+      "Jiashuo Yu",
+      "Yue Wu",
+      "Tianxiang Jiang",
+      "Songze Li",
+      "Kanghui Tian",
+      "Yicheng Xu",
+      "Yinan He",
+      "Kai Chen",
+      "Limin Wang",
+      "Yu Qiao",
+      "Yi Wang"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.12195",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12195",
+    "pdf_url": "https://arxiv.org/pdf/2606.12195",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.11869",
+    "title": "Agents All the Way Down; A Methodology for Building Custom AI Agents from Substrate to Production",
+    "url": "https://arxiv.org/abs/2606.11869",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Marc Alier Forment",
+      "Juanan Pereira",
+      "Francisco José García-Peñalvo",
+      "María José Casañ Guerrero"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2606.11869",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.11869",
+    "pdf_url": "https://arxiv.org/pdf/2606.11869",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2606.17076",
+    "title": "CMIP-Forge: An Agentic System that Retrieves, Computes, and Self-Reviews Climate Science",
+    "url": "https://arxiv.org/abs/2606.17076",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Dmitrii Pantiukhin",
+      "Boris Shapkin",
+      "Ivan Kuznetsov",
+      "Thomas Jung",
+      "Nikolay Koldunov"
+    ],
+    "categories": [
+      "physics.ao-ph",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.17076",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17076",
+    "pdf_url": "https://arxiv.org/pdf/2606.17076",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.11349",
+    "title": "Knowing When to Ask: Self-Gated Clarification for Hierarchical Language Agents",
+    "url": "https://arxiv.org/abs/2606.11349",
+    "published": "2026-06-09",
+    "updated": "2026-06-12",
+    "authors": [
+      "Aijing Gao",
+      "Yiming Kang",
+      "Mengdie Flora Wang",
+      "Jae Oh Woo"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.11349",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.11349",
+    "pdf_url": "https://arxiv.org/pdf/2606.11349",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.11078",
+    "title": "A History-Aware Visually Grounded Critic for Computer Use Agents",
+    "url": "https://arxiv.org/abs/2606.11078",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Jaewoo Lee",
+      "Zaid Khan",
+      "Archiki Prasad",
+      "Justin Chih-Yao Chen",
+      "Supriyo Chakraborty",
+      "Kartik Balasubramaniam",
+      "Sambit Sahu",
+      "Elias Stengel-Eskin",
+      "Hyunji Lee",
+      "Mohit Bansal"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.11078",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.11078",
+    "pdf_url": "https://arxiv.org/pdf/2606.11078",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.10423",
+    "title": "WebChallenger: A Reliable and Efficient Generalist Web Agent",
+    "url": "https://arxiv.org/abs/2606.10423",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Jayoo Hwang",
+      "Xiaowen Zhang",
+      "Vedant Padwal"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "computer-use",
+      "embodied-agent",
+      "memory",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.10423",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10423",
+    "pdf_url": "https://arxiv.org/pdf/2606.10423",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.10381",
+    "title": "Agentic Hybrid RAG for Evidence-Grounded Muon Collider Analysis",
+    "url": "https://arxiv.org/abs/2606.10381",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Ruobing Jiang",
+      "Dawei Fu",
+      "Cheng Jiang",
+      "Tianyi Yang",
+      "Zijian Wang",
+      "Youpeng Wu",
+      "Yong Ban",
+      "Yajun Mao",
+      "Qiang Li"
+    ],
+    "categories": [
+      "hep-ex",
+      "cs.AI",
+      "cs.CL",
+      "cs.IR",
+      "physics.ins-det"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.10381",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10381",
+    "pdf_url": "https://arxiv.org/pdf/2606.10381",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.09764",
+    "title": "iOSWorld: A Benchmark for Personally Intelligent Phone Agents",
+    "url": "https://arxiv.org/abs/2606.09764",
+    "published": "2026-06-08",
+    "updated": "2026-06-08",
+    "authors": [
+      "Lawrence Keunho Jang",
+      "Mareks Woodside",
+      "Geronimo Carom",
+      "Andrew Keunwoo Jang",
+      "Jing Yu Koh",
+      "Ruslan Salakhutdinov"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.09764",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09764",
+    "pdf_url": "https://arxiv.org/pdf/2606.09764",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.09399",
+    "title": "RunAgent SuperBrowser: A Theory of Autonomous Web Navigation Grounded in Human Browsing Behaviour",
+    "url": "https://arxiv.org/abs/2606.09399",
+    "published": "2026-06-08",
+    "updated": "2026-06-08",
+    "authors": [
+      "Radeen Mostafa",
+      "Sawradip Saha"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.09399",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09399",
+    "pdf_url": "https://arxiv.org/pdf/2606.09399",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.09549",
+    "title": "SecureClaw: Clawing Back Control of LLM Agents",
+    "url": "https://arxiv.org/abs/2606.09549",
+    "published": "2026-06-08",
+    "updated": "2026-06-08",
+    "authors": [
+      "Yuhan Ma",
+      "Stefan Schmid"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.09549",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09549",
+    "pdf_url": "https://arxiv.org/pdf/2606.09549",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.09071",
+    "title": "REFLECT: Intervention-Supported Error Attribution for Silent Failures in LLM Agent Traces",
+    "url": "https://arxiv.org/abs/2606.09071",
+    "published": "2026-06-08",
+    "updated": "2026-06-08",
+    "authors": [
+      "Xiaofeng Lin",
+      "Yingxu Wang",
+      "Tung Sum Thomas Kwok",
+      "Daniel Guo",
+      "Sahil Arun Nale",
+      "Charles Fleming",
+      "Guang Cheng"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.09071",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09071",
+    "pdf_url": "https://arxiv.org/pdf/2606.09071",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.05463",
+    "title": "PSEBench: A Controllable and Verifiable Benchmark for Evaluating LLMs in Patient Safety Event Triage",
+    "url": "https://arxiv.org/abs/2606.05463",
+    "published": "2026-06-03",
+    "updated": "2026-06-09",
+    "authors": [
+      "Keqi Han",
+      "Ryan Young",
+      "Annabel Strauss",
+      "Lindsey Hughes",
+      "Katharine M. Nesbitt",
+      "Nicole Schueler",
+      "Che Ngufor",
+      "Carl Yang",
+      "Yuan Xue",
+      "Zhijun Yin"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.05463",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.05463",
+    "pdf_url": "https://arxiv.org/pdf/2606.05463",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.03135",
+    "title": "Uncertainty-Aware Clarification in LLM Agents with Information Gain",
+    "url": "https://arxiv.org/abs/2606.03135",
+    "published": "2026-06-02",
+    "updated": "2026-06-02",
+    "authors": [
+      "Mengyi Deng",
+      "Zhiwei Li",
+      "Xin Li",
+      "Tingyu Zhu",
+      "Ying Zhao",
+      "Zhijiang Guo",
+      "Wei Wang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.03135",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.03135",
+    "pdf_url": "https://arxiv.org/pdf/2606.03135",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.04120",
+    "title": "SaliMory: Orchestrating Cognitive Memory for Conversational Agents",
+    "url": "https://arxiv.org/abs/2606.04120",
+    "published": "2026-06-02",
+    "updated": "2026-06-02",
+    "authors": [
+      "Kai Zhang",
+      "Xinyuan Zhang",
+      "Hongda Jiang",
+      "Shiun-Zu Kuo",
+      "Hyokun Yun",
+      "Ejaz Ahmed",
+      "Shereen Oraby",
+      "Ziyun Li",
+      "Sanat Sharma",
+      "Ann Lee",
+      "Ahmed A Aly",
+      "Anuj Kumar",
+      "Raffay Hamid",
+      "Xin Luna Dong"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.04120",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.04120",
+    "pdf_url": "https://arxiv.org/pdf/2606.04120",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.04296",
+    "title": "The Saturation Trap and the Subjectivity of Intervention Timing: Why Affect-Based Triggers and LLM Judges Fail to Time Interventions on Autonomous Agents",
+    "url": "https://arxiv.org/abs/2606.04296",
+    "published": "2026-06-02",
+    "updated": "2026-06-02",
+    "authors": [
+      "Manvendra Modgil"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.04296",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.04296",
+    "pdf_url": "https://arxiv.org/pdf/2606.04296",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.03108",
+    "title": "EvoTrainer: Co-Evolving LLM Policies and Training Harnesses for Autonomous Agentic Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2606.03108",
+    "published": "2026-06-02",
+    "updated": "2026-06-12",
+    "authors": [
+      "Guhong Chen",
+      "Yingcheng Shi",
+      "Yongbin Li",
+      "Binhua Li",
+      "Xander Xu",
+      "Hu Wei",
+      "Shiwen Ni",
+      "Min Yang",
+      "Jieping Ye"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "reasoning"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.03108",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.03108",
+    "pdf_url": "https://arxiv.org/pdf/2606.03108",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.02388",
+    "title": "Policy and World Modeling Co-Training for Language Agents",
+    "url": "https://arxiv.org/abs/2606.02388",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Ning Lu",
+      "Baijiong Lin",
+      "Shengcai Liu",
+      "Jiahao Wu",
+      "Haoze Lv",
+      "Yanbin Wei",
+      "Lingting Zhu",
+      "Shengju Qian",
+      "Xin Wang",
+      "Ying-Cong Chen",
+      "Qi Wang",
+      "Ke Tang"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.02388",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.02388",
+    "pdf_url": "https://arxiv.org/pdf/2606.02388",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.01815",
+    "title": "CRAB-Bench: Evaluating LLM Agents under Complex Task Dependencies and Human-aligned User Simulation",
+    "url": "https://arxiv.org/abs/2606.01815",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Danqing Wang",
+      "Akshay Sivaraman",
+      "Lei Li"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.01815",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.01815",
+    "pdf_url": "https://arxiv.org/pdf/2606.01815",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.02380",
+    "title": "SPADE-Bench: Evaluating Spontaneous Strategic Deception in Agents via Plan-Action Divergence",
+    "url": "https://arxiv.org/abs/2606.02380",
+    "published": "2026-06-01",
+    "updated": "2026-06-28",
+    "authors": [
+      "Yuyan Bu",
+      "Haowei Li",
+      "Qirui Zheng",
+      "Bowen Dong",
+      "Kaiyue Yang",
+      "Jiaming Ji",
+      "Yingshui Tan",
+      "Wenxin Li",
+      "Yaodong Yang",
+      "Juntao Dai"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2606.02380",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.02380",
+    "pdf_url": "https://arxiv.org/pdf/2606.02380",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.00914",
+    "title": "Adversarial Feeds Steer LLM Agent Decisions Against Their Defaults",
+    "url": "https://arxiv.org/abs/2606.00914",
+    "published": "2026-05-30",
+    "updated": "2026-05-30",
+    "authors": [
+      "Rana Muhammad Usman"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.00914",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.00914",
+    "pdf_url": "https://arxiv.org/pdf/2606.00914",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.00915",
+    "title": "Autonomous agentic design for photonics",
+    "url": "https://arxiv.org/abs/2606.00915",
+    "published": "2026-05-30",
+    "updated": "2026-05-30",
+    "authors": [
+      "Prashanta Kharel",
+      "Amin Khavasi",
+      "Xinzhong Chen",
+      "Tyler W. Hughes"
+    ],
+    "categories": [
+      "physics.optics"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.00915",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.00915",
+    "pdf_url": "https://arxiv.org/pdf/2606.00915",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.31278",
+    "title": "Industrializing Prediction-Powered Inference: The GLIDE Library for Reliable GenAI and Agentic Systems Evaluation",
+    "url": "https://arxiv.org/abs/2605.31278",
+    "published": "2026-05-29",
+    "updated": "2026-06-04",
+    "authors": [
+      "Grégoire Martinon",
+      "Ibrahim Merad",
+      "Mohammed Raki"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG",
+      "stat.ME"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2605.31278",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.31278",
+    "pdf_url": "https://arxiv.org/pdf/2605.31278",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2605.29676",
+    "title": "Notation Matters: A Benchmark Study of Token-Optimized Formats in Agentic AI Systems",
+    "url": "https://arxiv.org/abs/2605.29676",
+    "published": "2026-05-28",
+    "updated": "2026-06-17",
+    "authors": [
+      "Lorenz Kutschka",
+      "Bernhard Geiger"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2605.29676",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.29676",
+    "pdf_url": "https://arxiv.org/pdf/2605.29676",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2605.28046",
+    "title": "MemCog: From Memory-as-Tool to Memory-as-Cognition in Conversational Agents",
+    "url": "https://arxiv.org/abs/2605.28046",
+    "published": "2026-05-27",
+    "updated": "2026-05-27",
+    "authors": [
+      "Zihan Li",
+      "Xingyu Fan",
+      "Feifei Li",
+      "Wenhui Que"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.28046",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.28046",
+    "pdf_url": "https://arxiv.org/pdf/2605.28046",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.28607",
+    "title": "Adaptive Multimodal Agents-Based Framework for Automatic Workflow Execution",
+    "url": "https://arxiv.org/abs/2605.28607",
+    "published": "2026-05-27",
+    "updated": "2026-05-27",
+    "authors": [
+      "Susanna Cifani",
+      "Mario Luca Bernardi",
+      "Marta Cimitile"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "embodied-agent",
+      "multi-agent",
+      "planning",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.28607",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.28607",
+    "pdf_url": "https://arxiv.org/pdf/2605.28607",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.28120",
+    "title": "LegalGraphRAG: Multi-Agent Graph Retrieval-Augmented Generation for Reliable Legal Reasoning",
+    "url": "https://arxiv.org/abs/2605.28120",
+    "published": "2026-05-27",
+    "updated": "2026-05-27",
+    "authors": [
+      "Zerui Chen",
+      "Qinggang Zhang",
+      "Zhishang Xiang",
+      "Zhimin Wei",
+      "Linfeng Gao",
+      "Xiao Huang",
+      "Zhihong Zhang",
+      "Jinsong Su"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.28120",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.28120",
+    "pdf_url": "https://arxiv.org/pdf/2605.28120",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.28787",
+    "title": "Do Agents Need Semantic Metadata? A Comparative Study in Agentic Data Retrieval",
+    "url": "https://arxiv.org/abs/2605.28787",
+    "published": "2026-05-27",
+    "updated": "2026-05-27",
+    "authors": [
+      "Shiyu Chen",
+      "Tarfah Alrashed",
+      "Alon Halevy",
+      "Natasha Noy"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.28787",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.28787",
+    "pdf_url": "https://arxiv.org/pdf/2605.28787",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.27366",
+    "title": "MUSE-Autoskill: Self-Evolving Agents via Skill Creation, Memory, Management, and Evaluation",
+    "url": "https://arxiv.org/abs/2605.27366",
+    "published": "2026-05-26",
+    "updated": "2026-07-03",
+    "authors": [
+      "Huawei Lin",
+      "Peng Li",
+      "Jie Song",
+      "Fuxin Jiang",
+      "Tieying Zhang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.LG",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.27366",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.27366",
+    "pdf_url": "https://arxiv.org/pdf/2605.27366",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.26926",
+    "title": "From Norms to Indicators (N2I-RAG): An Agentic Retrieval-Augmented Generation Framework for Legal Indicator Computation",
+    "url": "https://arxiv.org/abs/2605.26926",
+    "published": "2026-05-26",
+    "updated": "2026-05-26",
+    "authors": [
+      "Youssef Al Mouatamid",
+      "Marie Bonnin",
+      "Jihad Zahir"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.26926",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.26926",
+    "pdf_url": "https://arxiv.org/pdf/2605.26926",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.27333",
+    "title": "FinHarness: An Inline Lifecycle Safety Harness for Finance LLM Agents",
+    "url": "https://arxiv.org/abs/2605.27333",
+    "published": "2026-05-26",
+    "updated": "2026-05-26",
+    "authors": [
+      "Haoxuan Jia",
+      "Yang Liu",
+      "Bin Chong",
+      "Yingguang Yang",
+      "Yancheng Chen",
+      "Jiayu Liang",
+      "Qian Li",
+      "Hanning Lu",
+      "Kefu Xu",
+      "Hao Zheng",
+      "Chongyang Zhang",
+      "Hao Peng",
+      "Philip S. Yu"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.27333",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.27333",
+    "pdf_url": "https://arxiv.org/pdf/2605.27333",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.26720",
+    "title": "Towards Feedback-to-Plan Decisions for Self-Evolving LLM Agents in CUDA Kernel Generation",
+    "url": "https://arxiv.org/abs/2605.26720",
+    "published": "2026-05-26",
+    "updated": "2026-05-26",
+    "authors": [
+      "Yee Hin Chong",
+      "Jiaming Wu",
+      "Youhui Zhang",
+      "Peng Qu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.26720",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.26720",
+    "pdf_url": "https://arxiv.org/pdf/2605.26720",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.26252",
+    "title": "Is Agent Memory a Database? Rethinking Data Foundations for Long-Term AI Agent Memory",
+    "url": "https://arxiv.org/abs/2605.26252",
+    "published": "2026-05-25",
+    "updated": "2026-05-25",
+    "authors": [
+      "Abdelghny Orogat",
+      "Essam Mansour"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.DB"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.26252",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.26252",
+    "pdf_url": "https://arxiv.org/pdf/2605.26252",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.26305",
+    "title": "Experiments in Agentic AI for Science",
+    "url": "https://arxiv.org/abs/2605.26305",
+    "published": "2026-05-25",
+    "updated": "2026-05-29",
+    "authors": [
+      "Judy Fox",
+      "Geoffrey Fox"
+    ],
+    "categories": [
+      "cs.AI",
+      "eess.SY",
+      "hep-ph"
+    ],
+    "topics": [
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm",
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.26305",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.26305",
+    "pdf_url": "https://arxiv.org/pdf/2605.26305",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.23636",
+    "title": "RF Instrument Agent (RFIA): Empowering RF Instruments with Natural Language Understanding, Scheduling and Execution of Complex Tasks",
+    "url": "https://arxiv.org/abs/2605.23636",
+    "published": "2026-05-22",
+    "updated": "2026-05-22",
+    "authors": [
+      "Chunhui Li",
+      "Wei Fan"
+    ],
+    "categories": [
+      "eess.SY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.23636",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.23636",
+    "pdf_url": "https://arxiv.org/pdf/2605.23636",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.22321",
+    "title": "Benchmarking Autonomous Agents against Temporal, Spatial, and Semantic Evasions",
+    "url": "https://arxiv.org/abs/2605.22321",
+    "published": "2026-05-21",
+    "updated": "2026-05-21",
+    "authors": [
+      "Jianan Ma",
+      "Xiaohu Du",
+      "Ruixiao Lin",
+      "Yaoxiang Bian",
+      "Jialuo Chen",
+      "Jingyi Wang",
+      "Xiaofang Yang",
+      "Shiwen Cui",
+      "Changhua Meng",
+      "Xinhao Deng",
+      "Zhen Wang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.22321",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.22321",
+    "pdf_url": "https://arxiv.org/pdf/2605.22321",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.21740",
+    "title": "SMDD-Bench: Can LLMs Solve Real-World Small Molecule Drug Design Tasks?",
+    "url": "https://arxiv.org/abs/2605.21740",
+    "published": "2026-05-20",
+    "updated": "2026-05-24",
+    "authors": [
+      "Kevin Han",
+      "Renfei Zhang",
+      "Kathy Wei",
+      "Hamed Mahdavi",
+      "Niloofar Mireshghallah",
+      "Amir Barati Farimani"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.21740",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.21740",
+    "pdf_url": "https://arxiv.org/pdf/2605.21740",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.20874",
+    "title": "Governance by Construction for Generalist Agents",
+    "url": "https://arxiv.org/abs/2605.20874",
+    "published": "2026-05-20",
+    "updated": "2026-05-20",
+    "authors": [
+      "Segev Shlomov",
+      "Iftach Shoham",
+      "Alon Oved",
+      "Ido Levy",
+      "Sami Marreed",
+      "Harold Ship",
+      "Offer Akrabi",
+      "Sergey Zeltyn",
+      "Avi Yaeli",
+      "Nir Mashkif"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.20874",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.20874",
+    "pdf_url": "https://arxiv.org/pdf/2605.20874",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.20306",
+    "title": "WildRoadBench: A Wild Aerial Road-Damage Grounding Benchmark for Vision-Language Models and Autonomous Agents",
+    "url": "https://arxiv.org/abs/2605.20306",
+    "published": "2026-05-19",
+    "updated": "2026-06-02",
+    "authors": [
+      "Bingnan Liu",
+      "Chenhang Cui",
+      "Rui Huang",
+      "Jiani Luo",
+      "Zhirong Shen",
+      "Tinghao Wang",
+      "Xiande Huang",
+      "Lingbei Meng",
+      "Fei Shen",
+      "An Zhang"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.20306",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.20306",
+    "pdf_url": "https://arxiv.org/pdf/2605.20306",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.18672",
+    "title": "Position: A Three-Layer Probabilistic Assume-Guarantee Architecture Is Structurally Required for Safe LLM Agent Deployment",
+    "url": "https://arxiv.org/abs/2605.18672",
+    "published": "2026-05-18",
+    "updated": "2026-05-18",
+    "authors": [
+      "S. Bensalem",
+      "Y. Dong",
+      "M. Franzle",
+      "X. Huang",
+      "J. Kroger",
+      "D. Nickovic",
+      "A. Nouri",
+      "R. Roy",
+      "C. Wu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.18672",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.18672",
+    "pdf_url": "https://arxiv.org/pdf/2605.18672",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.17348",
+    "title": "Taming \"Zombie'' Agents: A Markov State-Aware Framework for Resilient Multi-Agent Evolution",
+    "url": "https://arxiv.org/abs/2605.17348",
+    "published": "2026-05-17",
+    "updated": "2026-05-17",
+    "authors": [
+      "Taolin Zhang",
+      "Pukun Zhao",
+      "Qizhou Chen",
+      "Jiuheng Wan",
+      "Chen Chen",
+      "Xiaofeng He",
+      "Chengyu Wang",
+      "Richang Hong"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-safety",
+      "memory",
+      "multi-agent",
+      "reasoning"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.17348",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.17348",
+    "pdf_url": "https://arxiv.org/pdf/2605.17348",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.17453",
+    "title": "Trust No Tool: Evaluating and Defending LLM Agents under Untrusted Tool Feedback",
+    "url": "https://arxiv.org/abs/2605.17453",
+    "published": "2026-05-17",
+    "updated": "2026-05-17",
+    "authors": [
+      "Lecheng Yan",
+      "Ruizhe Li",
+      "Xicheng Han",
+      "Wenxi Li",
+      "Binwu Wang",
+      "Longyue Wang",
+      "Chenyang Lyu",
+      "Guanhua Chen"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.17453",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.17453",
+    "pdf_url": "https://arxiv.org/pdf/2605.17453",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.23986",
+    "title": "MemForest: An Efficient Agent Memory System with Hierarchical Temporal Indexing",
+    "url": "https://arxiv.org/abs/2605.23986",
+    "published": "2026-05-16",
+    "updated": "2026-05-16",
+    "authors": [
+      "Han Chen",
+      "Zining Zhang",
+      "Wenqi Pei",
+      "Bingsheng He",
+      "Ming Wu",
+      "Jason Zeng",
+      "Michael Heinrich",
+      "Wei Wu",
+      "Hongbao Zhang"
+    ],
+    "categories": [
+      "cs.DB",
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.23986",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.23986",
+    "pdf_url": "https://arxiv.org/pdf/2605.23986",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.28850",
+    "title": "Representation Signatures and Risk-Feedback Alignment in LLM Trading Agents",
+    "url": "https://arxiv.org/abs/2605.28850",
+    "published": "2026-05-16",
+    "updated": "2026-05-30",
+    "authors": [
+      "Weicheng Xue"
+    ],
+    "categories": [
+      "cs.LG",
+      "q-fin.CP"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.28850",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.28850",
+    "pdf_url": "https://arxiv.org/pdf/2605.28850",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.14460",
+    "title": "Exploiting LLM Agent Supply Chains via Payload-less Skills",
+    "url": "https://arxiv.org/abs/2605.14460",
+    "published": "2026-05-14",
+    "updated": "2026-05-14",
+    "authors": [
+      "Xinyu Liu",
+      "Yukai Zhao",
+      "Xing Hu",
+      "Xin Xia"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.14460",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.14460",
+    "pdf_url": "https://arxiv.org/pdf/2605.14460",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.14126",
+    "title": "Reinforcement Learning for Tool-Calling Agents in Fast Healthcare Interoperability Resources (FHIR)",
+    "url": "https://arxiv.org/abs/2605.14126",
+    "published": "2026-05-13",
+    "updated": "2026-05-13",
+    "authors": [
+      "Marius S. Knorr",
+      "Robert Müller",
+      "Jan P. Bremer",
+      "Nils Schweingruber"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.14126",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.14126",
+    "pdf_url": "https://arxiv.org/pdf/2605.14126",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.11882",
+    "title": "On-Policy Self-Evolution via Failure Trajectories for Agentic Safety Alignment",
+    "url": "https://arxiv.org/abs/2605.11882",
+    "published": "2026-05-12",
+    "updated": "2026-05-12",
+    "authors": [
+      "Bo Yin",
+      "Qi Li",
+      "Xinchao Wang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.11882",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.11882",
+    "pdf_url": "https://arxiv.org/pdf/2605.11882",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.11534",
+    "title": "PRISM: : Planning and Reasoning with Intent in Simulated Embodied Environments",
+    "url": "https://arxiv.org/abs/2605.11534",
+    "published": "2026-05-12",
+    "updated": "2026-05-12",
+    "authors": [
+      "Yunn Kang Lim",
+      "Pengzhan Sun",
+      "Ziyi Bai",
+      "Xun Xu",
+      "Angela Yao",
+      "Xulei Yang",
+      "Shijie Li"
+    ],
+    "categories": [
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.11534",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.11534",
+    "pdf_url": "https://arxiv.org/pdf/2605.11534",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.11388",
+    "title": "Deep Reasoning in General Purpose Agents via Structured Meta-Cognition",
+    "url": "https://arxiv.org/abs/2605.11388",
+    "published": "2026-05-12",
+    "updated": "2026-05-12",
+    "authors": [
+      "Dean Light",
+      "Michael Theologitis",
+      "Kshitish Ghate",
+      "Shuyue Stella Li",
+      "Benjamin Newman",
+      "Chirag Shah",
+      "Aylin Caliskan",
+      "Pang Wei Koh",
+      "Dan Suciu",
+      "Yulia Tsvetkov"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.11388",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.11388",
+    "pdf_url": "https://arxiv.org/pdf/2605.11388",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.11039",
+    "title": "The Granularity Mismatch in Agent Security: Argument-Level Provenance Solves Enforcement and Isolates the LLM Reasoning Bottleneck",
+    "url": "https://arxiv.org/abs/2605.11039",
+    "published": "2026-05-11",
+    "updated": "2026-05-11",
+    "authors": [
+      "Linfeng Fan",
+      "Ziwei Li",
+      "Yuan Tian",
+      "Yichen Wang",
+      "Rongsheng Li",
+      "Xiong Wang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.11039",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.11039",
+    "pdf_url": "https://arxiv.org/pdf/2605.11039",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.10763",
+    "title": "MATRA: Modeling the Attack Surface of Agentic AI Systems -- OpenClaw Case Study",
+    "url": "https://arxiv.org/abs/2605.10763",
+    "published": "2026-05-11",
+    "updated": "2026-05-11",
+    "authors": [
+      "Tim Van hamme",
+      "Thomas Vissers",
+      "Javier Carnerero-Cano",
+      "Mario Fritz",
+      "Emil C. Lupu",
+      "Lieven Desmet",
+      "Dinil Mon Divakaran"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.10763",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.10763",
+    "pdf_url": "https://arxiv.org/pdf/2605.10763",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.10365",
+    "title": "Agent-ValueBench: A Comprehensive Benchmark for Evaluating Agent Values",
+    "url": "https://arxiv.org/abs/2605.10365",
+    "published": "2026-05-11",
+    "updated": "2026-05-11",
+    "authors": [
+      "Haonan Dong",
+      "Qiguan Feng",
+      "Kehan Jiang",
+      "Haoran Ye",
+      "Xin Zhang",
+      "Guojie Song"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.10365",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.10365",
+    "pdf_url": "https://arxiv.org/pdf/2605.10365",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.09168",
+    "title": "CIVeX: Causal Intervention Verification for Language Agents",
+    "url": "https://arxiv.org/abs/2605.09168",
+    "published": "2026-05-09",
+    "updated": "2026-05-09",
+    "authors": [
+      "Fabio Rovai"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-safety",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.09168",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.09168",
+    "pdf_url": "https://arxiv.org/pdf/2605.09168",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.08763",
+    "title": "When LLMs Team Up: A Coordinated Attack Framework for Automated Cyber Intrusions",
+    "url": "https://arxiv.org/abs/2605.08763",
+    "published": "2026-05-09",
+    "updated": "2026-05-09",
+    "authors": [
+      "Minfeng Qi",
+      "Tianqing Zhu",
+      "Zijie Xu",
+      "Congcong Zhu",
+      "Qin Wang",
+      "Wanlei Zhou"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.08763",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.08763",
+    "pdf_url": "https://arxiv.org/pdf/2605.08763",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.06078",
+    "title": "Milestone-Guided Policy Learning for Long-Horizon Language Agents",
+    "url": "https://arxiv.org/abs/2605.06078",
+    "published": "2026-05-07",
+    "updated": "2026-05-07",
+    "authors": [
+      "Zixuan Wang",
+      "Yuchen Yan",
+      "Hongxing Li",
+      "Teng Pan",
+      "Dingming Li",
+      "Ruiqing Zhang",
+      "Weiming Lu",
+      "Jun Xiao",
+      "Yueting Zhuang",
+      "Yongliang Shen"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.06078",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.06078",
+    "pdf_url": "https://arxiv.org/pdf/2605.06078",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.03505",
+    "title": "LATS-RCA: Language Agent Tree Search for Root Cause Analysis in Microservices",
+    "url": "https://arxiv.org/abs/2605.03505",
+    "published": "2026-05-05",
+    "updated": "2026-06-12",
+    "authors": [
+      "Alexander Naakka",
+      "Yuqing Wang",
+      "Mika V Mäntylä"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "reasoning"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.03505",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.03505",
+    "pdf_url": "https://arxiv.org/pdf/2605.03505",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.04107",
+    "title": "TSCG: Deterministic Tool-Schema Compilation for Agentic LLM Deployments",
+    "url": "https://arxiv.org/abs/2605.04107",
+    "published": "2026-05-04",
+    "updated": "2026-05-04",
+    "authors": [
+      "Furkan Sakizli"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2605.04107",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.04107",
+    "pdf_url": "https://arxiv.org/pdf/2605.04107",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2605.05242",
+    "title": "Beyond Semantic Similarity: Rethinking Retrieval for Agentic Search via Direct Corpus Interaction",
+    "url": "https://arxiv.org/abs/2605.05242",
+    "published": "2026-05-03",
+    "updated": "2026-05-03",
+    "authors": [
+      "Zhuofeng Li",
+      "Haoxiang Zhang",
+      "Cong Wei",
+      "Pan Lu",
+      "Ping Nie",
+      "Yi Lu",
+      "Yuyang Bai",
+      "Shangbin Feng",
+      "Hangxiao Zhu",
+      "Ming Zhong",
+      "Yuyu Zhang",
+      "Jianwen Xie",
+      "Yejin Choi",
+      "James Zou",
+      "Jiawei Han",
+      "Wenhu Chen",
+      "Jimmy Lin",
+      "Dongfu Jiang",
+      "Yu Zhang"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.05242",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.05242",
+    "pdf_url": "https://arxiv.org/pdf/2605.05242",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.00081",
+    "title": "Alignment Contracts for Agentic Security Systems",
+    "url": "https://arxiv.org/abs/2605.00081",
+    "published": "2026-04-30",
+    "updated": "2026-04-30",
+    "authors": [
+      "Isaac David",
+      "Marco Guarnieri",
+      "Arthur Gervais"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.LO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.00081",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.00081",
+    "pdf_url": "https://arxiv.org/pdf/2605.00081",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.27699",
+    "title": "Bridging Values and Behavior: A Hierarchical Framework for Proactive Embodied Agents",
+    "url": "https://arxiv.org/abs/2604.27699",
+    "published": "2026-04-30",
+    "updated": "2026-04-30",
+    "authors": [
+      "Chunhui Zhang",
+      "Yuxuan Wang",
+      "Aoyang Qin",
+      "Yi-Long Lu",
+      "Kunlun Wu",
+      "Yizhou Wang",
+      "Wei Wang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2604.27699",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.27699",
+    "pdf_url": "https://arxiv.org/pdf/2604.27699",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2604.26274",
+    "title": "Enforcing Benign Trajectories: A Behavioral Firewall for Structured-Workflow AI Agents",
+    "url": "https://arxiv.org/abs/2604.26274",
+    "published": "2026-04-29",
+    "updated": "2026-04-29",
+    "authors": [
+      "Hung Dang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.26274",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.26274",
+    "pdf_url": "https://arxiv.org/pdf/2604.26274",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.24826",
+    "title": "A Comparative Evaluation of AI Agent Security Guardrails",
+    "url": "https://arxiv.org/abs/2604.24826",
+    "published": "2026-04-27",
+    "updated": "2026-04-27",
+    "authors": [
+      "Qi Li",
+      "Jiu Li",
+      "Pingtao Wei",
+      "Jianjun Xu",
+      "Xueyi Wei",
+      "Jiwei Shi",
+      "Xuan Zhang",
+      "Yanhui Yang",
+      "Xiaodong Hui",
+      "Peng Xu",
+      "Lingquan Zhou"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.24826",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.24826",
+    "pdf_url": "https://arxiv.org/pdf/2604.24826",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.13686",
+    "title": "Benchmarking Web Agent Safety under E-commerce Deceptive Interfaces",
+    "url": "https://arxiv.org/abs/2606.13686",
+    "published": "2026-04-26",
+    "updated": "2026-04-26",
+    "authors": [
+      "Zijing Shi",
+      "Meng Fang",
+      "Ling Chen"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.CY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2606.13686",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13686",
+    "pdf_url": "https://arxiv.org/pdf/2606.13686",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.19821",
+    "title": "JTPRO: A Joint Tool-Prompt Reflective Optimization Framework for Language Agents",
+    "url": "https://arxiv.org/abs/2604.19821",
+    "published": "2026-04-20",
+    "updated": "2026-04-20",
+    "authors": [
+      "Sandip Ghoshal",
+      "Anshul Mittal",
+      "Jyotika Singh",
+      "Miguel Ballesteros",
+      "Weiyi Sun",
+      "Fang Tu",
+      "Shailender Singh",
+      "Yassine Benajiba",
+      "Fahad Shah",
+      "Sujeeth Bharadwaj",
+      "Sujith Ravi",
+      "Dan Roth"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2604.19821",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.19821",
+    "pdf_url": "https://arxiv.org/pdf/2604.19821",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2604.18718",
+    "title": "Towards Optimal Agentic Architectures for Offensive Security Tasks",
+    "url": "https://arxiv.org/abs/2604.18718",
+    "published": "2026-04-20",
+    "updated": "2026-04-20",
+    "authors": [
+      "Isaac David",
+      "Arthur Gervais"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.18718",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.18718",
+    "pdf_url": "https://arxiv.org/pdf/2604.18718",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.12986",
+    "title": "Parallax: Why AI Agents That Think Must Never Act",
+    "url": "https://arxiv.org/abs/2604.12986",
+    "published": "2026-04-14",
+    "updated": "2026-04-14",
+    "authors": [
+      "Joel Fokou"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.12986",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.12986",
+    "pdf_url": "https://arxiv.org/pdf/2604.12986",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.06972",
+    "title": "Differentiable Environment-Trajectory Co-Optimization for Safe Multi-Agent Navigation",
+    "url": "https://arxiv.org/abs/2604.06972",
+    "published": "2026-04-08",
+    "updated": "2026-04-08",
+    "authors": [
+      "Zhan Gao",
+      "Gabriele Fadini",
+      "Stelian Coros",
+      "Amanda Prorok"
+    ],
+    "categories": [
+      "cs.RO",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "embodied-agent",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.06972",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.06972",
+    "pdf_url": "https://arxiv.org/pdf/2604.06972",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.04426",
+    "title": "ShieldNet: Network-Level Guardrails against Emerging Supply-Chain Injections in Agentic Systems",
+    "url": "https://arxiv.org/abs/2604.04426",
+    "published": "2026-04-06",
+    "updated": "2026-04-06",
+    "authors": [
+      "Zhuowen Yuan",
+      "Zhaorun Chen",
+      "Zhen Xiang",
+      "Nathaniel D. Bastian",
+      "Seyyed Hadi Hashemi",
+      "Chaowei Xiao",
+      "Wenbo Guo",
+      "Bo Li"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.04426",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.04426",
+    "pdf_url": "https://arxiv.org/pdf/2604.04426",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.03098",
+    "title": "Co-Evolution of Policy and Internal Reward for Language Agents",
+    "url": "https://arxiv.org/abs/2604.03098",
+    "published": "2026-04-03",
+    "updated": "2026-04-03",
+    "authors": [
+      "Xinyu Wang",
+      "Hanwei Wu",
+      "Jingwei Song",
+      "Shuyuan Zhang",
+      "Jiayi Zhang",
+      "Fanqi Kong",
+      "Tung Sum Thomas Kwok",
+      "Xiao-Wen Chang",
+      "Yuyu Luo",
+      "Chenglin Wu",
+      "Bang Liu"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2604.03098",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.03098",
+    "pdf_url": "https://arxiv.org/pdf/2604.03098",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.15309",
+    "title": "CCTU: A Benchmark for Tool Use under Complex Constraints",
+    "url": "https://arxiv.org/abs/2603.15309",
+    "published": "2026-03-16",
+    "updated": "2026-03-16",
+    "authors": [
+      "Junjie Ye",
+      "Guoqiang Zhang",
+      "Wenjie Fu",
+      "Tao Gui",
+      "Qi Zhang",
+      "Xuanjing Huang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2603.15309",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.15309",
+    "pdf_url": "https://arxiv.org/pdf/2603.15309",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2603.11890",
+    "title": "QUARE: Quality-Aware Requirements Analysis through Multi-Agent Dialectical Negotiation",
+    "url": "https://arxiv.org/abs/2603.11890",
+    "published": "2026-03-12",
+    "updated": "2026-06-05",
+    "authors": [
+      "Haowei Cheng",
+      "Milhan Kim",
+      "Foutse Khomh",
+      "Teeradaj Racharak",
+      "Nobukazu Yoshioka",
+      "Naoyasu Ubayashi",
+      "Hironori Washizaki"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "rag"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.11890",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.11890",
+    "pdf_url": "https://arxiv.org/pdf/2603.11890",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.07557",
+    "title": "AgentRaft: Automated Detection of Data Over-Exposure in LLM Agents",
+    "url": "https://arxiv.org/abs/2603.07557",
+    "published": "2026-03-08",
+    "updated": "2026-03-08",
+    "authors": [
+      "Yixi Lin",
+      "Jiangrong Wu",
+      "Yuhong Nan",
+      "Xueqiang Wang",
+      "Xinyuan Zhang",
+      "Zibin Zheng"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2603.07557",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.07557",
+    "pdf_url": "https://arxiv.org/pdf/2603.07557",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2603.05578",
+    "title": "Tool-Genesis: A Task-Driven Tool Creation Benchmark for Self-Evolving Language Agent",
+    "url": "https://arxiv.org/abs/2603.05578",
+    "published": "2026-03-05",
+    "updated": "2026-03-05",
+    "authors": [
+      "Bowei Xia",
+      "Mengkang Hu",
+      "Shijian Wang",
+      "Jiarui Jin",
+      "Wenxiang Jiao",
+      "Yuan Lu",
+      "Kexin Li",
+      "Ping Luo"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.05578",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.05578",
+    "pdf_url": "https://arxiv.org/pdf/2603.05578",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.01712",
+    "title": "FT-Dojo: Towards Autonomous LLM Fine-Tuning with Language Agents",
+    "url": "https://arxiv.org/abs/2603.01712",
+    "published": "2026-03-02",
+    "updated": "2026-05-20",
+    "authors": [
+      "Qizheng Li",
+      "Yifei Zhang",
+      "Xiao Yang",
+      "Xu Yang",
+      "Zhuo Wang",
+      "Weiqing Liu",
+      "Jiang Bian"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.01712",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.01712",
+    "pdf_url": "https://arxiv.org/pdf/2603.01712",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2602.13379",
+    "title": "Unsafer in Many Turns: Benchmarking and Defending Multi-Turn Safety Risks in Tool-Using Agents",
+    "url": "https://arxiv.org/abs/2602.13379",
+    "published": "2026-02-13",
+    "updated": "2026-06-10",
+    "authors": [
+      "Xu Li",
+      "Simon Yu",
+      "Minzhou Pan",
+      "Yiyou Sun",
+      "Bo Li",
+      "Dawn Song",
+      "Xue Lin",
+      "Weiyan Shi"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.CL",
+      "cs.LG",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.13379",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.13379",
+    "pdf_url": "https://arxiv.org/pdf/2602.13379",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2602.11749",
+    "title": "AIR: Improving Agent Safety through Incident Response",
+    "url": "https://arxiv.org/abs/2602.11749",
+    "published": "2026-02-12",
+    "updated": "2026-06-20",
+    "authors": [
+      "Zibo Xiao",
+      "Jun Sun",
+      "Junjie Chen"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.11749",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.11749",
+    "pdf_url": "https://arxiv.org/pdf/2602.11749",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2602.18456",
+    "title": "Beyond single-channel agentic benchmarking",
+    "url": "https://arxiv.org/abs/2602.18456",
+    "published": "2026-02-05",
+    "updated": "2026-02-05",
+    "authors": [
+      "Nelu D. Radpour"
+    ],
+    "categories": [
+      "cs.CY",
+      "cs.AI",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.18456",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.18456",
+    "pdf_url": "https://arxiv.org/pdf/2602.18456",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2601.14652",
+    "title": "MAS-Orchestra: Understanding and Improving Multi-Agent Reasoning Through Holistic Orchestration and Controlled Benchmarks",
+    "url": "https://arxiv.org/abs/2601.14652",
+    "published": "2026-01-21",
+    "updated": "2026-05-21",
+    "authors": [
+      "Zixuan Ke",
+      "Yifei Ming",
+      "Austin Xu",
+      "Ryan Chin",
+      "Xuan-Phi Nguyen",
+      "Prathyusha Jwalapuram",
+      "Jiayu Wang",
+      "Semih Yavuz",
+      "Caiming Xiong",
+      "Shafiq Joty"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent",
+      "reasoning"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2601.14652",
+    "source": "arxiv",
+    "source_id": "arxiv:2601.14652",
+    "pdf_url": "https://arxiv.org/pdf/2601.14652",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2512.23647",
+    "title": "Nested Browser-Use Learning for Agentic Information Seeking",
+    "url": "https://arxiv.org/abs/2512.23647",
+    "published": "2025-12-29",
+    "updated": "2025-12-29",
+    "authors": [
+      "Baixuan Li",
+      "Jialong Wu",
+      "Wenbiao Yin",
+      "Kuan Li",
+      "Zhongwang Zhang",
+      "Huifeng Yin",
+      "Zhengwei Tao",
+      "Liwen Zhang",
+      "Pengjun Xie",
+      "Jingren Zhou",
+      "Yong Jiang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.IR",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2512.23647",
+    "source": "arxiv",
+    "source_id": "arxiv:2512.23647",
+    "pdf_url": "https://arxiv.org/pdf/2512.23647",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2512.23611",
+    "title": "Close the Loop: Synthesizing Infinite Tool-Use Data via Multi-Agent Role-Playing",
+    "url": "https://arxiv.org/abs/2512.23611",
+    "published": "2025-12-29",
+    "updated": "2025-12-29",
+    "authors": [
+      "Yuwen Li",
+      "Wei Zhang",
+      "Zelong Huang",
+      "Mason Yang",
+      "Jiajun Wu",
+      "Shawn Guo",
+      "Huahao Hu",
+      "Lingyi Sun",
+      "Jian Yang",
+      "Mingjie Tang",
+      "Byran Dai"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2512.23611",
+    "source": "arxiv",
+    "source_id": "arxiv:2512.23611",
+    "pdf_url": "https://arxiv.org/pdf/2512.23611",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2512.02605",
+    "title": "IACT: A Self-Organizing Recursive Model for General AI Agents: A Technical White Paper on the Architecture Behind kragent.ai",
+    "url": "https://arxiv.org/abs/2512.02605",
+    "published": "2025-12-02",
+    "updated": "2025-12-02",
+    "authors": [
+      "Pengju Lu"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2512.02605",
+    "source": "arxiv",
+    "source_id": "arxiv:2512.02605",
+    "pdf_url": "https://arxiv.org/pdf/2512.02605",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2510.14548",
+    "title": "LLM Agents Beyond Utility: An Open-Ended Perspective",
+    "url": "https://arxiv.org/abs/2510.14548",
+    "published": "2025-10-16",
+    "updated": "2025-10-16",
+    "authors": [
+      "Asen Nachkov",
+      "Xi Wang",
+      "Luc Van Gool"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2510.14548",
+    "source": "arxiv",
+    "source_id": "arxiv:2510.14548",
+    "pdf_url": "https://arxiv.org/pdf/2510.14548",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2509.26553",
+    "title": "Towards Reliable Benchmarking: A Contamination Free, Controllable Evaluation Framework for Multi-step LLM Function Calling",
+    "url": "https://arxiv.org/abs/2509.26553",
+    "published": "2025-09-30",
+    "updated": "2026-02-06",
+    "authors": [
+      "Seiji Maekawa",
+      "Jackson Hassell",
+      "Pouya Pezeshkpour",
+      "Tom Mitchell",
+      "Estevam Hruschka"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.PL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2509.26553",
+    "source": "arxiv",
+    "source_id": "arxiv:2509.26553",
+    "pdf_url": "https://arxiv.org/pdf/2509.26553",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2509.14477",
+    "title": "Ticket-Bench: A Kickoff for Multilingual and Regionalized Agent Evaluation",
+    "url": "https://arxiv.org/abs/2509.14477",
+    "published": "2025-09-17",
+    "updated": "2025-09-17",
+    "authors": [
+      "Thales Sales Almeida",
+      "João Guilherme Alves Santos",
+      "Thiago Laitz",
+      "Giovana Kerche Bonás"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "reasoning"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2509.14477",
+    "source": "arxiv",
+    "source_id": "arxiv:2509.14477",
+    "pdf_url": "https://arxiv.org/pdf/2509.14477",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2509.02444",
+    "title": "AppCopilot: Toward General, Accurate, Long-Horizon, and Efficient Mobile Agent",
+    "url": "https://arxiv.org/abs/2509.02444",
+    "published": "2025-09-02",
+    "updated": "2025-10-17",
+    "authors": [
+      "Jingru Fan",
+      "Yufan Dang",
+      "Jingyao Wu",
+      "Huatao Li",
+      "Runde Yang",
+      "Xiyuan Yang",
+      "Yuheng Wang",
+      "Chen Qian"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.CV",
+      "cs.HC"
+    ],
+    "topics": [
+      "computer-use",
+      "memory",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 15,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2509.02444",
+    "source": "arxiv",
+    "source_id": "arxiv:2509.02444",
+    "pdf_url": "https://arxiv.org/pdf/2509.02444",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2607.06140",
+    "title": "CurateEvo: Data-Curation Evolving for Agentic Post-Training",
+    "url": "https://arxiv.org/abs/2607.06140",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Dingzirui Wang",
+      "Xuanliang Zhang",
+      "Keyan Xu",
+      "Qingfu Zhu",
+      "Wanxiang Che"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "memory",
+      "planning",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.06140",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06140",
+    "pdf_url": "https://arxiv.org/pdf/2607.06140",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.06452",
+    "title": "From Voting to Agent Collaboration: Answer-Type-Aware LLM Pipelines for BioASQ 14b",
+    "url": "https://arxiv.org/abs/2607.06452",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Taeyun Roh",
+      "Eunha Lee",
+      "Wonjune Jang",
+      "Sohyun Chung",
+      "Junha Jung",
+      "Jaewoo Kang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.06452",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06452",
+    "pdf_url": "https://arxiv.org/pdf/2607.06452",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.05772",
+    "title": "Detecting Vulnerability-Inducing Commits via Multi-Stage Reasoning with LLM-Based Agents",
+    "url": "https://arxiv.org/abs/2607.05772",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Liyou Chen",
+      "Hailong Sun",
+      "Xiang Gao",
+      "Yue Pan"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-safety",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.05772",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05772",
+    "pdf_url": "https://arxiv.org/pdf/2607.05772",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.05378",
+    "title": "CompactionRL: Reinforcement Learning with Context Compaction for Long-Horizon Agents",
+    "url": "https://arxiv.org/abs/2607.05378",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Yujiang Li",
+      "Zhenyu Hou",
+      "Yi Jing",
+      "Jie Tang",
+      "Yuxiao Dong"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "coding-agent",
+      "planning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent",
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.05378",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05378",
+    "pdf_url": "https://arxiv.org/pdf/2607.05378",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.05132",
+    "title": "When Agents Lie: Premeditation, Persistence, and Exploitation in Repeated Games",
+    "url": "https://arxiv.org/abs/2607.05132",
+    "published": "2026-07-06",
+    "updated": "2026-07-07",
+    "authors": [
+      "Jerick Shi",
+      "Terry Jingcheng Zhang",
+      "Bernhard Schölkopf",
+      "Vincent Conitzer",
+      "Zhijing Jin"
+    ],
+    "categories": [
+      "cs.CY",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm",
+      "llm-agent",
+      "planning-agent"
+    ],
+    "arxiv_id": "2607.05132",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05132",
+    "pdf_url": "https://arxiv.org/pdf/2607.05132",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2607.04963",
+    "title": "STAPO: Selective Trajectory-Aware Policy Optimization for LLM Agent Training",
+    "url": "https://arxiv.org/abs/2607.04963",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Qiuyi Qi",
+      "Tian Liang",
+      "Mutian Bao",
+      "Jinjian Zhang",
+      "Dongnan Liu",
+      "Wei Zhou",
+      "Linjian Mo",
+      "Ming Kong",
+      "Jie Liu",
+      "Feng Zhang",
+      "Qiang Zhu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.04963",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04963",
+    "pdf_url": "https://arxiv.org/pdf/2607.04963",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.05677",
+    "title": "From Conversation to Contribution: Characterizing Coding Agent in Open-Source Software",
+    "url": "https://arxiv.org/abs/2607.05677",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Zihan Fang",
+      "Yueke Zhang",
+      "Ningzhi Tang",
+      "Collin McMillan",
+      "Toby Jia-Jun Li",
+      "Yu Huang"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "multi-agent",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.05677",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05677",
+    "pdf_url": "https://arxiv.org/pdf/2607.05677",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.05188",
+    "title": "Latent Programming Horizons in Coding Agents",
+    "url": "https://arxiv.org/abs/2607.05188",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "André Silva",
+      "Han Tu",
+      "Martin Monperrus"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "reasoning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.05188",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05188",
+    "pdf_url": "https://arxiv.org/pdf/2607.05188",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.04623",
+    "title": "Can LLMs Really Recover Microservice Failures? A Recovery-Aware Evaluation of Diagnosis-to-Action Reasoning",
+    "url": "https://arxiv.org/abs/2607.04623",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Jiaxing Qi",
+      "Zhongzhi Luan",
+      "Hongyu Zhang",
+      "Shaohan Huang",
+      "Carol Fung",
+      "Yongxin Tong",
+      "Hailong Yang",
+      "Depei Qian"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.DC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2607.04623",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04623",
+    "pdf_url": "https://arxiv.org/pdf/2607.04623",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2607.04470",
+    "title": "Regime-Conditional Stabilisation of LLM-Augmented Cooperative Multi-Agent Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2607.04470",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Faid Keddouri",
+      "Sohaib Houhou",
+      "Aissa Boulmerka",
+      "Nadir Farhi"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "math.OC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.04470",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04470",
+    "pdf_url": "https://arxiv.org/pdf/2607.04470",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.03853",
+    "title": "CogRad: A Cognitively-Inspired Multi-Agent Framework for Radiology Report Generation",
+    "url": "https://arxiv.org/abs/2607.03853",
+    "published": "2026-07-04",
+    "updated": "2026-07-04",
+    "authors": [
+      "Saif Ur Rehman Khan",
+      "Hasaan Maqsood",
+      "Sebastian Vollmer",
+      "Andreas Dengel",
+      "Muhammad Nabeel Asim"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "reasoning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.03853",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03853",
+    "pdf_url": "https://arxiv.org/pdf/2607.03853",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.02879",
+    "title": "MedCalc-Pro: Solving Complex Medical Calculations with LLM Agents",
+    "url": "https://arxiv.org/abs/2607.02879",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Siran Zhao",
+      "Ruihui Hou",
+      "Ziyue Huai",
+      "Chennuo Zhang",
+      "Tong Ruan"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.02879",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02879",
+    "pdf_url": "https://arxiv.org/pdf/2607.02879",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.03423",
+    "title": "Securing Multi-Tool AI Agent Chains With Dynamic, Real-Time Compositional Policies",
+    "url": "https://arxiv.org/abs/2607.03423",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Chris Schneider",
+      "Kriti Faujdar",
+      "Philipp Schoenegger",
+      "Ben Bariach"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.03423",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03423",
+    "pdf_url": "https://arxiv.org/pdf/2607.03423",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.03162",
+    "title": "APeB: Benchmarking Personalization Ability of Large Language Model Agents",
+    "url": "https://arxiv.org/abs/2607.03162",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Garry Yang",
+      "Zizhe Chen",
+      "Xinru Chen",
+      "Yongqiang Chen",
+      "Jianxiang Wang",
+      "Deyu Zou",
+      "Linyi Ding",
+      "Jialiang Wu",
+      "Yunzhong He",
+      "Yu Gong",
+      "James Cheng",
+      "Huaixiao Tou"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.03162",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03162",
+    "pdf_url": "https://arxiv.org/pdf/2607.03162",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.02882",
+    "title": "Diagnosis-Driven Automatic Repair for Agentic Workflow via Symbolic Inference",
+    "url": "https://arxiv.org/abs/2607.02882",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Xuyan Ma",
+      "Yawen Wang",
+      "Junjie Wang",
+      "Xiaofei Xie",
+      "Boyu Wu",
+      "Mingyang Li",
+      "Dandan Wang",
+      "Qing Wang"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.02882",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02882",
+    "pdf_url": "https://arxiv.org/pdf/2607.02882",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.03628",
+    "title": "Swarm-Driven Multi-Agent Reasoning for Smart City Security",
+    "url": "https://arxiv.org/abs/2607.03628",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Saeid Jamshidi",
+      "Kawser Wazed Nafi",
+      "Carol Fung",
+      "Foutse Khomh"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-safety",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.03628",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03628",
+    "pdf_url": "https://arxiv.org/pdf/2607.03628",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.01600",
+    "title": "BOUNDARY_SYNC: Measuring Communication-Induced Representational Coupling in Multi-Agent LLM Systems",
+    "url": "https://arxiv.org/abs/2607.01600",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Zewen Liu"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.CL"
+    ],
+    "topics": [
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.01600",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01600",
+    "pdf_url": "https://arxiv.org/pdf/2607.01600",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.02210",
+    "title": "Criticality-Based Guard Rail Validation for AI Agent Decisions in Autonomous Telecom Networks",
+    "url": "https://arxiv.org/abs/2607.02210",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Ravi Kant Sharma"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.NI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.02210",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02210",
+    "pdf_url": "https://arxiv.org/pdf/2607.02210",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.01661",
+    "title": "Diverse Evidence, Better Forecasts: Multi-Agent Deliberation Under Information Asymmetry",
+    "url": "https://arxiv.org/abs/2607.01661",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Yuante Li",
+      "Yicheng Tao",
+      "Kate Zhang",
+      "Taozhi Wang",
+      "Gefei Gu",
+      "Yaxin Zhou"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "reasoning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.01661",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01661",
+    "pdf_url": "https://arxiv.org/pdf/2607.01661",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.00692",
+    "title": "Self-GC: Self-Governing Context for Long-Horizon LLM Agents",
+    "url": "https://arxiv.org/abs/2607.00692",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Xubin Hao",
+      "Hongjin Meng",
+      "Xin Yin",
+      "Jiawei Zhu",
+      "Chenpeng Cao"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "planning-agent"
+    ],
+    "arxiv_id": "2607.00692",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00692",
+    "pdf_url": "https://arxiv.org/pdf/2607.00692",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.00297",
+    "title": "EPC: A Standardized Protocol for Measuring Evaluator Preference Dynamics in LLM Agent Systems",
+    "url": "https://arxiv.org/abs/2607.00297",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Zewen Liu"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.00297",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00297",
+    "pdf_url": "https://arxiv.org/pdf/2607.00297",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.01523",
+    "title": "Multi-Head Recurrent Memory Agents",
+    "url": "https://arxiv.org/abs/2607.01523",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Jiatong Li",
+      "Samuel Yeh",
+      "Sharon Li"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2607.01523",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01523",
+    "pdf_url": "https://arxiv.org/pdf/2607.01523",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2607.01211",
+    "title": "Are Performance-Optimization Benchmarks Reliably Measuring Coding Agents?",
+    "url": "https://arxiv.org/abs/2607.01211",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Zhi Chen",
+      "Zhensu Sun",
+      "Yuling Shi",
+      "David Lo",
+      "Lingxiao Jiang"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.01211",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01211",
+    "pdf_url": "https://arxiv.org/pdf/2607.01211",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.00918",
+    "title": "From Personas to Plot: Character-Grounded Multi-Agent Story Generation for Long-Form Narratives",
+    "url": "https://arxiv.org/abs/2607.00918",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Aayush Aluru",
+      "Chloe Ho",
+      "Muhammad Hammouri",
+      "Kerry Luo",
+      "Myra Malik",
+      "Ryan Lagasse",
+      "Arjun Bahuguna",
+      "Vasu Sharma"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.00918",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00918",
+    "pdf_url": "https://arxiv.org/pdf/2607.00918",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.00604",
+    "title": "Vehicle Routing Problem Meets Large Language Models: An Overview and Perspectives",
+    "url": "https://arxiv.org/abs/2607.00604",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Xianchao Xiu",
+      "Chong Shen",
+      "Yanjiao Zhu",
+      "Wanquan Liu"
+    ],
+    "categories": [
+      "math.OC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.00604",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00604",
+    "pdf_url": "https://arxiv.org/pdf/2607.00604",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.00972",
+    "title": "Bayesian Uncertainty Propagation for Agentic RAG Pipelines: A Proof-of-Concept Study on Multi-Hop Question Answering",
+    "url": "https://arxiv.org/abs/2607.00972",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Louis Donaldson",
+      "Connor Walker",
+      "Koorosh Aslansefat",
+      "Yiannis Papadopoulos"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2607.00972",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00972",
+    "pdf_url": "https://arxiv.org/pdf/2607.00972",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2607.00422",
+    "title": "KidnapRAG: A Black-Box Attack for Hijacking Reasoning in Agentic Retrieval-Augmented Generation Systems",
+    "url": "https://arxiv.org/abs/2607.00422",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Chanwoo Choi",
+      "Euntae Kim",
+      "Kyuho Lee",
+      "Youngsam Chun",
+      "Jinhee Jeong",
+      "Eunmi Kim",
+      "Myunggyo Oh",
+      "Junseo Jang",
+      "Buru Chang"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2607.00422",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00422",
+    "pdf_url": "https://arxiv.org/pdf/2607.00422",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.31635",
+    "title": "A Tutorial on Autonomous Fault-Tolerant Control Using Knowledge-Grounded LLM Agents",
+    "url": "https://arxiv.org/abs/2606.31635",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Javal Vyas",
+      "Milapji Singh Gill",
+      "Artan Markaj",
+      "Felix Gehlhoff",
+      "Mehmet Mercangöz"
+    ],
+    "categories": [
+      "eess.SY",
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-safety",
+      "planning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.31635",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31635",
+    "pdf_url": "https://arxiv.org/pdf/2606.31635",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.31227",
+    "title": "Securing the AI Agent: A Unified Framework for Multi-Layer Agent Red Teaming",
+    "url": "https://arxiv.org/abs/2606.31227",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Yong Yang",
+      "Xing Zheng",
+      "Huiyu Wu",
+      "Huangsheng Cheng",
+      "Xiaorong Shi",
+      "Jing Guo",
+      "Bo Yang",
+      "Yi Zhou",
+      "Xiangfan Wu",
+      "Zonghao Ying"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.31227",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31227",
+    "pdf_url": "https://arxiv.org/pdf/2606.31227",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2607.00255",
+    "title": "SLM, LLM or Agentic AI? Toward Intelligent UAV-Enabled WPT Systems in Low-Altitude Economy Networks",
+    "url": "https://arxiv.org/abs/2607.00255",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Feibo Jiang",
+      "Li Dong",
+      "Lei Mao",
+      "Kezhi Wang",
+      "Xianbin Wang",
+      "Abbas Jamalipour"
+    ],
+    "categories": [
+      "cs.IT"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.00255",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00255",
+    "pdf_url": "https://arxiv.org/pdf/2607.00255",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.31648",
+    "title": "Think in English, Answer in Korean: Efficient Adaptation of Multilingual Tool-Using Agents",
+    "url": "https://arxiv.org/abs/2606.31648",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Utsav Garg",
+      "Sungjin Hong",
+      "Jason Jung",
+      "Justin Lee",
+      "Shaan Desai",
+      "Joon Hee Kim",
+      "Anirudh Shrinivason",
+      "Edmond Wen",
+      "Susie Park"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "memory",
+      "multi-agent",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "function-calling",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.31648",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31648",
+    "pdf_url": "https://arxiv.org/pdf/2606.31648",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.02577",
+    "title": "Benchmarking the Benchmarks: A Validity Audit of Tool-Calling Evaluation",
+    "url": "https://arxiv.org/abs/2607.02577",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Vishvesh Bhat",
+      "Jay Vaghasiya",
+      "Muhammad Ahmed Mohsin",
+      "Asad Aali"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2607.02577",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02577",
+    "pdf_url": "https://arxiv.org/pdf/2607.02577",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.31314",
+    "title": "A Novel Method for Differential-Algebraic Dynamic Model Discovery in Power Systems: An LLM-Based Multi-Agent Collaborative Framework",
+    "url": "https://arxiv.org/abs/2606.31314",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Xinming Wang",
+      "Fan Tang",
+      "Yingli Wei",
+      "Yakun He",
+      "Zhe Liu",
+      "Ping Jiang",
+      "Haoyu Wu",
+      "Zihan Guo",
+      "Chao Shen"
+    ],
+    "categories": [
+      "eess.SY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.31314",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31314",
+    "pdf_url": "https://arxiv.org/pdf/2606.31314",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.30840",
+    "title": "Contrastive Reflection for Iterative Prompt Optimization",
+    "url": "https://arxiv.org/abs/2606.30840",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Derek Koh",
+      "Jinghui Mo",
+      "Benjamin H. Le",
+      "Jiening Zhan",
+      "Baofen Zheng",
+      "Kevin Bevis",
+      "Nathaniel C. Owen",
+      "Lauren Elizabeth Charney",
+      "Wenqiong Liu",
+      "Jingwei Wu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.30840",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30840",
+    "pdf_url": "https://arxiv.org/pdf/2606.30840",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.30454",
+    "title": "Collective cooperation without individual fidelity in LLM agents",
+    "url": "https://arxiv.org/abs/2606.30454",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Henrique Ferraz de Arruda",
+      "Carlos Gracia Lázaro",
+      "Alberto Aleta",
+      "Yamir Moreno"
+    ],
+    "categories": [
+      "physics.soc-ph",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.30454",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30454",
+    "pdf_url": "https://arxiv.org/pdf/2606.30454",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.30005",
+    "title": "LLM Agents Are Latent Context Managers: Eliciting Self-Managed Context via a Proprioceptive Dashboard",
+    "url": "https://arxiv.org/abs/2606.30005",
+    "published": "2026-06-29",
+    "updated": "2026-07-05",
+    "authors": [
+      "Binyan Xu",
+      "Haitao Li",
+      "Kehuan Zhang"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "memory",
+      "planning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.30005",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30005",
+    "pdf_url": "https://arxiv.org/pdf/2606.30005",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.29762",
+    "title": "Do Recommendation Algorithms Work When Users Are LLM Agents? A Case Study on Moltbook",
+    "url": "https://arxiv.org/abs/2606.29762",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Daming Li",
+      "Simeng Han",
+      "Jialu Zhang"
+    ],
+    "categories": [
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.29762",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29762",
+    "pdf_url": "https://arxiv.org/pdf/2606.29762",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.30266",
+    "title": "Towards Continual Motion-Language Agents: LoRA Variants for Incremental Motion Understanding and Generation",
+    "url": "https://arxiv.org/abs/2606.30266",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Bertram Taetz",
+      "Hugo Albuquerque Cosme da Silva",
+      "Gabriele Bleser-Taetz"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm",
+      "language-agent"
+    ],
+    "arxiv_id": "2606.30266",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30266",
+    "pdf_url": "https://arxiv.org/pdf/2606.30266",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.30185",
+    "title": "Dynamo: Dynamic Skill-Tool Evolution for Vision-Language Agents",
+    "url": "https://arxiv.org/abs/2606.30185",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Yutao Sun",
+      "Yanting Miao",
+      "Hao-Xuan Ma",
+      "Mengyu Zhou",
+      "Mingshuai Chen",
+      "Tiancheng Zhao",
+      "Dexin Wang",
+      "Lei Lv",
+      "Li Xu",
+      "Xiaoxi Jiang",
+      "Guanjun Jiang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.30185",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30185",
+    "pdf_url": "https://arxiv.org/pdf/2606.30185",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.30755",
+    "title": "Understanding and Evaluating Claw-like Agent Security Through a Computer-Systems Lens",
+    "url": "https://arxiv.org/abs/2606.30755",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Peizhi Niu",
+      "Wenjie Qu",
+      "Shangding Gu",
+      "Tianneng Shi",
+      "Yuankai Li",
+      "Ahmad Tawaha",
+      "Hend Alzahrani",
+      "Vincent Siu",
+      "Boyi Li",
+      "Chenguang Wang",
+      "Jiaheng Zhang",
+      "Basel Alomair",
+      "Ming Jin",
+      "Muhao Chen",
+      "Chi Wang",
+      "Costas Spanos",
+      "Dawn Song"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.30755",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30755",
+    "pdf_url": "https://arxiv.org/pdf/2606.30755",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.29894",
+    "title": "SABER-Math: Automated Benchmark for Information Retrieval Evaluation in Mathematics",
+    "url": "https://arxiv.org/abs/2606.29894",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Nikolay Georgiev",
+      "Maria Drencheva",
+      "Kseniia Ibragimova",
+      "Ivo Petrov",
+      "Dimitar I. Dimitrov",
+      "Martin Vechev"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI",
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.29894",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29894",
+    "pdf_url": "https://arxiv.org/pdf/2606.29894",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.29961",
+    "title": "DuoMem: Towards Capable On-Device Memory Agents via Dual-Space Distillation",
+    "url": "https://arxiv.org/abs/2606.29961",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Peyman Hosseini",
+      "Ondrej Bohdal",
+      "Ahmed Alajrami",
+      "Andrea Maracani",
+      "Ignacio Castro",
+      "Matthew Purver",
+      "Mete Ozay",
+      "Savas Ozkan",
+      "Taha Ceritli"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.29961",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29961",
+    "pdf_url": "https://arxiv.org/pdf/2606.29961",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.30119",
+    "title": "On the Internet, Nobody Knows You're an LLM Bot: Unmasking Web Agents with Multi-Layer Fingerprinting",
+    "url": "https://arxiv.org/abs/2606.30119",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Iliana Fayolle",
+      "Sihem Bouhenniche",
+      "Samuel Pélissier",
+      "Pierre Laperdrix",
+      "Clémentine Maurice",
+      "Walter Rudametkin"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "embodied-agent",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.30119",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30119",
+    "pdf_url": "https://arxiv.org/pdf/2606.30119",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.30602",
+    "title": "MESA: Prioritizing Vulnerable Communication Channels for Securing Multi-Agent Systems",
+    "url": "https://arxiv.org/abs/2606.30602",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Kunyang Li",
+      "Kyle Domico",
+      "Jonathan Gregory",
+      "Patrick McDaniel"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.30602",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30602",
+    "pdf_url": "https://arxiv.org/pdf/2606.30602",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.29932",
+    "title": "SAGA: Scene-Aware, Goal-Evolving Agents for Long-Horizon CivRealm Strategy Planning",
+    "url": "https://arxiv.org/abs/2606.29932",
+    "published": "2026-06-29",
+    "updated": "2026-07-02",
+    "authors": [
+      "Tianyu Jin",
+      "Shuo Chen",
+      "Yida Wang",
+      "Liuyu Xiang",
+      "Yingzhuo Liu",
+      "Zhiyao Jiang",
+      "Yexin Li",
+      "Zhaofeng He"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.29932",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29932",
+    "pdf_url": "https://arxiv.org/pdf/2606.29932",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.29746",
+    "title": "DEEPMED Search: An Open-Source Agentic Platform for Medical Deep Research with Introspective Verification",
+    "url": "https://arxiv.org/abs/2606.29746",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Maolin Liu",
+      "Fanyu Xu",
+      "Ruoqing Xu",
+      "Jiahang Zhang",
+      "Hao Wang",
+      "Rui Wang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.HC"
+    ],
+    "topics": [
+      "computer-use",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.29746",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29746",
+    "pdf_url": "https://arxiv.org/pdf/2606.29746",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.29225",
+    "title": "PolicyGuard: A Dialogue-Grounded Sub-Agent Verifier for Policy Adherence in LLM Agents",
+    "url": "https://arxiv.org/abs/2606.29225",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Seongjae Kang",
+      "Taehyung Yu",
+      "Sung Ju Hwang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "computer-use",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.29225",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29225",
+    "pdf_url": "https://arxiv.org/pdf/2606.29225",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.29142",
+    "title": "Agent Security Meets Regulatory Reality -- A Practitioner Systematization of Autonomous-Agent Threats and Controls in Regulated Financial Systems",
+    "url": "https://arxiv.org/abs/2606.29142",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Krishna Mohan",
+      "Guda Nagavenkata Srinivasa"
+    ],
+    "categories": [
+      "cs.CY",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "autonomous-agent-llm",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.29142",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29142",
+    "pdf_url": "https://arxiv.org/pdf/2606.29142",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.28733",
+    "title": "Agentic Abstention: Do Agents Know When to Stop Instead of Act?",
+    "url": "https://arxiv.org/abs/2606.28733",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Han Luo",
+      "Bingbing Wen",
+      "Lucy Lu Wang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.28733",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28733",
+    "pdf_url": "https://arxiv.org/pdf/2606.28733",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.28679",
+    "title": "Capability Gates Are Not Authorization: Confused-Deputy Failures in LLM Agent Frameworks",
+    "url": "https://arxiv.org/abs/2606.28679",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "David Mellafe Zuvic"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.28679",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28679",
+    "pdf_url": "https://arxiv.org/pdf/2606.28679",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.29026",
+    "title": "Preventing Error Propagation in Multi-Agent AI through Runtime Monitoring",
+    "url": "https://arxiv.org/abs/2606.29026",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Shahnewaz Karim Sakib",
+      "Anindya Bijoy Das"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.ET"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "reasoning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.29026",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29026",
+    "pdf_url": "https://arxiv.org/pdf/2606.29026",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.28739",
+    "title": "Agent Safety Is Action Alignment",
+    "url": "https://arxiv.org/abs/2606.28739",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Shawn Li",
+      "Yue Zhao"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2606.28739",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28739",
+    "pdf_url": "https://arxiv.org/pdf/2606.28739",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.27632",
+    "title": "Yuvion LLM: An Adversarially-Aware Large Language Model for Content And AI Safety",
+    "url": "https://arxiv.org/abs/2606.27632",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Ting Ma",
+      "Xiufeng Huang",
+      "Benlei Cui",
+      "Xiaowen Xu",
+      "Shikai Qiu",
+      "Ruijie Jian",
+      "Hongxing Li",
+      "Guanghui Wang",
+      "Longtao Huang",
+      "Haiwen Hong",
+      "Haolei Xu",
+      "Wenjing Jiang",
+      "Ziwen Xu",
+      "Zhaoyu Fan",
+      "Shaoxuan He",
+      "Chuxi Xiao",
+      "Yujian Li",
+      "Xinyue Chen",
+      "Chunyang Chai",
+      "Wenxuan Liu",
+      "Ziheng Wang",
+      "Dongjie Zhang",
+      "Yangfan Zhou",
+      "Libin Dong",
+      "Yupeng Cao",
+      "Xiaoqian Xia",
+      "Jing Wang",
+      "Zhe Jiang",
+      "Zhenan Ye",
+      "Guang Yang",
+      "Bin Liu",
+      "Wei Peng",
+      "Ziqiang Zhu",
+      "Meihui Lian",
+      "Kaiwen Lv Kacuila",
+      "Haidong Ding",
+      "Bingyu Zhu",
+      "Yan Wang",
+      "Hai Zhao",
+      "Xuan Jin",
+      "Wei Zhao",
+      "Pengfei Sun",
+      "Wei Wang",
+      "Huiming Zhang",
+      "Bin Li",
+      "Hui Xue"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.27632",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27632",
+    "pdf_url": "https://arxiv.org/pdf/2606.27632",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.26806",
+    "title": "Memory Depth, Not Memory Access: Selective Parametric Consolidation for Long-Running Language Agents",
+    "url": "https://arxiv.org/abs/2606.26806",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Haoliang Han"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.26806",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26806",
+    "pdf_url": "https://arxiv.org/pdf/2606.26806",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.27154",
+    "title": "OpenRCA 2.0: From Outcome Labels to Causal Process Supervision",
+    "url": "https://arxiv.org/abs/2606.27154",
+    "published": "2026-06-25",
+    "updated": "2026-06-30",
+    "authors": [
+      "Aoyang Fang",
+      "Yifan Yang",
+      "Jin'ao Shang",
+      "Qisheng Lu",
+      "Junjielung Xu",
+      "Rui Wang",
+      "Songhan Zhang",
+      "Yuzhong Zhang",
+      "Boxi Yu",
+      "Pinjia He"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.27154",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27154",
+    "pdf_url": "https://arxiv.org/pdf/2606.27154",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.27492",
+    "title": "QueenBee Planner: Skill-Evolving Communication Topologies for Token-Efficient LLM Multi-Agent Systems",
+    "url": "https://arxiv.org/abs/2606.27492",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Congjia Tian",
+      "Yuhang Yao",
+      "Jiaming Cui"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.27492",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27492",
+    "pdf_url": "https://arxiv.org/pdf/2606.27492",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.26758",
+    "title": "EGG: An Expert-Guided Agent Framework for Kernel Generation",
+    "url": "https://arxiv.org/abs/2606.26758",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Yaochen Han",
+      "Ke Fan",
+      "Hongxu Jiang",
+      "Wanqi Xu",
+      "Weiyu Xie",
+      "Runhua Zhang",
+      "Chenhui Zhu",
+      "Yixiang Zhang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "computer-use",
+      "memory",
+      "multi-agent",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.26758",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26758",
+    "pdf_url": "https://arxiv.org/pdf/2606.26758",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.26205",
+    "title": "Knowledge-augmented Agentic AI for Mental Health Medication Information Seeking",
+    "url": "https://arxiv.org/abs/2606.26205",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Huizi Yu",
+      "Jian Liu",
+      "Wenkong Wang",
+      "Lingyao Li",
+      "Jiayan Zhou",
+      "Zhaoqian Xue",
+      "Xiang Li",
+      "Xinxin Lin",
+      "Zhiying Liang",
+      "Zhuoru Wu",
+      "Siyuan Ma",
+      "Xin Ma",
+      "Lizhou Fan"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.26205",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26205",
+    "pdf_url": "https://arxiv.org/pdf/2606.26205",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.25899",
+    "title": "Manipulation Is Task-Dependent: A Multi-Axis, Multi-Environment Evaluation of Frontier LLMs",
+    "url": "https://arxiv.org/abs/2606.25899",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Adeeb Zaman",
+      "Erik Nordby",
+      "Fred Heiding"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.25899",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25899",
+    "pdf_url": "https://arxiv.org/pdf/2606.25899",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.25484",
+    "title": "From Causal Discovery to Implementation: An Agentic AI Framework for E-Scooter Mobility Hub Planning Across 29 German Cities",
+    "url": "https://arxiv.org/abs/2606.25484",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Meng Jin",
+      "Melanie Handrich",
+      "Simone Martinenz",
+      "Nicholas Hoeser",
+      "Ziyue Li"
+    ],
+    "categories": [
+      "cs.CY",
+      "econ.GN",
+      "stat.AP"
+    ],
+    "topics": [
+      "coding-agent",
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.25484",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25484",
+    "pdf_url": "https://arxiv.org/pdf/2606.25484",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.26300",
+    "title": "The Verification Horizon: No Silver Bullet for Coding Agent Rewards",
+    "url": "https://arxiv.org/abs/2606.26300",
+    "published": "2026-06-24",
+    "updated": "2026-06-29",
+    "authors": [
+      "Binghai Wang",
+      "Chenlong Zhang",
+      "Dayiheng Liu",
+      "Jiajun Zhang",
+      "Jiawei Chen",
+      "Mingze Li",
+      "Mouxiang Chen",
+      "Rongyao Fang",
+      "Siyuan Zhang",
+      "Xuwu Wang",
+      "Yuheng Jing",
+      "Zeyao Ma",
+      "Zeyu Cui"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.26300",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26300",
+    "pdf_url": "https://arxiv.org/pdf/2606.26300",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.25705",
+    "title": "GUI agent: Guided Exploration of User-Sensitive Screens",
+    "url": "https://arxiv.org/abs/2606.25705",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Aradhana Nayak",
+      "Mussadiq Nazeer",
+      "Wang Peng",
+      "Feng Liu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.25705",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25705",
+    "pdf_url": "https://arxiv.org/pdf/2606.25705",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.25656",
+    "title": "Is GraphRAG Needed? From Basic RAG to Graph-/Agentic Solutions with Context Optimization",
+    "url": "https://arxiv.org/abs/2606.25656",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Long Chen",
+      "Ryan Razkenari",
+      "Yuxuan Zhou",
+      "Yuan Tian",
+      "Rahul Ghosh",
+      "Venkatesh Pappakrishnan",
+      "Disha Ahuja",
+      "Vidya Sagar Ravipati"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.25656",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25656",
+    "pdf_url": "https://arxiv.org/pdf/2606.25656",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.25189",
+    "title": "ActPlane: Programmable OS-Level Policy Enforcement for Agent Harnesses",
+    "url": "https://arxiv.org/abs/2606.25189",
+    "published": "2026-06-23",
+    "updated": "2026-06-30",
+    "authors": [
+      "Yusheng Zheng",
+      "Tianyuan Wu",
+      "Quanzhi Fu",
+      "Tong Yu",
+      "Wenan Mao",
+      "Tao Ma",
+      "Dan Williams",
+      "Wei Wang",
+      "Andi Quinn"
+    ],
+    "categories": [
+      "cs.OS"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "planning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.25189",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25189",
+    "pdf_url": "https://arxiv.org/pdf/2606.25189",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.24402",
+    "title": "Poisoned Playbooks: Demystifying Knowledge Poisoning Effects on AI Security Agents",
+    "url": "https://arxiv.org/abs/2606.24402",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Juho Park",
+      "Hyunmin Choi",
+      "Kevin Nam"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.24402",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24402",
+    "pdf_url": "https://arxiv.org/pdf/2606.24402",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.24235",
+    "title": "SP-Mind: An Autonomous Reasoning Agent for Spatial Proteomics Analysis",
+    "url": "https://arxiv.org/abs/2606.24235",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Yucheng Yuan",
+      "Yuanfeng Ji",
+      "Zhongxiao Li",
+      "Ruijiang Li"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.24235",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24235",
+    "pdf_url": "https://arxiv.org/pdf/2606.24235",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.25206",
+    "title": "RAVEN: Long-Horizon Reasoning & Navigation with a Visuo-Spatio-Temporal Memory",
+    "url": "https://arxiv.org/abs/2606.25206",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Yixun Hu",
+      "Zhicheng Zheng",
+      "Lihan Zha",
+      "Chunwei Xing",
+      "Rajdeep Singh",
+      "Omar Hossain",
+      "Antonio Loquercio",
+      "Dhruv Shah"
+    ],
+    "categories": [
+      "cs.RO",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.25206",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25206",
+    "pdf_url": "https://arxiv.org/pdf/2606.25206",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.24515",
+    "title": "Reinforcement Learning for Computer-Use Agents with Autonomous Evaluation",
+    "url": "https://arxiv.org/abs/2606.24515",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Marta Sumyk",
+      "Oleksandr Kosovan"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.24515",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24515",
+    "pdf_url": "https://arxiv.org/pdf/2606.24515",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.24694",
+    "title": "SupplyNet: Supporting Visual Exploratory Learning in Supply Chain via Contextual Multi-Agent Simulation",
+    "url": "https://arxiv.org/abs/2606.24694",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Yanjia Li",
+      "Kelcy Kexin Han",
+      "Tianrui Hu",
+      "Yi-Fan Cao",
+      "Huamin Qu",
+      "Sicheng Song"
+    ],
+    "categories": [
+      "cs.HC"
+    ],
+    "topics": [
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.24694",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24694",
+    "pdf_url": "https://arxiv.org/pdf/2606.24694",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.24976",
+    "title": "Diagnosing and Mitigating Compounding Failures in Agentic Persuasion via Taxonomic Strategy Retrieval",
+    "url": "https://arxiv.org/abs/2606.24976",
+    "published": "2026-06-23",
+    "updated": "2026-07-04",
+    "authors": [
+      "Sana Ayromlou",
+      "Purvi Sehgal",
+      "Pradyumna Narayana"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.24976",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24976",
+    "pdf_url": "https://arxiv.org/pdf/2606.24976",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.22737",
+    "title": "GroundEval: A Deterministic Replacement for LLM-as-Judge in Stateful Agent Evaluation",
+    "url": "https://arxiv.org/abs/2606.22737",
+    "published": "2026-06-22",
+    "updated": "2026-07-02",
+    "authors": [
+      "Jeffrey Flynt"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.22737",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22737",
+    "pdf_url": "https://arxiv.org/pdf/2606.22737",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.23283",
+    "title": "Towards Root Memories: Benchmarking and Enhancing Implicit Logical Memory Retrieval for Personalized LLMs",
+    "url": "https://arxiv.org/abs/2606.23283",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Hongxun Ding",
+      "Xiang Yu",
+      "Chengbing Wang",
+      "Jianfei Xiao",
+      "Keqin Bao",
+      "Wenjie Wang",
+      "Xiangnan He"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.23283",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23283",
+    "pdf_url": "https://arxiv.org/pdf/2606.23283",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.23195",
+    "title": "Memory Contagion: Cross-Temporal Propagation of Evaluator Bias via Agent Memory",
+    "url": "https://arxiv.org/abs/2606.23195",
+    "published": "2026-06-22",
+    "updated": "2026-06-24",
+    "authors": [
+      "Zewen Liu"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.23195",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23195",
+    "pdf_url": "https://arxiv.org/pdf/2606.23195",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.22864",
+    "title": "When AUC 0.998 Is Not Enough: A Candidate Evaluation Protocol for Hidden-State Probes of Indirect Prompt Injection in Multimodal Computer-Use Agents",
+    "url": "https://arxiv.org/abs/2606.22864",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Yanhang Li",
+      "Zhichao Fan",
+      "Zexin Zhuang"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.22864",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22864",
+    "pdf_url": "https://arxiv.org/pdf/2606.22864",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.22495",
+    "title": "Grounded Scaling: Why Agentic AI Needs Deterministic Environments",
+    "url": "https://arxiv.org/abs/2606.22495",
+    "published": "2026-06-21",
+    "updated": "2026-06-21",
+    "authors": [
+      "Liang Ding",
+      "Xintong Wang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "embodied-agent",
+      "multi-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.22495",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22495",
+    "pdf_url": "https://arxiv.org/pdf/2606.22495",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.22484",
+    "title": "Governed AI-Assisted Engineering: Graduated Human Oversight for Agentic Code Generation in Regulated Domains",
+    "url": "https://arxiv.org/abs/2606.22484",
+    "published": "2026-06-21",
+    "updated": "2026-07-04",
+    "authors": [
+      "Richard Kang"
+    ],
+    "categories": [
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.22484",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22484",
+    "pdf_url": "https://arxiv.org/pdf/2606.22484",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.22030",
+    "title": "Nous: A Predictive World Model for Long-Term Agent Memory",
+    "url": "https://arxiv.org/abs/2606.22030",
+    "published": "2026-06-20",
+    "updated": "2026-06-20",
+    "authors": [
+      "Pranav Singh"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.IR",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.22030",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22030",
+    "pdf_url": "https://arxiv.org/pdf/2606.22030",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.22151",
+    "title": "Novelty-Aware Agentic Retrieval: Comparing Research Contributions Through Structured Multi-Step Reasoning",
+    "url": "https://arxiv.org/abs/2606.22151",
+    "published": "2026-06-20",
+    "updated": "2026-06-20",
+    "authors": [
+      "Shou-Tzu Han"
+    ],
+    "categories": [
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.22151",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22151",
+    "pdf_url": "https://arxiv.org/pdf/2606.22151",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.21842",
+    "title": "Agent-Assisted Side-Channel Attacks on Non-Prefix KV Cache in RAG",
+    "url": "https://arxiv.org/abs/2606.21842",
+    "published": "2026-06-20",
+    "updated": "2026-06-20",
+    "authors": [
+      "He Sun",
+      "Shinan Liu",
+      "Siyuan Ma",
+      "Junhao Li",
+      "Mingjun Xiao",
+      "Wenhao Jiang"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.21842",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21842",
+    "pdf_url": "https://arxiv.org/pdf/2606.21842",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.21409",
+    "title": "Don't Blindly Trust It: How Unreliable Feedback Breaks Tool-Using LLM Agents",
+    "url": "https://arxiv.org/abs/2606.21409",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Chubin Zhang",
+      "Zhenglin Wan",
+      "Xingrui Yu",
+      "Pengfei Zhou",
+      "Wangbo Zhao",
+      "Jingxuan Wu",
+      "Yaxin Zhou",
+      "Ivor Tsang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.21409",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21409",
+    "pdf_url": "https://arxiv.org/pdf/2606.21409",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.21553",
+    "title": "Dissecting Agentic RAG: A Component Ablation for Multi-Hop QA with a Local 7B Model",
+    "url": "https://arxiv.org/abs/2606.21553",
+    "published": "2026-06-19",
+    "updated": "2026-06-19",
+    "authors": [
+      "Sheroz Shaikh"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.21553",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21553",
+    "pdf_url": "https://arxiv.org/pdf/2606.21553",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.20470",
+    "title": "Analyzing Defensive Misdirection Against Model-Guided Automated Attacks on Agentic AI Systems",
+    "url": "https://arxiv.org/abs/2606.20470",
+    "published": "2026-06-18",
+    "updated": "2026-06-26",
+    "authors": [
+      "Reza Soosahabi",
+      "Vivek Namsani"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.20470",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20470",
+    "pdf_url": "https://arxiv.org/pdf/2606.20470",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.19812",
+    "title": "Human-on-the-Loop Orchestration for AI-Assisted Legal Discovery",
+    "url": "https://arxiv.org/abs/2606.19812",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Anushree Sinha",
+      "Srivaths Ranganathan",
+      "Abhishek Dharmaratnakar",
+      "Debanshu Das"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag",
+      "reasoning",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.19812",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19812",
+    "pdf_url": "https://arxiv.org/pdf/2606.19812",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.20515",
+    "title": "S-Agent: Spatial Tool-Use Elicits Reasoning for Spatial Intelligence",
+    "url": "https://arxiv.org/abs/2606.20515",
+    "published": "2026-06-18",
+    "updated": "2026-06-28",
+    "authors": [
+      "Yalun Dai",
+      "Hao Li",
+      "Shulin Tian",
+      "Runmao Yao",
+      "Yuhao Dong",
+      "Fangzhou Hong",
+      "Zhaoxi Chen",
+      "Fangfu Liu",
+      "Baoliang Tian",
+      "Dingwen Zhang",
+      "Tao Wang",
+      "Kim-Hui Yap",
+      "Ziwei Liu"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.20515",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20515",
+    "pdf_url": "https://arxiv.org/pdf/2606.20515",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.20023",
+    "title": "When Lower Privileges Suffice: Investigating Over-Privileged Tool Selection in LLM Agents",
+    "url": "https://arxiv.org/abs/2606.20023",
+    "published": "2026-06-18",
+    "updated": "2026-07-07",
+    "authors": [
+      "Kaiyue Yang",
+      "Yuyan Bu",
+      "Jingwei Yi",
+      "Yuchi Wang",
+      "Biyu Zhou",
+      "Juntao Dai",
+      "Songlin Hu",
+      "Yaodong Yang"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.20023",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20023",
+    "pdf_url": "https://arxiv.org/pdf/2606.20023",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.20785",
+    "title": "Fara-1.5: Scalable Learning Environments for Computer Use Agents",
+    "url": "https://arxiv.org/abs/2606.20785",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Ahmed Awadallah",
+      "Sahil Gupta",
+      "Yash Lara",
+      "Yadong Lu",
+      "Hussein Mozannar",
+      "Akshay Nambi",
+      "Zach Nussbaum",
+      "Yash Pandya",
+      "Aravind Rajeswaran",
+      "Corby Rosset",
+      "Alexey Taymanov",
+      "Luiz do Valle",
+      "Vibhav Vineet",
+      "Spencer Whitehead",
+      "Andrew Zhao"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.20785",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20785",
+    "pdf_url": "https://arxiv.org/pdf/2606.20785",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.19930",
+    "title": "MobileForge: Annotation-Free Adaptation for Mobile GUI Agents with Hierarchical Feedback-Guided Policy Optimization",
+    "url": "https://arxiv.org/abs/2606.19930",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Guangyi Liu",
+      "Pengxiang Zhao",
+      "Gao Wu",
+      "Yiwen Yin",
+      "Mading Li",
+      "Liang Liu",
+      "Congxiao Liu",
+      "Zhang Qi",
+      "Mengyan Wang",
+      "Liang Guo",
+      "Yong Liu"
+    ],
+    "categories": [
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.19930",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19930",
+    "pdf_url": "https://arxiv.org/pdf/2606.19930",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.18671",
+    "title": "HANSEL: Extracting Breadcrumbs from Web Agent Trajectories for Interactive Verification",
+    "url": "https://arxiv.org/abs/2606.18671",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Yujin Zhang",
+      "Daye Nam"
+    ],
+    "categories": [
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "embodied-agent",
+      "planning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.18671",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18671",
+    "pdf_url": "https://arxiv.org/pdf/2606.18671",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.19063",
+    "title": "PYPILINE: Malicious PyPI Package Detection via Suspicious API Knowledge and Agent Workflow",
+    "url": "https://arxiv.org/abs/2606.19063",
+    "published": "2026-06-17",
+    "updated": "2026-06-26",
+    "authors": [
+      "Siyuan Pang",
+      "Yepeng Yao",
+      "Zhengwei Jiang",
+      "Zijing Fan",
+      "Haozhe Li",
+      "Baoxu Liu"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.19063",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19063",
+    "pdf_url": "https://arxiv.org/pdf/2606.19063",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.19409",
+    "title": "OpenRath: Session-Centered Runtime State for Agent Systems",
+    "url": "https://arxiv.org/abs/2606.19409",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Fukang Wen",
+      "Zhijie Wang",
+      "Ruilin Xu"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.PL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.19409",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19409",
+    "pdf_url": "https://arxiv.org/pdf/2606.19409",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.19613",
+    "title": "StaminaBench: Stress-Testing Coding Agents over 100 Interaction Turns",
+    "url": "https://arxiv.org/abs/2606.19613",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Vlad Sobal",
+      "Shuo Yang",
+      "Yuting Zhang",
+      "Wei Xia",
+      "Stefano Soatto"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.19613",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19613",
+    "pdf_url": "https://arxiv.org/pdf/2606.19613",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.20717",
+    "title": "MIRAGE: Stealthy Visual Prompt Injection for Vulnerability Detection in Web Agents",
+    "url": "https://arxiv.org/abs/2606.20717",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Xuelong Dai",
+      "Jianyu Ma",
+      "Boyang Ma",
+      "Biwei Yan",
+      "Yijun Yang",
+      "Yue Zhang"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.AI",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.20717",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20717",
+    "pdf_url": "https://arxiv.org/pdf/2606.20717",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.16111",
+    "title": "Towards Pareto-Optimal Tool-Integrated Agents with Pareto Ranking Policy Optimization",
+    "url": "https://arxiv.org/abs/2606.16111",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Junyi Li",
+      "Xiaowei Qian",
+      "Yingyi Zhang",
+      "Wenlin Zhang",
+      "Guojing Li",
+      "Sheng Zhang",
+      "Xiao Han",
+      "Yichao Wang",
+      "Xiangyu Zhao"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.16111",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16111",
+    "pdf_url": "https://arxiv.org/pdf/2606.16111",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.16748",
+    "title": "MyPCBench: A Benchmark for Personally Intelligent Computer-Use Agents",
+    "url": "https://arxiv.org/abs/2606.16748",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Lawrence Keunho Jang",
+      "Andrew Keunwoo Jang",
+      "Jing Yu Koh",
+      "Ruslan Salakhutdinov"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.16748",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16748",
+    "pdf_url": "https://arxiv.org/pdf/2606.16748",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.15591",
+    "title": "Agentic Retrieval and Reinforcement Learned Equation Chains: A Controlled Generation Framework for Complex and Novel Physics Word Problems",
+    "url": "https://arxiv.org/abs/2606.15591",
+    "published": "2026-06-14",
+    "updated": "2026-06-14",
+    "authors": [
+      "Tirthankar Mittra"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.15591",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15591",
+    "pdf_url": "https://arxiv.org/pdf/2606.15591",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.15152",
+    "title": "Can Agents Read the Room? Benchmarking Visual Social Intelligence in Multimodal Simulation",
+    "url": "https://arxiv.org/abs/2606.15152",
+    "published": "2026-06-13",
+    "updated": "2026-06-13",
+    "authors": [
+      "Shijun Wan",
+      "Xuehai Wu",
+      "Jiwen Zhang",
+      "Siyuan Wang",
+      "Zhongyu Wei"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.15152",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15152",
+    "pdf_url": "https://arxiv.org/pdf/2606.15152",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.15242",
+    "title": "Benign in Isolation, Harmful in Composition: Security Risks in Agent Skill Ecosystems",
+    "url": "https://arxiv.org/abs/2606.15242",
+    "published": "2026-06-13",
+    "updated": "2026-06-13",
+    "authors": [
+      "Yi Xie",
+      "Jiawei Du",
+      "Yu Cheng",
+      "Jiuan Zhou",
+      "Zhaoxia Yin"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.15242",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15242",
+    "pdf_url": "https://arxiv.org/pdf/2606.15242",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.14502",
+    "title": "From Chatbot to Digital Colleague: The Paradigm Shift Toward Persistent Autonomous AI",
+    "url": "https://arxiv.org/abs/2606.14502",
+    "published": "2026-06-12",
+    "updated": "2026-06-12",
+    "authors": [
+      "Yongheng Zhang",
+      "Ziang Liu",
+      "Jiaxuan Zhu",
+      "Shuai Wang",
+      "Xiangqi Chen",
+      "Haojing Huang",
+      "Jiayi Kuang",
+      "Siyu Chen",
+      "Ao Shen",
+      "Hao Wu",
+      "Qiufeng Wang",
+      "Qian-Wen Zhang",
+      "Junnan Dong",
+      "Wenhao Jiang",
+      "Ying Shen",
+      "Hai-Tao Zheng",
+      "Yinghui Li",
+      "Di Yin",
+      "Xing Sun",
+      "Philip S. Yu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.14502",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.14502",
+    "pdf_url": "https://arxiv.org/pdf/2606.14502",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.15017",
+    "title": "Are Online Skill and Memory Modules Always Worth Their Tokens? A Budget-Constrained Study of Web Agents",
+    "url": "https://arxiv.org/abs/2606.15017",
+    "published": "2026-06-12",
+    "updated": "2026-06-12",
+    "authors": [
+      "Sina Hajimiri",
+      "Masih Aminbeidokhti",
+      "Jose Dolz",
+      "Ismail Ben Ayed",
+      "Issam H. Laradji",
+      "Spandana Gella",
+      "Nicolas Gontier"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.15017",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15017",
+    "pdf_url": "https://arxiv.org/pdf/2606.15017",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.14574",
+    "title": "SIMMER: Benchmarking Latent Failures in LLM Executable Planning with a World Model",
+    "url": "https://arxiv.org/abs/2606.14574",
+    "published": "2026-06-12",
+    "updated": "2026-06-12",
+    "authors": [
+      "Xiaoxin Lu",
+      "Ranran Haoran Zhang",
+      "Rui Zhang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.14574",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.14574",
+    "pdf_url": "https://arxiv.org/pdf/2606.14574",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.13317",
+    "title": "SkillCAT: Contrastive Assessment and Topology-Aware Skill Self-Evolution for LLM Agents",
+    "url": "https://arxiv.org/abs/2606.13317",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Kunfeng Chen",
+      "Qihuang Zhong",
+      "Juhua Liu",
+      "Bo Du"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.13317",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13317",
+    "pdf_url": "https://arxiv.org/pdf/2606.13317",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.14805",
+    "title": "Knowledge-Based Zero-Replay Debugging of Multi-Agent LLM Traces",
+    "url": "https://arxiv.org/abs/2606.14805",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Dong Ho Kang",
+      "Hyeonjeong Cha",
+      "Daein Weon"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "memory",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.14805",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.14805",
+    "pdf_url": "https://arxiv.org/pdf/2606.14805",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.13602",
+    "title": "EpiBench: Verifiable Evaluation of AI Agents on Epigenomics Analysis",
+    "url": "https://arxiv.org/abs/2606.13602",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Harihara Muralidharan",
+      "Reema Baskar",
+      "Soo Hee Lee",
+      "Tim Proctor",
+      "Kenny Workman"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.13602",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13602",
+    "pdf_url": "https://arxiv.org/pdf/2606.13602",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.13192",
+    "title": "Reasoning for Mobile User Experience with Multimodal LLMs: Task, Benchmark, and Approach",
+    "url": "https://arxiv.org/abs/2606.13192",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Ruichao Mao",
+      "Zhou Fang",
+      "Teng Guo",
+      "Hao Yang",
+      "Yaping Li",
+      "Shaohua Peng",
+      "Maji Huang",
+      "Xiaoyu Lin",
+      "Shuoyang Liu",
+      "Xuepeng Li",
+      "Yuyu Zhang",
+      "Hai Rao"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.13192",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13192",
+    "pdf_url": "https://arxiv.org/pdf/2606.13192",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.12674",
+    "title": "Evoflux: Inference-Time Evolution of Executable Tool Workflows for Compact Agents",
+    "url": "https://arxiv.org/abs/2606.12674",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Kushal Raj Bhandari",
+      "Ling Yue",
+      "Ching-Yun Ko",
+      "Dhaval Patel",
+      "Shaowu Pan",
+      "Pin-Yu Chen",
+      "Jianxi Gao"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.12674",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12674",
+    "pdf_url": "https://arxiv.org/pdf/2606.12674",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2606.12384",
+    "title": "APPO: Agentic Procedural Policy Optimization",
+    "url": "https://arxiv.org/abs/2606.12384",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Xucong Wang",
+      "Ziyu Ma",
+      "Yong Wang",
+      "Yuxiang Ji",
+      "Shidong Yang",
+      "Guanhua Chen",
+      "Pengkun Wang",
+      "Xiangxiang Chu"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.12384",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12384",
+    "pdf_url": "https://arxiv.org/pdf/2606.12384",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.12563",
+    "title": "Arbor: Tree Search as a Cognition Layer for Autonomous Agents",
+    "url": "https://arxiv.org/abs/2606.12563",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Neha Prakriya",
+      "Chaojun Hou",
+      "Zheng Gong",
+      "Huasha Zhao",
+      "Xi Zhao",
+      "Mou Li",
+      "Zhenyu Gu",
+      "Emad Barsoum"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.12563",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12563",
+    "pdf_url": "https://arxiv.org/pdf/2606.12563",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.11079",
+    "title": "VISTA: A Versatile Interactive User Simulation Toolkit for Agent Evaluation",
+    "url": "https://arxiv.org/abs/2606.11079",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Yunan Lu",
+      "Ryan Shea",
+      "Yusen Zhang",
+      "Zhou Yu"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.11079",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.11079",
+    "pdf_url": "https://arxiv.org/pdf/2606.11079",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.10742",
+    "title": "MemVenom: Triggered Poisoning of Multimodal Memories in Web Agents",
+    "url": "https://arxiv.org/abs/2606.10742",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Yv Zhang",
+      "Hao Sun",
+      "Hao Fang",
+      "Kuofeng Gao",
+      "Fan Mo",
+      "Bin Chen",
+      "Shu-Tao Xia",
+      "Yaowei Wang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.10742",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10742",
+    "pdf_url": "https://arxiv.org/pdf/2606.10742",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.09774",
+    "title": "Auto-Configuring Scientific Simulators with Lightweight Coding-Agent Adapters",
+    "url": "https://arxiv.org/abs/2606.09774",
+    "published": "2026-06-08",
+    "updated": "2026-06-25",
+    "authors": [
+      "Matthew Ho",
+      "Brian Liu",
+      "Jixuan Chen",
+      "Audrey Wang",
+      "Lianhui Qin"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "memory",
+      "rag",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.09774",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09774",
+    "pdf_url": "https://arxiv.org/pdf/2606.09774",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.09198",
+    "title": "MASS: Deep Research for Social Sciences with Memory-Augmented Social Simulation",
+    "url": "https://arxiv.org/abs/2606.09198",
+    "published": "2026-06-08",
+    "updated": "2026-06-08",
+    "authors": [
+      "Yongrui Liu",
+      "Deyi Xiong"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "rag",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.09198",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09198",
+    "pdf_url": "https://arxiv.org/pdf/2606.09198",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.08790",
+    "title": "RAILS: Verification-Native Clearing For Agentic Commerce",
+    "url": "https://arxiv.org/abs/2606.08790",
+    "published": "2026-06-07",
+    "updated": "2026-06-07",
+    "authors": [
+      "Adrian de Valois-Franklin",
+      "Alex Bogdan"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CR",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.08790",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.08790",
+    "pdf_url": "https://arxiv.org/pdf/2606.08790",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.08625",
+    "title": "From Holistic Evaluation to Structured Criteria: Rubrics Across the Evolving LLM Landscape",
+    "url": "https://arxiv.org/abs/2606.08625",
+    "published": "2026-06-07",
+    "updated": "2026-07-01",
+    "authors": [
+      "Hao Chen",
+      "Ziyu Han",
+      "Yukun Yan",
+      "Qingfu Zhu",
+      "Maosong Sun",
+      "Wanxiang Che"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.08625",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.08625",
+    "pdf_url": "https://arxiv.org/pdf/2606.08625",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.07379",
+    "title": "Do Coding Agents Deceive Us? Detecting and Preventing Cheating via Capped Evaluation with Randomized Tests",
+    "url": "https://arxiv.org/abs/2606.07379",
+    "published": "2026-06-05",
+    "updated": "2026-06-08",
+    "authors": [
+      "Thanawat Lodkaew",
+      "Johannes Ackermann",
+      "Soichiro Nishimori",
+      "Nontawat Charoenphakdee",
+      "Masashi Sugiyama",
+      "Takashi Ishida"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL",
+      "stat.ME"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.07379",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.07379",
+    "pdf_url": "https://arxiv.org/pdf/2606.07379",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.05548",
+    "title": "ADK Arena: Evaluating Agent Development Kits via LLM-as-a-Developer",
+    "url": "https://arxiv.org/abs/2606.05548",
+    "published": "2026-06-04",
+    "updated": "2026-06-04",
+    "authors": [
+      "Jintao Huang",
+      "Xiaomin Li",
+      "Gaurav Mittal",
+      "Yu Hu"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.05548",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.05548",
+    "pdf_url": "https://arxiv.org/pdf/2606.05548",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.06473",
+    "title": "MLEvolve: A Self-Evolving Framework for Automated Machine Learning Algorithm Discovery",
+    "url": "https://arxiv.org/abs/2606.06473",
+    "published": "2026-06-04",
+    "updated": "2026-06-04",
+    "authors": [
+      "Shangheng Du",
+      "Xiangchao Yan",
+      "Jinxin Shi",
+      "Zongsheng Cao",
+      "Shiyang Feng",
+      "Zichen Liang",
+      "Boyuan Sun",
+      "Tianshuo Peng",
+      "Yifan Zhou",
+      "Xin Li",
+      "Jie Zhou",
+      "Liang He",
+      "Bo Zhang",
+      "Lei Bai"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.06473",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.06473",
+    "pdf_url": "https://arxiv.org/pdf/2606.06473",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.06462",
+    "title": "Benchmark Everything Everywhere All at Once",
+    "url": "https://arxiv.org/abs/2606.06462",
+    "published": "2026-06-04",
+    "updated": "2026-06-04",
+    "authors": [
+      "Shiyun Xiong",
+      "Dongming Wu",
+      "Peiwen Sun",
+      "Yuang Ai",
+      "Bokang Yang",
+      "Wencheng Han",
+      "Xiao-Hui Li",
+      "Xiangyu Yue"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.06462",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.06462",
+    "pdf_url": "https://arxiv.org/pdf/2606.06462",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.05263",
+    "title": "Policy-Conditioned Counterfactual Credit for Verifiable Reinforcement Learning of Long-Horizon Language Agents",
+    "url": "https://arxiv.org/abs/2606.05263",
+    "published": "2026-06-03",
+    "updated": "2026-06-03",
+    "authors": [
+      "Renwei Meng"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.05263",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.05263",
+    "pdf_url": "https://arxiv.org/pdf/2606.05263",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.04628",
+    "title": "RAMPART: Registry-based Agentic Memory with Priority-Aware Runtime Transformation",
+    "url": "https://arxiv.org/abs/2606.04628",
+    "published": "2026-06-03",
+    "updated": "2026-06-03",
+    "authors": [
+      "Nikodem Tomczak"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.MA"
+    ],
+    "topics": [
+      "memory"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.04628",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.04628",
+    "pdf_url": "https://arxiv.org/pdf/2606.04628",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.05414",
+    "title": "When Evidence is Sparse: Weakly Supervised Early Failure Alerting in Dialogs and LLM-Agent Trajectories",
+    "url": "https://arxiv.org/abs/2606.05414",
+    "published": "2026-06-03",
+    "updated": "2026-06-03",
+    "authors": [
+      "Avinash Baidya",
+      "Xinran Liang",
+      "Ruocheng Guo",
+      "Xiang Gao",
+      "Kamalika Das"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.HC",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.05414",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.05414",
+    "pdf_url": "https://arxiv.org/pdf/2606.05414",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.03329",
+    "title": "InfoMem: Training Long-Context Memory Agents with Answer-Conditioned Information Gain",
+    "url": "https://arxiv.org/abs/2606.03329",
+    "published": "2026-06-02",
+    "updated": "2026-06-02",
+    "authors": [
+      "Tiancheng Han",
+      "Yong Li",
+      "Wuzhou Yu",
+      "Qiaosheng Zhang",
+      "Wenqi Shao"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.03329",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.03329",
+    "pdf_url": "https://arxiv.org/pdf/2606.03329",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.02965",
+    "title": "What Benchmarks Don't Measure: The Case for Evaluating Abstention Competence in Autonomous Agents",
+    "url": "https://arxiv.org/abs/2606.02965",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Victor Ojewale",
+      "Suresh Venkatasubramanian"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.02965",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.02965",
+    "pdf_url": "https://arxiv.org/pdf/2606.02965",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.02404",
+    "title": "K-BrowseComp: A Web Browsing Agent Benchmark Grounded in Korean Contexts",
+    "url": "https://arxiv.org/abs/2606.02404",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Nahyun Lee",
+      "Dongkeun Yoon",
+      "Guijin Son",
+      "Geewook Kim",
+      "Dayoon Ko",
+      "Jeonghun Park",
+      "Haneul Yoo",
+      "Jaewon Cho",
+      "Junghun Park",
+      "Changyoon Lee",
+      "Kyochul Jang",
+      "Jaeyeon Kim",
+      "Eunsu Kim",
+      "Woojin Cho",
+      "Seungone Kim"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.02404",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.02404",
+    "pdf_url": "https://arxiv.org/pdf/2606.02404",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.09863",
+    "title": "From Confident Closing to Silent Failure: Characterizing False Success in LLM Agents",
+    "url": "https://arxiv.org/abs/2606.09863",
+    "published": "2026-06-01",
+    "updated": "2026-06-01",
+    "authors": [
+      "Laksh Advani"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.09863",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09863",
+    "pdf_url": "https://arxiv.org/pdf/2606.09863",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.00611",
+    "title": "TRACE: Trajectory Risk-Aware Compression for Long-Horizon Agent Safety",
+    "url": "https://arxiv.org/abs/2606.00611",
+    "published": "2026-05-30",
+    "updated": "2026-05-30",
+    "authors": [
+      "Zhepei Hong",
+      "Lin Wang",
+      "Liting Li",
+      "Haokai Ma",
+      "Junfeng Fang",
+      "Fei Shen",
+      "Dan Zhang",
+      "Xiang Wang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2606.00611",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.00611",
+    "pdf_url": "https://arxiv.org/pdf/2606.00611",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.00198",
+    "title": "BAGEN: Are LLM Agents Budget-Aware?",
+    "url": "https://arxiv.org/abs/2606.00198",
+    "published": "2026-05-29",
+    "updated": "2026-05-29",
+    "authors": [
+      "Yuxiang Lin",
+      "Zihan Wang",
+      "Mengyang Liu",
+      "Yuxuan Shan",
+      "Longju Bai",
+      "Junyao Zhang",
+      "Xing Jin",
+      "Boshan Chen",
+      "Jinyan Su",
+      "Xingyao Wang",
+      "Jiaxin Pei",
+      "Manling Li"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.00198",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.00198",
+    "pdf_url": "https://arxiv.org/pdf/2606.00198",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.29790",
+    "title": "Evolve as a Team: Collaborative Self-Evolution for LLM-based Multi-Agent Systems",
+    "url": "https://arxiv.org/abs/2605.29790",
+    "published": "2026-05-28",
+    "updated": "2026-05-28",
+    "authors": [
+      "Zhezheng Hao",
+      "Tianfu Wang",
+      "Huanshuo Dong",
+      "Ziyan Liu",
+      "Hong Wang",
+      "Xiankun Lin",
+      "Qiang Lin",
+      "Can Wang",
+      "Hande Dong",
+      "Jiawei Chen"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2605.29790",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.29790",
+    "pdf_url": "https://arxiv.org/pdf/2605.29790",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2605.29640",
+    "title": "VikingMem: A Memory Base Management System for Stateful LLM-based Applications",
+    "url": "https://arxiv.org/abs/2605.29640",
+    "published": "2026-05-28",
+    "updated": "2026-06-12",
+    "authors": [
+      "Jiajie Fu",
+      "Junwen Chen",
+      "Mengzhao Wang",
+      "Aoxiang He",
+      "Maojia Sheng",
+      "Xiangyu Ke",
+      "Yifan Zhu",
+      "Yunjun Gao"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.29640",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.29640",
+    "pdf_url": "https://arxiv.org/pdf/2605.29640",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.29801",
+    "title": "AgentDoG 1.5: A Lightweight and Scalable Alignment Framework for AI Agent Safety and Security",
+    "url": "https://arxiv.org/abs/2605.29801",
+    "published": "2026-05-28",
+    "updated": "2026-05-28",
+    "authors": [
+      "Dongrui Liu",
+      "Yu Li",
+      "Zhonghao Yang",
+      "Peng Wang",
+      "Guanxu Chen",
+      "Yuejin Xie",
+      "Qinghua Mao",
+      "Wanying Qu",
+      "Yanxu Zhu",
+      "Tianyi Zhou",
+      "Leitao Yuan",
+      "Zhijie Zheng",
+      "Qihao Lin",
+      "Yimin Wang",
+      "Haoyu Luo",
+      "Shuai Shao",
+      "Chen Qian",
+      "Qingyu Liu",
+      "Ling Tang",
+      "Ruiyang Qin",
+      "Qihan Ren",
+      "Junxiao Yang",
+      "Kun Wang",
+      "Zhiheng Xi",
+      "Linfeng Zhang",
+      "Ranjie Duan",
+      "Bo Zhang",
+      "Wenjie Wang",
+      "Wen Shen",
+      "Qiaosheng Zhang",
+      "Yan Teng",
+      "Chaochao Lu",
+      "Rui Mei",
+      "Man Li",
+      "Jialing Tao",
+      "Xi Lin",
+      "Tianhang Zheng",
+      "Yong Liu",
+      "Quanshi Zhang",
+      "Lei Zhu",
+      "Xingjun Ma",
+      "Junhua Liu",
+      "Hui Xue",
+      "Xiaoxiang Zuo",
+      "Xiangnan He",
+      "Chao Shen",
+      "Xianglong Liu",
+      "Minlie Huang",
+      "Jing Shao",
+      "Xia Hu"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.CR",
+      "cs.CV",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.29801",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.29801",
+    "pdf_url": "https://arxiv.org/pdf/2605.29801",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.30407",
+    "title": "Exploring Autonomous Agentic Data Engineering for Model Specialization",
+    "url": "https://arxiv.org/abs/2605.30407",
+    "published": "2026-05-28",
+    "updated": "2026-06-08",
+    "authors": [
+      "Yujie Luo",
+      "Xiangyuan Ru",
+      "Jingsheng Zheng",
+      "Jingjing Wang",
+      "Yuqi Zhu",
+      "Jintian Zhang",
+      "Runnan Fang",
+      "Kewei Xu",
+      "Ye Liu",
+      "Zheng Wei",
+      "Jiang Bian",
+      "Zang Li",
+      "Shumin Deng"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.IR",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.30407",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.30407",
+    "pdf_url": "https://arxiv.org/pdf/2605.30407",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.27825",
+    "title": "MRMMIA: Membership Inference Attacks on Memory in Chat Agents",
+    "url": "https://arxiv.org/abs/2605.27825",
+    "published": "2026-05-27",
+    "updated": "2026-05-27",
+    "authors": [
+      "Kai Chen",
+      "Yan Pang",
+      "Tianhao Wang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.27825",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.27825",
+    "pdf_url": "https://arxiv.org/pdf/2605.27825",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.28175",
+    "title": "Mixture-of-Experts Knowledge Graph Retrieval-Augmented Generation for Multi-Agent LLM-based Recommendation",
+    "url": "https://arxiv.org/abs/2605.28175",
+    "published": "2026-05-27",
+    "updated": "2026-05-29",
+    "authors": [
+      "Shijie Wang",
+      "Chengyi Liu",
+      "Yujuan Ding",
+      "Shanru Lin",
+      "See-Kiong Ng",
+      "Xu Xin",
+      "Wenqi Fan"
+    ],
+    "categories": [
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.28175",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.28175",
+    "pdf_url": "https://arxiv.org/pdf/2605.28175",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.27935",
+    "title": "Do Agents Think Deeper? A Mechanistic Investigation of Layer-Wise Dynamics in Sequential Planning",
+    "url": "https://arxiv.org/abs/2605.27935",
+    "published": "2026-05-27",
+    "updated": "2026-05-27",
+    "authors": [
+      "Zhenyu Cui",
+      "Xiangzhong Luo"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm",
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.27935",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.27935",
+    "pdf_url": "https://arxiv.org/pdf/2605.27935",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.25920",
+    "title": "Can LLMs Time Travel? Enhancing Temporal Consistency in Legal Agentic Search through Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2605.25920",
+    "published": "2026-05-25",
+    "updated": "2026-05-25",
+    "authors": [
+      "Wei Fan",
+      "Yining Zhou",
+      "Mufan Zhang",
+      "Yanbing Weng",
+      "Yiran HU",
+      "Tianshi Zheng",
+      "Baixuan Xu",
+      "Chunyang Li",
+      "Jianhui Yang",
+      "Haoran Li",
+      "Yangqiu Song"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.25920",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.25920",
+    "pdf_url": "https://arxiv.org/pdf/2605.25920",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.25393",
+    "title": "Decision-Making with Lightweight Confidence-Aware Language Model for Autonomous Driving",
+    "url": "https://arxiv.org/abs/2605.25393",
+    "published": "2026-05-25",
+    "updated": "2026-05-25",
+    "authors": [
+      "Ruoyu Yao",
+      "Ruiguo Zhong",
+      "Pei Liu",
+      "Mingxing Peng",
+      "Rui Yang",
+      "Jun Ma"
+    ],
+    "categories": [
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.25393",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.25393",
+    "pdf_url": "https://arxiv.org/pdf/2605.25393",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.25310",
+    "title": "Tool-Call Dependency Structure is Linearly Decodable in LLM Agent Residual Streams",
+    "url": "https://arxiv.org/abs/2605.25310",
+    "published": "2026-05-25",
+    "updated": "2026-05-25",
+    "authors": [
+      "Tianda Sun",
+      "Dimitar Kazakov"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.25310",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.25310",
+    "pdf_url": "https://arxiv.org/pdf/2605.25310",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.24309",
+    "title": "Reframing LLM Agent Security as an Agent-Human Interaction Problem",
+    "url": "https://arxiv.org/abs/2605.24309",
+    "published": "2026-05-23",
+    "updated": "2026-05-23",
+    "authors": [
+      "Peiran Wang",
+      "Ying Li",
+      "Yuan Tian"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.24309",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.24309",
+    "pdf_url": "https://arxiv.org/pdf/2605.24309",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.19604",
+    "title": "Formal Skill: Programmable Runtime Skills for Efficient and Accurate LLM Agents",
+    "url": "https://arxiv.org/abs/2605.19604",
+    "published": "2026-05-19",
+    "updated": "2026-05-19",
+    "authors": [
+      "Xi Zhang",
+      "Meijun Gao",
+      "Yuntian Zhao",
+      "Xinyu Tan",
+      "Yilun Yao",
+      "Feiyu Wang",
+      "Yanshu Wang",
+      "Dingsiyi",
+      "Tong Yang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2605.19604",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.19604",
+    "pdf_url": "https://arxiv.org/pdf/2605.19604",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2605.18502",
+    "title": "The distance-based formation controller design for multi-agent systems in port-Hamiltonian form",
+    "url": "https://arxiv.org/abs/2605.18502",
+    "published": "2026-05-18",
+    "updated": "2026-05-18",
+    "authors": [
+      "Jingyi Zhao",
+      "Yongxin Wu",
+      "Héctor García de Marina",
+      "Yuhu Wu",
+      "Yann Le Gorrec"
+    ],
+    "categories": [
+      "math.OC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.18502",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.18502",
+    "pdf_url": "https://arxiv.org/pdf/2605.18502",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.15701",
+    "title": "H-Mem: A Novel Memory Mechanism for Evolving and Retrieving Agent Memory via a Hybrid Structure",
+    "url": "https://arxiv.org/abs/2605.15701",
+    "published": "2026-05-15",
+    "updated": "2026-05-15",
+    "authors": [
+      "Jiawei Yu",
+      "Yixiang Fang",
+      "Xilin Liu",
+      "Yuchi Ma"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.15701",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.15701",
+    "pdf_url": "https://arxiv.org/pdf/2605.15701",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.15625",
+    "title": "ColPackAgent: Agent-Skill-Guided Hard-Particle Monte Carlo Workflows for Colloidal Packing",
+    "url": "https://arxiv.org/abs/2605.15625",
+    "published": "2026-05-15",
+    "updated": "2026-05-15",
+    "authors": [
+      "Lijie Ding",
+      "Changwoo Do"
+    ],
+    "categories": [
+      "cs.AI",
+      "cond-mat.soft"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.15625",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.15625",
+    "pdf_url": "https://arxiv.org/pdf/2605.15625",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.14290",
+    "title": "Web Agents Should Adopt the Plan-Then-Execute Paradigm",
+    "url": "https://arxiv.org/abs/2605.14290",
+    "published": "2026-05-14",
+    "updated": "2026-05-14",
+    "authors": [
+      "Julien Piet",
+      "Annabella Chow",
+      "Yiwei Hou",
+      "Muxi Lyu",
+      "Sylvie Venuto",
+      "Jinhao Zhu",
+      "Raluca Ada Popa",
+      "David Wagner"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.CL",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.14290",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.14290",
+    "pdf_url": "https://arxiv.org/pdf/2605.14290",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.13716",
+    "title": "SkillOps: Managing LLM Agent Skill Libraries as Self-Maintaining Software Ecosystems",
+    "url": "https://arxiv.org/abs/2605.13716",
+    "published": "2026-05-13",
+    "updated": "2026-05-13",
+    "authors": [
+      "Hongji Pu",
+      "Xinyuan Song",
+      "Liang Zhao"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.13716",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.13716",
+    "pdf_url": "https://arxiv.org/pdf/2605.13716",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.13618",
+    "title": "OpenAaaS: An Open Agent-as-a-Service Framework for Distributed Materials-Informatics Research",
+    "url": "https://arxiv.org/abs/2605.13618",
+    "published": "2026-05-13",
+    "updated": "2026-05-13",
+    "authors": [
+      "Peng Kang",
+      "Bixuan Li",
+      "Xiaoya Huang",
+      "Shuo Shi",
+      "Weiqiao Zhou",
+      "Zhen Li",
+      "Yu Liu",
+      "Lei Zheng"
+    ],
+    "categories": [
+      "cond-mat.mtrl-sci",
+      "cs.AI"
+    ],
+    "topics": [
+      "memory",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.13618",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.13618",
+    "pdf_url": "https://arxiv.org/pdf/2605.13618",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.07112",
+    "title": "Switchcraft: AI Model Router for Agentic Tool Calling",
+    "url": "https://arxiv.org/abs/2605.07112",
+    "published": "2026-05-08",
+    "updated": "2026-05-08",
+    "authors": [
+      "Sharad Agarwal",
+      "Pooria Namyar",
+      "Alec Wolman",
+      "Rahul Ambavat",
+      "Ankur Gupta",
+      "Qizheng Zhang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2605.07112",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.07112",
+    "pdf_url": "https://arxiv.org/pdf/2605.07112",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2605.06992",
+    "title": "Why Does Agentic Safety Fail to Generalize Across Tasks?",
+    "url": "https://arxiv.org/abs/2605.06992",
+    "published": "2026-05-07",
+    "updated": "2026-05-07",
+    "authors": [
+      "Yonatan Slutzky",
+      "Yotam Alexander",
+      "Tomer Slor",
+      "Yoav Nagel",
+      "Nadav Cohen"
+    ],
+    "categories": [
+      "cs.LG",
+      "stat.ML"
+    ],
+    "topics": [
+      "agent-safety",
+      "embodied-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.06992",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.06992",
+    "pdf_url": "https://arxiv.org/pdf/2605.06992",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.06957",
+    "title": "Learning and Reusing Policy Decompositions for Hierarchical Generalized Planning with LLM Agents",
+    "url": "https://arxiv.org/abs/2605.06957",
+    "published": "2026-05-07",
+    "updated": "2026-05-07",
+    "authors": [
+      "Shirin Sohrabi",
+      "Haritha Ananthakrishnan",
+      "Harsha Kokel",
+      "Kavitha Srinivas",
+      "Michael Katz"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.06957",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.06957",
+    "pdf_url": "https://arxiv.org/pdf/2605.06957",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.06737",
+    "title": "A Self-Healing Framework for Reliable LLM-Based Autonomous Agents",
+    "url": "https://arxiv.org/abs/2605.06737",
+    "published": "2026-05-07",
+    "updated": "2026-05-07",
+    "authors": [
+      "Cheonsu Jeong",
+      "Younggun Shin"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.06737",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.06737",
+    "pdf_url": "https://arxiv.org/pdf/2605.06737",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2604.27464",
+    "title": "Security Attack and Defense Strategies for Autonomous Agent Frameworks: A Layered Review with OpenClaw as a Case Study",
+    "url": "https://arxiv.org/abs/2604.27464",
+    "published": "2026-04-30",
+    "updated": "2026-04-30",
+    "authors": [
+      "Luyao Xu",
+      "Xiang Chen"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety",
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2604.27464",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.27464",
+    "pdf_url": "https://arxiv.org/pdf/2604.27464",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.28157",
+    "title": "FlashRT: Towards Computationally and Memory Efficient Red-Teaming for Prompt Injection and Knowledge Corruption",
+    "url": "https://arxiv.org/abs/2604.28157",
+    "published": "2026-04-30",
+    "updated": "2026-04-30",
+    "authors": [
+      "Yanting Wang",
+      "Chenlong Yin",
+      "Ying Chen",
+      "Jinyuan Jia"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2604.28157",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.28157",
+    "pdf_url": "https://arxiv.org/pdf/2604.28157",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2604.27859",
+    "title": "Rethinking Agentic Reinforcement Learning In Large Language Models",
+    "url": "https://arxiv.org/abs/2604.27859",
+    "published": "2026-04-30",
+    "updated": "2026-05-15",
+    "authors": [
+      "Fangming Cui",
+      "Ruixiao Zhu",
+      "Cheng Fang",
+      "Sunan Li",
+      "Jiahong Li"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.ET"
+    ],
+    "topics": [
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2604.27859",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.27859",
+    "pdf_url": "https://arxiv.org/pdf/2604.27859",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.20573",
+    "title": "AONA: A Comprehensive Architecture and Workflow Design for Global Agentic Collaboration",
+    "url": "https://arxiv.org/abs/2606.20573",
+    "published": "2026-04-30",
+    "updated": "2026-04-30",
+    "authors": [
+      "Jinliang Xu",
+      "Runkai Zhu",
+      "Bingqi Li",
+      "Fanjie Nie",
+      "Jin Li",
+      "Jiagui Xie"
+    ],
+    "categories": [
+      "cs.NI",
+      "cs.MA"
+    ],
+    "topics": [
+      "multi-agent",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.20573",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20573",
+    "pdf_url": "https://arxiv.org/pdf/2606.20573",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2604.25555",
+    "title": "From CRUD to Autonomous Agents: Formal Validation and Zero-Trust Security for Semantic Gateways in AI-Native Enterprise Systems",
+    "url": "https://arxiv.org/abs/2604.25555",
+    "published": "2026-04-28",
+    "updated": "2026-04-28",
+    "authors": [
+      "Ignacio Peyrano"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2604.25555",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.25555",
+    "pdf_url": "https://arxiv.org/pdf/2604.25555",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2604.21190",
+    "title": "SpatiO: Adaptive Test-Time Orchestration of Vision-Language Agents for Spatial Reasoning",
+    "url": "https://arxiv.org/abs/2604.21190",
+    "published": "2026-04-23",
+    "updated": "2026-06-27",
+    "authors": [
+      "Chan Yeong Hwang",
+      "Miso Choi",
+      "Sunghyun On",
+      "Jinkyu Kim",
+      "Jungbeom Lee"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "reasoning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2604.21190",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.21190",
+    "pdf_url": "https://arxiv.org/pdf/2604.21190",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2604.16706",
+    "title": "Evaluating Tool-Using Language Agents: Judge Reliability, Propagation Cascades, and Runtime Mitigation in AgentProp-Bench",
+    "url": "https://arxiv.org/abs/2604.16706",
+    "published": "2026-04-17",
+    "updated": "2026-04-17",
+    "authors": [
+      "Bhaskar Gurram"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2604.16706",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.16706",
+    "pdf_url": "https://arxiv.org/pdf/2604.16706",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2604.15579",
+    "title": "Don't Make Models Guess Security and Safety: Symbolic Guardrails for Domain-Specific AI Agents",
+    "url": "https://arxiv.org/abs/2604.15579",
+    "published": "2026-04-16",
+    "updated": "2026-07-05",
+    "authors": [
+      "Yining Hong",
+      "Yining She",
+      "Eunsuk Kang",
+      "Christopher S. Timperley",
+      "Christian Kästner"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.15579",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.15579",
+    "pdf_url": "https://arxiv.org/pdf/2604.15579",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.15415",
+    "title": "HarmfulSkillBench: How Do Harmful Skills Weaponize Your Agents?",
+    "url": "https://arxiv.org/abs/2604.15415",
+    "published": "2026-04-16",
+    "updated": "2026-04-16",
+    "authors": [
+      "Yukun Jiang",
+      "Yage Zhang",
+      "Michael Backes",
+      "Xinyue Shen",
+      "Yang Zhang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.15415",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.15415",
+    "pdf_url": "https://arxiv.org/pdf/2604.15415",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.14399",
+    "title": "SpaceMind: A Modular and Self-Evolving Embodied Vision-Language Agent Framework for Autonomous On-orbit Servicing",
+    "url": "https://arxiv.org/abs/2604.14399",
+    "published": "2026-04-15",
+    "updated": "2026-04-15",
+    "authors": [
+      "Aodi Wu",
+      "Haodong Han",
+      "Xubo Luo",
+      "Ruisuo Wang",
+      "Shan He",
+      "Xue Wan"
+    ],
+    "categories": [
+      "cs.RO",
+      "cs.AI",
+      "eess.SY"
+    ],
+    "topics": [
+      "embodied-agent",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2604.14399",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.14399",
+    "pdf_url": "https://arxiv.org/pdf/2604.14399",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2604.08388",
+    "title": "Awakening the Sleeping Agent: Lean-Specific Agentic Data Reactivates General Tool Use in Goedel Prover",
+    "url": "https://arxiv.org/abs/2604.08388",
+    "published": "2026-04-09",
+    "updated": "2026-04-09",
+    "authors": [
+      "Jui-Hui Chung",
+      "Hongzhou Lin",
+      "Lai Jiang",
+      "Shange Tang",
+      "Chi Jin"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2604.08388",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.08388",
+    "pdf_url": "https://arxiv.org/pdf/2604.08388",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2604.06762",
+    "title": "ARuleCon: Agentic Security Rule Conversion",
+    "url": "https://arxiv.org/abs/2604.06762",
+    "published": "2026-04-08",
+    "updated": "2026-04-08",
+    "authors": [
+      "Ming Xu",
+      "Hongtai Wang",
+      "Yanpei Guo",
+      "Zhengmin Yu",
+      "Weili Han",
+      "Hoon Wei Lim",
+      "Jin Song Dong",
+      "Jiaheng Zhang"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2604.06762",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.06762",
+    "pdf_url": "https://arxiv.org/pdf/2604.06762",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2604.02155",
+    "title": "Brief Is Better: Non-Monotonic Chain-of-Thought Budget Effects in Function-Calling Language Agents",
+    "url": "https://arxiv.org/abs/2604.02155",
+    "published": "2026-04-02",
+    "updated": "2026-04-02",
+    "authors": [
+      "Xuan Qi"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling",
+      "language-agent"
+    ],
+    "arxiv_id": "2604.02155",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.02155",
+    "pdf_url": "https://arxiv.org/pdf/2604.02155",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2603.27148",
+    "title": "SafetyDrift: Predicting When AI Agents Cross the Line Before They Actually Do",
+    "url": "https://arxiv.org/abs/2603.27148",
+    "published": "2026-03-28",
+    "updated": "2026-03-28",
+    "authors": [
+      "Aditya Dhodapkar",
+      "Farhaan Pishori"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.27148",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.27148",
+    "pdf_url": "https://arxiv.org/pdf/2603.27148",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.19469",
+    "title": "A Framework for Formalizing LLM Agent Security",
+    "url": "https://arxiv.org/abs/2603.19469",
+    "published": "2026-03-19",
+    "updated": "2026-03-19",
+    "authors": [
+      "Vincent Siu",
+      "Jingxuan He",
+      "Kyle Montgomery",
+      "Zhun Wang",
+      "Neil Gong",
+      "Chenguang Wang",
+      "Dawn Song"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "memory",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.19469",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.19469",
+    "pdf_url": "https://arxiv.org/pdf/2603.19469",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.11088",
+    "title": "The Attack and Defense Landscape of Agentic AI: A Comprehensive Survey",
+    "url": "https://arxiv.org/abs/2603.11088",
+    "published": "2026-03-11",
+    "updated": "2026-03-11",
+    "authors": [
+      "Juhee Kim",
+      "Xiaoyuan Liu",
+      "Zhun Wang",
+      "Shi Qiu",
+      "Bo Li",
+      "Wenbo Guo",
+      "Dawn Song"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.11088",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.11088",
+    "pdf_url": "https://arxiv.org/pdf/2603.11088",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.01438",
+    "title": "Enhancing Persona Following at Decoding Time via Dynamic Importance Estimation for Role-Playing Agents",
+    "url": "https://arxiv.org/abs/2603.01438",
+    "published": "2026-03-02",
+    "updated": "2026-03-02",
+    "authors": [
+      "Yuxin Liu",
+      "Mingye Zhu",
+      "Siyuan Liu",
+      "Bo Hu",
+      "Lei Zhang"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "planning",
+      "rag",
+      "world-model"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.01438",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.01438",
+    "pdf_url": "https://arxiv.org/pdf/2603.01438",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2602.23320",
+    "title": "ParamMem: Augmenting Language Agents with Parametric Reflective Memory",
+    "url": "https://arxiv.org/abs/2602.23320",
+    "published": "2026-02-26",
+    "updated": "2026-02-27",
+    "authors": [
+      "Tianjun Yao",
+      "Yongqiang Chen",
+      "Yujia Zheng",
+      "Pan Li",
+      "Zhiqiang Shen",
+      "Kun Zhang"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "reasoning"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2602.23320",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.23320",
+    "pdf_url": "https://arxiv.org/pdf/2602.23320",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2602.16931",
+    "title": "Narrow Fine-Tuning Erodes Safety Alignment in Vision-Language Agents",
+    "url": "https://arxiv.org/abs/2602.16931",
+    "published": "2026-02-18",
+    "updated": "2026-03-15",
+    "authors": [
+      "Idhant Gulati",
+      "Shivam Raval"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2602.16931",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.16931",
+    "pdf_url": "https://arxiv.org/pdf/2602.16931",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2602.07391",
+    "title": "NAAMSE: Framework for Evolutionary Security Evaluation of Agents",
+    "url": "https://arxiv.org/abs/2602.07391",
+    "published": "2026-02-07",
+    "updated": "2026-03-08",
+    "authors": [
+      "Kunal Pai",
+      "Parth Shah",
+      "Harshil Patel"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.07391",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.07391",
+    "pdf_url": "https://arxiv.org/pdf/2602.07391",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2601.05467",
+    "title": "STELP: Secure Transpilation and Execution of LLM-Generated Programs",
+    "url": "https://arxiv.org/abs/2601.05467",
+    "published": "2026-01-09",
+    "updated": "2026-01-15",
+    "authors": [
+      "Swapnil Shinde",
+      "Sahil Wadhwa",
+      "Andy Luo",
+      "Akshay Gupta",
+      "Mohammad Shahed Sorower"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2601.05467",
+    "source": "arxiv",
+    "source_id": "arxiv:2601.05467",
+    "pdf_url": "https://arxiv.org/pdf/2601.05467",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2510.26167",
+    "title": "ToolRM: Towards Agentic Tool-Use Reward Modeling",
+    "url": "https://arxiv.org/abs/2510.26167",
+    "published": "2025-10-30",
+    "updated": "2026-01-13",
+    "authors": [
+      "Renhao Li",
+      "Jianhong Tu",
+      "Yang Su",
+      "Yantao Liu",
+      "Fei Huang",
+      "Hamid Alinejad-Rokny",
+      "Derek F. Wong",
+      "Junyang Lin",
+      "Min Yang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2510.26167",
+    "source": "arxiv",
+    "source_id": "arxiv:2510.26167",
+    "pdf_url": "https://arxiv.org/pdf/2510.26167",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2510.22768",
+    "title": "Seeing is Believing? Evaluating Vision-Language Model Susceptibility in Agent-to-Agent Multimodal Persuasion",
+    "url": "https://arxiv.org/abs/2510.22768",
+    "published": "2025-10-26",
+    "updated": "2026-06-02",
+    "authors": [
+      "Haoyi Qiu",
+      "Yilun Zhou",
+      "Pranav Narayanan Venkit",
+      "Kung-Hsiang Huang",
+      "Jiaxin Zhang",
+      "Nanyun Peng",
+      "Chien-Sheng Wu"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 14,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2510.22768",
+    "source": "arxiv",
+    "source_id": "arxiv:2510.22768",
+    "pdf_url": "https://arxiv.org/pdf/2510.22768",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2607.05794",
+    "title": "From Passive Retrieval to Active Memory Navigation: Learning to Use Memory as a Structured Action Space",
+    "url": "https://arxiv.org/abs/2607.05794",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Yue Xu",
+      "Yutao Sun",
+      "Yihao Liu",
+      "Mengyu Zhou",
+      "Jiayi Qiao",
+      "Lu Ma",
+      "Kai Tang",
+      "Wenjie Wang",
+      "Xiaoxi Jiang",
+      "Guanjun Jiang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "memory",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2607.05794",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05794",
+    "pdf_url": "https://arxiv.org/pdf/2607.05794",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2607.06341",
+    "title": "Harnessing Code Agents for Automatic Software Verification",
+    "url": "https://arxiv.org/abs/2607.06341",
+    "published": "2026-07-07",
+    "updated": "2026-07-07",
+    "authors": [
+      "Shuangxiang Kan",
+      "Shuanglong Kan",
+      "Sebastian Ertel"
+    ],
+    "categories": [
+      "cs.FL",
+      "cs.AI",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.06341",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.06341",
+    "pdf_url": "https://arxiv.org/pdf/2607.06341",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.05001",
+    "title": "TACTIC-KG: Toward Small Agent Teams for Cyber Threat Intelligence Knowledge Graph Construction",
+    "url": "https://arxiv.org/abs/2607.05001",
+    "published": "2026-07-06",
+    "updated": "2026-07-07",
+    "authors": [
+      "Mouhamed Amine Bouchiha",
+      "Gregory Blanc"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.LG",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.05001",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05001",
+    "pdf_url": "https://arxiv.org/pdf/2607.05001",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.05666",
+    "title": "What Do AI Agents Actually Change? An Empirical Taxonomy of Mutation Patterns in Performance-Improving Pull Requests",
+    "url": "https://arxiv.org/abs/2607.05666",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Illia Dovhoshliubnyi",
+      "Nima Soroush",
+      "Ashkan Sami",
+      "Alexander Brownlee"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "coding-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.05666",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05666",
+    "pdf_url": "https://arxiv.org/pdf/2607.05666",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.05518",
+    "title": "aiAuthZ: Off-Host, Identity-Bound Authorization for AI Agents",
+    "url": "https://arxiv.org/abs/2607.05518",
+    "published": "2026-07-06",
+    "updated": "2026-07-06",
+    "authors": [
+      "Sai Varun Kodathala"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.05518",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05518",
+    "pdf_url": "https://arxiv.org/pdf/2607.05518",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.04697",
+    "title": "AI Agent Pull Requests on GitHub: Frequency, Structure, and Merge Conflict Rates",
+    "url": "https://arxiv.org/abs/2607.04697",
+    "published": "2026-07-06",
+    "updated": "2026-07-07",
+    "authors": [
+      "George Xu",
+      "Arjun Subramanian",
+      "Nithilan Karthik"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.04697",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04697",
+    "pdf_url": "https://arxiv.org/pdf/2607.04697",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.05391",
+    "title": "LLM-as-a-Verifier: A General-Purpose Verification Framework",
+    "url": "https://arxiv.org/abs/2607.05391",
+    "published": "2026-07-06",
+    "updated": "2026-07-07",
+    "authors": [
+      "Jacky Kwok",
+      "Shulu Li",
+      "Pranav Atreya",
+      "Yuejiang Liu",
+      "Yixing Jiang",
+      "Chelsea Finn",
+      "Marco Pavone",
+      "Ion Stoica",
+      "Azalia Mirhoseini"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.LG",
+      "cs.MA",
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "embodied-agent",
+      "reasoning"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.05391",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.05391",
+    "pdf_url": "https://arxiv.org/pdf/2607.05391",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.04334",
+    "title": "Do GUI Agents Believe Their Eyes? Diagnosing State-Belief Reliance on Pixels versus Structure",
+    "url": "https://arxiv.org/abs/2607.04334",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Guijia Zhang",
+      "Harry Yang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2607.04334",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04334",
+    "pdf_url": "https://arxiv.org/pdf/2607.04334",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2607.04394",
+    "title": "MechMath Agent Team: LLM Driven Agents for Mathematical Research",
+    "url": "https://arxiv.org/abs/2607.04394",
+    "published": "2026-07-05",
+    "updated": "2026-07-05",
+    "authors": [
+      "Yichuan Cao",
+      "Ruichen Qiu",
+      "Junqi Liu",
+      "Jiaqi Wang",
+      "Dakai Guo",
+      "Ruyong Feng",
+      "Lihong Zhi",
+      "Xiao-Shan Gao"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.SC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "rag",
+      "reasoning"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2607.04394",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.04394",
+    "pdf_url": "https://arxiv.org/pdf/2607.04394",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2607.02942",
+    "title": "A Workflow-Aware Serving Layer for Agentic Applications",
+    "url": "https://arxiv.org/abs/2607.02942",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Jiayi Qian",
+      "Zishen Wan",
+      "Hanchen Yang",
+      "Chun Tao",
+      "Souvik Kundu",
+      "Tushar Krishna"
+    ],
+    "categories": [
+      "cs.DC",
+      "cs.MA"
+    ],
+    "topics": [
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.02942",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02942",
+    "pdf_url": "https://arxiv.org/pdf/2607.02942",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.03105",
+    "title": "ORBIT-Q: Dual-axis benchmarking of autonomous agents in scientific quantum programming",
+    "url": "https://arxiv.org/abs/2607.03105",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Shi-Xin Zhang",
+      "Yu-Qin Chen"
+    ],
+    "categories": [
+      "quant-ph"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.03105",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03105",
+    "pdf_url": "https://arxiv.org/pdf/2607.03105",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2607.03316",
+    "title": "Is Agentic Code Review Helpful? Mining Developers' Feedback to CodeRabbit Reviews in the Wild",
+    "url": "https://arxiv.org/abs/2607.03316",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Hong Yi Lin",
+      "Mingzhao Liang",
+      "Kla Tantithamthavorn",
+      "Patanamon Thongtanunam"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "coding-agent",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2607.03316",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03316",
+    "pdf_url": "https://arxiv.org/pdf/2607.03316",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2607.03220",
+    "title": "CONTRA: Red-Teaming Configurations of Personalizable Agents",
+    "url": "https://arxiv.org/abs/2607.03220",
+    "published": "2026-07-03",
+    "updated": "2026-07-03",
+    "authors": [
+      "Jonathan Nöther",
+      "Adish Singla",
+      "Goran Radanovic"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2607.03220",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.03220",
+    "pdf_url": "https://arxiv.org/pdf/2607.03220",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2607.01812",
+    "title": "TO-Master: an LLM-agent framework for automated topology optimization",
+    "url": "https://arxiv.org/abs/2607.01812",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Haoju Lin",
+      "Wenchang Zhang",
+      "Weipeng Xu",
+      "Xiang Li",
+      "Tian Xu",
+      "Tianju Xue"
+    ],
+    "categories": [
+      "cs.CE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.01812",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01812",
+    "pdf_url": "https://arxiv.org/pdf/2607.01812",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.02453",
+    "title": "Adoption and Ecosystem Health: A Longitudinal Analysis of Open-Source Multi-Agent Frameworks",
+    "url": "https://arxiv.org/abs/2607.02453",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Xi Zhang",
+      "Papi Menon",
+      "Vivian Chu",
+      "Koray Cosguner"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.02453",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02453",
+    "pdf_url": "https://arxiv.org/pdf/2607.02453",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.02245",
+    "title": "Copewell: A Multi-Agent Swarm Architecture for Equitable Mental Wellness Support",
+    "url": "https://arxiv.org/abs/2607.02245",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Seren Yenikent",
+      "Jack Vinijtrongjit",
+      "Katherine Ng"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CY",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.02245",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02245",
+    "pdf_url": "https://arxiv.org/pdf/2607.02245",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.02381",
+    "title": "HULAT2 at MER-TRANS 2026: Governed Multi-Agent Simplification for Spanish Easy-to-Read Generation",
+    "url": "https://arxiv.org/abs/2607.02381",
+    "published": "2026-07-02",
+    "updated": "2026-07-02",
+    "authors": [
+      "Lourdes Moreno",
+      "Paloma Martínez",
+      "Marco Antonio Sanchez-Escudero",
+      "Miguel Domínguez-Gómez"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2607.02381",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.02381",
+    "pdf_url": "https://arxiv.org/pdf/2607.02381",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.01531",
+    "title": "OPINE-World: Programmatic World Modeling with Ontology-error-Prioritized Interactive Exploration",
+    "url": "https://arxiv.org/abs/2607.01531",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "David Courtis",
+      "Wenhao Li",
+      "Scott Sanner"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent",
+      "planning-agent"
+    ],
+    "arxiv_id": "2607.01531",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01531",
+    "pdf_url": "https://arxiv.org/pdf/2607.01531",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.01047",
+    "title": "Conversable Complexity: Agentic LLM Collectives as Interpretable Substrates",
+    "url": "https://arxiv.org/abs/2607.01047",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Elias Najarro",
+      "Ane Espeseth",
+      "Eleni Nisioti",
+      "Sebastian Risi",
+      "Stefano Nichele"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "memory",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2607.01047",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01047",
+    "pdf_url": "https://arxiv.org/pdf/2607.01047",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2607.01510",
+    "title": "Janus: a Playground for User-Involved Agentic Permission Management",
+    "url": "https://arxiv.org/abs/2607.01510",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Natalie Grace Brigham",
+      "Eugene Bagdasarian",
+      "Tadayoshi Kohno",
+      "Franziska Roesner"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.01510",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01510",
+    "pdf_url": "https://arxiv.org/pdf/2607.01510",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.00407",
+    "title": "Personalization as Inverse Planning: Learning Latent Design Intents for Agentic Slide Generation via Structural Denoising",
+    "url": "https://arxiv.org/abs/2607.00407",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Tianci Liu",
+      "Zihan Dong",
+      "Linjun Zhang",
+      "Haoyu Wang",
+      "jing Gao",
+      "Emre Kiciman",
+      "Ranveer Chandra",
+      "Wei-Ting Chen"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "multi-agent",
+      "planning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2607.00407",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00407",
+    "pdf_url": "https://arxiv.org/pdf/2607.00407",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2607.01366",
+    "title": "Auto-FL-Research: Agentic Search for Federated Learning Algorithms",
+    "url": "https://arxiv.org/abs/2607.01366",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Holger R. Roth",
+      "Ziyue Xu",
+      "Chester Chen",
+      "Daguang Xu",
+      "Peter Cnudde",
+      "Andrew Feng"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.01366",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.01366",
+    "pdf_url": "https://arxiv.org/pdf/2607.01366",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2607.00990",
+    "title": "SWE-Doctor: Guiding Software Engineering Agents with Runtime Diagnosis from Multi-Faceted Bug Reproduction Tests",
+    "url": "https://arxiv.org/abs/2607.00990",
+    "published": "2026-07-01",
+    "updated": "2026-07-01",
+    "authors": [
+      "Yaoqi Guo",
+      "Yang Liu",
+      "Jie M. Zhang",
+      "Yun Ma",
+      "Yiling Lou",
+      "Zhenpeng Chen"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "rag"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2607.00990",
+    "source": "arxiv",
+    "source_id": "arxiv:2607.00990",
+    "pdf_url": "https://arxiv.org/pdf/2607.00990",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.31471",
+    "title": "Think While You Map: Asynchronous Vision-Language Agents for Incremental 3D Scene Graphs",
+    "url": "https://arxiv.org/abs/2606.31471",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Deniz Bickici",
+      "Michael Pabst",
+      "Shohei Mori",
+      "Dieter Schmalstieg"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.31471",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31471",
+    "pdf_url": "https://arxiv.org/pdf/2606.31471",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.31831",
+    "title": "An Agentic AI Framework to Accelerate Scientific Discovery in Plant Phenotyping",
+    "url": "https://arxiv.org/abs/2606.31831",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Renan Souza",
+      "Daniel Rosendo",
+      "Kelsey Carter",
+      "John Lagergren",
+      "Frédéric Suter",
+      "Shelaine L. Curd",
+      "Gerald A. Tuskan",
+      "Rafael Ferreira da Silva",
+      "David Weston"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "planning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.31831",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31831",
+    "pdf_url": "https://arxiv.org/pdf/2606.31831",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.31916",
+    "title": "Theory of Mind and Persuasion Beyond Conversation: Assessing the Capacity of LLMs to Induce Belief States via Planning and Action",
+    "url": "https://arxiv.org/abs/2606.31916",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Ben Slater",
+      "Matteo G. Mecattaf",
+      "Lucy G. Cheke",
+      "John Burden",
+      "Winnie Street"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.31916",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31916",
+    "pdf_url": "https://arxiv.org/pdf/2606.31916",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.31767",
+    "title": "JETO-Bench: A Reproducible Benchmark for Execution Time Improvement Patches in Java",
+    "url": "https://arxiv.org/abs/2606.31767",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Khashayar Etemadi",
+      "Zhendong Su"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.31767",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31767",
+    "pdf_url": "https://arxiv.org/pdf/2606.31767",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.31665",
+    "title": "ForecastAgentSearch: Towards a Multi-Expert Agent Search System for Geopolitical Event Forecasting",
+    "url": "https://arxiv.org/abs/2606.31665",
+    "published": "2026-06-30",
+    "updated": "2026-06-30",
+    "authors": [
+      "Miaomiao Cai",
+      "He Chang",
+      "Yunshan Ma",
+      "See-kiong Ng"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "planning",
+      "rag"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.31665",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.31665",
+    "pdf_url": "https://arxiv.org/pdf/2606.31665",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.29719",
+    "title": "A Diagnostic Framework and Multi-Evaluator Audit of Evaluator-Driven Preference Dynamics in Self-Adapting LLM Agents",
+    "url": "https://arxiv.org/abs/2606.29719",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Liu Zewen"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "llm-agent"
+    ],
+    "arxiv_id": "2606.29719",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29719",
+    "pdf_url": "https://arxiv.org/pdf/2606.29719",
+    "primary_query": "llm-agent"
+  },
+  {
+    "id": "2606.29745",
+    "title": "ECHO: Learning Epistemically Adaptive Language Agents with Turn-Level Credit",
+    "url": "https://arxiv.org/abs/2606.29745",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Abhijnan Nath",
+      "Nikhil Krishnaswamy"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.29745",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29745",
+    "pdf_url": "https://arxiv.org/pdf/2606.29745",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.30970",
+    "title": "Behavioral Governance for Autonomous AI Agents: The AgentBound Framework",
+    "url": "https://arxiv.org/abs/2606.30970",
+    "published": "2026-06-29",
+    "updated": "2026-07-01",
+    "authors": [
+      "Anuj Kaul",
+      "Qianlong Lan",
+      "Pranay Gupta"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.30970",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30970",
+    "pdf_url": "https://arxiv.org/pdf/2606.30970",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.29788",
+    "title": "MemLeak: Diagnosing Information Leaks in Multimodal Agent Memory",
+    "url": "https://arxiv.org/abs/2606.29788",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Kuan Wang",
+      "Chao Zhang"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory",
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.29788",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29788",
+    "pdf_url": "https://arxiv.org/pdf/2606.29788",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.30616",
+    "title": "Scaling the Horizon, Not the Parameters: Reaching Trillion-Parameter Performance with a 35B Agent",
+    "url": "https://arxiv.org/abs/2606.30616",
+    "published": "2026-06-29",
+    "updated": "2026-06-29",
+    "authors": [
+      "Lei Bai",
+      "Zongsheng Cao",
+      "Yang Chen",
+      "Zhiyao Cui",
+      "Shangheng Du",
+      "Yue Fan",
+      "Shiyang Feng",
+      "Zijie Guo",
+      "Haonan He",
+      "Liang He",
+      "Xiaohan He",
+      "Shuyue Hu",
+      "Yusong Hu",
+      "Songtao Huang",
+      "Yichen Jiang",
+      "Hao Li",
+      "Xin Li",
+      "Dahua Lin",
+      "Weihao Lin",
+      "Fenghua Ling",
+      "Dongrui Liu",
+      "Zhuo Liu",
+      "Runmin Ma",
+      "Chunjiang Mu",
+      "Haoyang Peng",
+      "Tianshuo Peng",
+      "Jinxin Shi",
+      "Luohe Shi",
+      "Boyuan Sun",
+      "Zelin Tan",
+      "Shengji Tang",
+      "Qianyi Wang",
+      "Yiming Wu",
+      "Yi Xie",
+      "Xiangchao Yan",
+      "Jingqi Ye",
+      "Peng Ye",
+      "Fangchen Yu",
+      "Jiakang Yuan",
+      "Bihao Zhan",
+      "Bo Zhang",
+      "Chen Zhang",
+      "Shufei Zhang",
+      "Shuaiyu Zhang",
+      "Wenlong Zhang",
+      "Yiqun Zhang",
+      "Junpeng Zhao",
+      "Zhijie Zhong",
+      "Bowen Zhou",
+      "Yuhao Zhou"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.30616",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30616",
+    "pdf_url": "https://arxiv.org/pdf/2606.30616",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.30111",
+    "title": "Automating the Design of Embodied Agent Architectures",
+    "url": "https://arxiv.org/abs/2606.30111",
+    "published": "2026-06-29",
+    "updated": "2026-07-03",
+    "authors": [
+      "Jian Zhou",
+      "Sihao Lin",
+      "Jin Li",
+      "Shuai Fu",
+      "Gengze Zhou",
+      "Qi Wu"
+    ],
+    "categories": [
+      "cs.RO",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "embodied-agent",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.30111",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.30111",
+    "pdf_url": "https://arxiv.org/pdf/2606.30111",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.29495",
+    "title": "Cognitive World Models for Process-Level Social Influence Evaluation",
+    "url": "https://arxiv.org/abs/2606.29495",
+    "published": "2026-06-28",
+    "updated": "2026-06-28",
+    "authors": [
+      "Minghui Ma",
+      "Bin Guo",
+      "Han Wang",
+      "Mengqi Chen",
+      "Jingqi Liu",
+      "Yan Liu",
+      "Zhiwen Yu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent",
+      "world-model"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.29495",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.29495",
+    "pdf_url": "https://arxiv.org/pdf/2606.29495",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.28896",
+    "title": "A Task-Driven and Quality-Assured Agent Framework for SAR Data Generation",
+    "url": "https://arxiv.org/abs/2606.28896",
+    "published": "2026-06-27",
+    "updated": "2026-06-27",
+    "authors": [
+      "Xuanting Wu",
+      "Fan Zhanga",
+      "Fei Ma",
+      "Ling Guan",
+      "Guochun Ma",
+      "Yongsheng Zhou"
+    ],
+    "categories": [
+      "eess.IV",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.28896",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28896",
+    "pdf_url": "https://arxiv.org/pdf/2606.28896",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.28279",
+    "title": "Agentic Hardware Design as Repository-Level Code Evolution",
+    "url": "https://arxiv.org/abs/2606.28279",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Cunxi Yu",
+      "Chenhui Deng",
+      "Nathaniel Pinckney",
+      "Brucek Khailany"
+    ],
+    "categories": [
+      "cs.AR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.28279",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28279",
+    "pdf_url": "https://arxiv.org/pdf/2606.28279",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.28430",
+    "title": "Building to the Test: Coding Agents Deliver What You Check, Not What You Requested",
+    "url": "https://arxiv.org/abs/2606.28430",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Yanuo Ma",
+      "Ben Kereopa-Yorke",
+      "Ben Schultz"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.28430",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28430",
+    "pdf_url": "https://arxiv.org/pdf/2606.28430",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.28187",
+    "title": "GBC: Gradient-Based Connections for Optimizing Multi-Agent Systems",
+    "url": "https://arxiv.org/abs/2606.28187",
+    "published": "2026-06-26",
+    "updated": "2026-06-26",
+    "authors": [
+      "Xiaocheng Yang",
+      "Abdulrahman Alrabah",
+      "Dilek Hakkani-Tür",
+      "Gokhan Tur"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "multi-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "multi-agent-llm"
+    ],
+    "arxiv_id": "2606.28187",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28187",
+    "pdf_url": "https://arxiv.org/pdf/2606.28187",
+    "primary_query": "multi-agent-llm"
+  },
+  {
+    "id": "2606.27416",
+    "title": "Glite ARF: Verifier-Driven Research with Parallel LLM Coding Agents",
+    "url": "https://arxiv.org/abs/2606.27416",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Vassili Philippov",
+      "Pavel Katunin",
+      "Dmitry Andreev",
+      "Igor Ostanin",
+      "Anton Nikolaev"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.SE"
+    ],
+    "topics": [
+      "coding-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.27416",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.27416",
+    "pdf_url": "https://arxiv.org/pdf/2606.27416",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.26649",
+    "title": "Autoformalization of Agent Instructions into Policy-as-Code",
+    "url": "https://arxiv.org/abs/2606.26649",
+    "published": "2026-06-25",
+    "updated": "2026-06-25",
+    "authors": [
+      "Adam Mondl",
+      "Matthew Maisel",
+      "John H. Brock"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2606.26649",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26649",
+    "pdf_url": "https://arxiv.org/pdf/2606.26649",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.26057",
+    "title": "The Unfireable Safety Kernel: Execution-Time AI Alignment for AI Agents and Other Escapable AI Systems",
+    "url": "https://arxiv.org/abs/2606.26057",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Seth Dobrin",
+      "Łukasz Chmiel"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CR",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.26057",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26057",
+    "pdf_url": "https://arxiv.org/pdf/2606.26057",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.26356",
+    "title": "Instruction Bleed: Cross-Module Interference in Prompt-Composed Agentic Systems",
+    "url": "https://arxiv.org/abs/2606.26356",
+    "published": "2026-06-24",
+    "updated": "2026-06-24",
+    "authors": [
+      "Ching-Yu Lin",
+      "Yifan Liu"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.IR",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.26356",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.26356",
+    "pdf_url": "https://arxiv.org/pdf/2606.26356",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.25519",
+    "title": "Quantization Inflates Reasoning: Token Inflation as a Hidden Cost of Low-Bit Reasoning Models",
+    "url": "https://arxiv.org/abs/2606.25519",
+    "published": "2026-06-24",
+    "updated": "2026-06-29",
+    "authors": [
+      "Xinyu Lian",
+      "Walid Krichene",
+      "Beichen Huang",
+      "Masahiro Tanaka",
+      "Olatunji Ruwase",
+      "Li Zhang",
+      "Minjia Zhang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.25519",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25519",
+    "pdf_url": "https://arxiv.org/pdf/2606.25519",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.25195",
+    "title": "SoK: AI Secure Code Generation: Progress, Pitfalls, and Paths Forward",
+    "url": "https://arxiv.org/abs/2606.25195",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Rupam Patir",
+      "Keyan Guo",
+      "Haipeng Cai",
+      "Hongxin Hu"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.25195",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25195",
+    "pdf_url": "https://arxiv.org/pdf/2606.25195",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.24416",
+    "title": "Agentic AI for Bilevel Long-Term Optimization of Policy-Driven Physical Layer Systems",
+    "url": "https://arxiv.org/abs/2606.24416",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Bingnan Xiao",
+      "Chenhao Yang",
+      "Wei Ni",
+      "Xin Wang",
+      "Tony Q. S. Quek"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "multi-agent",
+      "rag"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.24416",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24416",
+    "pdf_url": "https://arxiv.org/pdf/2606.24416",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.24855",
+    "title": "OpenThoughts-Agent: Data Recipes for Agentic Models",
+    "url": "https://arxiv.org/abs/2606.24855",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Negin Raoof",
+      "Richard Zhuang",
+      "Marianna Nezhurina",
+      "Etash Guha",
+      "Atula Tejaswi",
+      "Ryan Marten",
+      "Charlie F. Ruan",
+      "Tyler Griggs",
+      "Alexander Glenn Shaw",
+      "Hritik Bansal",
+      "E. Kelly Buchanan",
+      "Artem Gazizov",
+      "Reinhard Heckel",
+      "Chinmay Hegde",
+      "Sankalp Jajee",
+      "Daanish Khazi",
+      "Emmanouil Koukoumidis",
+      "Xiangyi Li",
+      "Hange Liu",
+      "Shlok Natarajan",
+      "Harsh Raj",
+      "Nicholas Roberts",
+      "Ethan Shen",
+      "Nishad Singhi",
+      "Michael Siu",
+      "Ashima Suvarna",
+      "Hanwen Xing",
+      "Patrick Yubeaton",
+      "Robert Zhang",
+      "Leon Liangyu Chen",
+      "Xiaokun Chen",
+      "Steven Dillmann",
+      "Saadia Gabriel",
+      "Xunyi Jiang",
+      "Anurag Kashyap",
+      "Boxuan Li",
+      "Yein Park",
+      "Minh Pham",
+      "Sujay Sanghavi",
+      "Lin Shi",
+      "Ke Sun",
+      "Yixin Wang",
+      "Zhiwei Xu",
+      "Erica Zhang",
+      "Siyan Zhao",
+      "Wanjia Zhao",
+      "Jenia Jitsev",
+      "Alex Dimakis",
+      "Benjamin Feuer",
+      "Ludwig Schmidt"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.24855",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.24855",
+    "pdf_url": "https://arxiv.org/pdf/2606.24855",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.25191",
+    "title": "To Isolate or to Score? Model-Adaptive Assessment for Cost-Efficient Multi-Agent RAG",
+    "url": "https://arxiv.org/abs/2606.25191",
+    "published": "2026-06-23",
+    "updated": "2026-06-23",
+    "authors": [
+      "Jungseob Lee",
+      "Chanjun Park",
+      "Heuiseok Lim"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag",
+      "reasoning"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.25191",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.25191",
+    "pdf_url": "https://arxiv.org/pdf/2606.25191",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.23130",
+    "title": "Understanding the (In)Security of Vibe-Coded Applications",
+    "url": "https://arxiv.org/abs/2606.23130",
+    "published": "2026-06-22",
+    "updated": "2026-06-23",
+    "authors": [
+      "Junquan Deng",
+      "Zhiyu Fan",
+      "Ruijie Meng"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "memory",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.23130",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.23130",
+    "pdf_url": "https://arxiv.org/pdf/2606.23130",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.22953",
+    "title": "Plans Don't Persist: Why Context Management Is Load Bearing for LLM Agents",
+    "url": "https://arxiv.org/abs/2606.22953",
+    "published": "2026-06-22",
+    "updated": "2026-06-22",
+    "authors": [
+      "Aman Mehta",
+      "Anupam Datta"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.22953",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.22953",
+    "pdf_url": "https://arxiv.org/pdf/2606.22953",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.21836",
+    "title": "AgentDSE: Reasoning-Augmented Architectural Design Space Exploration",
+    "url": "https://arxiv.org/abs/2606.21836",
+    "published": "2026-06-20",
+    "updated": "2026-06-20",
+    "authors": [
+      "Chenyu Wang",
+      "Jiahe Caroline Shi",
+      "David Kong",
+      "Duane S. Boning",
+      "Zishen Wan",
+      "Yilun Du",
+      "Vijay Janapa Reddi"
+    ],
+    "categories": [
+      "cs.AR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "reasoning"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.21836",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21836",
+    "pdf_url": "https://arxiv.org/pdf/2606.21836",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.21401",
+    "title": "SwarmX: Agentic Scheduling for Low-Latency Agentic Systems",
+    "url": "https://arxiv.org/abs/2606.21401",
+    "published": "2026-06-19",
+    "updated": "2026-06-28",
+    "authors": [
+      "Yeqi Huang",
+      "Yanwei Ye",
+      "Guomin Chen",
+      "Wenhao Su",
+      "Bin Gong",
+      "Jialian Li",
+      "Zhan Lu",
+      "Yangshen Deng",
+      "Xuan Sun",
+      "Le Xu",
+      "Luo Mai"
+    ],
+    "categories": [
+      "cs.DC",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai"
+    ],
+    "arxiv_id": "2606.21401",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21401",
+    "pdf_url": "https://arxiv.org/pdf/2606.21401",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.21228",
+    "title": "Sakana Fugu Technical Report",
+    "url": "https://arxiv.org/abs/2606.21228",
+    "published": "2026-06-19",
+    "updated": "2026-06-23",
+    "authors": [
+      "Yujin Tang",
+      "Edoardo Cetin",
+      "Jinglue Xu",
+      "Qi Sun",
+      "Stefan Nielsen",
+      "Vincent Richard",
+      "Haruto Goda",
+      "Iaroslav Tymchenko",
+      "Nhan Nguyen",
+      "Hyunin Lee",
+      "Mari Ashiga",
+      "Shashank Kotyan",
+      "So Kuroki",
+      "Tarin Clanuwat"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "coding-agent",
+      "multi-agent",
+      "rag",
+      "reasoning"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "coding-agent"
+    ],
+    "arxiv_id": "2606.21228",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.21228",
+    "pdf_url": "https://arxiv.org/pdf/2606.21228",
+    "primary_query": "coding-agent"
+  },
+  {
+    "id": "2606.20510",
+    "title": "Efficient and Sound Probabilistic Verification for AI Agents",
+    "url": "https://arxiv.org/abs/2606.20510",
+    "published": "2026-06-18",
+    "updated": "2026-06-18",
+    "authors": [
+      "Alaia Solko-Breslin",
+      "Pramod Kaushik Mudrakarta",
+      "Mihai Christodorescu",
+      "Somesh Jha",
+      "Krishnamurthy Dj Dvijotham"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent"
+    ],
+    "arxiv_id": "2606.20510",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.20510",
+    "pdf_url": "https://arxiv.org/pdf/2606.20510",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.19242",
+    "title": "Runtime Compliance Verification for AI Agents",
+    "url": "https://arxiv.org/abs/2606.19242",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Nafiseh Kahani",
+      "Masoud Barati",
+      "Diana Addae"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "ai-agent",
+      "function-calling",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.19242",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.19242",
+    "pdf_url": "https://arxiv.org/pdf/2606.19242",
+    "primary_query": "ai-agent"
+  },
+  {
+    "id": "2606.28374",
+    "title": "Recursive Self-Evolving Agents via Held-Out Selection",
+    "url": "https://arxiv.org/abs/2606.28374",
+    "published": "2026-06-17",
+    "updated": "2026-06-17",
+    "authors": [
+      "Michael Nguyen",
+      "Quoc Nguyen",
+      "Paul Vuong"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.28374",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.28374",
+    "pdf_url": "https://arxiv.org/pdf/2606.28374",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.18363",
+    "title": "Guava: An Effective and Universal Harness for Embodied Manipulation",
+    "url": "https://arxiv.org/abs/2606.18363",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Haowen Liu",
+      "Xirui Li",
+      "Shaoxiong Yao",
+      "Peng Shi",
+      "Tianyi Zhou",
+      "Jia-Bin Huang",
+      "Furong Huang",
+      "Jiayuan Mao"
+    ],
+    "categories": [
+      "cs.RO",
+      "cs.AI"
+    ],
+    "topics": [
+      "embodied-agent",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agentic-ai",
+      "tool-use"
+    ],
+    "arxiv_id": "2606.18363",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18363",
+    "pdf_url": "https://arxiv.org/pdf/2606.18363",
+    "primary_query": "agentic-ai"
+  },
+  {
+    "id": "2606.17453",
+    "title": "MapSatisfyBench: Benchmarking Satisfaction-Aware Map Agents through Behavior-Grounded Implicit Decision Factors",
+    "url": "https://arxiv.org/abs/2606.17453",
+    "published": "2026-06-16",
+    "updated": "2026-06-17",
+    "authors": [
+      "Lubin Bai",
+      "Mengyu Cao",
+      "Sixue Wang",
+      "Zhongwei Wan",
+      "Yue Pan",
+      "Jiale Hou",
+      "Xiang Li",
+      "Xiuyuan Zhang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.17453",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17453",
+    "pdf_url": "https://arxiv.org/pdf/2606.17453",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.18023",
+    "title": "LoopCoder-v2: Only Loop Once for Efficient Test-Time Computation Scaling",
+    "url": "https://arxiv.org/abs/2606.18023",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Jian Yang",
+      "Shawn Guo",
+      "Wei Zhang",
+      "Tianyu Zheng",
+      "Yaxin Du",
+      "Haau-Sing Li",
+      "Jiajun Wu",
+      "Yue Song",
+      "Yan Xing",
+      "Qingsong Cai",
+      "Zelong Huang",
+      "Chuan Hao",
+      "Ran Tao",
+      "Xianglong Liu",
+      "Wayne Xin Zhao",
+      "Mingjie Tang",
+      "Weifeng Lv",
+      "Ming Zhou",
+      "Bryan Dai"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "memory",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.18023",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18023",
+    "pdf_url": "https://arxiv.org/pdf/2606.18023",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.17383",
+    "title": "Model Validation of Agentic AI Systems: A POMDP-Based Framework for Belief-State, Forecast, and Policy Validation",
+    "url": "https://arxiv.org/abs/2606.17383",
+    "published": "2026-06-16",
+    "updated": "2026-06-16",
+    "authors": [
+      "Matthew Francis Dixon"
+    ],
+    "categories": [
+      "q-fin.RM",
+      "cs.AI",
+      "cs.LG",
+      "stat.ML"
+    ],
+    "topics": [
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.17383",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17383",
+    "pdf_url": "https://arxiv.org/pdf/2606.17383",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.16813",
+    "title": "GIST-CMTF: Goal-State Inference for Causal Minimal Tool Filtering in LLM Agents",
+    "url": "https://arxiv.org/abs/2606.16813",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Rahul Suresh Babu",
+      "Rohit Shukla"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.16813",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16813",
+    "pdf_url": "https://arxiv.org/pdf/2606.16813",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.16839",
+    "title": "Towards LLM Accelerated Rapid Reviews for Software Tool Discovery -- Case for Log Anomaly Detection",
+    "url": "https://arxiv.org/abs/2606.16839",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Jesse Nyyssölä",
+      "Hamza Bin Mazhar",
+      "Alexander Bakhtin",
+      "Matteo Esposito",
+      "Nana Reinikainen",
+      "Yuqing Wang",
+      "Ying Song",
+      "Davide Taibi",
+      "Mika Mäntylä"
+    ],
+    "categories": [
+      "cs.SE"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.16839",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16839",
+    "pdf_url": "https://arxiv.org/pdf/2606.16839",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.16481",
+    "title": "Steering Emotional Dynamics for Art Therapy: Controllable Narrative Script Generation through Hierarchically Guided LLM Agents",
+    "url": "https://arxiv.org/abs/2606.16481",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Suqing Wang",
+      "Qinghai Miao",
+      "Chao Guo",
+      "Yisheng Lv"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.16481",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.16481",
+    "pdf_url": "https://arxiv.org/pdf/2606.16481",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.17368",
+    "title": "Distributed General-Purpose Agent Networks: Architecture, Key Mechanisms, and Prototypes",
+    "url": "https://arxiv.org/abs/2606.17368",
+    "published": "2026-06-15",
+    "updated": "2026-06-15",
+    "authors": [
+      "Shengli Zhang",
+      "Deen Ma",
+      "Zibin Lin",
+      "Taotao Wang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.NI"
+    ],
+    "topics": [
+      "computer-use",
+      "multi-agent",
+      "planning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.17368",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.17368",
+    "pdf_url": "https://arxiv.org/pdf/2606.17368",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.15709",
+    "title": "AI-Driven Framework for Adaptive Water Network Management with Proof-of-Concept Implementation: Addressing Non-Revenue Water in Jordan",
+    "url": "https://arxiv.org/abs/2606.15709",
+    "published": "2026-06-14",
+    "updated": "2026-06-14",
+    "authors": [
+      "Mohammed Fasha",
+      "Nahel Al-Maayta",
+      "Bilal Sowan",
+      "Mohammad Athamneh",
+      "Husam Barham"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "rag",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling",
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.15709",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15709",
+    "pdf_url": "https://arxiv.org/pdf/2606.15709",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2606.15874",
+    "title": "LLM-as-Code: Agentic Programming for Agent Harness",
+    "url": "https://arxiv.org/abs/2606.15874",
+    "published": "2026-06-14",
+    "updated": "2026-06-22",
+    "authors": [
+      "Junjia Qi",
+      "Zichuan Fu",
+      "Jingtong Gao",
+      "Wenlin Zhang",
+      "Hanyu Yan",
+      "Xian Wu",
+      "Xiangyu Zhao"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.SE"
+    ],
+    "topics": [
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.15874",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15874",
+    "pdf_url": "https://arxiv.org/pdf/2606.15874",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.15906",
+    "title": "MAGE-RAG: Multigranular Adaptive Graph Evidence for Agentic Multimodal RAG in Long-Document QA",
+    "url": "https://arxiv.org/abs/2606.15906",
+    "published": "2026-06-14",
+    "updated": "2026-06-14",
+    "authors": [
+      "Yilong Zuo",
+      "Xunkai Li",
+      "Jing Yuan",
+      "Qiangqiang Dai",
+      "Hongchao Qin",
+      "Ronghua Li"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI",
+      "cs.CL",
+      "cs.DB",
+      "cs.MM"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "rag"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.15906",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15906",
+    "pdf_url": "https://arxiv.org/pdf/2606.15906",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.15994",
+    "title": "Agentic Framework for Deep Learning workload migration via In-Context Learning",
+    "url": "https://arxiv.org/abs/2606.15994",
+    "published": "2026-06-14",
+    "updated": "2026-06-14",
+    "authors": [
+      "Qiyue Liang",
+      "Steven Ingram",
+      "George Vanica",
+      "Andi Gavrilescu",
+      "Newfel Harrat",
+      "Hassan Sipra",
+      "Sethuraman Sankaran"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-safety",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.15994",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15994",
+    "pdf_url": "https://arxiv.org/pdf/2606.15994",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.15034",
+    "title": "OSGuard: A Benchmark for Safety in Computer-Use Agents",
+    "url": "https://arxiv.org/abs/2606.15034",
+    "published": "2026-06-13",
+    "updated": "2026-06-13",
+    "authors": [
+      "Mina Mohammadmirzaei",
+      "Jeffrey Flanigan"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.15034",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.15034",
+    "pdf_url": "https://arxiv.org/pdf/2606.15034",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.12837",
+    "title": "LoHoSearch: Benchmarking Long-Horizon Search Agents Beyond the Human Difficulty Ceiling",
+    "url": "https://arxiv.org/abs/2606.12837",
+    "published": "2026-06-11",
+    "updated": "2026-06-17",
+    "authors": [
+      "Jiarui Zhao",
+      "Rongzhi Zhang",
+      "Lingchuan Liu",
+      "Hao Yang",
+      "Xunliang Cai",
+      "Xi Su"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.12837",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12837",
+    "pdf_url": "https://arxiv.org/pdf/2606.12837",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.13663",
+    "title": "HyperTool: Beyond Step-Wise Tool Calls for Tool-Augmented Agents",
+    "url": "https://arxiv.org/abs/2606.13663",
+    "published": "2026-06-11",
+    "updated": "2026-06-11",
+    "authors": [
+      "Yaxin Du",
+      "Yifan Zhou",
+      "Yujie Ge",
+      "Jiajun Wang",
+      "Xianghe Pang",
+      "Shuo Tang",
+      "Tuney Zheng",
+      "Bryan Dai",
+      "Jian Yang",
+      "Siheng Chen"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.13663",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.13663",
+    "pdf_url": "https://arxiv.org/pdf/2606.13663",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.12634",
+    "title": "Keep Policy Gradient in Charge: Sibling-Guided Credit Distillation for Long-Horizon Tool-Use Agents",
+    "url": "https://arxiv.org/abs/2606.12634",
+    "published": "2026-06-10",
+    "updated": "2026-06-29",
+    "authors": [
+      "Tianyu Ding",
+      "Jianhong Xin",
+      "Juan Pablo De la Cruz Weinstein"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "tool-use"
+    ],
+    "arxiv_id": "2606.12634",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.12634",
+    "pdf_url": "https://arxiv.org/pdf/2606.12634",
+    "primary_query": "tool-use"
+  },
+  {
+    "id": "2606.11688",
+    "title": "Goal-Autopilot: A Verifiable Anti-Fabrication Firewall for Unattended Long-Horizon Agents",
+    "url": "https://arxiv.org/abs/2606.11688",
+    "published": "2026-06-10",
+    "updated": "2026-06-10",
+    "authors": [
+      "Youwang Deng"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "rag"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.11688",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.11688",
+    "pdf_url": "https://arxiv.org/pdf/2606.11688",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.10394",
+    "title": "STAGE-Claw: Automated State-based Agent Benchmarking for Realistic Scenarios",
+    "url": "https://arxiv.org/abs/2606.10394",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Sirui Liang",
+      "Bohan Yu",
+      "Peiyu Wang",
+      "Shiguang Guo",
+      "Wenxing Hu",
+      "Pengfei Cao",
+      "Jian Zhao",
+      "Cao Liu",
+      "Ke Zeng",
+      "Xunliang Cai",
+      "Kang Liu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.10394",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10394",
+    "pdf_url": "https://arxiv.org/pdf/2606.10394",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.10532",
+    "title": "ActiveMem: Distributed Active Memory for Long-Horizon LLM Reasoning",
+    "url": "https://arxiv.org/abs/2606.10532",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Yunhan Jiang",
+      "Wenbin Duan",
+      "Shasha Guo",
+      "Liang Pang",
+      "Xiaoqian Sun",
+      "Huawei Shen"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.10532",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.10532",
+    "pdf_url": "https://arxiv.org/pdf/2606.10532",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.11176",
+    "title": "Data Journalist Agent: Transforming Data into Verifiable Multimodal Stories",
+    "url": "https://arxiv.org/abs/2606.11176",
+    "published": "2026-06-09",
+    "updated": "2026-06-09",
+    "authors": [
+      "Kevin Qinghong Lin",
+      "Batu EI",
+      "Yuhong Shi",
+      "Pan Lu",
+      "Philip Torr",
+      "James Zou"
+    ],
+    "categories": [
+      "cs.CV",
+      "cs.CL",
+      "cs.CY",
+      "cs.HC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "multi-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.11176",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.11176",
+    "pdf_url": "https://arxiv.org/pdf/2606.11176",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.08960",
+    "title": "Hardening Agent Benchmarks with Adversarial Hacker-Fixer Loops",
+    "url": "https://arxiv.org/abs/2606.08960",
+    "published": "2026-06-08",
+    "updated": "2026-06-08",
+    "authors": [
+      "Ziqian Zhong",
+      "Ivgeni Segal",
+      "Ivan Bercovich",
+      "Shashwat Saxena",
+      "Kexun Zhang",
+      "Aditi Raghunathan"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.LG",
+      "cs.MA"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "reasoning"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.08960",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.08960",
+    "pdf_url": "https://arxiv.org/pdf/2606.08960",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.09447",
+    "title": "AliyunConsoleAgent: Training Web Agents in Real-World Cloud Environments via Distillation and Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2606.09447",
+    "published": "2026-06-08",
+    "updated": "2026-06-08",
+    "authors": [
+      "Bojie Rong",
+      "Zheyu Shen",
+      "Qiaoping Wang",
+      "Pengfei Kang",
+      "Yang Xu",
+      "Yawen Wei",
+      "Hanyu Wu",
+      "Zhi Zhao",
+      "Leihao Pei",
+      "Linquan Jiang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.09447",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09447",
+    "pdf_url": "https://arxiv.org/pdf/2606.09447",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.09426",
+    "title": "WeaveBench: A Long-Horizon, Real-World Benchmark for Computer-Use Agents with Hybrid Interfaces",
+    "url": "https://arxiv.org/abs/2606.09426",
+    "published": "2026-06-08",
+    "updated": "2026-07-06",
+    "authors": [
+      "Wanli Li",
+      "Bowen Zhou",
+      "Yunyao Yu",
+      "Zhou Xu",
+      "Yifan Yang",
+      "Dongsheng Li",
+      "Caihua Shan"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "web-gui-agent"
+    ],
+    "arxiv_id": "2606.09426",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09426",
+    "pdf_url": "https://arxiv.org/pdf/2606.09426",
+    "primary_query": "web-gui-agent"
+  },
+  {
+    "id": "2606.09316",
+    "title": "Anything2Skill: Compiling External Knowledge into Reusable Skills for Agents",
+    "url": "https://arxiv.org/abs/2606.09316",
+    "published": "2026-06-08",
+    "updated": "2026-06-19",
+    "authors": [
+      "Qianjun Pan",
+      "Yutao Yang",
+      "Junsong Li",
+      "Jie Zhou",
+      "Kai Chen",
+      "Xin Li",
+      "Qin Chen",
+      "Liang He"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "memory",
+      "planning",
+      "rag",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.09316",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09316",
+    "pdf_url": "https://arxiv.org/pdf/2606.09316",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.09961",
+    "title": "3SPO: State-Score-Supervised Policy Optimization for LLM Agents",
+    "url": "https://arxiv.org/abs/2606.09961",
+    "published": "2026-06-08",
+    "updated": "2026-06-08",
+    "authors": [
+      "Yu Han",
+      "Kailing Li",
+      "Yang Jiao",
+      "Yulin Dai",
+      "Yuqian Fu",
+      "Linhai Zhuo",
+      "Tianwen Qian"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.09961",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.09961",
+    "pdf_url": "https://arxiv.org/pdf/2606.09961",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.08172",
+    "title": "The Governance of Human-LLM Interaction: Safety Gating, Civility Steering, and Affective Default Lock-In",
+    "url": "https://arxiv.org/abs/2606.08172",
+    "published": "2026-06-06",
+    "updated": "2026-06-06",
+    "authors": [
+      "Manuele Reani",
+      "Hongjian Zhang",
+      "Hongyu Tian"
+    ],
+    "categories": [
+      "cs.HC",
+      "cs.AI",
+      "cs.CY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "planning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.08172",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.08172",
+    "pdf_url": "https://arxiv.org/pdf/2606.08172",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.08162",
+    "title": "Silent Failure in LLM Agent Systems: The Entropy Principle and the Inevitable Disorder of Autonomous Agents",
+    "url": "https://arxiv.org/abs/2606.08162",
+    "published": "2026-06-06",
+    "updated": "2026-06-06",
+    "authors": [
+      "Dexing Liu"
+    ],
+    "categories": [
+      "cs.MA"
+    ],
+    "topics": [
+      "memory",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.08162",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.08162",
+    "pdf_url": "https://arxiv.org/pdf/2606.08162",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.07836",
+    "title": "Agentic multi-fidelity learning of quasiparticle and excitonic properties",
+    "url": "https://arxiv.org/abs/2606.07836",
+    "published": "2026-06-05",
+    "updated": "2026-06-05",
+    "authors": [
+      "Arnab Neogi",
+      "Aaron Forde",
+      "Christopher A. Lane",
+      "Sergei Tretiak",
+      "Jian-Xin Zhu"
+    ],
+    "categories": [
+      "cond-mat.mtrl-sci",
+      "cond-mat.stat-mech",
+      "cs.AI",
+      "physics.comp-ph",
+      "quant-ph"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.07836",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.07836",
+    "pdf_url": "https://arxiv.org/pdf/2606.07836",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.18272",
+    "title": "Mitigating Anchoring Bias in LLM-Based Agents for Energy-Efficient 6G Autonomous Networks",
+    "url": "https://arxiv.org/abs/2606.18272",
+    "published": "2026-06-05",
+    "updated": "2026-06-18",
+    "authors": [
+      "Hatim Chergui",
+      "Claudia Carballo González",
+      "Farhad Rezazadeh",
+      "Merouane Debbah"
+    ],
+    "categories": [
+      "cs.NI",
+      "cs.AI",
+      "eess.SY"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "multi-agent",
+      "reasoning"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.18272",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.18272",
+    "pdf_url": "https://arxiv.org/pdf/2606.18272",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2606.05658",
+    "title": "Agent-Orchestrated Adaptive RAG: A Comparative Study on Structured and Multi-Hop Retrieval",
+    "url": "https://arxiv.org/abs/2606.05658",
+    "published": "2026-06-04",
+    "updated": "2026-06-04",
+    "authors": [
+      "Anuj Maharjan",
+      "Devinder Kaur",
+      "Richard Molyet"
+    ],
+    "categories": [
+      "cs.IR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "reasoning"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.05658",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.05658",
+    "pdf_url": "https://arxiv.org/pdf/2606.05658",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.05622",
+    "title": "AdaPlanBench: Evaluating Adaptive Planning in Large Language Model Agents under World and User Constraints",
+    "url": "https://arxiv.org/abs/2606.05622",
+    "published": "2026-06-04",
+    "updated": "2026-06-04",
+    "authors": [
+      "Jiayu Liu",
+      "Cheng Qian",
+      "Zhenhailong Wang",
+      "Bingxuan Li",
+      "Jiateng Liu",
+      "Heng Wang",
+      "Jeonghwan Kim",
+      "Yumeng Wang",
+      "Xiusi Chen",
+      "Yi R. Fung",
+      "Heng Ji"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2606.05622",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.05622",
+    "pdf_url": "https://arxiv.org/pdf/2606.05622",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2606.05436",
+    "title": "Ten Headache Specialists versus Artificial Intelligence for Clinical Literature Summarization: A Critical Evaluation and Comparison",
+    "url": "https://arxiv.org/abs/2606.05436",
+    "published": "2026-06-03",
+    "updated": "2026-06-03",
+    "authors": [
+      "Alejandro Lozano",
+      "Keiko Ihara",
+      "Ping-Hao Yang",
+      "Carrie E. Robertson",
+      "Jennifer Stern",
+      "Allan Purdy",
+      "Hsiangkuo Yuan",
+      "Pengfei Zhang",
+      "Yulia Orlova",
+      "Olga Fermo",
+      "Jennifer Hranilovich",
+      "Fred Cohen",
+      "Todd J. Schwedt",
+      "Jenelle A. Jindal",
+      "Serena Yeung-Levy",
+      "Chia-Chun Chiang"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.05436",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.05436",
+    "pdf_url": "https://arxiv.org/pdf/2606.05436",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2606.03544",
+    "title": "SAGE: A Quantitative Evaluation of Socialized Evolution in Agent Ecosystems",
+    "url": "https://arxiv.org/abs/2606.03544",
+    "published": "2026-06-02",
+    "updated": "2026-06-02",
+    "authors": [
+      "Linyue Pan",
+      "Yaoming Zhu",
+      "Lin Qiu",
+      "Xuezhi Cao",
+      "Xunliang Cai"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2606.03544",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.03544",
+    "pdf_url": "https://arxiv.org/pdf/2606.03544",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2606.03157",
+    "title": "ClinicalMC: A Benchmark for Multi-Course Clinical Decision-Making with Large Language Models",
+    "url": "https://arxiv.org/abs/2606.03157",
+    "published": "2026-06-02",
+    "updated": "2026-06-02",
+    "authors": [
+      "Ruihui Hou",
+      "Siyi Zhu",
+      "Ziyue Huai",
+      "Guangya Yu",
+      "Yongqi Fan",
+      "Chunming Wang",
+      "Tong Ruan"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "rag"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.03157",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.03157",
+    "pdf_url": "https://arxiv.org/pdf/2606.03157",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.01961",
+    "title": "AutoMedBench: Towards Medical AutoResearch with Agentic AI Models",
+    "url": "https://arxiv.org/abs/2606.01961",
+    "published": "2026-06-01",
+    "updated": "2026-06-03",
+    "authors": [
+      "Junqi Liu",
+      "Selena Song",
+      "Yuhan Wang",
+      "Jiawei Mao",
+      "Hardy Chen",
+      "Xiaoke Huang",
+      "Tianhao Qi",
+      "Pengfei Guo",
+      "Yucheng Tang",
+      "Yufan He",
+      "Can Zhao",
+      "Andriy Myronenko",
+      "Dong Yang",
+      "Daguang Xu",
+      "Yuyin Zhou"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.01961",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.01961",
+    "pdf_url": "https://arxiv.org/pdf/2606.01961",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.01185",
+    "title": "\"Skill issues'': data-centric optimization of lakehouse agents",
+    "url": "https://arxiv.org/abs/2606.01185",
+    "published": "2026-05-31",
+    "updated": "2026-05-31",
+    "authors": [
+      "Nicole Rose Schneider",
+      "Davide Ghilardi",
+      "Giacomo Piccinini",
+      "Jacopo Tagliabue"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2606.01185",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.01185",
+    "pdf_url": "https://arxiv.org/pdf/2606.01185",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2606.01138",
+    "title": "memorywire: A Vendor-Neutral Wire Format for Agent Memory Operations",
+    "url": "https://arxiv.org/abs/2606.01138",
+    "published": "2026-05-31",
+    "updated": "2026-06-03",
+    "authors": [
+      "Thamilvendhan Munirathinam"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.DC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2606.01138",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.01138",
+    "pdf_url": "https://arxiv.org/pdf/2606.01138",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.01166",
+    "title": "BraveGuard: From Open-World Threats to Safer Computer-Use Agents",
+    "url": "https://arxiv.org/abs/2606.01166",
+    "published": "2026-05-31",
+    "updated": "2026-06-02",
+    "authors": [
+      "Yunhao Feng",
+      "Xiaohu Du",
+      "Xinhao Deng",
+      "Yifan Ding",
+      "Ming Wen",
+      "Yixu Wang",
+      "Yuxiang Xie",
+      "Baihui Zheng",
+      "Yingshui Tan",
+      "Yige Li",
+      "Yutao Wu",
+      "Kerui Cao",
+      "Wenke Huang",
+      "Yanming Guo",
+      "Xingjun Ma",
+      "Yu-Gang Jiang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2606.01166",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.01166",
+    "pdf_url": "https://arxiv.org/pdf/2606.01166",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2606.00644",
+    "title": "ForeSci: Evaluating LLM Agents for Forward-Looking AI Research Judgment",
+    "url": "https://arxiv.org/abs/2606.00644",
+    "published": "2026-05-30",
+    "updated": "2026-06-04",
+    "authors": [
+      "Qiuyu Tian",
+      "Haojie Yin",
+      "Yingce Xia",
+      "Youyong Kong",
+      "Zequn Liu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2606.00644",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.00644",
+    "pdf_url": "https://arxiv.org/pdf/2606.00644",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.31308",
+    "title": "TraceGraph: Shared Decision Landscapes for Diagnosing and Improving Agent Trajectories",
+    "url": "https://arxiv.org/abs/2605.31308",
+    "published": "2026-05-29",
+    "updated": "2026-05-29",
+    "authors": [
+      "Junjie Nian",
+      "Kang Chen",
+      "Ge Zhang",
+      "Yixin Cao",
+      "Yugang Jiang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "embodied-agent",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation"
+    ],
+    "arxiv_id": "2605.31308",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.31308",
+    "pdf_url": "https://arxiv.org/pdf/2605.31308",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2605.31075",
+    "title": "Task-Focused Memorization for Multimodal Agents",
+    "url": "https://arxiv.org/abs/2605.31075",
+    "published": "2026-05-29",
+    "updated": "2026-05-29",
+    "authors": [
+      "Tao Zou",
+      "Yichen He",
+      "Tian Qiu",
+      "Yuan Lin",
+      "Hang Li"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "embodied-agent",
+      "memory"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.31075",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.31075",
+    "pdf_url": "https://arxiv.org/pdf/2605.31075",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2605.31268",
+    "title": "Mellum2 Technical Report",
+    "url": "https://arxiv.org/abs/2605.31268",
+    "published": "2026-05-29",
+    "updated": "2026-05-29",
+    "authors": [
+      "Marko Kojic",
+      "Ivan Bondyrev",
+      "Aral de Moor",
+      "Joseph Shtok",
+      "Petr Borovlev",
+      "Kseniia Lysaniuk",
+      "Madeeswaran Kannan",
+      "Ivan Dolgov",
+      "Nikita Pavlichenko"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2605.31268",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.31268",
+    "pdf_url": "https://arxiv.org/pdf/2605.31268",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2605.29653",
+    "title": "PTCG-Bench: Can LLM Agents Master Pokémon Trading Card Game?",
+    "url": "https://arxiv.org/abs/2605.29653",
+    "published": "2026-05-28",
+    "updated": "2026-05-28",
+    "authors": [
+      "Dongdong Hua",
+      "Yifei Sun",
+      "Renhong Huang",
+      "Feng Gao",
+      "Chunping Wang",
+      "Yang Yang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-evaluation",
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.29653",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.29653",
+    "pdf_url": "https://arxiv.org/pdf/2605.29653",
+    "primary_query": "agent-evaluation"
+  },
+  {
+    "id": "2605.29630",
+    "title": "Entity-Collision: A Stratified Protocol for Attributing Retrieval Lift in Agent Memory",
+    "url": "https://arxiv.org/abs/2605.29630",
+    "published": "2026-05-28",
+    "updated": "2026-05-28",
+    "authors": [
+      "Youwang Deng"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-memory"
+    ],
+    "arxiv_id": "2605.29630",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.29630",
+    "pdf_url": "https://arxiv.org/pdf/2605.29630",
+    "primary_query": "agent-memory"
+  },
+  {
+    "id": "2606.07591",
+    "title": "ResearchClawBench: A Benchmark for End-to-End Autonomous Scientific Research",
+    "url": "https://arxiv.org/abs/2606.07591",
+    "published": "2026-05-28",
+    "updated": "2026-07-03",
+    "authors": [
+      "Wanghan Xu",
+      "Shuo Li",
+      "Tianlin Ye",
+      "Qinglong Cao",
+      "Yixin Chen",
+      "Hengjian Gao",
+      "Yiheng Wang",
+      "Qi Li",
+      "Kun Li",
+      "Sheng Xu",
+      "Shengdu Chai",
+      "Fangchen Yu",
+      "Xiangyu Zhao",
+      "Zhangrui Zhao",
+      "Weijie Ma",
+      "Zijie Guo",
+      "Koutian Wu",
+      "Haoyu Zhou",
+      "Haoxiang Yin",
+      "Lixue Cheng",
+      "Chaofan Hu",
+      "Haoxuan Li",
+      "Lu Mi",
+      "Xuxuan Xie",
+      "Yifan Zhou",
+      "Ruizhe Chen",
+      "Zhiwang Zhou",
+      "Xingjian Guo",
+      "Yuhao Zhou",
+      "Xuming He",
+      "Shengyuan Xu",
+      "Xinyu Gu",
+      "Jiamin Wu",
+      "Mianxin Liu",
+      "Chunfeng Song",
+      "Fenghua Ling",
+      "Dongzhan Zhou",
+      "Shixiang Tang",
+      "Yuqiang Li",
+      "Mao Su",
+      "Peng Ye",
+      "Siqi Sun",
+      "Bin Wang",
+      "Xue Yang",
+      "Zhenfei Yin",
+      "Tianfan Fu",
+      "Guangtao Zhai",
+      "Wanli Ouyang",
+      "Bo Zhang",
+      "Lei Bai",
+      "Wenlong Zhang"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "rag"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2606.07591",
+    "source": "arxiv",
+    "source_id": "arxiv:2606.07591",
+    "pdf_url": "https://arxiv.org/pdf/2606.07591",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.28617",
+    "title": "LACUNA: Safe Agents as Recursive Program Holes",
+    "url": "https://arxiv.org/abs/2605.28617",
+    "published": "2026-05-27",
+    "updated": "2026-05-27",
+    "authors": [
+      "Yaoyu Zhao",
+      "Yichen Xu",
+      "Oliver Bračevac",
+      "Cao Nguyen Pham",
+      "Frank Zhengqing Wu",
+      "Martin Odersky"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.PL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.28617",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.28617",
+    "pdf_url": "https://arxiv.org/pdf/2605.28617",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2605.28424",
+    "title": "Skill0.5: Joint Skill Internalization and Utilization for Out-of-Distribution Generalization in Agentic Reinforcement Learning",
+    "url": "https://arxiv.org/abs/2605.28424",
+    "published": "2026-05-27",
+    "updated": "2026-05-27",
+    "authors": [
+      "Jiapeng Zhu",
+      "Jianxiang Yu",
+      "Yibo Zhao",
+      "Chengcheng Han",
+      "Qi Gu",
+      "Xunliang Cai",
+      "Xiang Li",
+      "Weining Qian"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-safety",
+      "memory"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.28424",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.28424",
+    "pdf_url": "https://arxiv.org/pdf/2605.28424",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.26497",
+    "title": "Aligning Provenance with Authorization: A Dual-Graph Defense for LLM Agents",
+    "url": "https://arxiv.org/abs/2605.26497",
+    "published": "2026-05-26",
+    "updated": "2026-05-26",
+    "authors": [
+      "Peiran Wang",
+      "Ying Li",
+      "Yuan Tian"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-safety",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2605.26497",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.26497",
+    "pdf_url": "https://arxiv.org/pdf/2605.26497",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2605.27123",
+    "title": "Rethinking Agentic RAG: Toward LLM-Driven Logical Retrieval Beyond Embeddings",
+    "url": "https://arxiv.org/abs/2605.27123",
+    "published": "2026-05-26",
+    "updated": "2026-05-26",
+    "authors": [
+      "Yuqi Zeng",
+      "Qixiang Deng",
+      "Yulei Wan",
+      "Ruiquan Jiang",
+      "Xiaoqing Zheng",
+      "Xuanjing Huang"
+    ],
+    "categories": [
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "rag"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.27123",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.27123",
+    "pdf_url": "https://arxiv.org/pdf/2605.27123",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.26165",
+    "title": "Tool-Schema Compression Enables Agentic RAG Under Constrained Context Budgets",
+    "url": "https://arxiv.org/abs/2605.26165",
+    "published": "2026-05-24",
+    "updated": "2026-05-24",
+    "authors": [
+      "Furkan Sakizli"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "rag-agent"
+    ],
+    "arxiv_id": "2605.26165",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.26165",
+    "pdf_url": "https://arxiv.org/pdf/2605.26165",
+    "primary_query": "rag-agent"
+  },
+  {
+    "id": "2605.23899",
+    "title": "From Raw Experience to Skill Consumption: A Systematic Study of Model-Generated Agent Skills",
+    "url": "https://arxiv.org/abs/2605.23899",
+    "published": "2026-05-22",
+    "updated": "2026-05-22",
+    "authors": [
+      "Zisu Huang",
+      "Jingwen Xu",
+      "Yifan Yang",
+      "Ziyang Gong",
+      "Qihao Yang",
+      "Muzhao Tian",
+      "Xiaohua Wang",
+      "Changze Lv",
+      "Xuemei Gao",
+      "Qi Dai",
+      "Bei Liu",
+      "Kai Qiu",
+      "Xue Yang",
+      "Dongdong Chen",
+      "Xiaoqing Zheng",
+      "Chong Luo"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "computer-use",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.23899",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.23899",
+    "pdf_url": "https://arxiv.org/pdf/2605.23899",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.17075",
+    "title": "A Red Teaming Framework for Evaluating Robustness of AI-enabled Security Orchestration, Automation, and Response Systems",
+    "url": "https://arxiv.org/abs/2605.17075",
+    "published": "2026-05-16",
+    "updated": "2026-05-16",
+    "authors": [
+      "Ayan Javeed Shaikh",
+      "Nathaniel D. Bastian",
+      "Ankit Shah"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use",
+      "workflow-agent",
+      "world-model"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.17075",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.17075",
+    "pdf_url": "https://arxiv.org/pdf/2605.17075",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.14322",
+    "title": "Are Agents Ready to Teach? A Multi-Stage Benchmark for Real-World Teaching Workflows",
+    "url": "https://arxiv.org/abs/2605.14322",
+    "published": "2026-05-14",
+    "updated": "2026-05-20",
+    "authors": [
+      "Zixin Chen",
+      "Peng Liu",
+      "Rui Sheng",
+      "Haobo Li",
+      "Jianhong Tu",
+      "Xiaodong Deng",
+      "Kashun Shum",
+      "Dayiheng Liu",
+      "Huamin Qu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.14322",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.14322",
+    "pdf_url": "https://arxiv.org/pdf/2605.14322",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.11928",
+    "title": "When Simulation Lies: A Sim-to-Real Benchmark and Domain-Randomized RL Recipe for Tool-Use Agents",
+    "url": "https://arxiv.org/abs/2605.11928",
+    "published": "2026-05-12",
+    "updated": "2026-05-12",
+    "authors": [
+      "Xiaolin Zhou",
+      "Aojie Yuan",
+      "Zheng Luo",
+      "Zipeng Ling",
+      "Xixiao Pan",
+      "Yicheng Gao",
+      "Haiyue Zhang",
+      "Jiate Li",
+      "Shuli Jiang",
+      "Prince Zizhuang Wang",
+      "Zixuan Zhu",
+      "Jinbo Liu",
+      "Ryan A. Rossi",
+      "Hua Wei",
+      "Xiyang Hu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling",
+      "language-agent"
+    ],
+    "arxiv_id": "2605.11928",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.11928",
+    "pdf_url": "https://arxiv.org/pdf/2605.11928",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2605.10870",
+    "title": "Remember the Decision, Not the Description: A Rate-Distortion Framework for Agent Memory",
+    "url": "https://arxiv.org/abs/2605.10870",
+    "published": "2026-05-11",
+    "updated": "2026-05-11",
+    "authors": [
+      "Mingxi Zou",
+      "Zhihan Guo",
+      "Langzhang Liang",
+      "Zhuo Wang",
+      "Qifan Wang",
+      "Qingsong Wen",
+      "Irwin King",
+      "Lizhen Qu",
+      "Zenglin Xu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "memory",
+      "planning"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2605.10870",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.10870",
+    "pdf_url": "https://arxiv.org/pdf/2605.10870",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2605.08876",
+    "title": "OTora: A Unified Red Teaming Framework for Reasoning-Level Denial-of-Service in LLM Agents",
+    "url": "https://arxiv.org/abs/2605.08876",
+    "published": "2026-05-09",
+    "updated": "2026-06-07",
+    "authors": [
+      "Xinyu Li",
+      "Ronghui Mu",
+      "Lin Li",
+      "Tianjin Huang",
+      "Gaojie Jin"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "computer-use",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2605.08876",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.08876",
+    "pdf_url": "https://arxiv.org/pdf/2605.08876",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2605.03328",
+    "title": "LLM-ADAM: A Generalizable LLM Agent Framework for Pre-Print Anomaly Detection in Additive Manufacturing",
+    "url": "https://arxiv.org/abs/2605.03328",
+    "published": "2026-05-05",
+    "updated": "2026-05-05",
+    "authors": [
+      "Ahmadreza Eslaminia",
+      "Chuhan Cai",
+      "Cameron Smith",
+      "Ruo-Syuan Mei",
+      "Shichen Li",
+      "Rajiv Malhotra",
+      "Klara Nahrstedt",
+      "Chenhui Shao"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "planning-agent"
+    ],
+    "arxiv_id": "2605.03328",
+    "source": "arxiv",
+    "source_id": "arxiv:2605.03328",
+    "pdf_url": "https://arxiv.org/pdf/2605.03328",
+    "primary_query": "planning-agent"
+  },
+  {
+    "id": "2604.27092",
+    "title": "End-to-end autonomous scientific discovery on a real optical platform",
+    "url": "https://arxiv.org/abs/2604.27092",
+    "published": "2026-04-29",
+    "updated": "2026-04-29",
+    "authors": [
+      "Shuxing Yang",
+      "Fujia Chen",
+      "Rui Zhao",
+      "Junyao Wu",
+      "Yize Wang",
+      "Haiyao Luo",
+      "Ning Han",
+      "Qiaolu Chen",
+      "Yuze Hu",
+      "Wenhao Li",
+      "Mingzhu Li",
+      "Hongsheng Chen",
+      "Yihao Yang"
+    ],
+    "categories": [
+      "cs.AI",
+      "physics.optics"
+    ],
+    "topics": [
+      "memory",
+      "planning",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "autonomous-agent-llm"
+    ],
+    "arxiv_id": "2604.27092",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.27092",
+    "pdf_url": "https://arxiv.org/pdf/2604.27092",
+    "primary_query": "autonomous-agent-llm"
+  },
+  {
+    "id": "2604.20994",
+    "title": "Breaking MCP with Function Hijacking Attacks: Novel Threats for Function Calling and Agentic Models",
+    "url": "https://arxiv.org/abs/2604.20994",
+    "published": "2026-04-22",
+    "updated": "2026-04-22",
+    "authors": [
+      "Yannis Belkhiter",
+      "Giulio Zizzo",
+      "Sergio Maffeis",
+      "Seshu Tirupathi",
+      "John D. Kelleher"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-safety",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2604.20994",
+    "source": "arxiv",
+    "source_id": "arxiv:2604.20994",
+    "pdf_url": "https://arxiv.org/pdf/2604.20994",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2603.28900",
+    "title": "Robust Multi-Agent Reinforcement Learning for Small UAS Separation Assurance under GPS Degradation and Spoofing",
+    "url": "https://arxiv.org/abs/2603.28900",
+    "published": "2026-03-30",
+    "updated": "2026-03-30",
+    "authors": [
+      "Alex Zongo",
+      "Filippos Fotiadis",
+      "Ufuk Topcu",
+      "Peng Wei"
+    ],
+    "categories": [
+      "cs.RO",
+      "cs.AI",
+      "cs.LG",
+      "eess.SY"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "multi-agent",
+      "world-model"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.28900",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.28900",
+    "pdf_url": "https://arxiv.org/pdf/2603.28900",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.25353",
+    "title": "SafeGuard ASF: SR Agentic Humanoid Robot System for Autonomous Industrial Safety",
+    "url": "https://arxiv.org/abs/2603.25353",
+    "published": "2026-03-26",
+    "updated": "2026-03-26",
+    "authors": [
+      "Thanh Nguyen Canh",
+      "Thang Tran Viet",
+      "Thanh Tuan Tran",
+      "Ben Wei Lim"
+    ],
+    "categories": [
+      "cs.RO"
+    ],
+    "topics": [
+      "agent-safety",
+      "embodied-agent",
+      "reasoning",
+      "tool-use",
+      "world-model"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.25353",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.25353",
+    "pdf_url": "https://arxiv.org/pdf/2603.25353",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.19684",
+    "title": "TSegAgent: Zero-Shot Tooth Segmentation via Geometry-Aware Vision-Language Agents",
+    "url": "https://arxiv.org/abs/2603.19684",
+    "published": "2026-03-20",
+    "updated": "2026-06-23",
+    "authors": [
+      "Shaojie Zhuang",
+      "Lu Yin",
+      "Guangshun Wei",
+      "Yunpeng Li",
+      "Xilu Wang",
+      "Yuanfeng Zhou"
+    ],
+    "categories": [
+      "cs.CV"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "coding-agent",
+      "rag",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.19684",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.19684",
+    "pdf_url": "https://arxiv.org/pdf/2603.19684",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.17392",
+    "title": "Agentic Cognitive Profiling: Realigning Automated Alzheimer's Disease Detection with Clinical Construct Validity",
+    "url": "https://arxiv.org/abs/2603.17392",
+    "published": "2026-03-18",
+    "updated": "2026-03-18",
+    "authors": [
+      "Jiawen Kang",
+      "Kun Li",
+      "Dongrui Han",
+      "Jinchao Li",
+      "Junan Li",
+      "Lingwei Meng",
+      "Xixin Wu",
+      "Helen Meng"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.IR",
+      "q-bio.NC"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2603.17392",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.17392",
+    "pdf_url": "https://arxiv.org/pdf/2603.17392",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2603.15666",
+    "title": "Compiled Memory: Not More Information, but More Precise Instructions for Language Agents",
+    "url": "https://arxiv.org/abs/2603.15666",
+    "published": "2026-03-12",
+    "updated": "2026-03-12",
+    "authors": [
+      "James Rhodes",
+      "George Kang"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "memory",
+      "rag"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.15666",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.15666",
+    "pdf_url": "https://arxiv.org/pdf/2603.15666",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2603.00801",
+    "title": "The Synthetic Web: Adversarially-Curated Mini-Internets for Diagnosing Epistemic Weaknesses of Language Agents",
+    "url": "https://arxiv.org/abs/2603.00801",
+    "published": "2026-02-28",
+    "updated": "2026-02-28",
+    "authors": [
+      "Shrey Shah",
+      "Levent Ozgur"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.IR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "embodied-agent",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2603.00801",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.00801",
+    "pdf_url": "https://arxiv.org/pdf/2603.00801",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2602.21127",
+    "title": "\"Are You Sure?\": An Empirical Study of Human Perception Vulnerability in LLM-Driven Agentic Systems",
+    "url": "https://arxiv.org/abs/2602.21127",
+    "published": "2026-02-24",
+    "updated": "2026-02-24",
+    "authors": [
+      "Xinfeng Li",
+      "Shenyu Dai",
+      "Kelong Zheng",
+      "Yue Xiao",
+      "Gelei Deng",
+      "Wei Dong",
+      "Xiaofeng Wang"
+    ],
+    "categories": [
+      "cs.HC",
+      "cs.AI",
+      "cs.CR",
+      "cs.SI"
+    ],
+    "topics": [
+      "agent-safety",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.21127",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.21127",
+    "pdf_url": "https://arxiv.org/pdf/2602.21127",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2603.00131",
+    "title": "Thought Virus: Viral Misalignment via Subliminal Prompting in Multi-Agent Systems",
+    "url": "https://arxiv.org/abs/2603.00131",
+    "published": "2026-02-23",
+    "updated": "2026-02-23",
+    "authors": [
+      "Moritz Weckbecker",
+      "Jonas Müller",
+      "Ben Hagag",
+      "Michael Mulet"
+    ],
+    "categories": [
+      "cs.MA",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-safety",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2603.00131",
+    "source": "arxiv",
+    "source_id": "arxiv:2603.00131",
+    "pdf_url": "https://arxiv.org/pdf/2603.00131",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2602.19008",
+    "title": "Capable but Unreliable: Canonical Path Deviation as a Causal Mechanism of Agent Failure in Long-Horizon Tasks",
+    "url": "https://arxiv.org/abs/2602.19008",
+    "published": "2026-02-22",
+    "updated": "2026-02-22",
+    "authors": [
+      "Wilson Y. Lee"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "language-agent"
+    ],
+    "arxiv_id": "2602.19008",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.19008",
+    "pdf_url": "https://arxiv.org/pdf/2602.19008",
+    "primary_query": "language-agent"
+  },
+  {
+    "id": "2602.14234",
+    "title": "REDSearcher: A Scalable and Cost-Efficient Framework for Long-Horizon Search Agents",
+    "url": "https://arxiv.org/abs/2602.14234",
+    "published": "2026-02-15",
+    "updated": "2026-02-15",
+    "authors": [
+      "Zheng Chu",
+      "Xiao Wang",
+      "Jack Hong",
+      "Huiming Fan",
+      "Yuqi Huang",
+      "Yue Yang",
+      "Guohai Xu",
+      "Chenxiao Zhao",
+      "Cheng Xiang",
+      "Shengchao Hu",
+      "Dongdong Kuang",
+      "Ming Liu",
+      "Bing Qin",
+      "Xing Yu"
+    ],
+    "categories": [
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "planning",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2602.14234",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.14234",
+    "pdf_url": "https://arxiv.org/pdf/2602.14234",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2602.14281",
+    "title": "MCPShield: A Security Cognition Layer for Adaptive Trust Calibration in Model Context Protocol Agents",
+    "url": "https://arxiv.org/abs/2602.14281",
+    "published": "2026-02-15",
+    "updated": "2026-02-24",
+    "authors": [
+      "Zhenhong Zhou",
+      "Yuanhe Zhang",
+      "Hongwei Cai",
+      "Moayad Aloqaily",
+      "Ouns Bouachir",
+      "Linsey Pang",
+      "Prakhar Mehrotra",
+      "Kun Wang",
+      "Qingsong Wen"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-safety",
+      "computer-use",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.14281",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.14281",
+    "pdf_url": "https://arxiv.org/pdf/2602.14281",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2602.13665",
+    "title": "HyFunc: Accelerating LLM-based Function Calls for Agentic AI through Hybrid-Model Cascade and Dynamic Templating",
+    "url": "https://arxiv.org/abs/2602.13665",
+    "published": "2026-02-14",
+    "updated": "2026-02-14",
+    "authors": [
+      "Weibin Liao",
+      "Jian-guang Lou",
+      "Haoyi Xiong"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2602.13665",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.13665",
+    "pdf_url": "https://arxiv.org/pdf/2602.13665",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2602.08082",
+    "title": "Spectral Guardrails for Agents in the Wild: Detecting Tool Use Hallucinations via Attention Topology",
+    "url": "https://arxiv.org/abs/2602.08082",
+    "published": "2026-02-08",
+    "updated": "2026-02-08",
+    "authors": [
+      "Valentin Noël"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI",
+      "eess.SP"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.08082",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.08082",
+    "pdf_url": "https://arxiv.org/pdf/2602.08082",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2602.10133",
+    "title": "AgentTrace: A Structured Logging Framework for Agent System Observability",
+    "url": "https://arxiv.org/abs/2602.10133",
+    "published": "2026-02-07",
+    "updated": "2026-02-07",
+    "authors": [
+      "Adam AlSayyad",
+      "Kelvin Yuxiang Huang",
+      "Richik Pal"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.10133",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.10133",
+    "pdf_url": "https://arxiv.org/pdf/2602.10133",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2602.05386",
+    "title": "Spider-Sense: Intrinsic Risk Sensing for Efficient Agent Defense with Hierarchical Adaptive Screening",
+    "url": "https://arxiv.org/abs/2602.05386",
+    "published": "2026-02-05",
+    "updated": "2026-02-06",
+    "authors": [
+      "Zhenxiong Yu",
+      "Zhi Yang",
+      "Zhiheng Jin",
+      "Shuhe Wang",
+      "Heng Zhang",
+      "Yanlin Fei",
+      "Lingfeng Zeng",
+      "Fangqi Lou",
+      "Shuo Zhang",
+      "Tu Hu",
+      "Jingping Liu",
+      "Rongze Chen",
+      "Xingyu Zhu",
+      "Kunyi Wang",
+      "Chaofa Yuan",
+      "Xin Guo",
+      "Zhaowei Liu",
+      "Feipeng Zhang",
+      "Jie Huang",
+      "Huacan Wang",
+      "Ronghao Chen",
+      "Liwen Zhang"
+    ],
+    "categories": [
+      "cs.CR",
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.05386",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.05386",
+    "pdf_url": "https://arxiv.org/pdf/2602.05386",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2602.03117",
+    "title": "AgentDyn: Are Your Agent Security Defenses Deployable in Real-World Dynamic Environments?",
+    "url": "https://arxiv.org/abs/2602.03117",
+    "published": "2026-02-03",
+    "updated": "2026-05-07",
+    "authors": [
+      "Hao Li",
+      "Ruoyao Wen",
+      "Shanghao Shi",
+      "Ning Zhang",
+      "Yevgeniy Vorobeychik",
+      "Chaowei Xiao"
+    ],
+    "categories": [
+      "cs.CR"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "planning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "agent-safety"
+    ],
+    "arxiv_id": "2602.03117",
+    "source": "arxiv",
+    "source_id": "arxiv:2602.03117",
+    "pdf_url": "https://arxiv.org/pdf/2602.03117",
+    "primary_query": "agent-safety"
+  },
+  {
+    "id": "2601.12988",
+    "title": "PaperGuide: Making Small Language-Model Paper-Reading Agents More Efficient",
+    "url": "https://arxiv.org/abs/2601.12988",
+    "published": "2026-01-19",
+    "updated": "2026-01-19",
+    "authors": [
+      "Zijian Wang",
+      "Tiancheng Huang",
+      "Hanqi Li",
+      "Da Ma",
+      "Lu Chen",
+      "Kai Yu"
+    ],
+    "categories": [
+      "cs.LG"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "planning",
+      "reasoning",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2601.12988",
+    "source": "arxiv",
+    "source_id": "arxiv:2601.12988",
+    "pdf_url": "https://arxiv.org/pdf/2601.12988",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2601.06606",
+    "title": "CEDAR: Context Engineering for Agentic Data Science",
+    "url": "https://arxiv.org/abs/2601.06606",
+    "published": "2026-01-10",
+    "updated": "2026-04-22",
+    "authors": [
+      "Rishiraj Saha Roy",
+      "Chris Hinze",
+      "Luzian Hahn",
+      "Fabian Kuech"
+    ],
+    "categories": [
+      "cs.LG",
+      "cs.AI"
+    ],
+    "topics": [
+      "planning",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2601.06606",
+    "source": "arxiv",
+    "source_id": "arxiv:2601.06606",
+    "pdf_url": "https://arxiv.org/pdf/2601.06606",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2512.23747",
+    "title": "State-of-the-art Small Language Coder Model: Mify-Coder",
+    "url": "https://arxiv.org/abs/2512.23747",
+    "published": "2025-12-26",
+    "updated": "2025-12-26",
+    "authors": [
+      "Abhinav Parmar",
+      "Abhisek Panigrahi",
+      "Abhishek Kumar Dwivedi",
+      "Abhishek Bhattacharya",
+      "Adarsh Ramachandra",
+      "Aditya Choudhary",
+      "Aditya Garg",
+      "Aditya Raj",
+      "Alankrit Bhatt",
+      "Alpesh Yadav",
+      "Anant Vishnu",
+      "Ananthu Pillai",
+      "Ankush Kumar",
+      "Aryan Patnaik",
+      "Aswatha Narayanan S",
+      "Avanish Raj Singh",
+      "Bhavya Shree Gadda",
+      "Brijesh Pankajbhai Kachhadiya",
+      "Buggala Jahnavi",
+      "Chidurala Nithin Krishna",
+      "Chintan Shah",
+      "Chunduru Akshaya",
+      "Debarshi Banerjee",
+      "Debrup Dey",
+      "Deepa R.",
+      "Deepika B G",
+      "Faiz ur Rahman",
+      "Gagan Gayari",
+      "Gudhi Jagadeesh Kumar Naidu",
+      "Gursimar Singh",
+      "Harshal Tyagi",
+      "Harshini K",
+      "James Mani Vathalloor",
+      "Jayarama Nettar",
+      "Jayashree Gajjam",
+      "Joe Walter Sugil George",
+      "Kamalakara Sri Krishna Tadepalli",
+      "Kamalkumar Rathinasamy",
+      "Karan Chaurasia",
+      "Karthikeyan S",
+      "Kashish Arora",
+      "Kaushal Desai",
+      "Khushboo Buwade",
+      "Kiran Manjrekar",
+      "Malikireddy Venkata Sai Likhitha",
+      "Manjunath A",
+      "Mitali Mahavir Bedmutha",
+      "Mohammed Rafee Tarafdar",
+      "Nikhil Tiwari",
+      "Nikitha K Gigi",
+      "Pavan Ravikumar",
+      "Pendyala Swarnanjali",
+      "Piyush Anand",
+      "Prakash Chandrasekar",
+      "Prasanna Bhalchandra Gawade",
+      "Prasanth Sivan",
+      "Preeti Khurana",
+      "Priyanshi Babbar",
+      "Rajab Ali Mondal",
+      "Rajesh Kumar Vissapragada",
+      "Rajeshwari Ganesan",
+      "Rajeswari Koppisetti",
+      "Ramjee R.",
+      "Ramkumar Thiruppathisamy",
+      "Rani G. S.",
+      "S Reka",
+      "Samarth Gupta",
+      "Sandeep Reddy Kothakota",
+      "Sarathy K",
+      "Sathyanarayana Sampath Kumar",
+      "Saurabh Kumar",
+      "Shashank Khasare",
+      "Shenbaga Devi Venkatesh Kumar",
+      "Shiva Rama Krishna Parvatham",
+      "Shoeb Shaikh",
+      "Shrishanmathi A",
+      "Shubham Pathak",
+      "Sree Samhita Koppaka",
+      "Sreenivasa Raghavan K S",
+      "Sreeram Venkatasubramanian",
+      "Suprabha Desai Bojja",
+      "Swetha R",
+      "Syed Ahmed",
+      "Chinmai Harshitha Thota",
+      "Tushar Yadav",
+      "Veeravelly Kusumitha",
+      "V V S S Prasanth Patnaik",
+      "Vidya Sri Sesetti",
+      "Vijayakeerthi K",
+      "Vikram Raj Bakshi",
+      "Vinay K K",
+      "Vinoth Kumar Loganathan",
+      "Vipin Tiwari",
+      "Vivek Kumar Shrivastav",
+      "V Venkata Sri Datta Charan",
+      "Wasim Akhtar Khan"
+    ],
+    "categories": [
+      "cs.SE",
+      "cs.AI",
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "agent-safety",
+      "coding-agent",
+      "computer-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2512.23747",
+    "source": "arxiv",
+    "source_id": "arxiv:2512.23747",
+    "pdf_url": "https://arxiv.org/pdf/2512.23747",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2510.24645",
+    "title": "FunReason-MT Technical Report: Advanced Data Synthesis Solution for Real-world Multi-Turn Tool-use",
+    "url": "https://arxiv.org/abs/2510.24645",
+    "published": "2025-10-28",
+    "updated": "2025-11-16",
+    "authors": [
+      "Zengzhuang Xu",
+      "Bingguang Hao",
+      "Zechuan Wang",
+      "Yuntao Wen",
+      "Xinyi Xu",
+      "Yang Liu",
+      "Long Chen",
+      "Dong Wang",
+      "Maolin Wang",
+      "Tong Zhao",
+      "Yicheng Chen",
+      "Cunyin Peng",
+      "Jinjie Gu",
+      "Leilei Gan",
+      "Xiangyu Zhao",
+      "Chenyi Zhuang",
+      "Shi Gu"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "multi-agent",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2510.24645",
+    "source": "arxiv",
+    "source_id": "arxiv:2510.24645",
+    "pdf_url": "https://arxiv.org/pdf/2510.24645",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2510.04206",
+    "title": "AgentRL: Scaling Agentic Reinforcement Learning with a Multi-Turn, Multi-Task Framework",
+    "url": "https://arxiv.org/abs/2510.04206",
+    "published": "2025-10-05",
+    "updated": "2025-10-05",
+    "authors": [
+      "Hanchen Zhang",
+      "Xiao Liu",
+      "Bowen Lv",
+      "Xueqiao Sun",
+      "Bohao Jing",
+      "Iat Long Iong",
+      "Zhenyu Hou",
+      "Zehan Qi",
+      "Hanyu Lai",
+      "Yifan Xu",
+      "Rui Lu",
+      "Hongning Wang",
+      "Jie Tang",
+      "Yuxiao Dong"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2510.04206",
+    "source": "arxiv",
+    "source_id": "arxiv:2510.04206",
+    "pdf_url": "https://arxiv.org/pdf/2510.04206",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2509.13311",
+    "title": "Towards General Agentic Intelligence via Environment Scaling",
+    "url": "https://arxiv.org/abs/2509.13311",
+    "published": "2025-09-16",
+    "updated": "2025-09-16",
+    "authors": [
+      "Runnan Fang",
+      "Shihao Cai",
+      "Baixuan Li",
+      "Jialong Wu",
+      "Guangyu Li",
+      "Wenbiao Yin",
+      "Xinyu Wang",
+      "Xiaobin Wang",
+      "Liangcai Su",
+      "Zhen Zhang",
+      "Shibin Wu",
+      "Zhengwei Tao",
+      "Yong Jiang",
+      "Pengjun Xie",
+      "Fei Huang",
+      "Jingren Zhou"
+    ],
+    "categories": [
+      "cs.CL"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2509.13311",
+    "source": "arxiv",
+    "source_id": "arxiv:2509.13311",
+    "pdf_url": "https://arxiv.org/pdf/2509.13311",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2509.02494",
+    "title": "GridMind: LLMs-Powered Agents for Power System Analysis and Operations",
+    "url": "https://arxiv.org/abs/2509.02494",
+    "published": "2025-09-02",
+    "updated": "2025-09-02",
+    "authors": [
+      "Hongwei Jin",
+      "Kibaek Kim",
+      "Jonghwan Kwon"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "multi-agent",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2509.02494",
+    "source": "arxiv",
+    "source_id": "arxiv:2509.02494",
+    "pdf_url": "https://arxiv.org/pdf/2509.02494",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2508.17094",
+    "title": "PowerChain: A Verifiable Agentic AI System for Automating Distribution Grid Analyses",
+    "url": "https://arxiv.org/abs/2508.17094",
+    "published": "2025-08-23",
+    "updated": "2025-10-21",
+    "authors": [
+      "Emmanuel O. Badmus",
+      "Peng Sang",
+      "Dimitrios Stamoulis",
+      "Amritanshu Pandey"
+    ],
+    "categories": [
+      "cs.AI",
+      "eess.SY"
+    ],
+    "topics": [
+      "planning",
+      "rag",
+      "reasoning",
+      "tool-use",
+      "workflow-agent"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2508.17094",
+    "source": "arxiv",
+    "source_id": "arxiv:2508.17094",
+    "pdf_url": "https://arxiv.org/pdf/2508.17094",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2508.12685",
+    "title": "ToolACE-MT: Non-Autoregressive Generation for Agentic Multi-Turn Interaction",
+    "url": "https://arxiv.org/abs/2508.12685",
+    "published": "2025-08-18",
+    "updated": "2026-02-13",
+    "authors": [
+      "Xingshan Zeng",
+      "Weiwen Liu",
+      "Lingzhi Wang",
+      "Liangyou Li",
+      "Fei Mi",
+      "Yasheng Wang",
+      "Lifeng Shang",
+      "Xin Jiang",
+      "Qun Liu"
+    ],
+    "categories": [
+      "cs.CL",
+      "cs.AI",
+      "cs.LG"
+    ],
+    "topics": [
+      "tool-use",
+      "world-model"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2508.12685",
+    "source": "arxiv",
+    "source_id": "arxiv:2508.12685",
+    "pdf_url": "https://arxiv.org/pdf/2508.12685",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2507.20666",
+    "title": "MIMII-Agent: Leveraging LLMs with Function Calling for Relative Evaluation of Anomalous Sound Detection",
+    "url": "https://arxiv.org/abs/2507.20666",
+    "published": "2025-07-28",
+    "updated": "2025-07-28",
+    "authors": [
+      "Harsh Purohit",
+      "Tomoya Nishida",
+      "Kota Dohi",
+      "Takashi Endo",
+      "Yohei Kawaguchi"
+    ],
+    "categories": [
+      "eess.AS",
+      "cs.AI",
+      "cs.LG",
+      "cs.SD"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "rag",
+      "tool-use"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2507.20666",
+    "source": "arxiv",
+    "source_id": "arxiv:2507.20666",
+    "pdf_url": "https://arxiv.org/pdf/2507.20666",
+    "primary_query": "function-calling"
+  },
+  {
+    "id": "2507.20395",
+    "title": "MazeEval: A Benchmark for Testing Sequential Decision-Making in Language Models",
+    "url": "https://arxiv.org/abs/2507.20395",
+    "published": "2025-07-27",
+    "updated": "2025-07-27",
+    "authors": [
+      "Hafsteinn Einarsson"
+    ],
+    "categories": [
+      "cs.AI"
+    ],
+    "topics": [
+      "agent-evaluation",
+      "computer-use",
+      "embodied-agent",
+      "reasoning"
+    ],
+    "score": 13,
+    "relevance": "high",
+    "matched_queries": [
+      "function-calling"
+    ],
+    "arxiv_id": "2507.20395",
+    "source": "arxiv",
+    "source_id": "arxiv:2507.20395",
+    "pdf_url": "https://arxiv.org/pdf/2507.20395",
+    "primary_query": "function-calling"
+  }
+]
diff --git a/data/index.json b/data/index.json
index 3a53aa3..6252b01 100644
--- a/data/index.json
+++ b/data/index.json
@@ -301,6 +301,1245 @@
       "relevance": "medium"
     }
   },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2507-20395-mazeeval-a-benchmark-for-testing-sequential-decision-making-in-language-models.md",
+    "title": "\"MazeEval: A Benchmark for Testing Sequential Decision-Making in Language Models\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MazeEval: A Benchmark for Testing Sequential Decision-Making in Language Models\"",
+      "authors": "Hafsteinn Einarsson",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2507.20395",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-07-27",
+      "updated_at": "2025-07-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "embodied-agent",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2507-20666-mimii-agent-leveraging-llms-with-function-calling-for-relative-evaluation-of-ano.md",
+    "title": "\"MIMII-Agent: Leveraging LLMs with Function Calling for Relative Evaluation of Anomalous Sound Detection\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MIMII-Agent: Leveraging LLMs with Function Calling for Relative Evaluation of Anomalous Sound Detection\"",
+      "authors": "Harsh Purohit, Tomoya Nishida, Kota Dohi, Takashi Endo, Yohei Kawaguchi",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2507.20666",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-07-28",
+      "updated_at": "2025-07-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "eess.AS",
+        "cs.AI",
+        "cs.LG",
+        "cs.SD"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2508-07575-mcptoolbench-a-large-scale-ai-agent-model-context-protocol-mcp-tool-use-benchmar.md",
+    "title": "\"MCPToolBench++: A Large Scale AI Agent Model Context Protocol MCP Tool Use Benchmark\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MCPToolBench++: A Large Scale AI Agent Model Context Protocol MCP Tool Use Benchmark\"",
+      "authors": "Shiqing Fan, Xichen Ding, Liang Zhang, Linjian Mo",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2508.07575",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-08-11",
+      "updated_at": "2025-08-11",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "21",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2508-11027-hell-or-high-water-evaluating-agentic-recovery-from-external-failures.md",
+    "title": "\"Hell or High Water: Evaluating Agentic Recovery from External Failures\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Hell or High Water: Evaluating Agentic Recovery from External Failures\"",
+      "authors": "Andrew Wang, Sophia Hager, Adi Asija, Daniel Khashabi, Nicholas Andrews",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2508.11027",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-08-14",
+      "updated_at": "2025-08-14",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2508-12685-toolace-mt-non-autoregressive-generation-for-agentic-multi-turn-interaction.md",
+    "title": "\"ToolACE-MT: Non-Autoregressive Generation for Agentic Multi-Turn Interaction\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ToolACE-MT: Non-Autoregressive Generation for Agentic Multi-Turn Interaction\"",
+      "authors": "Xingshan Zeng, Weiwen Liu, Lingzhi Wang, Liangyou Li, Fei Mi, Yasheng Wang, Lifeng Shang, Xin Jiang, et al.",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2508.12685",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-08-18",
+      "updated_at": "2026-02-13",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2508-17094-powerchain-a-verifiable-agentic-ai-system-for-automating-distribution-grid-analy.md",
+    "title": "\"PowerChain: A Verifiable Agentic AI System for Automating Distribution Grid Analyses\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"PowerChain: A Verifiable Agentic AI System for Automating Distribution Grid Analyses\"",
+      "authors": "Emmanuel O. Badmus, Peng Sang, Dimitrios Stamoulis, Amritanshu Pandey",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2508.17094",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-08-23",
+      "updated_at": "2025-10-21",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "eess.SY"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2509-02444-appcopilot-toward-general-accurate-long-horizon-and-efficient-mobile-agent.md",
+    "title": "\"AppCopilot: Toward General, Accurate, Long-Horizon, and Efficient Mobile Agent\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AppCopilot: Toward General, Accurate, Long-Horizon, and Efficient Mobile Agent\"",
+      "authors": "Jingru Fan, Yufan Dang, Jingyao Wu, Huatao Li, Runde Yang, Xiyuan Yang, Yuheng Wang, Chen Qian",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2509.02444",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-09-02",
+      "updated_at": "2025-10-17",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "computer-use",
+        "memory",
+        "multi-agent",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL",
+        "cs.CV",
+        "cs.HC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2509-02494-gridmind-llms-powered-agents-for-power-system-analysis-and-operations.md",
+    "title": "\"GridMind: LLMs-Powered Agents for Power System Analysis and Operations\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"GridMind: LLMs-Powered Agents for Power System Analysis and Operations\"",
+      "authors": "Hongwei Jin, Kibaek Kim, Jonghwan Kwon",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2509.02494",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-09-02",
+      "updated_at": "2025-09-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2509-08863-geojson-agents-a-multi-agent-llm-architecture-for-geospatial-analysis-function-c.md",
+    "title": "\"GeoJSON Agents:A Multi-Agent LLM Architecture for Geospatial Analysis-Function Calling vs Code Generation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"GeoJSON Agents:A Multi-Agent LLM Architecture for Geospatial Analysis-Function Calling vs Code Generation\"",
+      "authors": "Qianqian Luo, Qingming Lin, Liuchang Xu, Sensen Wu, Ruichen Mao, Chao Wang, Hailin Feng, Bo Huang, et al.",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2509.08863",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-09-10",
+      "updated_at": "2025-12-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "planning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2509-10769-agentarch-a-comprehensive-benchmark-to-evaluate-agent-architectures-in-enterpris.md",
+    "title": "\"AgentArch: A Comprehensive Benchmark to Evaluate Agent Architectures in Enterprise\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AgentArch: A Comprehensive Benchmark to Evaluate Agent Architectures in Enterprise\"",
+      "authors": "Tara Bogavelli, Roshnee Sharma, Hari Subramani",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2509.10769",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-09-13",
+      "updated_at": "2026-01-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "multi-agent",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2509-13311-towards-general-agentic-intelligence-via-environment-scaling.md",
+    "title": "Towards General Agentic Intelligence via Environment Scaling",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Towards General Agentic Intelligence via Environment Scaling",
+      "authors": "Runnan Fang, Shihao Cai, Baixuan Li, Jialong Wu, Guangyu Li, Wenbiao Yin, Xinyu Wang, Xiaobin Wang, et al.",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2509.13311",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-09-16",
+      "updated_at": "2025-09-16",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2509-14477-ticket-bench-a-kickoff-for-multilingual-and-regionalized-agent-evaluation.md",
+    "title": "\"Ticket-Bench: A Kickoff for Multilingual and Regionalized Agent Evaluation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Ticket-Bench: A Kickoff for Multilingual and Regionalized Agent Evaluation\"",
+      "authors": "Thales Sales Almeida, João Guilherme Alves Santos, Thiago Laitz, Giovana Kerche Bonás",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2509.14477",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-09-17",
+      "updated_at": "2025-09-17",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2509-20998-core-full-path-evaluation-of-llm-agents-beyond-final-state.md",
+    "title": "\"CORE: Full-Path Evaluation of LLM Agents Beyond Final State\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"CORE: Full-Path Evaluation of LLM Agents Beyond Final State\"",
+      "authors": "Panagiotis Michelakis, Yiannis Hadjiyiannis, Dimitrios Stamoulis",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2509.20998",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-09-25",
+      "updated_at": "2025-09-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2509-26553-towards-reliable-benchmarking-a-contamination-free-controllable-evaluation-frame.md",
+    "title": "\"Towards Reliable Benchmarking: A Contamination Free, Controllable Evaluation Framework for Multi-step LLM Function Calling\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Towards Reliable Benchmarking: A Contamination Free, Controllable Evaluation Framework for Multi-step LLM Function Calling\"",
+      "authors": "Seiji Maekawa, Jackson Hassell, Pouya Pezeshkpour, Tom Mitchell, Estevam Hruschka",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2509.26553",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-09-30",
+      "updated_at": "2026-02-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.PL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2510-03847-small-language-models-for-agentic-systems-a-survey-of-architectures-capabilities.md",
+    "title": "\"Small Language Models for Agentic Systems: A Survey of Architectures, Capabilities, and Deployment Trade offs\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Small Language Models for Agentic Systems: A Survey of Architectures, Capabilities, and Deployment Trade offs\"",
+      "authors": "Raghav Sharma, Manan Mehta",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2510.03847",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-10-04",
+      "updated_at": "2025-10-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "computer-use",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2510-04206-agentrl-scaling-agentic-reinforcement-learning-with-a-multi-turn-multi-task-fram.md",
+    "title": "\"AgentRL: Scaling Agentic Reinforcement Learning with a Multi-Turn, Multi-Task Framework\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AgentRL: Scaling Agentic Reinforcement Learning with a Multi-Turn, Multi-Task Framework\"",
+      "authors": "Hanchen Zhang, Xiao Liu, Bowen Lv, Xueqiao Sun, Bohao Jing, Iat Long Iong, Zhenyu Hou, Zehan Qi, et al.",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2510.04206",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-10-05",
+      "updated_at": "2025-10-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2510-14548-llm-agents-beyond-utility-an-open-ended-perspective.md",
+    "title": "\"LLM Agents Beyond Utility: An Open-Ended Perspective\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"LLM Agents Beyond Utility: An Open-Ended Perspective\"",
+      "authors": "Asen Nachkov, Xi Wang, Luc Van Gool",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2510.14548",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-10-16",
+      "updated_at": "2025-10-16",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "memory",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2510-18586-tokencake-a-kv-cache-centric-serving-framework-for-llm-based-multi-agent-applica.md",
+    "title": "\"TokenCake: A KV-Cache-centric Serving Framework for LLM-based Multi-Agent Applications\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"TokenCake: A KV-Cache-centric Serving Framework for LLM-based Multi-Agent Applications\"",
+      "authors": "Zhuohang Bian, Feiyang Wu, Zhuoran Li, Teng Ma, Youwei Zhuo",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2510.18586",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-10-21",
+      "updated_at": "2026-05-20",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "multi-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.DC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2510-21524-eu-agent-bench-measuring-illegal-behavior-of-llm-agents-under-eu-law.md",
+    "title": "\"EU-Agent-Bench: Measuring Illegal Behavior of LLM Agents Under EU Law\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"EU-Agent-Bench: Measuring Illegal Behavior of LLM Agents Under EU Law\"",
+      "authors": "Ilija Lichkovski, Alexander Müller, Mariam Ibrahim, Tiwai Mhundwa",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2510.21524",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-10-24",
+      "updated_at": "2025-10-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2510-22768-seeing-is-believing-evaluating-vision-language-model-susceptibility-in-agent-to-.md",
+    "title": "Seeing is Believing? Evaluating Vision-Language Model Susceptibility in Agent-to-Agent Multimodal Persuasion",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Seeing is Believing? Evaluating Vision-Language Model Susceptibility in Agent-to-Agent Multimodal Persuasion",
+      "authors": "Haoyi Qiu, Yilun Zhou, Pranav Narayanan Venkit, Kung-Hsiang Huang, Jiaxin Zhang, Nanyun Peng, Chien-Sheng Wu",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2510.22768",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-10-26",
+      "updated_at": "2026-06-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2510-24645-funreason-mt-technical-report-advanced-data-synthesis-solution-for-real-world-mu.md",
+    "title": "\"FunReason-MT Technical Report: Advanced Data Synthesis Solution for Real-world Multi-Turn Tool-use\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"FunReason-MT Technical Report: Advanced Data Synthesis Solution for Real-world Multi-Turn Tool-use\"",
+      "authors": "Zengzhuang Xu, Bingguang Hao, Zechuan Wang, Yuntao Wen, Xinyi Xu, Yang Liu, Long Chen, Dong Wang, et al.",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2510.24645",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-10-28",
+      "updated_at": "2025-11-16",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "multi-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2510-26167-toolrm-towards-agentic-tool-use-reward-modeling.md",
+    "title": "\"ToolRM: Towards Agentic Tool-Use Reward Modeling\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ToolRM: Towards Agentic Tool-Use Reward Modeling\"",
+      "authors": "Renhao Li, Jianhong Tu, Yang Su, Yantao Liu, Fei Huang, Hamid Alinejad-Rokny, Derek F. Wong, Junyang Lin, et al.",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2510.26167",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-10-30",
+      "updated_at": "2026-01-13",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2511-04847-test-time-adaptation-for-llm-agents-via-environment-interaction.md",
+    "title": "Test-Time Adaptation for LLM Agents via Environment Interaction",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Test-Time Adaptation for LLM Agents via Environment Interaction",
+      "authors": "Arthur Chen, Zuxin Liu, Jianguo Zhang, Akshara Prabhakar, Zhiwei Liu, Shelby Heinecke, Silvio Savarese, Victor Zhong, et al.",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2511.04847",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-11-06",
+      "updated_at": "2026-02-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "embodied-agent",
+        "rag",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2511-11169-refine-and-align-confidence-calibration-through-multi-agent-interaction-in-vqa.md",
+    "title": "\"Refine and Align: Confidence Calibration through Multi-Agent Interaction in VQA\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Refine and Align: Confidence Calibration through Multi-Agent Interaction in VQA\"",
+      "authors": "Ayush Pandey, Jai Bardhan, Ishita Jain, Ramya S Hebbalaguppe, Rohan Raju Dhanakshirur, Lovekesh Vig",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2511.11169",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-11-14",
+      "updated_at": "2025-11-14",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "embodied-agent",
+        "multi-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV",
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2511-15203-taxonomy-evaluation-and-exploitation-of-ipi-centric-llm-agent-defense-frameworks.md",
+    "title": "Taxonomy, Evaluation and Exploitation of IPI-Centric LLM Agent Defense Frameworks",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Taxonomy, Evaluation and Exploitation of IPI-Centric LLM Agent Defense Frameworks",
+      "authors": "Zimo Ji, Xunguang Wang, Zongjie Li, Pingchuan Ma, Yudong Gao, Daoyuan Wu, Xincheng Yan, Tian Tian, et al.",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2511.15203",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-11-19",
+      "updated_at": "2025-11-19",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2511-22138-tinyllm-evaluation-and-optimization-of-small-language-models-for-agentic-tasks-o.md",
+    "title": "\"TinyLLM: Evaluation and Optimization of Small Language Models for Agentic Tasks on Edge Devices\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"TinyLLM: Evaluation and Optimization of Small Language Models for Agentic Tasks on Edge Devices\"",
+      "authors": "Mohd Ariful Haque, Fahad Rahman, Kishor Datta Gupta, Khalil Shujaee, Roy George",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2511.22138",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-11-27",
+      "updated_at": "2025-11-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2512-02605-iact-a-self-organizing-recursive-model-for-general-ai-agents-a-technical-white-p.md",
+    "title": "\"IACT: A Self-Organizing Recursive Model for General AI Agents: A Technical White Paper on the Architecture Behind kragent.ai\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"IACT: A Self-Organizing Recursive Model for General AI Agents: A Technical White Paper on the Architecture Behind kragent.ai\"",
+      "authors": "Pengju Lu",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2512.02605",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-12-02",
+      "updated_at": "2025-12-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.MA",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2512-11682-medai-evaluating-txagent-s-therapeutic-agentic-reasoning-in-the-neurips-cure-ben.md",
+    "title": "\"MedAI: Evaluating TxAgent's Therapeutic Agentic Reasoning in the NeurIPS CURE-Bench Competition\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MedAI: Evaluating TxAgent's Therapeutic Agentic Reasoning in the NeurIPS CURE-Bench Competition\"",
+      "authors": "Tim Cofala, Christian Kalfar, Jingge Xiao, Johanna Schrader, Michelle Tang, Wolfgang Nejdl",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2512.11682",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-12-12",
+      "updated_at": "2026-06-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2512-23611-close-the-loop-synthesizing-infinite-tool-use-data-via-multi-agent-role-playing.md",
+    "title": "\"Close the Loop: Synthesizing Infinite Tool-Use Data via Multi-Agent Role-Playing\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Close the Loop: Synthesizing Infinite Tool-Use Data via Multi-Agent Role-Playing\"",
+      "authors": "Yuwen Li, Wei Zhang, Zelong Huang, Mason Yang, Jiajun Wu, Shawn Guo, Huahao Hu, Lingyi Sun, et al.",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2512.23611",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-12-29",
+      "updated_at": "2025-12-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2512-23647-nested-browser-use-learning-for-agentic-information-seeking.md",
+    "title": "Nested Browser-Use Learning for Agentic Information Seeking",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Nested Browser-Use Learning for Agentic Information Seeking",
+      "authors": "Baixuan Li, Jialong Wu, Wenbiao Yin, Kuan Li, Zhongwang Zhang, Huifeng Yin, Zhengwei Tao, Liwen Zhang, et al.",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2512.23647",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-12-29",
+      "updated_at": "2025-12-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI",
+        "cs.IR",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2025-2512-23747-state-of-the-art-small-language-coder-model-mify-coder.md",
+    "title": "\"State-of-the-art Small Language Coder Model: Mify-Coder\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"State-of-the-art Small Language Coder Model: Mify-Coder\"",
+      "authors": "Abhinav Parmar, Abhisek Panigrahi, Abhishek Kumar Dwivedi, Abhishek Bhattacharya, Adarsh Ramachandra, Aditya Choudhary, Aditya Garg, Aditya Raj, et al.",
+      "year": "2025",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2512.23747",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2025-12-26",
+      "updated_at": "2025-12-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "computer-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "function-calling"
+    }
+  },
   {
     "collection": "papers",
     "path": "papers/items/2025-vijayvargiya-openagentsafety.md",
@@ -345,6 +1584,37873 @@
       "related_projects": []
     }
   },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2601-00268-beyond-perfect-apis-a-comprehensive-evaluation-of-llm-agents-under-real-world-ap.md",
+    "title": "\"Beyond Perfect APIs: A Comprehensive Evaluation of LLM Agents Under Real-World API Complexity\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Beyond Perfect APIs: A Comprehensive Evaluation of LLM Agents Under Real-World API Complexity\"",
+      "authors": "Doyoung Kim, Zhiwei Ren, Jie Hao, Zhongkai Sun, Lichao Wang, Xiyao Ma, Zack Ye, Xu Han, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2601.00268",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-01-01",
+      "updated_at": "2026-01-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2601-05467-stelp-secure-transpilation-and-execution-of-llm-generated-programs.md",
+    "title": "\"STELP: Secure Transpilation and Execution of LLM-Generated Programs\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"STELP: Secure Transpilation and Execution of LLM-Generated Programs\"",
+      "authors": "Swapnil Shinde, Sahil Wadhwa, Andy Luo, Akshay Gupta, Mohammad Shahed Sorower",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2601.05467",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-01-09",
+      "updated_at": "2026-01-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2601-06007-don-t-break-the-cache-an-evaluation-of-prompt-caching-for-long-horizon-agentic-t.md",
+    "title": "\"Don't Break the Cache: An Evaluation of Prompt Caching for Long-Horizon Agentic Tasks\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Don't Break the Cache: An Evaluation of Prompt Caching for Long-Horizon Agentic Tasks\"",
+      "authors": "Elias Lumer, Faheem Nizar, Akshaya Jangiti, Kevin Frank, Anmol Gulati, Mandar Phadate, Vamse Kumar Subbiah",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2601.06007",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-01-09",
+      "updated_at": "2026-01-31",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2601-06606-cedar-context-engineering-for-agentic-data-science.md",
+    "title": "\"CEDAR: Context Engineering for Agentic Data Science\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"CEDAR: Context Engineering for Agentic Data Science\"",
+      "authors": "Rishiraj Saha Roy, Chris Hinze, Luzian Hahn, Fabian Kuech",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2601.06606",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-01-10",
+      "updated_at": "2026-04-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "planning",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2601-12988-paperguide-making-small-language-model-paper-reading-agents-more-efficient.md",
+    "title": "\"PaperGuide: Making Small Language-Model Paper-Reading Agents More Efficient\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"PaperGuide: Making Small Language-Model Paper-Reading Agents More Efficient\"",
+      "authors": "Zijian Wang, Tiancheng Huang, Hanqi Li, Da Ma, Lu Chen, Kai Yu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2601.12988",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-01-19",
+      "updated_at": "2026-01-19",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2601-14652-mas-orchestra-understanding-and-improving-multi-agent-reasoning-through-holistic.md",
+    "title": "\"MAS-Orchestra: Understanding and Improving Multi-Agent Reasoning Through Holistic Orchestration and Controlled Benchmarks\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MAS-Orchestra: Understanding and Improving Multi-Agent Reasoning Through Holistic Orchestration and Controlled Benchmarks\"",
+      "authors": "Zixuan Ke, Yifei Ming, Austin Xu, Ryan Chin, Xuan-Phi Nguyen, Prathyusha Jwalapuram, Jiayu Wang, Semih Yavuz, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2601.14652",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-01-21",
+      "updated_at": "2026-05-21",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "multi-agent",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2602-03117-agentdyn-are-your-agent-security-defenses-deployable-in-real-world-dynamic-envir.md",
+    "title": "\"AgentDyn: Are Your Agent Security Defenses Deployable in Real-World Dynamic Environments?\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AgentDyn: Are Your Agent Security Defenses Deployable in Real-World Dynamic Environments?\"",
+      "authors": "Hao Li, Ruoyao Wen, Shanghao Shi, Ning Zhang, Yevgeniy Vorobeychik, Chaowei Xiao",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2602.03117",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-02-03",
+      "updated_at": "2026-05-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2602-03224-tame-a-trustworthy-test-time-evolution-of-agent-memory-with-systematic-benchmark.md",
+    "title": "\"TAME: A Trustworthy Test-Time Evolution of Agent Memory with Systematic Benchmarking\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"TAME: A Trustworthy Test-Time Evolution of Agent Memory with Systematic Benchmarking\"",
+      "authors": "Yu Cheng, Yongkang Hu, Jiuan Zhou, Yushuo Zhang, Yihang Chen, Huichi Zhou, Mingang Chen, Zhizhong Zhang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2602.03224",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-02-03",
+      "updated_at": "2026-06-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "memory",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2602-03786-aorchestra-automating-sub-agent-creation-for-agentic-orchestration.md",
+    "title": "\"AOrchestra: Automating Sub-Agent Creation for Agentic Orchestration\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AOrchestra: Automating Sub-Agent Creation for Agentic Orchestration\"",
+      "authors": "Jianhao Ruan, Zhihao Xu, Yiran Peng, Fashen Ren, Zhaoyang Yu, Xinbing Liang, Jinyu Xiang, Yongru Chen, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2602.03786",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-02-03",
+      "updated_at": "2026-02-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "planning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2602-05115-socialveil-probing-social-intelligence-of-language-agents-under-communication-ba.md",
+    "title": "\"SocialVeil: Probing Social Intelligence of Language Agents under Communication Barriers\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SocialVeil: Probing Social Intelligence of Language Agents under Communication Barriers\"",
+      "authors": "Keyang Xuan, Pengda Wang, Chongrui Ye, Haofei Yu, Tal August, Jiaxuan You",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2602.05115",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-02-04",
+      "updated_at": "2026-02-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2602-05302-piearena-ranking-and-profiling-language-agents-in-realistic-negotiation-scenario.md",
+    "title": "\"PieArena: Ranking and Profiling Language Agents in Realistic Negotiation Scenarios\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"PieArena: Ranking and Profiling Language Agents in Realistic Negotiation Scenarios\"",
+      "authors": "Chris Zhu, Sasha Cui, Will Sanok Dufallo, Runzhi Jin, Zhen Xu, Linjun Zhang, Daylian Cain",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2602.05302",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-02-05",
+      "updated_at": "2026-06-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2602-05386-spider-sense-intrinsic-risk-sensing-for-efficient-agent-defense-with-hierarchica.md",
+    "title": "\"Spider-Sense: Intrinsic Risk Sensing for Efficient Agent Defense with Hierarchical Adaptive Screening\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Spider-Sense: Intrinsic Risk Sensing for Efficient Agent Defense with Hierarchical Adaptive Screening\"",
+      "authors": "Zhenxiong Yu, Zhi Yang, Zhiheng Jin, Shuhe Wang, Heng Zhang, Yanlin Fei, Lingfeng Zeng, Fangqi Lou, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2602.05386",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-02-05",
+      "updated_at": "2026-02-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2602-07391-naamse-framework-for-evolutionary-security-evaluation-of-agents.md",
+    "title": "\"NAAMSE: Framework for Evolutionary Security Evaluation of Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"NAAMSE: Framework for Evolutionary Security Evaluation of Agents\"",
+      "authors": "Kunal Pai, Parth Shah, Harshil Patel",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2602.07391",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-02-07",
+      "updated_at": "2026-03-08",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2602-07652-agent-fence-mapping-security-vulnerabilities-across-deep-research-agents.md",
+    "title": "\"Agent-Fence: Mapping Security Vulnerabilities Across Deep Research Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Agent-Fence: Mapping Security Vulnerabilities Across Deep Research Agents\"",
+      "authors": "Sai Puppala, Ismail Hossain, Md Jahangir Alam, Yoonpyo Lee, Jay Yoo, Tanzim Ahad, Syed Bahauddin Alam, Sajedul Talukder",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2602.07652",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-02-07",
+      "updated_at": "2026-02-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2602-07962-loca-bench-benchmarking-language-agents-under-controllable-and-extreme-context-g.md",
+    "title": "\"LOCA-bench: Benchmarking Language Agents Under Controllable and Extreme Context Growth\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"LOCA-bench: Benchmarking Language Agents Under Controllable and Extreme Context Growth\"",
+      "authors": "Weihao Zeng, Yuzhen Huang, Junxian He",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2602.07962",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-02-08",
+      "updated_at": "2026-02-08",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2602-08082-spectral-guardrails-for-agents-in-the-wild-detecting-tool-use-hallucinations-via.md",
+    "title": "\"Spectral Guardrails for Agents in the Wild: Detecting Tool Use Hallucinations via Attention Topology\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Spectral Guardrails for Agents in the Wild: Detecting Tool Use Hallucinations via Attention Topology\"",
+      "authors": "Valentin Noël",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2602.08082",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-02-08",
+      "updated_at": "2026-02-08",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI",
+        "eess.SP"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2602-08412-from-assistant-to-double-agent-formalizing-and-benchmarking-attacks-on-openclaw-.md",
+    "title": "\"From Assistant to Double Agent: Formalizing and Benchmarking Attacks on OpenClaw for Personalized Local AI Agent\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"From Assistant to Double Agent: Formalizing and Benchmarking Attacks on OpenClaw for Personalized Local AI Agent\"",
+      "authors": "Yuhang Wang, Feiming Xu, Zheng Lin, Guangyu He, Yuzhe Huang, Haichang Gao, Zhenxing Niu, Shiguo Lian, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2602.08412",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-02-09",
+      "updated_at": "2026-02-11",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "21",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2602-10133-agenttrace-a-structured-logging-framework-for-agent-system-observability.md",
+    "title": "\"AgentTrace: A Structured Logging Framework for Agent System Observability\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AgentTrace: A Structured Logging Framework for Agent System Observability\"",
+      "authors": "Adam AlSayyad, Kelvin Yuxiang Huang, Richik Pal",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2602.10133",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-02-07",
+      "updated_at": "2026-02-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2602-11749-air-improving-agent-safety-through-incident-response.md",
+    "title": "\"AIR: Improving Agent Safety through Incident Response\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AIR: Improving Agent Safety through Incident Response\"",
+      "authors": "Zibo Xiao, Jun Sun, Junjie Chen",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2602.11749",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-02-12",
+      "updated_at": "2026-06-20",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2602-13379-unsafer-in-many-turns-benchmarking-and-defending-multi-turn-safety-risks-in-tool.md",
+    "title": "\"Unsafer in Many Turns: Benchmarking and Defending Multi-Turn Safety Risks in Tool-Using Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Unsafer in Many Turns: Benchmarking and Defending Multi-Turn Safety Risks in Tool-Using Agents\"",
+      "authors": "Xu Li, Simon Yu, Minzhou Pan, Yiyou Sun, Bo Li, Dawn Song, Xue Lin, Weiyan Shi",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2602.13379",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-02-13",
+      "updated_at": "2026-06-10",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI",
+        "cs.CL",
+        "cs.LG",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2602-13530-remem-reasoning-with-episodic-memory-in-language-agent.md",
+    "title": "\"REMem: Reasoning with Episodic Memory in Language Agent\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"REMem: Reasoning with Episodic Memory in Language Agent\"",
+      "authors": "Yiheng Shu, Saisri Padmaja Jonnalagedda, Xiang Gao, Bernal Jiménez Gutiérrez, Weijian Qi, Kamalika Das, Huan Sun, Yu Su",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2602.13530",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-02-13",
+      "updated_at": "2026-02-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2602-13665-hyfunc-accelerating-llm-based-function-calls-for-agentic-ai-through-hybrid-model.md",
+    "title": "\"HyFunc: Accelerating LLM-based Function Calls for Agentic AI through Hybrid-Model Cascade and Dynamic Templating\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"HyFunc: Accelerating LLM-based Function Calls for Agentic AI through Hybrid-Model Cascade and Dynamic Templating\"",
+      "authors": "Weibin Liao, Jian-guang Lou, Haoyi Xiong",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2602.13665",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-02-14",
+      "updated_at": "2026-02-14",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2602-14234-redsearcher-a-scalable-and-cost-efficient-framework-for-long-horizon-search-agen.md",
+    "title": "\"REDSearcher: A Scalable and Cost-Efficient Framework for Long-Horizon Search Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"REDSearcher: A Scalable and Cost-Efficient Framework for Long-Horizon Search Agents\"",
+      "authors": "Zheng Chu, Xiao Wang, Jack Hong, Huiming Fan, Yuqi Huang, Yue Yang, Guohai Xu, Chenxiao Zhao, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2602.14234",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-02-15",
+      "updated_at": "2026-02-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2602-14281-mcpshield-a-security-cognition-layer-for-adaptive-trust-calibration-in-model-con.md",
+    "title": "\"MCPShield: A Security Cognition Layer for Adaptive Trust Calibration in Model Context Protocol Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MCPShield: A Security Cognition Layer for Adaptive Trust Calibration in Model Context Protocol Agents\"",
+      "authors": "Zhenhong Zhou, Yuanhe Zhang, Hongwei Cai, Moayad Aloqaily, Ouns Bouachir, Linsey Pang, Prakhar Mehrotra, Kun Wang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2602.14281",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-02-15",
+      "updated_at": "2026-02-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "computer-use",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2602-16931-narrow-fine-tuning-erodes-safety-alignment-in-vision-language-agents.md",
+    "title": "Narrow Fine-Tuning Erodes Safety Alignment in Vision-Language Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Narrow Fine-Tuning Erodes Safety Alignment in Vision-Language Agents",
+      "authors": "Idhant Gulati, Shivam Raval",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2602.16931",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-02-18",
+      "updated_at": "2026-03-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2602-18456-beyond-single-channel-agentic-benchmarking.md",
+    "title": "Beyond single-channel agentic benchmarking",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Beyond single-channel agentic benchmarking",
+      "authors": "Nelu D. Radpour",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2602.18456",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-02-05",
+      "updated_at": "2026-02-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CY",
+        "cs.AI",
+        "cs.HC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2602-19008-capable-but-unreliable-canonical-path-deviation-as-a-causal-mechanism-of-agent-f.md",
+    "title": "\"Capable but Unreliable: Canonical Path Deviation as a Causal Mechanism of Agent Failure in Long-Horizon Tasks\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Capable but Unreliable: Canonical Path Deviation as a Causal Mechanism of Agent Failure in Long-Horizon Tasks\"",
+      "authors": "Wilson Y. Lee",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2602.19008",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-02-22",
+      "updated_at": "2026-02-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2602-21127-are-you-sure-an-empirical-study-of-human-perception-vulnerability-in-llm-driven-.md",
+    "title": "\"\\\"Are You Sure?\\\": An Empirical Study of Human Perception Vulnerability in LLM-Driven Agentic Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"\\\"Are You Sure?\\\": An Empirical Study of Human Perception Vulnerability in LLM-Driven Agentic Systems\"",
+      "authors": "Xinfeng Li, Shenyu Dai, Kelong Zheng, Yue Xiao, Gelei Deng, Wei Dong, Xiaofeng Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2602.21127",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-02-24",
+      "updated_at": "2026-02-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.HC",
+        "cs.AI",
+        "cs.CR",
+        "cs.SI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2602-23320-parammem-augmenting-language-agents-with-parametric-reflective-memory.md",
+    "title": "\"ParamMem: Augmenting Language Agents with Parametric Reflective Memory\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ParamMem: Augmenting Language Agents with Parametric Reflective Memory\"",
+      "authors": "Tianjun Yao, Yongqiang Chen, Yujia Zheng, Pan Li, Zhiqiang Shen, Kun Zhang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2602.23320",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-02-26",
+      "updated_at": "2026-02-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-00131-thought-virus-viral-misalignment-via-subliminal-prompting-in-multi-agent-systems.md",
+    "title": "\"Thought Virus: Viral Misalignment via Subliminal Prompting in Multi-Agent Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Thought Virus: Viral Misalignment via Subliminal Prompting in Multi-Agent Systems\"",
+      "authors": "Moritz Weckbecker, Jonas Müller, Ben Hagag, Michael Mulet",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.00131",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-02-23",
+      "updated_at": "2026-02-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "multi-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-00623-tracesir-a-multi-agent-framework-for-structured-analysis-and-reporting-of-agenti.md",
+    "title": "\"TraceSIR: A Multi-Agent Framework for Structured Analysis and Reporting of Agentic Execution Traces\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"TraceSIR: A Multi-Agent Framework for Structured Analysis and Reporting of Agentic Execution Traces\"",
+      "authors": "Shu-Xun Yang, Cunxiang Wang, Haoke Zhang, Wenbo Yu, Lindong Wu, Jiayi Gui, Dayong Yang, Yukuo Cen, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.00623",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-02-28",
+      "updated_at": "2026-02-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "multi-agent",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-00801-the-synthetic-web-adversarially-curated-mini-internets-for-diagnosing-epistemic-.md",
+    "title": "\"The Synthetic Web: Adversarially-Curated Mini-Internets for Diagnosing Epistemic Weaknesses of Language Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"The Synthetic Web: Adversarially-Curated Mini-Internets for Diagnosing Epistemic Weaknesses of Language Agents\"",
+      "authors": "Shrey Shah, Levent Ozgur",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.00801",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-02-28",
+      "updated_at": "2026-02-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "embodied-agent",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.IR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-01438-enhancing-persona-following-at-decoding-time-via-dynamic-importance-estimation-f.md",
+    "title": "Enhancing Persona Following at Decoding Time via Dynamic Importance Estimation for Role-Playing Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Enhancing Persona Following at Decoding Time via Dynamic Importance Estimation for Role-Playing Agents",
+      "authors": "Yuxin Liu, Mingye Zhu, Siyuan Liu, Bo Hu, Lei Zhang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.01438",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-03-02",
+      "updated_at": "2026-03-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "coding-agent",
+        "computer-use",
+        "planning",
+        "rag",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-01712-ft-dojo-towards-autonomous-llm-fine-tuning-with-language-agents.md",
+    "title": "\"FT-Dojo: Towards Autonomous LLM Fine-Tuning with Language Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"FT-Dojo: Towards Autonomous LLM Fine-Tuning with Language Agents\"",
+      "authors": "Qizheng Li, Yifei Zhang, Xiao Yang, Xu Yang, Zhuo Wang, Weiqing Liu, Jiang Bian",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.01712",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-03-02",
+      "updated_at": "2026-05-20",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "planning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-02711-a-natural-language-agentic-approach-to-study-affective-polarization.md",
+    "title": "A Natural Language Agentic Approach to Study Affective Polarization",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "A Natural Language Agentic Approach to Study Affective Polarization",
+      "authors": "Stephanie Anneris Malvicini, Ewelina Gajewska, Arda Derbent, Katarzyna Budzynska, Jarosław A. Chudziak, Maria Vanina Martinez",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.02711",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-03-03",
+      "updated_at": "2026-03-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "multi-agent",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-03515-the-controllability-trap-a-governance-framework-for-military-ai-agents.md",
+    "title": "\"The Controllability Trap: A Governance Framework for Military AI Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"The Controllability Trap: A Governance Framework for Military AI Agents\"",
+      "authors": "Subramanyam Sahoo",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.03515",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-03-03",
+      "updated_at": "2026-03-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CY",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-03680-mage-meta-reinforcement-learning-for-language-agents-toward-strategic-exploratio.md",
+    "title": "\"MAGE: Meta-Reinforcement Learning for Language Agents toward Strategic Exploration and Exploitation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MAGE: Meta-Reinforcement Learning for Language Agents toward Strategic Exploration and Exploitation\"",
+      "authors": "Lu Yang, Zelai Xu, Minyang Xie, Jiaxuan Gao, Zhao Shok, Yu Wang, Yi Wu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.03680",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-03-04",
+      "updated_at": "2026-03-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "memory",
+        "multi-agent",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-05553-eigendata-a-self-evolving-multi-agent-platform-for-function-calling-data-synthes.md",
+    "title": "\"EigenData: A Self-Evolving Multi-Agent Platform for Function-Calling Data Synthesis, Auditing, and Repair\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"EigenData: A Self-Evolving Multi-Agent Platform for Function-Calling Data Synthesis, Auditing, and Repair\"",
+      "authors": "Jiaao Chen, Jingyuan Qi, Mingye Gao, Wei-Chen Wang, Hanrui Wang, Di Jin",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.05553",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-03-05",
+      "updated_at": "2026-03-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "multi-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-05578-tool-genesis-a-task-driven-tool-creation-benchmark-for-self-evolving-language-ag.md",
+    "title": "\"Tool-Genesis: A Task-Driven Tool Creation Benchmark for Self-Evolving Language Agent\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Tool-Genesis: A Task-Driven Tool Creation Benchmark for Self-Evolving Language Agent\"",
+      "authors": "Bowei Xia, Mengkang Hu, Shijian Wang, Jiarui Jin, Wenxiang Jiao, Yuan Lu, Kexin Li, Ping Luo",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.05578",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-03-05",
+      "updated_at": "2026-03-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-07496-from-thinker-to-society-security-in-hierarchical-autonomy-evolution-of-ai-agents.md",
+    "title": "\"From Thinker to Society: Security in Hierarchical Autonomy Evolution of AI Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"From Thinker to Society: Security in Hierarchical Autonomy Evolution of AI Agents\"",
+      "authors": "Xiaolei Zhang, Lu Zhou, Xiaogang Xu, Jiafei Wu, Tianyu Du, Heqing Huang, Hao Peng, Zhe Liu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.07496",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-03-08",
+      "updated_at": "2026-03-21",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "multi-agent",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-07557-agentraft-automated-detection-of-data-over-exposure-in-llm-agents.md",
+    "title": "\"AgentRaft: Automated Detection of Data Over-Exposure in LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AgentRaft: Automated Detection of Data Over-Exposure in LLM Agents\"",
+      "authors": "Yixi Lin, Jiangrong Wu, Yuhong Nan, Xueqiang Wang, Xinyuan Zhang, Zibin Zheng",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.07557",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-03-08",
+      "updated_at": "2026-03-08",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-07980-onemillion-bench-how-far-are-language-agents-from-human-experts.md",
+    "title": "\"\\\\$OneMillion-Bench: How Far are Language Agents from Human Experts?\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"\\\\$OneMillion-Bench: How Far are Language Agents from Human Experts?\"",
+      "authors": "Qianyu Yang, Yang Liu, Jiaqi Li, Jun Bai, Hao Chen, Kaiyuan Chen, Tiliang Duan, Jiayun Dong, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.07980",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-03-09",
+      "updated_at": "2026-03-09",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-08721-kernelcraft-benchmarking-for-agentic-close-to-metal-kernel-generation-on-emergin.md",
+    "title": "\"KernelCraft: Benchmarking for Agentic Close-to-Metal Kernel Generation on Emerging Hardware\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"KernelCraft: Benchmarking for Agentic Close-to-Metal Kernel Generation on Emerging Hardware\"",
+      "authors": "Jiayi Nie, Haoran Wu, Yao Lai, Zeyu Cao, Cheng Zhang, Binglei Lou, Erwei Wang, Jianyi Cheng, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.08721",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-02-10",
+      "updated_at": "2026-05-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "reasoning",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AR",
+        "cs.LG",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-09002-security-considerations-for-multi-agent-systems.md",
+    "title": "Security Considerations for Multi-agent Systems",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Security Considerations for Multi-agent Systems",
+      "authors": "Tam Nguyen, Moses Ndebugre, Dheeraj Arremsetty",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.09002",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-03-09",
+      "updated_at": "2026-04-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "memory",
+        "multi-agent",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-10492-human-ai-co-reasoning-for-clinical-diagnosis-with-evidence-integrated-language-a.md",
+    "title": "Human-AI Co-reasoning for Clinical Diagnosis with Evidence-Integrated Language Agent",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Human-AI Co-reasoning for Clinical Diagnosis with Evidence-Integrated Language Agent",
+      "authors": "Zhongzhen Huang, Yan Ling, Hong Chen, Ye Feng, Li Wu, Linjie Mu, Shaoting Zhang, Xiaofan Zhang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.10492",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-03-11",
+      "updated_at": "2026-03-18",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "rag",
+        "reasoning",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-11088-the-attack-and-defense-landscape-of-agentic-ai-a-comprehensive-survey.md",
+    "title": "\"The Attack and Defense Landscape of Agentic AI: A Comprehensive Survey\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"The Attack and Defense Landscape of Agentic AI: A Comprehensive Survey\"",
+      "authors": "Juhee Kim, Xiaoyuan Liu, Zhun Wang, Shi Qiu, Bo Li, Wenbo Guo, Dawn Song",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.11088",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-03-11",
+      "updated_at": "2026-03-11",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-11890-quare-quality-aware-requirements-analysis-through-multi-agent-dialectical-negoti.md",
+    "title": "\"QUARE: Quality-Aware Requirements Analysis through Multi-Agent Dialectical Negotiation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"QUARE: Quality-Aware Requirements Analysis through Multi-Agent Dialectical Negotiation\"",
+      "authors": "Haowei Cheng, Milhan Kim, Foutse Khomh, Teeradaj Racharak, Nobukazu Yoshioka, Naoyasu Ubayashi, Hironori Washizaki",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.11890",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-03-12",
+      "updated_at": "2026-06-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-15309-cctu-a-benchmark-for-tool-use-under-complex-constraints.md",
+    "title": "\"CCTU: A Benchmark for Tool Use under Complex Constraints\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"CCTU: A Benchmark for Tool Use under Complex Constraints\"",
+      "authors": "Junjie Ye, Guoqiang Zhang, Wenjie Fu, Tao Gui, Qi Zhang, Xuanjing Huang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.15309",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-03-16",
+      "updated_at": "2026-03-16",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-15666-compiled-memory-not-more-information-but-more-precise-instructions-for-language-.md",
+    "title": "\"Compiled Memory: Not More Information, but More Precise Instructions for Language Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Compiled Memory: Not More Information, but More Precise Instructions for Language Agents\"",
+      "authors": "James Rhodes, George Kang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.15666",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-03-12",
+      "updated_at": "2026-03-12",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "memory",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-16734-differential-harm-propensity-in-personalized-llm-agents-the-curious-case-of-ment.md",
+    "title": "\"Differential Harm Propensity in Personalized LLM Agents: The Curious Case of Mental Health Disclosure\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Differential Harm Propensity in Personalized LLM Agents: The Curious Case of Mental Health Disclosure\"",
+      "authors": "Caglar Yildirim",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.16734",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-03-17",
+      "updated_at": "2026-03-17",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-17392-agentic-cognitive-profiling-realigning-automated-alzheimer-s-disease-detection-w.md",
+    "title": "\"Agentic Cognitive Profiling: Realigning Automated Alzheimer's Disease Detection with Clinical Construct Validity\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Agentic Cognitive Profiling: Realigning Automated Alzheimer's Disease Detection with Clinical Construct Validity\"",
+      "authors": "Jiawen Kang, Kun Li, Dongrui Han, Jinchao Li, Junan Li, Lingwei Meng, Xixin Wu, Helen Meng",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.17392",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-03-18",
+      "updated_at": "2026-03-18",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA",
+        "cs.IR",
+        "q-bio.NC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-18245-who-tests-the-testers-systematic-enumeration-and-coverage-audit-of-llm-agent-too.md",
+    "title": "Who Tests the Testers? Systematic Enumeration and Coverage Audit of LLM Agent Tool Call Safety",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Who Tests the Testers? Systematic Enumeration and Coverage Audit of LLM Agent Tool Call Safety",
+      "authors": "Xuan Chen, Lu Yan, Ruqi Zhang, Xiangyu Zhang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.18245",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-03-18",
+      "updated_at": "2026-03-18",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-19469-a-framework-for-formalizing-llm-agent-security.md",
+    "title": "A Framework for Formalizing LLM Agent Security",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "A Framework for Formalizing LLM Agent Security",
+      "authors": "Vincent Siu, Jingxuan He, Kyle Montgomery, Zhun Wang, Neil Gong, Chenguang Wang, Dawn Song",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.19469",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-03-19",
+      "updated_at": "2026-03-19",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "memory",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-19684-tsegagent-zero-shot-tooth-segmentation-via-geometry-aware-vision-language-agents.md",
+    "title": "\"TSegAgent: Zero-Shot Tooth Segmentation via Geometry-Aware Vision-Language Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"TSegAgent: Zero-Shot Tooth Segmentation via Geometry-Aware Vision-Language Agents\"",
+      "authors": "Shaojie Zhuang, Lu Yin, Guangshun Wei, Yunpeng Li, Xilu Wang, Yuanfeng Zhou",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.19684",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-03-20",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-21357-agenther-hindsight-experience-replay-for-llm-agent-trajectory-relabeling.md",
+    "title": "\"AgentHER: Hindsight Experience Replay for LLM Agent Trajectory Relabeling\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AgentHER: Hindsight Experience Replay for LLM Agent Trajectory Relabeling\"",
+      "authors": "Liang Ding",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.21357",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-03-22",
+      "updated_at": "2026-05-10",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "computer-use",
+        "memory",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-21564-toward-a-theory-of-hierarchical-memory-for-language-agents.md",
+    "title": "Toward a Theory of Hierarchical Memory for Language Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Toward a Theory of Hierarchical Memory for Language Agents",
+      "authors": "Yashar Talebirad, Ali Parsaee, Csongor Y. Szepesvari, Amirhossein Nadiri, Osmar Zaiane",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.21564",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-03-23",
+      "updated_at": "2026-03-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.IR",
+        "cs.AI",
+        "cs.IT",
+        "cs.SI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-24257-memory-augmented-vision-language-agents-for-persistent-and-semantically-consiste.md",
+    "title": "Memory-Augmented Vision-Language Agents for Persistent and Semantically Consistent Object Captioning",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Memory-Augmented Vision-Language Agents for Persistent and Semantically Consistent Object Captioning",
+      "authors": "Tommaso Galliena, Stefano Rosa, Tommaso Apicella, Pietro Morerio, Alessio Del Bue, Lorenzo Natale",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.24257",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-03-25",
+      "updated_at": "2026-03-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "embodied-agent",
+        "memory"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-25353-safeguard-asf-sr-agentic-humanoid-robot-system-for-autonomous-industrial-safety.md",
+    "title": "\"SafeGuard ASF: SR Agentic Humanoid Robot System for Autonomous Industrial Safety\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SafeGuard ASF: SR Agentic Humanoid Robot System for Autonomous Industrial Safety\"",
+      "authors": "Thanh Nguyen Canh, Thang Tran Viet, Thanh Tuan Tran, Ben Wei Lim",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.25353",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-03-26",
+      "updated_at": "2026-03-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "embodied-agent",
+        "reasoning",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.RO"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-27148-safetydrift-predicting-when-ai-agents-cross-the-line-before-they-actually-do.md",
+    "title": "\"SafetyDrift: Predicting When AI Agents Cross the Line Before They Actually Do\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SafetyDrift: Predicting When AI Agents Cross the Line Before They Actually Do\"",
+      "authors": "Aditya Dhodapkar, Farhaan Pishori",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.27148",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-03-28",
+      "updated_at": "2026-03-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-28166-evaluating-privilege-usage-of-agents-with-real-world-tools.md",
+    "title": "Evaluating Privilege Usage of Agents with Real-World Tools",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Evaluating Privilege Usage of Agents with Real-World Tools",
+      "authors": "Quan Zhang, Lianhang Fu, Lvsi Lian, Gwihwan Go, Yujue Wang, Chijin Zhou, Yu Jiang, Geguang Pu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.28166",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-03-30",
+      "updated_at": "2026-04-20",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-28428-synergy-a-next-generation-general-purpose-agent-for-open-agentic-web.md",
+    "title": "\"Synergy: A Next-Generation General-Purpose Agent for Open Agentic Web\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Synergy: A Next-Generation General-Purpose Agent for Open Agentic Web\"",
+      "authors": "Xiaohang Nie, Zihan Guo, Kezhuo Yang, Zhichong Zheng, Bochen Ge, Shuai Pan, Zeyi Chen, Youling Xiang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.28428",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-03-30",
+      "updated_at": "2026-03-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "coding-agent",
+        "embodied-agent",
+        "memory",
+        "multi-agent",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CY",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2603-28900-robust-multi-agent-reinforcement-learning-for-small-uas-separation-assurance-und.md",
+    "title": "Robust Multi-Agent Reinforcement Learning for Small UAS Separation Assurance under GPS Degradation and Spoofing",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Robust Multi-Agent Reinforcement Learning for Small UAS Separation Assurance under GPS Degradation and Spoofing",
+      "authors": "Alex Zongo, Filippos Fotiadis, Ufuk Topcu, Peng Wei",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2603.28900",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-03-30",
+      "updated_at": "2026-03-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.RO",
+        "cs.AI",
+        "cs.LG",
+        "eess.SY"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-02022-atbench-a-diverse-and-realistic-agent-trajectory-benchmark-for-safety-evaluation.md",
+    "title": "\"ATBench: A Diverse and Realistic Agent Trajectory Benchmark for Safety Evaluation and Diagnosis\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ATBench: A Diverse and Realistic Agent Trajectory Benchmark for Safety Evaluation and Diagnosis\"",
+      "authors": "Yu Li, Haoyu Luo, Yuejin Xie, Yuqian Fu, Zhonghao Yang, Shuai Shao, Qihan Ren, Wanying Qu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.02022",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-02",
+      "updated_at": "2026-05-13",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-02155-brief-is-better-non-monotonic-chain-of-thought-budget-effects-in-function-callin.md",
+    "title": "\"Brief Is Better: Non-Monotonic Chain-of-Thought Budget Effects in Function-Calling Language Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Brief Is Better: Non-Monotonic Chain-of-Thought Budget Effects in Function-Calling Language Agents\"",
+      "authors": "Xuan Qi",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.02155",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-02",
+      "updated_at": "2026-04-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "function-calling, language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-03098-co-evolution-of-policy-and-internal-reward-for-language-agents.md",
+    "title": "Co-Evolution of Policy and Internal Reward for Language Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Co-Evolution of Policy and Internal Reward for Language Agents",
+      "authors": "Xinyu Wang, Hanwei Wu, Jingwei Song, Shuyuan Zhang, Jiayi Zhang, Fanqi Kong, Tung Sum Thomas Kwok, Xiao-Wen Chang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.03098",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-03",
+      "updated_at": "2026-04-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-03242-draft-task-decoupled-latent-reasoning-for-agent-safety.md",
+    "title": "\"DRAFT: Task Decoupled Latent Reasoning for Agent Safety\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"DRAFT: Task Decoupled Latent Reasoning for Agent Safety\"",
+      "authors": "Lin Wang, Junfeng Fang, Dan Zhang, Fei Shen, Xiang Wang, Tat-Seng Chua",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.03242",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-02-11",
+      "updated_at": "2026-02-11",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-04131-profile-then-reason-bounded-semantic-complexity-for-tool-augmented-language-agen.md",
+    "title": "\"Profile-Then-Reason: Bounded Semantic Complexity for Tool-Augmented Language Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Profile-Then-Reason: Bounded Semantic Complexity for Tool-Augmented Language Agents\"",
+      "authors": "Paulo Akira F. Enabe",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.04131",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-05",
+      "updated_at": "2026-04-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-04426-shieldnet-network-level-guardrails-against-emerging-supply-chain-injections-in-a.md",
+    "title": "\"ShieldNet: Network-Level Guardrails against Emerging Supply-Chain Injections in Agentic Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ShieldNet: Network-Level Guardrails against Emerging Supply-Chain Injections in Agentic Systems\"",
+      "authors": "Zhuowen Yuan, Zhaorun Chen, Zhen Xiang, Nathaniel D. Bastian, Seyyed Hadi Hashemi, Chaowei Xiao, Wenbo Guo, Bo Li",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.04426",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-06",
+      "updated_at": "2026-04-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-06762-arulecon-agentic-security-rule-conversion.md",
+    "title": "\"ARuleCon: Agentic Security Rule Conversion\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ARuleCon: Agentic Security Rule Conversion\"",
+      "authors": "Ming Xu, Hongtai Wang, Yanpei Guo, Zhengmin Yu, Weili Han, Hoon Wei Lim, Jin Song Dong, Jiaheng Zhang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.06762",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-08",
+      "updated_at": "2026-04-08",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-06972-differentiable-environment-trajectory-co-optimization-for-safe-multi-agent-navig.md",
+    "title": "Differentiable Environment-Trajectory Co-Optimization for Safe Multi-Agent Navigation",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Differentiable Environment-Trajectory Co-Optimization for Safe Multi-Agent Navigation",
+      "authors": "Zhan Gao, Gabriele Fadini, Stelian Coros, Amanda Prorok",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.06972",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-08",
+      "updated_at": "2026-04-08",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "embodied-agent",
+        "multi-agent",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.RO",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-08388-awakening-the-sleeping-agent-lean-specific-agentic-data-reactivates-general-tool.md",
+    "title": "\"Awakening the Sleeping Agent: Lean-Specific Agentic Data Reactivates General Tool Use in Goedel Prover\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Awakening the Sleeping Agent: Lean-Specific Agentic Data Reactivates General Tool Use in Goedel Prover\"",
+      "authors": "Jui-Hui Chung, Hongzhou Lin, Lai Jiang, Shange Tang, Chi Jin",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.08388",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-09",
+      "updated_at": "2026-04-09",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-10577-the-blind-spot-of-agent-safety-how-benign-user-instructions-expose-critical-vuln.md",
+    "title": "\"The Blind Spot of Agent Safety: How Benign User Instructions Expose Critical Vulnerabilities in Computer-Use Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"The Blind Spot of Agent Safety: How Benign User Instructions Expose Critical Vulnerabilities in Computer-Use Agents\"",
+      "authors": "Xuwei Ding, Skylar Zhai, Linxin Song, Jiate Li, Taiwei Shi, Nicholas Meade, Siva Reddy, Jian Kang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.10577",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-12",
+      "updated_at": "2026-04-17",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-11557-unitoolcall-unifying-tool-use-representation-data-and-evaluation-for-llm-agents.md",
+    "title": "\"UniToolCall: Unifying Tool-Use Representation, Data, and Evaluation for LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"UniToolCall: Unifying Tool-Use Representation, Data, and Evaluation for LLM Agents\"",
+      "authors": "Yijuan Liang, Xinghao Chen, Yifan Ge, Ziyi Wu, Hao Wu, Changyu Zeng, Wei Xing, Xiaoyu Shen",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.11557",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-13",
+      "updated_at": "2026-05-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-12986-parallax-why-ai-agents-that-think-must-never-act.md",
+    "title": "\"Parallax: Why AI Agents That Think Must Never Act\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Parallax: Why AI Agents That Think Must Never Act\"",
+      "authors": "Joel Fokou",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.12986",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-14",
+      "updated_at": "2026-04-14",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-13298-can-agents-secure-hardware-evaluating-agentic-llm-driven-obfuscation-for-ip-prot.md",
+    "title": "Can Agents Secure Hardware? Evaluating Agentic LLM-Driven Obfuscation for IP Protection",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Can Agents Secure Hardware? Evaluating Agentic LLM-Driven Obfuscation for IP Protection",
+      "authors": "Sujan Ghimire, Parsa Mirfasihi, Muhtasim Alam Chowdhury, Veeramani Pugazhenthi, Harish Kumar Dharavath, Farshad Firouzi, Rozhin Yasaei, Pratik Satam, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.13298",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-14",
+      "updated_at": "2026-04-14",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-13536-don-t-let-ai-agents-yolo-your-files-shifting-information-and-control-to-filesyst.md",
+    "title": "\"Don't Let AI Agents YOLO Your Files: Shifting Information and Control to Filesystems for Agent Safety and Autonomy\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Don't Let AI Agents YOLO Your Files: Shifting Information and Control to Filesystems for Agent Safety and Autonomy\"",
+      "authors": "Shawn Wanxiang Zhong, Junxuan Liao, Jing Liu, Mai Zheng, Andrea C. Arpaci-Dusseau, Remzi H. Arpaci-Dusseau",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.13536",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-15",
+      "updated_at": "2026-04-16",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.OS"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-13954-hintbench-horizon-agent-intrinsic-non-attack-trajectory-benchmark.md",
+    "title": "\"HINTBench: Horizon-agent Intrinsic Non-attack Trajectory Benchmark\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"HINTBench: Horizon-agent Intrinsic Non-attack Trajectory Benchmark\"",
+      "authors": "Jiacheng Wang, Jinchang Hou, Fabian Wang, Ping Jian, Chenfu Bao, Zhonghou Lv",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.13954",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-15",
+      "updated_at": "2026-04-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-14399-spacemind-a-modular-and-self-evolving-embodied-vision-language-agent-framework-f.md",
+    "title": "\"SpaceMind: A Modular and Self-Evolving Embodied Vision-Language Agent Framework for Autonomous On-orbit Servicing\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SpaceMind: A Modular and Self-Evolving Embodied Vision-Language Agent Framework for Autonomous On-orbit Servicing\"",
+      "authors": "Aodi Wu, Haodong Han, Xubo Luo, Ruisuo Wang, Shan He, Xue Wan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.14399",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-15",
+      "updated_at": "2026-04-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "embodied-agent",
+        "reasoning",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.RO",
+        "cs.AI",
+        "eess.SY"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-15415-harmfulskillbench-how-do-harmful-skills-weaponize-your-agents.md",
+    "title": "\"HarmfulSkillBench: How Do Harmful Skills Weaponize Your Agents?\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"HarmfulSkillBench: How Do Harmful Skills Weaponize Your Agents?\"",
+      "authors": "Yukun Jiang, Yage Zhang, Michael Backes, Xinyue Shen, Yang Zhang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.15415",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-16",
+      "updated_at": "2026-04-16",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-15579-don-t-make-models-guess-security-and-safety-symbolic-guardrails-for-domain-speci.md",
+    "title": "\"Don't Make Models Guess Security and Safety: Symbolic Guardrails for Domain-Specific AI Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Don't Make Models Guess Security and Safety: Symbolic Guardrails for Domain-Specific AI Agents\"",
+      "authors": "Yining Hong, Yining She, Eunsuk Kang, Christopher S. Timperley, Christian Kästner",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.15579",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-16",
+      "updated_at": "2026-07-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI",
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-16706-evaluating-tool-using-language-agents-judge-reliability-propagation-cascades-and.md",
+    "title": "\"Evaluating Tool-Using Language Agents: Judge Reliability, Propagation Cascades, and Runtime Mitigation in AgentProp-Bench\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Evaluating Tool-Using Language Agents: Judge Reliability, Propagation Cascades, and Runtime Mitigation in AgentProp-Bench\"",
+      "authors": "Bhaskar Gurram",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.16706",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-17",
+      "updated_at": "2026-04-17",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-17562-safeagent-a-runtime-protection-architecture-for-agentic-systems.md",
+    "title": "\"SafeAgent: A Runtime Protection Architecture for Agentic Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SafeAgent: A Runtime Protection Architecture for Agentic Systems\"",
+      "authors": "Hailin Liu, Eugene Ilyushin, Jie Ni, Min Zhu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.17562",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-19",
+      "updated_at": "2026-04-19",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "memory",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-18658-owner-harm-a-missing-threat-model-for-ai-agent-safety.md",
+    "title": "\"Owner-Harm: A Missing Threat Model for AI Agent Safety\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Owner-Harm: A Missing Threat Model for AI Agent Safety\"",
+      "authors": "Dongcheng Zhang, Yiqing Jiang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.18658",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-20",
+      "updated_at": "2026-04-20",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-18718-towards-optimal-agentic-architectures-for-offensive-security-tasks.md",
+    "title": "Towards Optimal Agentic Architectures for Offensive Security Tasks",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Towards Optimal Agentic Architectures for Offensive Security Tasks",
+      "authors": "Isaac David, Arthur Gervais",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.18718",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-20",
+      "updated_at": "2026-04-20",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-18847-human-guided-harm-recovery-for-computer-use-agents.md",
+    "title": "Human-Guided Harm Recovery for Computer Use Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Human-Guided Harm Recovery for Computer Use Agents",
+      "authors": "Christy Li, Sky CH-Wang, Andi Peng, Andreea Bobu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.18847",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-20",
+      "updated_at": "2026-05-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "memory",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-19821-jtpro-a-joint-tool-prompt-reflective-optimization-framework-for-language-agents.md",
+    "title": "\"JTPRO: A Joint Tool-Prompt Reflective Optimization Framework for Language Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"JTPRO: A Joint Tool-Prompt Reflective Optimization Framework for Language Agents\"",
+      "authors": "Sandip Ghoshal, Anshul Mittal, Jyotika Singh, Miguel Ballesteros, Weiyi Sun, Fang Tu, Shailender Singh, Yassine Benajiba, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.19821",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-20",
+      "updated_at": "2026-04-20",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-19844-if-you-re-waiting-for-a-sign-that-might-not-be-it-mitigating-trust-boundary-conf.md",
+    "title": "\"If you're waiting for a sign... that might not be it! Mitigating Trust Boundary Confusion from Visual Injections on Vision-Language Agentic Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"If you're waiting for a sign... that might not be it! Mitigating Trust Boundary Confusion from Visual Injections on Vision-Language Agentic Systems\"",
+      "authors": "Jiamin Chang, Minhui Xue, Ruoxi Sun, Shuchao Pang, Salil S. Kanhere, Hammond Pearce",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.19844",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-21",
+      "updated_at": "2026-04-21",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "embodied-agent",
+        "multi-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-20994-breaking-mcp-with-function-hijacking-attacks-novel-threats-for-function-calling-.md",
+    "title": "\"Breaking MCP with Function Hijacking Attacks: Novel Threats for Function Calling and Agentic Models\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Breaking MCP with Function Hijacking Attacks: Novel Threats for Function Calling and Agentic Models\"",
+      "authors": "Yannis Belkhiter, Giulio Zizzo, Sergio Maffeis, Seshu Tirupathi, John D. Kelleher",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.20994",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-22",
+      "updated_at": "2026-04-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-21190-spatio-adaptive-test-time-orchestration-of-vision-language-agents-for-spatial-re.md",
+    "title": "\"SpatiO: Adaptive Test-Time Orchestration of Vision-Language Agents for Spatial Reasoning\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SpatiO: Adaptive Test-Time Orchestration of Vision-Language Agents for Spatial Reasoning\"",
+      "authors": "Chan Yeong Hwang, Miso Choi, Sunghyun On, Jinkyu Kim, Jungbeom Lee",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.21190",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-23",
+      "updated_at": "2026-06-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-22879-beyond-single-agent-alignment-preventing-context-fragmented-violations-in-multi-.md",
+    "title": "\"Beyond Single-Agent Alignment: Preventing Context-Fragmented Violations in Multi-Agent Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Beyond Single-Agent Alignment: Preventing Context-Fragmented Violations in Multi-Agent Systems\"",
+      "authors": "Jie Wu, Ming Gong",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.22879",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-24",
+      "updated_at": "2026-04-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "rag",
+        "tool-use",
+        "workflow-agent",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA",
+        "cs.AI",
+        "cs.CR",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-23210-discovering-agentic-safety-specifications-from-1-bit-danger-signals.md",
+    "title": "Discovering Agentic Safety Specifications from 1-Bit Danger Signals",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Discovering Agentic Safety Specifications from 1-Bit Danger Signals",
+      "authors": "Víctor Gallego",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.23210",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-25",
+      "updated_at": "2026-04-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-23374-ghost-in-the-agent-redefining-information-flow-tracking-for-llm-agents.md",
+    "title": "\"Ghost in the Agent: Redefining Information Flow Tracking for LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Ghost in the Agent: Redefining Information Flow Tracking for LLM Agents\"",
+      "authors": "Yuandao Cai, Wensheng Tang, Cheng Wen, Shengchao Qin",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.23374",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-25",
+      "updated_at": "2026-04-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-23459-architecture-matters-for-multi-agent-security.md",
+    "title": "Architecture Matters for Multi-Agent Security",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Architecture Matters for Multi-Agent Security",
+      "authors": "Ben Hagag, William L. Anderson, Christian Schroeder de Witt, Sarah Scheffler",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.23459",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-25",
+      "updated_at": "2026-04-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "memory",
+        "multi-agent",
+        "planning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA",
+        "cs.CR",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-24212-empowering-autonomous-debugging-agents-with-efficient-dynamic-analysis.md",
+    "title": "Empowering Autonomous Debugging Agents with Efficient Dynamic Analysis",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Empowering Autonomous Debugging Agents with Efficient Dynamic Analysis",
+      "authors": "Jiahong Xiang, Xiaoyang Xu, Xiaopan Chu, Hongliang Tian, Yuqun Zhang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.24212",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-27",
+      "updated_at": "2026-04-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "embodied-agent",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-24826-a-comparative-evaluation-of-ai-agent-security-guardrails.md",
+    "title": "A Comparative Evaluation of AI Agent Security Guardrails",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "A Comparative Evaluation of AI Agent Security Guardrails",
+      "authors": "Qi Li, Jiu Li, Pingtao Wei, Jianjun Xu, Xueyi Wei, Jiwei Shi, Xuan Zhang, Yanhui Yang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.24826",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-27",
+      "updated_at": "2026-04-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-25135-fama-failure-aware-meta-agentic-framework-for-open-source-llms-in-interactive-to.md",
+    "title": "\"FAMA: Failure-Aware Meta-Agentic Framework for Open-Source LLMs in Interactive Tool Use Environments\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"FAMA: Failure-Aware Meta-Agentic Framework for Open-Source LLMs in Interactive Tool Use Environments\"",
+      "authors": "Amir Saeidi, Venkatesh Mishra, Souradeep Mukhopadhyay, Gaowen Liu, Ali Payani, Jayanth Srinivasa, Chitta Baral",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.25135",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-28",
+      "updated_at": "2026-04-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-25318-cutscene-agent-an-llm-agent-framework-for-automated-3d-cutscene-generation.md",
+    "title": "\"Cutscene Agent: An LLM Agent Framework for Automated 3D Cutscene Generation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Cutscene Agent: An LLM Agent Framework for Automated 3D Cutscene Generation\"",
+      "authors": "Lanshan He, Haozhou Pang, Qi Gan, Xin Shen, Ziwei Zhang, Yibo Liu, Gang Fang, Bo Liu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.25318",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-28",
+      "updated_at": "2026-04-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.GR",
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-25555-from-crud-to-autonomous-agents-formal-validation-and-zero-trust-security-for-sem.md",
+    "title": "\"From CRUD to Autonomous Agents: Formal Validation and Zero-Trust Security for Semantic Gateways in AI-Native Enterprise Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"From CRUD to Autonomous Agents: Formal Validation and Zero-Trust Security for Semantic Gateways in AI-Native Enterprise Systems\"",
+      "authors": "Ignacio Peyrano",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.25555",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-28",
+      "updated_at": "2026-04-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-26274-enforcing-benign-trajectories-a-behavioral-firewall-for-structured-workflow-ai-a.md",
+    "title": "\"Enforcing Benign Trajectories: A Behavioral Firewall for Structured-Workflow AI Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Enforcing Benign Trajectories: A Behavioral Firewall for Structured-Workflow AI Agents\"",
+      "authors": "Hung Dang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.26274",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-29",
+      "updated_at": "2026-04-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-26959-careguardai-context-aware-multi-agent-guardrails-for-clinical-safety-hallucinati.md",
+    "title": "\"CareGuardAI: Context-Aware Multi-Agent Guardrails for Clinical Safety & Hallucination Mitigation in Patient-Facing LLMs\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"CareGuardAI: Context-Aware Multi-Agent Guardrails for Clinical Safety & Hallucination Mitigation in Patient-Facing LLMs\"",
+      "authors": "Elham Nasarian, Abhilash Neog, Kwok-Leung Tsui, Niyousha HosseiniChimeh",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.26959",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-07",
+      "updated_at": "2026-04-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CY",
+        "cs.AI",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-27092-end-to-end-autonomous-scientific-discovery-on-a-real-optical-platform.md",
+    "title": "End-to-end autonomous scientific discovery on a real optical platform",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "End-to-end autonomous scientific discovery on a real optical platform",
+      "authors": "Shuxing Yang, Fujia Chen, Rui Zhao, Junyao Wu, Yize Wang, Haiyao Luo, Ning Han, Qiaolu Chen, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.27092",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-29",
+      "updated_at": "2026-04-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "memory",
+        "planning",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "physics.optics"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-27464-security-attack-and-defense-strategies-for-autonomous-agent-frameworks-a-layered.md",
+    "title": "\"Security Attack and Defense Strategies for Autonomous Agent Frameworks: A Layered Review with OpenClaw as a Case Study\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Security Attack and Defense Strategies for Autonomous Agent Frameworks: A Layered Review with OpenClaw as a Case Study\"",
+      "authors": "Luyao Xu, Xiang Chen",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.27464",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-30",
+      "updated_at": "2026-04-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-safety, autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-27699-bridging-values-and-behavior-a-hierarchical-framework-for-proactive-embodied-age.md",
+    "title": "\"Bridging Values and Behavior: A Hierarchical Framework for Proactive Embodied Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Bridging Values and Behavior: A Hierarchical Framework for Proactive Embodied Agents\"",
+      "authors": "Chunhui Zhang, Yuxuan Wang, Aoyang Qin, Yi-Long Lu, Kunlun Wu, Yizhou Wang, Wei Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.27699",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-30",
+      "updated_at": "2026-04-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "embodied-agent",
+        "memory",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-27859-rethinking-agentic-reinforcement-learning-in-large-language-models.md",
+    "title": "Rethinking Agentic Reinforcement Learning In Large Language Models",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Rethinking Agentic Reinforcement Learning In Large Language Models",
+      "authors": "Fangming Cui, Ruixiao Zhu, Cheng Fang, Sunan Li, Jiahong Li",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.27859",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-30",
+      "updated_at": "2026-05-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "memory",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.ET"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2604-28157-flashrt-towards-computationally-and-memory-efficient-red-teaming-for-prompt-inje.md",
+    "title": "\"FlashRT: Towards Computationally and Memory Efficient Red-Teaming for Prompt Injection and Knowledge Corruption\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"FlashRT: Towards Computationally and Memory Efficient Red-Teaming for Prompt Injection and Knowledge Corruption\"",
+      "authors": "Yanting Wang, Chenlong Yin, Ying Chen, Jinyuan Jia",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2604.28157",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-30",
+      "updated_at": "2026-04-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-00081-alignment-contracts-for-agentic-security-systems.md",
+    "title": "Alignment Contracts for Agentic Security Systems",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Alignment Contracts for Agentic Security Systems",
+      "authors": "Isaac David, Marco Guarnieri, Arthur Gervais",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.00081",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-30",
+      "updated_at": "2026-04-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.LO"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-00741-self-adaptive-multi-agent-llm-based-security-pattern-selection-for-iot-systems.md",
+    "title": "Self-Adaptive Multi-Agent LLM-Based Security Pattern Selection for IoT Systems",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Self-Adaptive Multi-Agent LLM-Based Security Pattern Selection for IoT Systems",
+      "authors": "Saeid Jamshidi, Foutse Khomh, Carol Fung, Kawser Wazed Nafi",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.00741",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-01",
+      "updated_at": "2026-05-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-00845-graph-query-generation-with-constraint-guided-large-language-agents.md",
+    "title": "Graph Query Generation with Constraint-guided Large Language Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Graph Query Generation with Constraint-guided Large Language Agents",
+      "authors": "Mengying Wang, Nicolaas Jedema, Rahul Pandey, RaviKiran Krishnan, Jens Lehmann, Yinghui Wu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.00845",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-09",
+      "updated_at": "2026-04-09",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "reasoning",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.DB",
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-01101-virtual-speech-therapist-a-clinician-in-the-loop-ai-speech-therapy-agent-for-per.md",
+    "title": "\"Virtual Speech Therapist: A Clinician-in-the-Loop AI Speech Therapy Agent for Personalized and Supervised Therapy\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Virtual Speech Therapist: A Clinician-in-the-Loop AI Speech Therapy Agent for Personalized and Supervised Therapy\"",
+      "authors": "Shakeel Sheikh, Patrick Marmaroli, MD Sahidullah, Slim Ouni, Fabrice Hirsch, Goncalo Leal, Bjorn W Schuller",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.01101",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-01",
+      "updated_at": "2026-06-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "multi-agent",
+        "planning",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL",
+        "cs.SD",
+        "eess.AS"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-01644-toward-a-principled-framework-for-agent-safety-measurement.md",
+    "title": "Toward a Principled Framework for Agent Safety Measurement",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Toward a Principled Framework for Agent Safety Measurement",
+      "authors": "Shuyi Lin, Anshuman Suri, Alina Oprea, Cheng Tan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.01644",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-02",
+      "updated_at": "2026-05-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-02240-physicianbench-evaluating-llm-agents-in-real-world-ehr-environments.md",
+    "title": "\"PhysicianBench: Evaluating LLM Agents in Real-World EHR Environments\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"PhysicianBench: Evaluating LLM Agents in Real-World EHR Environments\"",
+      "authors": "Ruoqi Liu, Imran Q. Mohiuddin, Austin J. Schoeffler, Kavita Renduchintala, Ashwin Nayak, Prasantha L. Vemu, Shivam C. Vedak, Kameron C. Black, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.02240",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-04",
+      "updated_at": "2026-05-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-03242-enhancing-agent-safety-judgment-controlled-benchmark-rewriting-and-analogical-re.md",
+    "title": "\"Enhancing Agent Safety Judgment: Controlled Benchmark Rewriting and Analogical Reasoning for Deceptive Out-of-Distribution Scenarios\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Enhancing Agent Safety Judgment: Controlled Benchmark Rewriting and Analogical Reasoning for Deceptive Out-of-Distribution Scenarios\"",
+      "authors": "Zuoyu Zhang, Yancheng Zhu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.03242",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-05",
+      "updated_at": "2026-05-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "multi-agent",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "22",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-03312-memflow-intent-driven-memory-orchestration-for-small-language-model-agents.md",
+    "title": "\"MemFlow: Intent-Driven Memory Orchestration for Small Language Model Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MemFlow: Intent-Driven Memory Orchestration for Small Language Model Agents\"",
+      "authors": "Jiayi Chen, Yingcong Li, Guiling Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.03312",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-05",
+      "updated_at": "2026-05-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-03328-llm-adam-a-generalizable-llm-agent-framework-for-pre-print-anomaly-detection-in-.md",
+    "title": "\"LLM-ADAM: A Generalizable LLM Agent Framework for Pre-Print Anomaly Detection in Additive Manufacturing\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"LLM-ADAM: A Generalizable LLM Agent Framework for Pre-Print Anomaly Detection in Additive Manufacturing\"",
+      "authors": "Ahmadreza Eslaminia, Chuhan Cai, Cameron Smith, Ruo-Syuan Mei, Shichen Li, Rajiv Malhotra, Klara Nahrstedt, Chenhui Shao",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.03328",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-05",
+      "updated_at": "2026-05-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-03505-lats-rca-language-agent-tree-search-for-root-cause-analysis-in-microservices.md",
+    "title": "\"LATS-RCA: Language Agent Tree Search for Root Cause Analysis in Microservices\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"LATS-RCA: Language Agent Tree Search for Root Cause Analysis in Microservices\"",
+      "authors": "Alexander Naakka, Yuqing Wang, Mika V Mäntylä",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.03505",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-05",
+      "updated_at": "2026-06-12",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-04107-tscg-deterministic-tool-schema-compilation-for-agentic-llm-deployments.md",
+    "title": "\"TSCG: Deterministic Tool-Schema Compilation for Agentic LLM Deployments\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"TSCG: Deterministic Tool-Schema Compilation for Agentic LLM Deployments\"",
+      "authors": "Furkan Sakizli",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.04107",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-04",
+      "updated_at": "2026-05-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-04808-decodingtrust-agent-platform-dtap-a-controllable-and-interactive-red-teaming-pla.md",
+    "title": "\"DecodingTrust-Agent Platform (DTap): A Controllable and Interactive Red-Teaming Platform for AI Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"DecodingTrust-Agent Platform (DTap): A Controllable and Interactive Red-Teaming Platform for AI Agents\"",
+      "authors": "Zhaorun Chen, Xun Liu, Haibo Tong, Chengquan Guo, Yuzhou Nie, Jiawei Zhang, Mintong Kang, Chejian Xu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.04808",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-06",
+      "updated_at": "2026-05-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "planning",
+        "tool-use",
+        "workflow-agent",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-05242-beyond-semantic-similarity-rethinking-retrieval-for-agentic-search-via-direct-co.md",
+    "title": "\"Beyond Semantic Similarity: Rethinking Retrieval for Agentic Search via Direct Corpus Interaction\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Beyond Semantic Similarity: Rethinking Retrieval for Agentic Search via Direct Corpus Interaction\"",
+      "authors": "Zhuofeng Li, Haoxiang Zhang, Cong Wei, Pan Lu, Ping Nie, Yi Lu, Yuyang Bai, Shangbin Feng, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.05242",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-03",
+      "updated_at": "2026-05-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.IR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-05704-safeharbor-hierarchical-memory-augmented-guardrail-for-llm-agent-safety.md",
+    "title": "\"SafeHarbor: Hierarchical Memory-Augmented Guardrail for LLM Agent Safety\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SafeHarbor: Hierarchical Memory-Augmented Guardrail for LLM Agent Safety\"",
+      "authors": "Zhe Liu, Zonghao Ying, Wenxin Zhang, Quanchen Zou, Deyue Zhang, Dongdong Yang, Xiangzheng Zhang, Hao Peng",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.05704",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-07",
+      "updated_at": "2026-05-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "computer-use",
+        "memory",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "22",
+      "collection_queries": "agent-safety, autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-05716-more-is-not-always-better-cross-component-interference-in-llm-agent-scaffolding.md",
+    "title": "\"More Is Not Always Better: Cross-Component Interference in LLM Agent Scaffolding\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"More Is Not Always Better: Cross-Component Interference in LLM Agent Scaffolding\"",
+      "authors": "Ming Liu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.05716",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-07",
+      "updated_at": "2026-05-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-06078-milestone-guided-policy-learning-for-long-horizon-language-agents.md",
+    "title": "Milestone-Guided Policy Learning for Long-Horizon Language Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Milestone-Guided Policy Learning for Long-Horizon Language Agents",
+      "authors": "Zixuan Wang, Yuchen Yan, Hongxing Li, Teng Pan, Dingming Li, Ruiqing Zhang, Weiming Lu, Jun Xiao, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.06078",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-07",
+      "updated_at": "2026-05-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-06713-agentic-ai-and-the-industrialization-of-cyber-offense-forecast-consequences-and-.md",
+    "title": "\"Agentic AI and the Industrialization of Cyber Offense: Forecast, Consequences, and Defensive Priorities for Enterprises and the Mittelstand\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Agentic AI and the Industrialization of Cyber Offense: Forecast, Consequences, and Defensive Priorities for Enterprises and the Mittelstand\"",
+      "authors": "Christopher Koch",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.06713",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-06",
+      "updated_at": "2026-05-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "computer-use",
+        "planning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI",
+        "cs.HC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-safety, planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-06716-from-storage-to-experience-a-survey-on-the-evolution-of-llm-agent-memory-mechani.md",
+    "title": "\"From Storage to Experience: A Survey on the Evolution of LLM Agent Memory Mechanisms\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"From Storage to Experience: A Survey on the Evolution of LLM Agent Memory Mechanisms\"",
+      "authors": "Jinghao Luo, Yuchen Tian, Chuxue Cao, Ziyang Luo, Hongzhan Lin, Kaixin Li, Chuyi Kong, Ruichao Yang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.06716",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-07",
+      "updated_at": "2026-05-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "memory",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-06737-a-self-healing-framework-for-reliable-llm-based-autonomous-agents.md",
+    "title": "A Self-Healing Framework for Reliable LLM-Based Autonomous Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "A Self-Healing Framework for Reliable LLM-Based Autonomous Agents",
+      "authors": "Cheonsu Jeong, Younggun Shin",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.06737",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-07",
+      "updated_at": "2026-05-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "multi-agent",
+        "planning",
+        "reasoning",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-06812-towards-security-auditable-llm-agents-a-unified-graph-representation.md",
+    "title": "\"Towards Security-Auditable LLM Agents: A Unified Graph Representation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Towards Security-Auditable LLM Agents: A Unified Graph Representation\"",
+      "authors": "Chaofan Li, Lyuye Zhang, Jintao Zhai, Siyue Feng, Xichun Yang, Huahao Wang, Shihan Dou, Yu Ji, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.06812",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-07",
+      "updated_at": "2026-05-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "multi-agent",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "21",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-06869-agentick-a-unified-benchmark-for-general-sequential-decision-making-agents.md",
+    "title": "\"Agentick: A Unified Benchmark for General Sequential Decision-Making Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Agentick: A Unified Benchmark for General Sequential Decision-Making Agents\"",
+      "authors": "Roger Creus Castanyer, Pablo Samuel Castro, Glen Berseth",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.06869",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-07",
+      "updated_at": "2026-05-12",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "multi-agent",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "23",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-06890-beyond-the-black-box-interpretability-of-agentic-ai-tool-use.md",
+    "title": "\"Beyond the Black Box: Interpretability of Agentic AI Tool Use\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Beyond the Black Box: Interpretability of Agentic AI Tool Use\"",
+      "authors": "Hariom Tatsat, Ariye Shater",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.06890",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-07",
+      "updated_at": "2026-07-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-06957-learning-and-reusing-policy-decompositions-for-hierarchical-generalized-planning.md",
+    "title": "Learning and Reusing Policy Decompositions for Hierarchical Generalized Planning with LLM Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Learning and Reusing Policy Decompositions for Hierarchical Generalized Planning with LLM Agents",
+      "authors": "Shirin Sohrabi, Haritha Ananthakrishnan, Harsha Kokel, Kavitha Srinivas, Michael Katz",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.06957",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-07",
+      "updated_at": "2026-05-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-06992-why-does-agentic-safety-fail-to-generalize-across-tasks.md",
+    "title": "Why Does Agentic Safety Fail to Generalize Across Tasks?",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Why Does Agentic Safety Fail to Generalize Across Tasks?",
+      "authors": "Yonatan Slutzky, Yotam Alexander, Tomer Slor, Yoav Nagel, Nadav Cohen",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.06992",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-07",
+      "updated_at": "2026-05-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "embodied-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "stat.ML"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-07112-switchcraft-ai-model-router-for-agentic-tool-calling.md",
+    "title": "\"Switchcraft: AI Model Router for Agentic Tool Calling\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Switchcraft: AI Model Router for Agentic Tool Calling\"",
+      "authors": "Sharad Agarwal, Pooria Namyar, Alec Wolman, Rahul Ambavat, Ankur Gupta, Qizheng Zhang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.07112",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-08",
+      "updated_at": "2026-05-08",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-07251-can-agents-price-a-reaction-evaluating-llms-on-chemical-cost-reasoning.md",
+    "title": "Can Agents Price a Reaction? Evaluating LLMs on Chemical Cost Reasoning",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Can Agents Price a Reaction? Evaluating LLMs on Chemical Cost Reasoning",
+      "authors": "Yuyang Wu, Yue Huang, Shuaike Shen, Xujian Wang, Shuhao Zhang, Qiyao Xue, Weichen Liu, Runtian Gao, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.07251",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-08",
+      "updated_at": "2026-05-08",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-07830-cybiasbench-benchmarking-bias-in-llm-agents-for-cyber-attack-scenarios.md",
+    "title": "\"CyBiasBench: Benchmarking Bias in LLM Agents for Cyber-Attack Scenarios\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"CyBiasBench: Benchmarking Bias in LLM Agents for Cyber-Attack Scenarios\"",
+      "authors": "Taein Lim, Seongyong Ju, Munhyeok Kim, Hyunjun Kim, Hoki Kim",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.07830",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-08",
+      "updated_at": "2026-05-08",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-08374-memq-integrating-q-learning-into-self-evolving-memory-agents-over-provenance-dag.md",
+    "title": "\"MemQ: Integrating Q-Learning into Self-Evolving Memory Agents over Provenance DAGs\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MemQ: Integrating Q-Learning into Self-Evolving Memory Agents over Provenance DAGs\"",
+      "authors": "Junwei Liao, Haoting Shi, Ruiwen Zhou, Jiaqian Wang, Shengtao Zhang, Wei Zhang, Ying Wen, Zhiyu Li, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.08374",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-08",
+      "updated_at": "2026-05-14",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "embodied-agent",
+        "memory",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-08442-defense-effectiveness-across-architectural-layers-a-mechanistic-evaluation-of-pe.md",
+    "title": "\"Defense effectiveness across architectural layers: a mechanistic evaluation of persistent memory attacks on stateful LLM agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Defense effectiveness across architectural layers: a mechanistic evaluation of persistent memory attacks on stateful LLM agents\"",
+      "authors": "Jun Wen Leong",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.08442",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-08",
+      "updated_at": "2026-07-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "memory",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "23",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-08763-when-llms-team-up-a-coordinated-attack-framework-for-automated-cyber-intrusions.md",
+    "title": "\"When LLMs Team Up: A Coordinated Attack Framework for Automated Cyber Intrusions\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"When LLMs Team Up: A Coordinated Attack Framework for Automated Cyber Intrusions\"",
+      "authors": "Minfeng Qi, Tianqing Zhu, Zijie Xu, Congcong Zhu, Qin Wang, Wanlei Zhou",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.08763",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-09",
+      "updated_at": "2026-05-09",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "planning",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-08876-otora-a-unified-red-teaming-framework-for-reasoning-level-denial-of-service-in-l.md",
+    "title": "\"OTora: A Unified Red Teaming Framework for Reasoning-Level Denial-of-Service in LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"OTora: A Unified Red Teaming Framework for Reasoning-Level Denial-of-Service in LLM Agents\"",
+      "authors": "Xinyu Li, Ronghui Mu, Lin Li, Tianjin Huang, Gaojie Jin",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.08876",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-09",
+      "updated_at": "2026-06-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "computer-use",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-08964-trustworthy-ai-ensuring-reliability-and-accountability-from-models-to-agents.md",
+    "title": "\"Trustworthy AI: Ensuring Reliability and Accountability from Models to Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Trustworthy AI: Ensuring Reliability and Accountability from Models to Agents\"",
+      "authors": "Carol Xuan Long",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.08964",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-09",
+      "updated_at": "2026-05-09",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "multi-agent",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-09168-civex-causal-intervention-verification-for-language-agents.md",
+    "title": "\"CIVeX: Causal Intervention Verification for Language Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"CIVeX: Causal Intervention Verification for Language Agents\"",
+      "authors": "Fabio Rovai",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.09168",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-09",
+      "updated_at": "2026-05-09",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-09692-causal-state-binding-predicts-action-control-in-language-agents.md",
+    "title": "Causal state binding predicts action control in language agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Causal state binding predicts action control in language agents",
+      "authors": "Xiao Jia",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.09692",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-10",
+      "updated_at": "2026-06-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "memory",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-10365-agent-valuebench-a-comprehensive-benchmark-for-evaluating-agent-values.md",
+    "title": "\"Agent-ValueBench: A Comprehensive Benchmark for Evaluating Agent Values\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Agent-ValueBench: A Comprehensive Benchmark for Evaluating Agent Values\"",
+      "authors": "Haonan Dong, Qiguan Feng, Kehan Jiang, Haoran Ye, Xin Zhang, Guojie Song",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.10365",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-11",
+      "updated_at": "2026-05-11",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-10763-matra-modeling-the-attack-surface-of-agentic-ai-systems-openclaw-case-study.md",
+    "title": "\"MATRA: Modeling the Attack Surface of Agentic AI Systems -- OpenClaw Case Study\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MATRA: Modeling the Attack Surface of Agentic AI Systems -- OpenClaw Case Study\"",
+      "authors": "Tim Van hamme, Thomas Vissers, Javier Carnerero-Cano, Mario Fritz, Emil C. Lupu, Lieven Desmet, Dinil Mon Divakaran",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.10763",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-11",
+      "updated_at": "2026-05-11",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-10779-litmus-benchmarking-behavioral-jailbreaks-of-llm-agents-in-real-os-environments.md",
+    "title": "\"LITMUS: Benchmarking Behavioral Jailbreaks of LLM Agents in Real OS Environments\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"LITMUS: Benchmarking Behavioral Jailbreaks of LLM Agents in Real OS Environments\"",
+      "authors": "Chiyu Zhang, Huiqin Yang, Bendong Jiang, Xiaolei Zhang, Yiran Zhao, Ruyi Chen, Lu Zhou, Xiaogang Xu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.10779",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-11",
+      "updated_at": "2026-05-11",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-10870-remember-the-decision-not-the-description-a-rate-distortion-framework-for-agent-.md",
+    "title": "\"Remember the Decision, Not the Description: A Rate-Distortion Framework for Agent Memory\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Remember the Decision, Not the Description: A Rate-Distortion Framework for Agent Memory\"",
+      "authors": "Mingxi Zou, Zhihan Guo, Langzhang Liang, Zhuo Wang, Qifan Wang, Qingsong Wen, Irwin King, Lizhen Qu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.10870",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-11",
+      "updated_at": "2026-05-11",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-11039-the-granularity-mismatch-in-agent-security-argument-level-provenance-solves-enfo.md",
+    "title": "\"The Granularity Mismatch in Agent Security: Argument-Level Provenance Solves Enforcement and Isolates the LLM Reasoning Bottleneck\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"The Granularity Mismatch in Agent Security: Argument-Level Provenance Solves Enforcement and Isolates the LLM Reasoning Bottleneck\"",
+      "authors": "Linfeng Fan, Ziwei Li, Yuan Tian, Yichen Wang, Rongsheng Li, Xiong Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.11039",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-11",
+      "updated_at": "2026-05-11",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-11225-pivot-bridging-planning-and-execution-in-llm-agents-via-trajectory-refinement.md",
+    "title": "\"PIVOT: Bridging Planning and Execution in LLM Agents via Trajectory Refinement\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"PIVOT: Bridging Planning and Execution in LLM Agents via Trajectory Refinement\"",
+      "authors": "Tuo Zhang, Alin-Ionut Popa, Yan Xu, Rui Song, Dimitrios Dimitriadis",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.11225",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-11",
+      "updated_at": "2026-05-11",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.LG",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "autonomous-agent-llm, planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-11388-deep-reasoning-in-general-purpose-agents-via-structured-meta-cognition.md",
+    "title": "Deep Reasoning in General Purpose Agents via Structured Meta-Cognition",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Deep Reasoning in General Purpose Agents via Structured Meta-Cognition",
+      "authors": "Dean Light, Michael Theologitis, Kshitish Ghate, Shuyue Stella Li, Benjamin Newman, Chirag Shah, Aylin Caliskan, Pang Wei Koh, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.11388",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-12",
+      "updated_at": "2026-05-12",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-11534-prism-planning-and-reasoning-with-intent-in-simulated-embodied-environments.md",
+    "title": "\"PRISM: : Planning and Reasoning with Intent in Simulated Embodied Environments\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"PRISM: : Planning and Reasoning with Intent in Simulated Embodied Environments\"",
+      "authors": "Yunn Kang Lim, Pengzhan Sun, Ziyi Bai, Xun Xu, Angela Yao, Xulei Yang, Shijie Li",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.11534",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-12",
+      "updated_at": "2026-05-12",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "embodied-agent",
+        "memory",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.RO"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-11633-can-llm-agents-respond-to-disasters-benchmarking-heterogeneous-geospatial-reason.md",
+    "title": "Can LLM Agents Respond to Disasters? Benchmarking Heterogeneous Geospatial Reasoning in Emergency Operations",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Can LLM Agents Respond to Disasters? Benchmarking Heterogeneous Geospatial Reasoning in Emergency Operations",
+      "authors": "Junjue Wang, Weihao Xuan, Heli Qi, Pengyu Dai, Kunyi Liu, Hongruixuan Chen, Zhuo Zheng, Junshi Xia, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.11633",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-12",
+      "updated_at": "2026-05-12",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "21",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-11882-on-policy-self-evolution-via-failure-trajectories-for-agentic-safety-alignment.md",
+    "title": "On-Policy Self-Evolution via Failure Trajectories for Agentic Safety Alignment",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "On-Policy Self-Evolution via Failure Trajectories for Agentic Safety Alignment",
+      "authors": "Bo Yin, Qi Li, Xinchao Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.11882",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-12",
+      "updated_at": "2026-05-12",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-11928-when-simulation-lies-a-sim-to-real-benchmark-and-domain-randomized-rl-recipe-for.md",
+    "title": "\"When Simulation Lies: A Sim-to-Real Benchmark and Domain-Randomized RL Recipe for Tool-Use Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"When Simulation Lies: A Sim-to-Real Benchmark and Domain-Randomized RL Recipe for Tool-Use Agents\"",
+      "authors": "Xiaolin Zhou, Aojie Yuan, Zheng Luo, Zipeng Ling, Xixiao Pan, Yicheng Gao, Haiyue Zhang, Jiate Li, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.11928",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-12",
+      "updated_at": "2026-05-12",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "function-calling, language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-11946-counterfactual-trace-auditing-of-llm-agent-skills.md",
+    "title": "Counterfactual Trace Auditing of LLM Agent Skills",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Counterfactual Trace Auditing of LLM Agent Skills",
+      "authors": "Xiaolin Zhou, Jinbo Liu, Li Li, Ryan A. Rossi, Xiyang Hu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.11946",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-12",
+      "updated_at": "2026-05-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "planning",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-12015-skillsafetybench-evaluating-agent-safety-under-skill-facing-attack-surfaces.md",
+    "title": "\"SkillSafetyBench: Evaluating Agent Safety under Skill-Facing Attack Surfaces\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SkillSafetyBench: Evaluating Agent Safety under Skill-Facing Attack Surfaces\"",
+      "authors": "Chang Jin, An Wang, Zeming Wei, Kai Wang, Biaojie Zeng, Qiaosheng Zhang, Chao Yang, Jingjing Qu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.12015",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-12",
+      "updated_at": "2026-05-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "memory",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI",
+        "cs.CL",
+        "cs.LG",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-12061-sage-a-self-evolving-agentic-graph-memory-engine-for-structure-aware-associative.md",
+    "title": "\"SAGE: A Self-Evolving Agentic Graph-Memory Engine for Structure-Aware Associative Memory\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SAGE: A Self-Evolving Agentic Graph-Memory Engine for Structure-Aware Associative Memory\"",
+      "authors": "Juntong Wang, Haoyue Zhao, guanghui Pan, Xiyuan Wang, Yanbo Wang, Qiyan Deng, Muhan Zhang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.12061",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-12",
+      "updated_at": "2026-05-12",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-12260-prism-pareto-efficient-retrieval-over-intent-aware-structured-memory-for-long-ho.md",
+    "title": "\"PRISM: Pareto-Efficient Retrieval over Intent-Aware Structured Memory for Long-Horizon Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"PRISM: Pareto-Efficient Retrieval over Intent-Aware Structured Memory for Long-Horizon Agents\"",
+      "authors": "Jingyi Peng, Zhongwei Wan, Weiting Liu, Qiuzhuang Sun",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.12260",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-12",
+      "updated_at": "2026-05-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-13481-personalai-2-0-enhancing-knowledge-graph-traversal-retrieval-with-planning-mecha.md",
+    "title": "\"PersonalAI 2.0: Enhancing knowledge graph traversal/retrieval with planning mechanism for Personalized LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"PersonalAI 2.0: Enhancing knowledge graph traversal/retrieval with planning mechanism for Personalized LLM Agents\"",
+      "authors": "Mikhail Menschikov, Matvey Iskornev, Alexander Kharitonov, Alina Bogdanova, Mikhail Belkin, Ekaterina Lisitsyna, Artyom Sosedka, Victoria Dochkina, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.13481",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-13",
+      "updated_at": "2026-05-13",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-13542-realicu-do-llm-agents-understand-long-context-icu-data-a-benchmark-beyond-behavi.md",
+    "title": "\"RealICU: Do LLM Agents Understand Long-Context ICU Data? A Benchmark Beyond Behavior Imitation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"RealICU: Do LLM Agents Understand Long-Context ICU Data? A Benchmark Beyond Behavior Imitation\"",
+      "authors": "Chengzhi Shen, Weixiang Shen, Tobias Susetzky, Chen, Chen, Jun Li, Yuyuan Liu, Xuepeng Zhang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.13542",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-13",
+      "updated_at": "2026-05-13",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL",
+        "cs.LG",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-13618-openaaas-an-open-agent-as-a-service-framework-for-distributed-materials-informat.md",
+    "title": "\"OpenAaaS: An Open Agent-as-a-Service Framework for Distributed Materials-Informatics Research\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"OpenAaaS: An Open Agent-as-a-Service Framework for Distributed Materials-Informatics Research\"",
+      "authors": "Peng Kang, Bixuan Li, Xiaoya Huang, Shuo Shi, Weiqiao Zhou, Zhen Li, Yu Liu, Lei Zheng",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.13618",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-13",
+      "updated_at": "2026-05-13",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "memory",
+        "multi-agent",
+        "planning",
+        "rag",
+        "reasoning",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cond-mat.mtrl-sci",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-13716-skillops-managing-llm-agent-skill-libraries-as-self-maintaining-software-ecosyst.md",
+    "title": "\"SkillOps: Managing LLM Agent Skill Libraries as Self-Maintaining Software Ecosystems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SkillOps: Managing LLM Agent Skill Libraries as Self-Maintaining Software Ecosystems\"",
+      "authors": "Hongji Pu, Xinyuan Song, Liang Zhao",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.13716",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-13",
+      "updated_at": "2026-05-13",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-14126-reinforcement-learning-for-tool-calling-agents-in-fast-healthcare-interoperabili.md",
+    "title": "Reinforcement Learning for Tool-Calling Agents in Fast Healthcare Interoperability Resources (FHIR)",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Reinforcement Learning for Tool-Calling Agents in Fast Healthcare Interoperability Resources (FHIR)",
+      "authors": "Marius S. Knorr, Robert Müller, Jan P. Bremer, Nils Schweingruber",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.14126",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-13",
+      "updated_at": "2026-05-13",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-14290-web-agents-should-adopt-the-plan-then-execute-paradigm.md",
+    "title": "Web Agents Should Adopt the Plan-Then-Execute Paradigm",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Web Agents Should Adopt the Plan-Then-Execute Paradigm",
+      "authors": "Julien Piet, Annabella Chow, Yiwei Hou, Muxi Lyu, Sylvie Venuto, Jinhao Zhu, Raluca Ada Popa, David Wagner",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.14290",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-14",
+      "updated_at": "2026-05-14",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI",
+        "cs.CL",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-14322-are-agents-ready-to-teach-a-multi-stage-benchmark-for-real-world-teaching-workfl.md",
+    "title": "Are Agents Ready to Teach? A Multi-Stage Benchmark for Real-World Teaching Workflows",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Are Agents Ready to Teach? A Multi-Stage Benchmark for Real-World Teaching Workflows",
+      "authors": "Zixin Chen, Peng Liu, Rui Sheng, Haobo Li, Jianhong Tu, Xiaodong Deng, Kashun Shum, Dayiheng Liu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.14322",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-14",
+      "updated_at": "2026-05-20",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-14421-memlineage-lineage-guided-enforcement-for-llm-agent-memory.md",
+    "title": "\"MemLineage: Lineage-Guided Enforcement for LLM Agent Memory\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MemLineage: Lineage-Guided Enforcement for LLM Agent Memory\"",
+      "authors": "Ciyan Ouyang, Rui Hou",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.14421",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-14",
+      "updated_at": "2026-05-14",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-14460-exploiting-llm-agent-supply-chains-via-payload-less-skills.md",
+    "title": "Exploiting LLM Agent Supply Chains via Payload-less Skills",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Exploiting LLM Agent Supply Chains via Payload-less Skills",
+      "authors": "Xinyu Liu, Yukai Zhao, Xing Hu, Xin Xia",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.14460",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-14",
+      "updated_at": "2026-05-14",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-14498-groupmembench-benchmarking-llm-agent-memory-in-multi-party-conversations.md",
+    "title": "\"GroupMemBench: Benchmarking LLM Agent Memory in Multi-Party Conversations\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"GroupMemBench: Benchmarking LLM Agent Memory in Multi-Party Conversations\"",
+      "authors": "Jingbo Yang, Kwei-Herng Lai, Xiaowen Wang, Shiyu Chang, Yaar Harari, Evgeniy Gabrilovich",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.14498",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-14",
+      "updated_at": "2026-05-16",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "21",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-14527-lang2mlip-end-to-end-language-to-machine-learning-interatomic-potential-developm.md",
+    "title": "\"Lang2MLIP: End-to-End Language-to-Machine Learning Interatomic Potential Development with Autonomous Agentic Workflows\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Lang2MLIP: End-to-End Language-to-Machine Learning Interatomic Potential Development with Autonomous Agentic Workflows\"",
+      "authors": "Wenwen Li, Yuki Orimo, Nontawat Charoenphakdee",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.14527",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-14",
+      "updated_at": "2026-05-14",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "tool-use",
+        "workflow-agent",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cond-mat.mtrl-sci",
+        "physics.comp-ph"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-14892-beyond-individual-intelligence-surveying-collaboration-failure-attribution-and-s.md",
+    "title": "\"Beyond Individual Intelligence: Surveying Collaboration, Failure Attribution, and Self-Evolution in LLM-based Multi-Agent Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Beyond Individual Intelligence: Surveying Collaboration, Failure Attribution, and Self-Evolution in LLM-based Multi-Agent Systems\"",
+      "authors": "Shihao Qi, Jie Ma, Rui Xing, Wei Guo, Xiao Huang, Zhitao Gao, Jianhao Deng, Jun Liu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.14892",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-14",
+      "updated_at": "2026-05-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "multi-agent",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-14906-memlens-benchmarking-multimodal-long-term-memory-in-large-vision-language-models.md",
+    "title": "\"MemLens: Benchmarking Multimodal Long-Term Memory in Large Vision-Language Models\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MemLens: Benchmarking Multimodal Long-Term Memory in Large Vision-Language Models\"",
+      "authors": "Xiyu Ren, Zhaowei Wang, Yiming Du, Zhongwei Xie, Chi Liu, Xinlin Yang, Haoyue Feng, Wenjun Pan, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.14906",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-14",
+      "updated_at": "2026-05-14",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-14932-toward-securing-ai-agents-like-operating-systems.md",
+    "title": "Toward Securing AI Agents Like Operating Systems",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Toward Securing AI Agents Like Operating Systems",
+      "authors": "Lukas Pirch, Micha Horlboge, Patrick Großmann, Syeda Mahnur Asif, Klim Kireev, Thorsten Holz, Konrad Rieck",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.14932",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-14",
+      "updated_at": "2026-05-14",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-15040-orchard-an-open-source-agentic-modeling-framework.md",
+    "title": "\"Orchard: An Open-Source Agentic Modeling Framework\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Orchard: An Open-Source Agentic Modeling Framework\"",
+      "authors": "Baolin Peng, Wenlin Yao, Qianhui Wu, Hao Cheng, Xiao Yu, Rui Yang, Tao Ge, Alessandro Sordoni, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.15040",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-14",
+      "updated_at": "2026-05-21",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "computer-use",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-15128-memeye-a-visual-centric-evaluation-framework-for-multimodal-agent-memory.md",
+    "title": "\"MemEye: A Visual-Centric Evaluation Framework for Multimodal Agent Memory\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MemEye: A Visual-Centric Evaluation Framework for Multimodal Agent Memory\"",
+      "authors": "Minghao Guo, Qingyue Jiao, Zeru Shi, Yihao Quan, Boxuan Zhang, Danrui Li, Liwei Che, Wujiang Xu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.15128",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-14",
+      "updated_at": "2026-05-14",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV",
+        "cs.CL",
+        "cs.IR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-15206-agentstop-terminating-local-ai-agents-early-to-save-energy-in-consumer-devices.md",
+    "title": "\"AgentStop: Terminating Local AI Agents Early to Save Energy in Consumer Devices\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AgentStop: Terminating Local AI Agents Early to Save Energy in Consumer Devices\"",
+      "authors": "Dzung Pham, Kleomenis Katevas, Ali Shahin Shamsabadi, Hamed Haddadi",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.15206",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-01",
+      "updated_at": "2026-05-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI",
+        "cs.DC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-15625-colpackagent-agent-skill-guided-hard-particle-monte-carlo-workflows-for-colloida.md",
+    "title": "\"ColPackAgent: Agent-Skill-Guided Hard-Particle Monte Carlo Workflows for Colloidal Packing\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ColPackAgent: Agent-Skill-Guided Hard-Particle Monte Carlo Workflows for Colloidal Packing\"",
+      "authors": "Lijie Ding, Changwoo Do",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.15625",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-15",
+      "updated_at": "2026-05-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "tool-use",
+        "workflow-agent",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cond-mat.soft"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-15701-h-mem-a-novel-memory-mechanism-for-evolving-and-retrieving-agent-memory-via-a-hy.md",
+    "title": "\"H-Mem: A Novel Memory Mechanism for Evolving and Retrieving Agent Memory via a Hybrid Structure\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"H-Mem: A Novel Memory Mechanism for Evolving and Retrieving Agent Memory via a Hybrid Structure\"",
+      "authors": "Jiawei Yu, Yixiang Fang, Xilin Liu, Yuchi Ma",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.15701",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-15",
+      "updated_at": "2026-05-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-15710-smmbench-a-benchmark-for-source-distributed-multimodal-agent-memory.md",
+    "title": "\"SMMBench: A Benchmark for Source-Distributed Multimodal Agent Memory\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SMMBench: A Benchmark for Source-Distributed Multimodal Agent Memory\"",
+      "authors": "Huacan Chai, Yukai Wang, Yingxuan Yang, Dan Peng, Yuanyi Song, Zhihui Fu, Weiwen Liu, Jianghao Lin, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.15710",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-15",
+      "updated_at": "2026-05-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-15759-dimmem-dimensional-structuring-for-efficient-long-term-agent-memory.md",
+    "title": "\"DimMem: Dimensional Structuring for Efficient Long-Term Agent Memory\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"DimMem: Dimensional Structuring for Efficient Long-Term Agent Memory\"",
+      "authors": "Wentao Qiu, Haotian Hu, Fanyi Wang, Jinwei Kong, Yu Zhang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.15759",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-15",
+      "updated_at": "2026-05-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-16233-forge-self-evolving-agent-memory-with-no-weight-updates-via-population-broadcast.md",
+    "title": "\"FORGE: Self-Evolving Agent Memory With No Weight Updates via Population Broadcast\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"FORGE: Self-Evolving Agent Memory With No Weight Updates via Population Broadcast\"",
+      "authors": "Igor Bogdanov, Chung-Horng Lung, Thomas Kunz, Jie Gao, Adrian Taylor, Marzia Zaman",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.16233",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-15",
+      "updated_at": "2026-05-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL",
+        "cs.LG",
+        "cs.MA",
+        "eess.SY"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-16481-visual-agentic-memory-enabling-online-long-video-understanding-via-online-indexi.md",
+    "title": "\"Visual Agentic Memory: Enabling Online Long Video Understanding via Online Indexing, Hierarchical Memory, and Agentic Retrieval\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Visual Agentic Memory: Enabling Online Long Video Understanding via Online Indexing, Hierarchical Memory, and Agentic Retrieval\"",
+      "authors": "Aiden Yiliu Li, Nels Numan, Anthony Steed",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.16481",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-15",
+      "updated_at": "2026-05-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-16821-multi-paradigm-agent-interaction-in-practice-a-systematic-analysis-of-generator-.md",
+    "title": "\"Multi-Paradigm Agent Interaction in Practice:A Systematic Analysis of Generator-Evaluator, ReAct Loop,and Adversarial Evaluation in the buddyMe Framework\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Multi-Paradigm Agent Interaction in Practice:A Systematic Analysis of Generator-Evaluator, ReAct Loop,and Adversarial Evaluation in the buddyMe Framework\"",
+      "authors": "Xiaohua Wang, Chao Han, Kai Yu, XiaoLiang Xu, Liang Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.16821",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-16",
+      "updated_at": "2026-05-16",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "multi-agent",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-17075-a-red-teaming-framework-for-evaluating-robustness-of-ai-enabled-security-orchest.md",
+    "title": "A Red Teaming Framework for Evaluating Robustness of AI-enabled Security Orchestration, Automation, and Response Systems",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "A Red Teaming Framework for Evaluating Robustness of AI-enabled Security Orchestration, Automation, and Response Systems",
+      "authors": "Ayan Javeed Shaikh, Nathaniel D. Bastian, Ankit Shah",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.17075",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-16",
+      "updated_at": "2026-05-16",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "tool-use",
+        "workflow-agent",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-17348-taming-zombie-agents-a-markov-state-aware-framework-for-resilient-multi-agent-ev.md",
+    "title": "\"Taming \\\"Zombie'' Agents: A Markov State-Aware Framework for Resilient Multi-Agent Evolution\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Taming \\\"Zombie'' Agents: A Markov State-Aware Framework for Resilient Multi-Agent Evolution\"",
+      "authors": "Taolin Zhang, Pukun Zhao, Qizhou Chen, Jiuheng Wan, Chen Chen, Xiaofeng He, Chengyu Wang, Richang Hong",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.17348",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-17",
+      "updated_at": "2026-05-17",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "memory",
+        "multi-agent",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-17453-trust-no-tool-evaluating-and-defending-llm-agents-under-untrusted-tool-feedback.md",
+    "title": "\"Trust No Tool: Evaluating and Defending LLM Agents under Untrusted Tool Feedback\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Trust No Tool: Evaluating and Defending LLM Agents under Untrusted Tool Feedback\"",
+      "authors": "Lecheng Yan, Ruizhe Li, Xicheng Han, Wenxi Li, Binwu Wang, Longyue Wang, Chenyang Lyu, Guanhua Chen",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.17453",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-17",
+      "updated_at": "2026-05-17",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-17625-episodic-semantic-memory-architecture-for-long-horizon-scientific-agents.md",
+    "title": "Episodic-Semantic Memory Architecture for Long-Horizon Scientific Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Episodic-Semantic Memory Architecture for Long-Horizon Scientific Agents",
+      "authors": "Nikola Milosevic",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.17625",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-17",
+      "updated_at": "2026-05-17",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-18284-commitdistill-a-lightweight-knowledge-centric-memory-layer-for-software-reposito.md",
+    "title": "\"CommitDistill: A Lightweight Knowledge-Centric Memory Layer for Software Repositories\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"CommitDistill: A Lightweight Knowledge-Centric Memory Layer for Software Repositories\"",
+      "authors": "Divya Chukkapalli, Thejesh Avula, Aditya Aggarwal, Harsimran Singh, Amith Tallanki",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.18284",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-18",
+      "updated_at": "2026-05-18",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "computer-use",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-18502-the-distance-based-formation-controller-design-for-multi-agent-systems-in-port-h.md",
+    "title": "The distance-based formation controller design for multi-agent systems in port-Hamiltonian form",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "The distance-based formation controller design for multi-agent systems in port-Hamiltonian form",
+      "authors": "Jingyi Zhao, Yongxin Wu, Héctor García de Marina, Yuhu Wu, Yann Le Gorrec",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.18502",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-18",
+      "updated_at": "2026-05-18",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "math.OC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-18652-mementogui-learning-agentic-multimodal-memory-control-for-long-horizon-gui-agent.md",
+    "title": "\"MementoGUI: Learning Agentic Multimodal Memory Control for Long-Horizon GUI Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MementoGUI: Learning Agentic Multimodal Memory Control for Long-Horizon GUI Agents\"",
+      "authors": "Ziyun Zeng, Hang Hua, Bocheng Zou, Mu Cai, Rogerio Feris, Jiebo Luo",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.18652",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-18",
+      "updated_at": "2026-05-18",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "22",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-18672-position-a-three-layer-probabilistic-assume-guarantee-architecture-is-structural.md",
+    "title": "\"Position: A Three-Layer Probabilistic Assume-Guarantee Architecture Is Structurally Required for Safe LLM Agent Deployment\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Position: A Three-Layer Probabilistic Assume-Guarantee Architecture Is Structurally Required for Safe LLM Agent Deployment\"",
+      "authors": "S. Bensalem, Y. Dong, M. Franzle, X. Huang, J. Kroger, D. Nickovic, A. Nouri, R. Roy, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.18672",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-18",
+      "updated_at": "2026-05-18",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "multi-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-18930-oep-poisoning-self-evolving-llm-agents-via-locally-correct-but-non-transferable-.md",
+    "title": "\"OEP: Poisoning Self-Evolving LLM Agents via Locally Correct but Non-Transferable Experiences\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"OEP: Poisoning Self-Evolving LLM Agents via Locally Correct but Non-Transferable Experiences\"",
+      "authors": "Kaixiang Wang, Jiong Lou, Zhaojiacheng Zhou, Jie Li",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.18930",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-18",
+      "updated_at": "2026-05-18",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-19604-formal-skill-programmable-runtime-skills-for-efficient-and-accurate-llm-agents.md",
+    "title": "\"Formal Skill: Programmable Runtime Skills for Efficient and Accurate LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Formal Skill: Programmable Runtime Skills for Efficient and Accurate LLM Agents\"",
+      "authors": "Xi Zhang, Meijun Gao, Yuntian Zhao, Xinyu Tan, Yilun Yao, Feiyu Wang, Yanshu Wang, Dingsiyi, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.19604",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-19",
+      "updated_at": "2026-05-19",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-19952-rethinking-how-to-remember-beyond-atomic-facts-in-lifelong-llm-agent-memory.md",
+    "title": "\"Rethinking How to Remember: Beyond Atomic Facts in Lifelong LLM Agent Memory\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Rethinking How to Remember: Beyond Atomic Facts in Lifelong LLM Agent Memory\"",
+      "authors": "Jingwei Sun, Jianing Zhu, Jiangchao Yao, Tongliang Liu, Bo Han",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.19952",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-19",
+      "updated_at": "2026-05-19",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-20306-wildroadbench-a-wild-aerial-road-damage-grounding-benchmark-for-vision-language-.md",
+    "title": "\"WildRoadBench: A Wild Aerial Road-Damage Grounding Benchmark for Vision-Language Models and Autonomous Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"WildRoadBench: A Wild Aerial Road-Damage Grounding Benchmark for Vision-Language Models and Autonomous Agents\"",
+      "authors": "Bingnan Liu, Chenhang Cui, Rui Huang, Jiani Luo, Zhirong Shen, Tinghao Wang, Xiande Huang, Lingbei Meng, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.20306",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-19",
+      "updated_at": "2026-06-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-20315-mix-quant-quantized-prefilling-precise-decoding-for-agentic-llms.md",
+    "title": "\"Mix-Quant: Quantized Prefilling, Precise Decoding for Agentic LLMs\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Mix-Quant: Quantized Prefilling, Precise Decoding for Agentic LLMs\"",
+      "authors": "Haiquan Lu, Zigeng Chen, Gongfan Fang, Xinyin Ma, Xinchao Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.20315",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-19",
+      "updated_at": "2026-05-19",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "memory",
+        "planning",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "21",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-20616-auto-dreamer-learning-offline-memory-consolidation-for-language-agents.md",
+    "title": "\"Auto-Dreamer: Learning Offline Memory Consolidation for Language Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Auto-Dreamer: Learning Offline Memory Consolidation for Language Agents\"",
+      "authors": "Chongrui Ye, Yuxiang Liu, Yu Wang, Haofei Yu, Yining Zhao, Ge Liu, Julian McAuley, Jiaxuan You",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.20616",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-20",
+      "updated_at": "2026-05-20",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-memory, language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-20833-memgym-a-long-horizon-memory-environment-for-llm-agents.md",
+    "title": "\"MemGym: a Long-Horizon Memory Environment for LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MemGym: a Long-Horizon Memory Environment for LLM Agents\"",
+      "authors": "Wujiang Xu, Yu Wang, Kai Mei, Kaiqu Liang, Zhenting Wang, Mingyu Jin, Han Zhang, Shi-Xiong Zhang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.20833",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-20",
+      "updated_at": "2026-05-20",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "computer-use",
+        "embodied-agent",
+        "memory",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "26",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-20874-governance-by-construction-for-generalist-agents.md",
+    "title": "Governance by Construction for Generalist Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Governance by Construction for Generalist Agents",
+      "authors": "Segev Shlomov, Iftach Shoham, Alon Oved, Ido Levy, Sami Marreed, Harold Ship, Offer Akrabi, Sergey Zeltyn, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.20874",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-20",
+      "updated_at": "2026-05-20",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "computer-use",
+        "planning",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-21240-apex-autonomous-policy-exploration-for-self-evolving-llm-agents.md",
+    "title": "\"APEX: Autonomous Policy Exploration for Self-Evolving LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"APEX: Autonomous Policy Exploration for Self-Evolving LLM Agents\"",
+      "authors": "Yibo Li, Jiashuo Yang, Zhi Zheng, Zhiyuan Hu, Yuan Sui, Shizun Wang, Yufei He, Bryan Hooi",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.21240",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-20",
+      "updated_at": "2026-05-20",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-21740-smdd-bench-can-llms-solve-real-world-small-molecule-drug-design-tasks.md",
+    "title": "\"SMDD-Bench: Can LLMs Solve Real-World Small Molecule Drug Design Tasks?\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SMDD-Bench: Can LLMs Solve Real-World Small Molecule Drug Design Tasks?\"",
+      "authors": "Kevin Han, Renfei Zhang, Kathy Wei, Hamed Mahdavi, Niloofar Mireshghallah, Amir Barati Farimani",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.21740",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-20",
+      "updated_at": "2026-05-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-22154-idlespec-exploiting-idle-time-via-speculative-planning-for-llm-agents.md",
+    "title": "\"IdleSpec: Exploiting Idle Time via Speculative Planning for LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"IdleSpec: Exploiting Idle Time via Speculative Planning for LLM Agents\"",
+      "authors": "Daewon Choi, Kyunghyun Park, Woomin Song, Saket Dingliwal, Sai Muralidhar Jayanthi, Jinwoo Shin, Aram Galstyan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.22154",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-21",
+      "updated_at": "2026-05-21",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-22321-benchmarking-autonomous-agents-against-temporal-spatial-and-semantic-evasions.md",
+    "title": "Benchmarking Autonomous Agents against Temporal, Spatial, and Semantic Evasions",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Benchmarking Autonomous Agents against Temporal, Spatial, and Semantic Evasions",
+      "authors": "Jianan Ma, Xiaohu Du, Ruixiao Lin, Yaoxiang Bian, Jialuo Chen, Jingyi Wang, Xiaofang Yang, Shiwen Cui, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.22321",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-21",
+      "updated_at": "2026-05-21",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-22643-boiling-the-frog-a-multi-turn-benchmark-for-agentic-safety.md",
+    "title": "\"Boiling the Frog: A Multi-Turn Benchmark for Agentic Safety\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Boiling the Frog: A Multi-Turn Benchmark for Agentic Safety\"",
+      "authors": "Piercosma Bisconti, Matteo Prandi, Federico Pierucci, Federico Sartore, Enrico Panai, Laura Caroli, Yue Zhu, Adam Leon Smith, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.22643",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-21",
+      "updated_at": "2026-05-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-23067-what-training-data-teaches-rl-memory-agents-an-empirical-study-of-curriculum-eff.md",
+    "title": "\"What Training Data Teaches RL Memory Agents: An Empirical Study of Curriculum Effects in Memory-Augmented QA\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"What Training Data Teaches RL Memory Agents: An Empirical Study of Curriculum Effects in Memory-Augmented QA\"",
+      "authors": "Xinjie He, Zhiyuan Lin, Su Liu, Jialun Wu, Qiyang Xie, Weikai Zhou, Shuai Xiao",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.23067",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-21",
+      "updated_at": "2026-05-21",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-23574-push-your-agent-measuring-and-enforcing-quantitative-goal-persistence-in-long-ho.md",
+    "title": "\"Push Your Agent: Measuring and Enforcing Quantitative Goal Persistence in Long-Horizon LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Push Your Agent: Measuring and Enforcing Quantitative Goal Persistence in Long-Horizon LLM Agents\"",
+      "authors": "Yuandao Cai, Yuzhang Zhu, Liyou Gao, Wensheng Tang, Shengchao Qin",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.23574",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-22",
+      "updated_at": "2026-05-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-23636-rf-instrument-agent-rfia-empowering-rf-instruments-with-natural-language-underst.md",
+    "title": "\"RF Instrument Agent (RFIA): Empowering RF Instruments with Natural Language Understanding, Scheduling and Execution of Complex Tasks\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"RF Instrument Agent (RFIA): Empowering RF Instruments with Natural Language Understanding, Scheduling and Execution of Complex Tasks\"",
+      "authors": "Chunhui Li, Wei Fan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.23636",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-22",
+      "updated_at": "2026-05-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "eess.SY"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-23723-memaudit-post-hoc-auditing-of-poisoned-agent-memory-via-causal-attribution-and-s.md",
+    "title": "\"MemAudit: Post-hoc Auditing of Poisoned Agent Memory via Causal Attribution and Structural Anomaly Detection\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MemAudit: Post-hoc Auditing of Poisoned Agent Memory via Causal Attribution and Structural Anomaly Detection\"",
+      "authors": "Zhewen Tan, Yilun Yao, Huiyan Jin, Wenhan Yu, Guoan Wang, Mengyuan Fan, liang lu, Feng Liu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.23723",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-22",
+      "updated_at": "2026-05-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-23899-from-raw-experience-to-skill-consumption-a-systematic-study-of-model-generated-a.md",
+    "title": "\"From Raw Experience to Skill Consumption: A Systematic Study of Model-Generated Agent Skills\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"From Raw Experience to Skill Consumption: A Systematic Study of Model-Generated Agent Skills\"",
+      "authors": "Zisu Huang, Jingwen Xu, Yifan Yang, Ziyang Gong, Qihao Yang, Muzhao Tian, Xiaohua Wang, Changze Lv, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.23899",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-22",
+      "updated_at": "2026-05-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "computer-use",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-23986-memforest-an-efficient-agent-memory-system-with-hierarchical-temporal-indexing.md",
+    "title": "\"MemForest: An Efficient Agent Memory System with Hierarchical Temporal Indexing\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MemForest: An Efficient Agent Memory System with Hierarchical Temporal Indexing\"",
+      "authors": "Han Chen, Zining Zhang, Wenqi Pei, Bingsheng He, Ming Wu, Jason Zeng, Michael Heinrich, Wei Wu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.23986",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-16",
+      "updated_at": "2026-05-16",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.DB",
+        "cs.AI",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-24069-when-the-manual-lies-a-realistic-benchmark-to-evaluate-mcp-poisoning-attacks-for.md",
+    "title": "\"When the Manual Lies: A Realistic Benchmark to Evaluate MCP Poisoning Attacks for LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"When the Manual Lies: A Realistic Benchmark to Evaluate MCP Poisoning Attacks for LLM Agents\"",
+      "authors": "Shi Liu, Xuehai Tang, Xikang Yang, Liang Lin, Biyu Zhou, Wenjie Xiao, Wantao Liu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.24069",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-22",
+      "updated_at": "2026-05-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-24216-agent-tom-learning-to-monitor-autonomous-llm-agents-via-theory-of-mind-reasoning.md",
+    "title": "\"Agent-ToM: Learning to Monitor Autonomous LLM Agents via Theory-of-Mind Reasoning\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Agent-ToM: Learning to Monitor Autonomous LLM Agents via Theory-of-Mind Reasoning\"",
+      "authors": "Nesreen K. Ahmed, Nima Nafisi",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.24216",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-22",
+      "updated_at": "2026-05-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "memory",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI",
+        "cs.CL",
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-24219-beyond-final-answers-auditing-trajectory-level-hallucinations-in-multi-agent-ind.md",
+    "title": "\"Beyond Final Answers: Auditing Trajectory-Level Hallucinations in Multi-Agent Industrial Workflows\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Beyond Final Answers: Auditing Trajectory-Level Hallucinations in Multi-Agent Industrial Workflows\"",
+      "authors": "Harshada Badave, Santosh Borse, Andrea Gomez, Harshitha Narahari, Sara Carter, Vishwa Bhatt, Aishani Rachakonda, Shuxin Lin, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.24219",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-22",
+      "updated_at": "2026-05-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-24220-polar-agentic-rl-on-any-harness-at-scale.md",
+    "title": "\"Polar: Agentic RL on Any Harness at Scale\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Polar: Agentic RL on Any Harness at Scale\"",
+      "authors": "Binfeng Xu, Hao Zhang, Shaokun Zhang, Songyang Han, Mingjie Liu, Jian Hu, Shizhe Diao, Zhenghui Jin, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.24220",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-22",
+      "updated_at": "2026-05-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "multi-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.DC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-24309-reframing-llm-agent-security-as-an-agent-human-interaction-problem.md",
+    "title": "Reframing LLM Agent Security as an Agent-Human Interaction Problem",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Reframing LLM Agent Security as an Agent-Human Interaction Problem",
+      "authors": "Peiran Wang, Ying Li, Yuan Tian",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.24309",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-23",
+      "updated_at": "2026-05-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-24659-iterinject-indirect-prompt-injection-against-llm-agents-via-feedback-guided-iter.md",
+    "title": "\"IterInject: Indirect Prompt Injection Against LLM Agents via Feedback-Guided Iterative Optimization\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"IterInject: Indirect Prompt Injection Against LLM Agents via Feedback-Guided Iterative Optimization\"",
+      "authors": "Zixuan Chen, Jiaxiang Chen, Li Luo, Ke Xu, Xiaoxiang Huang, Tanfeng Sun, Xinghao Jiang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.24659",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-23",
+      "updated_at": "2026-05-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "coding-agent",
+        "computer-use",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-24812-core-code-collaborative-reinforcement-learning-for-code-generation.md",
+    "title": "\"CoRe-Code: Collaborative Reinforcement Learning for Code Generation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"CoRe-Code: Collaborative Reinforcement Learning for Code Generation\"",
+      "authors": "Zhihao Dou, Qinjian Zhao, Zhongwei Wan, Xiaoyu Xia, Sumon Biswas",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.24812",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-24",
+      "updated_at": "2026-05-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "memory",
+        "multi-agent",
+        "planning",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-25141-llm-agent-based-renewable-energy-forecasting-using-edge-and-iot-data-a-review-of.md",
+    "title": "LLM Agent Based Renewable Energy Forecasting Using Edge and IoT Data A Review of Solar Wind Weather and Grid Aware Decision Support",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "LLM Agent Based Renewable Energy Forecasting Using Edge and IoT Data A Review of Solar Wind Weather and Grid Aware Decision Support",
+      "authors": "Pavan Manjunath, Thomas Pruefer",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.25141",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-24",
+      "updated_at": "2026-05-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-25200-grouptravelbench-benchmarking-llm-agents-on-multi-person-travel-planning.md",
+    "title": "\"GroupTravelBench: Benchmarking LLM Agents on Multi-Person Travel Planning\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"GroupTravelBench: Benchmarking LLM Agents on Multi-Person Travel Planning\"",
+      "authors": "Xiang Cheng, Yulan Hu, Lulu Zheng, Zheng Pan, Xin Li, Yong Liu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.25200",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-24",
+      "updated_at": "2026-06-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-25310-tool-call-dependency-structure-is-linearly-decodable-in-llm-agent-residual-strea.md",
+    "title": "Tool-Call Dependency Structure is Linearly Decodable in LLM Agent Residual Streams",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Tool-Call Dependency Structure is Linearly Decodable in LLM Agent Residual Streams",
+      "authors": "Tianda Sun, Dimitar Kazakov",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.25310",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-25",
+      "updated_at": "2026-05-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-25393-decision-making-with-lightweight-confidence-aware-language-model-for-autonomous-.md",
+    "title": "Decision-Making with Lightweight Confidence-Aware Language Model for Autonomous Driving",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Decision-Making with Lightweight Confidence-Aware Language Model for Autonomous Driving",
+      "authors": "Ruoyu Yao, Ruiguo Zhong, Pei Liu, Mingxing Peng, Rui Yang, Jun Ma",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.25393",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-25",
+      "updated_at": "2026-05-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.RO"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-25435-security-of-openclaw-agents-fundamentals-attacks-and-countermeasures.md",
+    "title": "\"Security of OpenClaw Agents: Fundamentals, Attacks, and Countermeasures\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Security of OpenClaw Agents: Fundamentals, Attacks, and Countermeasures\"",
+      "authors": "Yuntao Wang, Jianle Ba, Han Liu, Yanghe Pan, Jintao Wei, Zhou Su, Tom H. Luan, Linkang Du",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.25435",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-25",
+      "updated_at": "2026-05-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "computer-use",
+        "memory",
+        "multi-agent",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-25920-can-llms-time-travel-enhancing-temporal-consistency-in-legal-agentic-search-thro.md",
+    "title": "Can LLMs Time Travel? Enhancing Temporal Consistency in Legal Agentic Search through Reinforcement Learning",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Can LLMs Time Travel? Enhancing Temporal Consistency in Legal Agentic Search through Reinforcement Learning",
+      "authors": "Wei Fan, Yining Zhou, Mufan Zhang, Yanbing Weng, Yiran HU, Tianshi Zheng, Baixuan Xu, Chunyang Li, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.25920",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-25",
+      "updated_at": "2026-05-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-26165-tool-schema-compression-enables-agentic-rag-under-constrained-context-budgets.md",
+    "title": "Tool-Schema Compression Enables Agentic RAG Under Constrained Context Budgets",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Tool-Schema Compression Enables Agentic RAG Under Constrained Context Budgets",
+      "authors": "Furkan Sakizli",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.26165",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-24",
+      "updated_at": "2026-05-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-26252-is-agent-memory-a-database-rethinking-data-foundations-for-long-term-ai-agent-me.md",
+    "title": "Is Agent Memory a Database? Rethinking Data Foundations for Long-Term AI Agent Memory",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Is Agent Memory a Database? Rethinking Data Foundations for Long-Term AI Agent Memory",
+      "authors": "Abdelghny Orogat, Essam Mansour",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.26252",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-25",
+      "updated_at": "2026-05-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.DB"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-26305-experiments-in-agentic-ai-for-science.md",
+    "title": "Experiments in Agentic AI for Science",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Experiments in Agentic AI for Science",
+      "authors": "Judy Fox, Geoffrey Fox",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.26305",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-25",
+      "updated_at": "2026-05-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "eess.SY",
+        "hep-ph"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "autonomous-agent-llm, rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-26497-aligning-provenance-with-authorization-a-dual-graph-defense-for-llm-agents.md",
+    "title": "\"Aligning Provenance with Authorization: A Dual-Graph Defense for LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Aligning Provenance with Authorization: A Dual-Graph Defense for LLM Agents\"",
+      "authors": "Peiran Wang, Ying Li, Yuan Tian",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.26497",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-26",
+      "updated_at": "2026-05-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-26720-towards-feedback-to-plan-decisions-for-self-evolving-llm-agents-in-cuda-kernel-g.md",
+    "title": "Towards Feedback-to-Plan Decisions for Self-Evolving LLM Agents in CUDA Kernel Generation",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Towards Feedback-to-Plan Decisions for Self-Evolving LLM Agents in CUDA Kernel Generation",
+      "authors": "Yee Hin Chong, Jiaming Wu, Youhui Zhang, Peng Qu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.26720",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-26",
+      "updated_at": "2026-05-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-26926-from-norms-to-indicators-n2i-rag-an-agentic-retrieval-augmented-generation-frame.md",
+    "title": "\"From Norms to Indicators (N2I-RAG): An Agentic Retrieval-Augmented Generation Framework for Legal Indicator Computation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"From Norms to Indicators (N2I-RAG): An Agentic Retrieval-Augmented Generation Framework for Legal Indicator Computation\"",
+      "authors": "Youssef Al Mouatamid, Marie Bonnin, Jihad Zahir",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.26926",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-26",
+      "updated_at": "2026-05-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-27123-rethinking-agentic-rag-toward-llm-driven-logical-retrieval-beyond-embeddings.md",
+    "title": "\"Rethinking Agentic RAG: Toward LLM-Driven Logical Retrieval Beyond Embeddings\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Rethinking Agentic RAG: Toward LLM-Driven Logical Retrieval Beyond Embeddings\"",
+      "authors": "Yuqi Zeng, Qixiang Deng, Yulei Wan, Ruiquan Jiang, Xiaoqing Zheng, Xuanjing Huang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.27123",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-26",
+      "updated_at": "2026-05-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.IR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-27134-scaling-benchmarking-and-reasoning-of-vision-language-agents-for-mobile-gui-navi.md",
+    "title": "Scaling, Benchmarking, and Reasoning of Vision-Language Agents for Mobile GUI Navigation",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Scaling, Benchmarking, and Reasoning of Vision-Language Agents for Mobile GUI Navigation",
+      "authors": "Heng Qu, Yike Liu, Renren Jin, Wenzong Zhang, Pengzhi Gao, Wei Liu, Jian Luan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.27134",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-26",
+      "updated_at": "2026-05-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "embodied-agent",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-27240-enpmr-bench-benchmarking-proactive-memory-retrieval-for-emotional-support-agents.md",
+    "title": "\"ENPMR-Bench: Benchmarking Proactive Memory Retrieval for Emotional Support Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ENPMR-Bench: Benchmarking Proactive Memory Retrieval for Emotional Support Agents\"",
+      "authors": "Xing Fu, Yulin Hu, Mengtong Ji, Haozhen Li, Yixin Sun, Weixiang Zhao, Yanyan Zhao, Bing Qin",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.27240",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-26",
+      "updated_at": "2026-05-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-27333-finharness-an-inline-lifecycle-safety-harness-for-finance-llm-agents.md",
+    "title": "\"FinHarness: An Inline Lifecycle Safety Harness for Finance LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"FinHarness: An Inline Lifecycle Safety Harness for Finance LLM Agents\"",
+      "authors": "Haoxuan Jia, Yang Liu, Bin Chong, Yingguang Yang, Yancheng Chen, Jiayu Liang, Qian Li, Hanning Lu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.27333",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-26",
+      "updated_at": "2026-05-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-27366-muse-autoskill-self-evolving-agents-via-skill-creation-memory-management-and-eva.md",
+    "title": "\"MUSE-Autoskill: Self-Evolving Agents via Skill Creation, Memory, Management, and Evaluation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MUSE-Autoskill: Self-Evolving Agents via Skill Creation, Memory, Management, and Evaluation\"",
+      "authors": "Huawei Lin, Peng Li, Jie Song, Fuxin Jiang, Tieying Zhang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.27366",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-26",
+      "updated_at": "2026-07-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL",
+        "cs.LG",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-27690-traces-proactive-safety-auditing-for-multi-turn-llm-agents-via-trajectory-state-.md",
+    "title": "\"TRACES: Proactive Safety Auditing for Multi-Turn LLM Agents via Trajectory-State Modeling\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"TRACES: Proactive Safety Auditing for Multi-Turn LLM Agents via Trajectory-State Modeling\"",
+      "authors": "Jiaqian Li, Yanshu Li, Boxuan Zhang, Ruixiang Tang, Kuan-Hao Huang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.27690",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-26",
+      "updated_at": "2026-05-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-27762-peam-parametric-embodied-agent-memory-through-contrastive-internalization-of-exp.md",
+    "title": "\"PEAM: Parametric Embodied Agent Memory through Contrastive Internalization of Experience in Minecraft\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"PEAM: Parametric Embodied Agent Memory through Contrastive Internalization of Experience in Minecraft\"",
+      "authors": "Yuchen Guo, Junli Gong, Weicheng Wang, Hongmin Cai, Yiu-ming Cheung, Weifeng Su",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.27762",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-26",
+      "updated_at": "2026-06-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "embodied-agent",
+        "memory",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-27825-mrmmia-membership-inference-attacks-on-memory-in-chat-agents.md",
+    "title": "\"MRMMIA: Membership Inference Attacks on Memory in Chat Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MRMMIA: Membership Inference Attacks on Memory in Chat Agents\"",
+      "authors": "Kai Chen, Yan Pang, Tianhao Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.27825",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-27",
+      "updated_at": "2026-05-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-27935-do-agents-think-deeper-a-mechanistic-investigation-of-layer-wise-dynamics-in-seq.md",
+    "title": "Do Agents Think Deeper? A Mechanistic Investigation of Layer-Wise Dynamics in Sequential Planning",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Do Agents Think Deeper? A Mechanistic Investigation of Layer-Wise Dynamics in Sequential Planning",
+      "authors": "Zhenyu Cui, Xiangzhong Luo",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.27935",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-27",
+      "updated_at": "2026-05-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "autonomous-agent-llm, planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-28046-memcog-from-memory-as-tool-to-memory-as-cognition-in-conversational-agents.md",
+    "title": "\"MemCog: From Memory-as-Tool to Memory-as-Cognition in Conversational Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MemCog: From Memory-as-Tool to Memory-as-Cognition in Conversational Agents\"",
+      "authors": "Zihan Li, Xingyu Fan, Feifei Li, Wenhui Que",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.28046",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-27",
+      "updated_at": "2026-05-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "embodied-agent",
+        "memory",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-28120-legalgraphrag-multi-agent-graph-retrieval-augmented-generation-for-reliable-lega.md",
+    "title": "\"LegalGraphRAG: Multi-Agent Graph Retrieval-Augmented Generation for Reliable Legal Reasoning\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"LegalGraphRAG: Multi-Agent Graph Retrieval-Augmented Generation for Reliable Legal Reasoning\"",
+      "authors": "Zerui Chen, Qinggang Zhang, Zhishang Xiang, Zhimin Wei, Linfeng Gao, Xiao Huang, Zhihong Zhang, Jinsong Su",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.28120",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-27",
+      "updated_at": "2026-05-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-28175-mixture-of-experts-knowledge-graph-retrieval-augmented-generation-for-multi-agen.md",
+    "title": "Mixture-of-Experts Knowledge Graph Retrieval-Augmented Generation for Multi-Agent LLM-based Recommendation",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Mixture-of-Experts Knowledge Graph Retrieval-Augmented Generation for Multi-Agent LLM-based Recommendation",
+      "authors": "Shijie Wang, Chengyi Liu, Yujuan Ding, Shanru Lin, See-Kiong Ng, Xu Xin, Wenqi Fan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.28175",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-27",
+      "updated_at": "2026-05-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.IR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-28424-skill0-5-joint-skill-internalization-and-utilization-for-out-of-distribution-gen.md",
+    "title": "\"Skill0.5: Joint Skill Internalization and Utilization for Out-of-Distribution Generalization in Agentic Reinforcement Learning\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Skill0.5: Joint Skill Internalization and Utilization for Out-of-Distribution Generalization in Agentic Reinforcement Learning\"",
+      "authors": "Jiapeng Zhu, Jianxiang Yu, Yibo Zhao, Chengcheng Han, Qi Gu, Xunliang Cai, Xiang Li, Weining Qian",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.28424",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-27",
+      "updated_at": "2026-05-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "memory"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-28607-adaptive-multimodal-agents-based-framework-for-automatic-workflow-execution.md",
+    "title": "Adaptive Multimodal Agents-Based Framework for Automatic Workflow Execution",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Adaptive Multimodal Agents-Based Framework for Automatic Workflow Execution",
+      "authors": "Susanna Cifani, Mario Luca Bernardi, Marta Cimitile",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.28607",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-27",
+      "updated_at": "2026-05-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "embodied-agent",
+        "multi-agent",
+        "planning",
+        "rag",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-28617-lacuna-safe-agents-as-recursive-program-holes.md",
+    "title": "\"LACUNA: Safe Agents as Recursive Program Holes\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"LACUNA: Safe Agents as Recursive Program Holes\"",
+      "authors": "Yaoyu Zhao, Yichen Xu, Oliver Bračevac, Cao Nguyen Pham, Frank Zhengqing Wu, Martin Odersky",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.28617",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-27",
+      "updated_at": "2026-05-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.PL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-28787-do-agents-need-semantic-metadata-a-comparative-study-in-agentic-data-retrieval.md",
+    "title": "Do Agents Need Semantic Metadata? A Comparative Study in Agentic Data Retrieval",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Do Agents Need Semantic Metadata? A Comparative Study in Agentic Data Retrieval",
+      "authors": "Shiyu Chen, Tarfah Alrashed, Alon Halevy, Natasha Noy",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.28787",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-27",
+      "updated_at": "2026-05-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.IR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-28835-genesisfunc-multi-agent-data-generation-for-accurate-and-generalizable-function-.md",
+    "title": "\"GenesisFunc: Multi-Agent Data Generation for Accurate and Generalizable Function-Calling\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"GenesisFunc: Multi-Agent Data Generation for Accurate and Generalizable Function-Calling\"",
+      "authors": "Hao-Xiang Xu, Chong Deng, Jiaqing Liu, Wen Wang, Qian Chen, Lujia Bao, Xiangang Li, Zhen-Hua Ling",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.28835",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-10",
+      "updated_at": "2026-04-10",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-28850-representation-signatures-and-risk-feedback-alignment-in-llm-trading-agents.md",
+    "title": "Representation Signatures and Risk-Feedback Alignment in LLM Trading Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Representation Signatures and Risk-Feedback Alignment in LLM Trading Agents",
+      "authors": "Weicheng Xue",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.28850",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-16",
+      "updated_at": "2026-05-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "coding-agent",
+        "memory",
+        "planning",
+        "reasoning",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "q-fin.CP"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-29341-worldmemarena-evaluating-multimodal-agent-memory-through-action-world-interactio.md",
+    "title": "\"WorldMemArena: Evaluating Multimodal Agent Memory Through Action-World Interaction\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"WorldMemArena: Evaluating Multimodal Agent Memory Through Action-World Interaction\"",
+      "authors": "Chengzhi Liu, Yuzhe Yang, Sophia Xiao Pu, Yepeng Liu, Lin Long, Yichen Guo, Nuo Chen, Zhaotian Weng, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.29341",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-28",
+      "updated_at": "2026-06-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "agent-memory, rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-29630-entity-collision-a-stratified-protocol-for-attributing-retrieval-lift-in-agent-m.md",
+    "title": "\"Entity-Collision: A Stratified Protocol for Attributing Retrieval Lift in Agent Memory\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Entity-Collision: A Stratified Protocol for Attributing Retrieval Lift in Agent Memory\"",
+      "authors": "Youwang Deng",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.29630",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-28",
+      "updated_at": "2026-05-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI",
+        "cs.IR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-29640-vikingmem-a-memory-base-management-system-for-stateful-llm-based-applications.md",
+    "title": "\"VikingMem: A Memory Base Management System for Stateful LLM-based Applications\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"VikingMem: A Memory Base Management System for Stateful LLM-based Applications\"",
+      "authors": "Jiajie Fu, Junwen Chen, Mengzhao Wang, Aoxiang He, Maojia Sheng, Xiangyu Ke, Yifan Zhu, Yunjun Gao",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.29640",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-28",
+      "updated_at": "2026-06-12",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-29653-ptcg-bench-can-llm-agents-master-pok-mon-trading-card-game.md",
+    "title": "\"PTCG-Bench: Can LLM Agents Master Pokémon Trading Card Game?\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"PTCG-Bench: Can LLM Agents Master Pokémon Trading Card Game?\"",
+      "authors": "Dongdong Hua, Yifei Sun, Renhong Huang, Feng Gao, Chunping Wang, Yang Yang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.29653",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-28",
+      "updated_at": "2026-05-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-evaluation, autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-29676-notation-matters-a-benchmark-study-of-token-optimized-formats-in-agentic-ai-syst.md",
+    "title": "\"Notation Matters: A Benchmark Study of Token-Optimized Formats in Agentic AI Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Notation Matters: A Benchmark Study of Token-Optimized Formats in Agentic AI Systems\"",
+      "authors": "Lorenz Kutschka, Bernhard Geiger",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.29676",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-28",
+      "updated_at": "2026-06-17",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-29790-evolve-as-a-team-collaborative-self-evolution-for-llm-based-multi-agent-systems.md",
+    "title": "\"Evolve as a Team: Collaborative Self-Evolution for LLM-based Multi-Agent Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Evolve as a Team: Collaborative Self-Evolution for LLM-based Multi-Agent Systems\"",
+      "authors": "Zhezheng Hao, Tianfu Wang, Huanshuo Dong, Ziyan Liu, Hong Wang, Xiankun Lin, Qiang Lin, Can Wang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.29790",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-28",
+      "updated_at": "2026-05-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "planning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-29801-agentdog-1-5-a-lightweight-and-scalable-alignment-framework-for-ai-agent-safety-.md",
+    "title": "\"AgentDoG 1.5: A Lightweight and Scalable Alignment Framework for AI Agent Safety and Security\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AgentDoG 1.5: A Lightweight and Scalable Alignment Framework for AI Agent Safety and Security\"",
+      "authors": "Dongrui Liu, Yu Li, Zhonghao Yang, Peng Wang, Guanxu Chen, Yuejin Xie, Qinghua Mao, Wanying Qu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.29801",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-28",
+      "updated_at": "2026-05-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "computer-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL",
+        "cs.CR",
+        "cs.CV",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-29861-towards-verifiable-multimodal-deep-research-a-multi-agent-harness-for-interleave.md",
+    "title": "\"Towards Verifiable Multimodal Deep Research: A Multi-Agent Harness for Interleaved Report Generation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Towards Verifiable Multimodal Deep Research: A Multi-Agent Harness for Interleaved Report Generation\"",
+      "authors": "Chenghao Zhang, Guanting Dong, Yufan Liu, Tong Zhao, Xiaoxi Li, Zhicheng Dou",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.29861",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-28",
+      "updated_at": "2026-06-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "multi-agent",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "22",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-29960-hijacking-agent-memory-stealthy-trojan-attacks-through-conversational-interactio.md",
+    "title": "\"Hijacking Agent Memory: Stealthy Trojan Attacks Through Conversational Interaction\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Hijacking Agent Memory: Stealthy Trojan Attacks Through Conversational Interaction\"",
+      "authors": "Hongtao Wang, Se Yang, Yu Chen, Puzhuo Liu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.29960",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-28",
+      "updated_at": "2026-05-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-30058-heart-bench-do-llm-agents-exhibit-human-like-psychology.md",
+    "title": "\"HEART-Bench: Do LLM Agents Exhibit Human-like Psychology?\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"HEART-Bench: Do LLM Agents Exhibit Human-like Psychology?\"",
+      "authors": "Weihan Peng, Chenxu Zhang, Qianao Wang, Yuling Shi, Heng Lian, Qihong Mao, Jiahao Pang, Chunliang Feng, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.30058",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-28",
+      "updated_at": "2026-05-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-30090-directorbench-diagnosing-long-form-video-generation-with-personalized-multi-agen.md",
+    "title": "\"DirectorBench: Diagnosing Long-Form Video Generation with Personalized Multi-Agent Evaluation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"DirectorBench: Diagnosing Long-Form Video Generation with Personalized Multi-Agent Evaluation\"",
+      "authors": "Jiamin Chen, Qianben Chen, Jiawen Zhang, Yidi Wu, Yuchen Li, Xiaokun Zhang, Wangchunshu Zhou, Chen Ma",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.30090",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-28",
+      "updated_at": "2026-05-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.CV"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-30407-exploring-autonomous-agentic-data-engineering-for-model-specialization.md",
+    "title": "Exploring Autonomous Agentic Data Engineering for Model Specialization",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Exploring Autonomous Agentic Data Engineering for Model Specialization",
+      "authors": "Yujie Luo, Xiangyuan Ru, Jingsheng Zheng, Jingjing Wang, Yuqi Zhu, Jintian Zhang, Runnan Fang, Kewei Xu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.30407",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-28",
+      "updated_at": "2026-06-08",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI",
+        "cs.IR",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-30604-an-organization-scoped-llm-agent-runtime-architecture-for-regulated-cybersecurit.md",
+    "title": "An Organization-Scoped LLM Agent Runtime Architecture for Regulated Cybersecurity Operations",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "An Organization-Scoped LLM Agent Runtime Architecture for Regulated Cybersecurity Operations",
+      "authors": "George Fatouros, Georgios Makridis, George Kousiouris, John Soldatos, Dimosthenis Kyriazis",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.30604",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-28",
+      "updated_at": "2026-05-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "planning",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI",
+        "cs.CL",
+        "cs.IR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-30690-elasticmem-latent-memory-as-a-learnable-resource-for-llm-agents.md",
+    "title": "\"ElasticMem: Latent Memory as a Learnable Resource for LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ElasticMem: Latent Memory as a Learnable Resource for LLM Agents\"",
+      "authors": "Tao Feng, Chongrui Ye, Tianyang Luo, Jingjun Xu, Xueqiang Xu, Haozhen Zhang, Ge Liu, Jiaxuan You",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.30690",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-29",
+      "updated_at": "2026-05-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "embodied-agent",
+        "memory",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-30711-sage-a-novelty-gate-for-efficient-memory-evolution-in-agentic-llms.md",
+    "title": "\"SAGE: A Novelty Gate for Efficient Memory Evolution in Agentic LLMs\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SAGE: A Novelty Gate for Efficient Memory Evolution in Agentic LLMs\"",
+      "authors": "Sijia Wang, Dhanajit Brahma, Ricardo Henao",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.30711",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-29",
+      "updated_at": "2026-06-18",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI",
+        "cs.LG",
+        "stat.ML"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-30858-forecastcompass-guiding-agentic-forecasting-with-adaptive-factor-memory.md",
+    "title": "\"ForecastCompass: Guiding Agentic Forecasting with Adaptive Factor Memory\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ForecastCompass: Guiding Agentic Forecasting with Adaptive Factor Memory\"",
+      "authors": "Yurui Chang, Yongkang Du, Yuanpu Cao, Jinghui Chen, Lu Lin",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.30858",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-29",
+      "updated_at": "2026-05-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-30883-trace-task-aware-adaptive-self-evolving-agentic-jailbreaking.md",
+    "title": "\"TRACE: Task-Aware Adaptive Self-Evolving Agentic Jailbreaking\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"TRACE: Task-Aware Adaptive Self-Evolving Agentic Jailbreaking\"",
+      "authors": "Churui Zeng, Weiwei Qi, Kedong Xiu, Tianhang Zheng, Chaochao Lu, Liang He, Zhan Qin, Kui Ren",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.30883",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-29",
+      "updated_at": "2026-05-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "computer-use",
+        "planning",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-30907-bluefin-benchmarking-llm-agents-on-financial-spreadsheets.md",
+    "title": "\"BlueFin: Benchmarking LLM Agents on Financial Spreadsheets\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"BlueFin: Benchmarking LLM Agents on Financial Spreadsheets\"",
+      "authors": "Srivatsa Kundurthy, Clara Na, Colton Moraine, Anoushka Mohta, Case Winter, George Fang, John Ling, Emma Strubell, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.30907",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-29",
+      "updated_at": "2026-05-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI",
+        "cs.CL",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-30947-extending-ai-for-research-to-the-humanities-a-multi-agent-framework-for-evidence.md",
+    "title": "\"Extending AI for Research to the Humanities: A Multi-Agent Framework for Evidence-Grounded Scholarship\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Extending AI for Research to the Humanities: A Multi-Agent Framework for Evidence-Grounded Scholarship\"",
+      "authors": "Yating Pan, Jiajun Zhang, Jun Wang, Qi Su",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.30947",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-29",
+      "updated_at": "2026-06-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-31075-task-focused-memorization-for-multimodal-agents.md",
+    "title": "Task-Focused Memorization for Multimodal Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Task-Focused Memorization for Multimodal Agents",
+      "authors": "Tao Zou, Yichen He, Tian Qiu, Yuan Lin, Hang Li",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.31075",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-29",
+      "updated_at": "2026-05-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "embodied-agent",
+        "memory"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-31268-mellum2-technical-report.md",
+    "title": "Mellum2 Technical Report",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Mellum2 Technical Report",
+      "authors": "Marko Kojic, Ivan Bondyrev, Aral de Moor, Joseph Shtok, Petr Borovlev, Kseniia Lysaniuk, Madeeswaran Kannan, Ivan Dolgov, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.31268",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-29",
+      "updated_at": "2026-05-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-31278-industrializing-prediction-powered-inference-the-glide-library-for-reliable-gena.md",
+    "title": "\"Industrializing Prediction-Powered Inference: The GLIDE Library for Reliable GenAI and Agentic Systems Evaluation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Industrializing Prediction-Powered Inference: The GLIDE Library for Reliable GenAI and Agentic Systems Evaluation\"",
+      "authors": "Grégoire Martinon, Ibrahim Merad, Mohammed Raki",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.31278",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-29",
+      "updated_at": "2026-06-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.LG",
+        "stat.ME"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-31308-tracegraph-shared-decision-landscapes-for-diagnosing-and-improving-agent-traject.md",
+    "title": "\"TraceGraph: Shared Decision Landscapes for Diagnosing and Improving Agent Trajectories\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"TraceGraph: Shared Decision Landscapes for Diagnosing and Improving Agent Trajectories\"",
+      "authors": "Junjie Nian, Kang Chen, Ge Zhang, Yixin Cao, Yugang Jiang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.31308",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-29",
+      "updated_at": "2026-05-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "computer-use",
+        "embodied-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2605-31377-dynatree-dynamic-agentic-retrieval-tree-for-time-sensitive-news-retrieval.md",
+    "title": "\"DynaTree: Dynamic Agentic Retrieval Tree for Time-Sensitive News Retrieval\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"DynaTree: Dynamic Agentic Retrieval Tree for Time-Sensitive News Retrieval\"",
+      "authors": "Siyuan Qi, Xinyuan Wang, Yingxuan Yang, Haochuan Guo, Jianghao Lin, Weiwen Liu, Yong Yu, Weinan Zhang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2605.31377",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-29",
+      "updated_at": "2026-05-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.IR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-00198-bagen-are-llm-agents-budget-aware.md",
+    "title": "\"BAGEN: Are LLM Agents Budget-Aware?\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"BAGEN: Are LLM Agents Budget-Aware?\"",
+      "authors": "Yuxiang Lin, Zihan Wang, Mengyang Liu, Yuxuan Shan, Longju Bai, Junyao Zhang, Xing Jin, Boshan Chen, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.00198",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-29",
+      "updated_at": "2026-05-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-00341-rogue-misaligned-agent-behavior-arising-from-ordinary-computer-use.md",
+    "title": "\"ROGUE: Misaligned Agent Behavior Arising from Ordinary Computer Use\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ROGUE: Misaligned Agent Behavior Arising from Ordinary Computer Use\"",
+      "authors": "Jeremy Tien, Abishek Anand, Yu-Rou Tuan, Yuchen Shen, J. Zico Kolter, Aran Nayebi",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.00341",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-29",
+      "updated_at": "2026-05-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-00610-memgraphrag-memory-based-multi-agent-system-for-graph-retrieval-augmented-genera.md",
+    "title": "\"MemGraphRAG: Memory-based Multi-Agent System for Graph Retrieval-Augmented Generation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MemGraphRAG: Memory-based Multi-Agent System for Graph Retrieval-Augmented Generation\"",
+      "authors": "Chuanjie Wu, Zhishang Xiang, Yunbo Tang, Zerui Chen, Qinggang Zhang, Jinsong Su",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.00610",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-30",
+      "updated_at": "2026-05-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "multi-agent",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.IR",
+        "cs.AI",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-00611-trace-trajectory-risk-aware-compression-for-long-horizon-agent-safety.md",
+    "title": "\"TRACE: Trajectory Risk-Aware Compression for Long-Horizon Agent Safety\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"TRACE: Trajectory Risk-Aware Compression for Long-Horizon Agent Safety\"",
+      "authors": "Zhepei Hong, Lin Wang, Liting Li, Haokai Ma, Junfeng Fang, Fei Shen, Dan Zhang, Xiang Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.00611",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-30",
+      "updated_at": "2026-05-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-00619-mempro-agentic-memory-systems-as-evolvable-programs.md",
+    "title": "\"MemPro: Agentic Memory Systems as Evolvable Programs\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MemPro: Agentic Memory Systems as Evolvable Programs\"",
+      "authors": "Qingshan Liu, Guoqing Wang, Wen Wu, Jingqi Huang, Xinqi Tao, Dejia Song, Jie Zhou, Liang He",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.00619",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-30",
+      "updated_at": "2026-05-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "planning",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-00644-foresci-evaluating-llm-agents-for-forward-looking-ai-research-judgment.md",
+    "title": "\"ForeSci: Evaluating LLM Agents for Forward-Looking AI Research Judgment\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ForeSci: Evaluating LLM Agents for Forward-Looking AI Research Judgment\"",
+      "authors": "Qiuyu Tian, Haojie Yin, Yingce Xia, Youyong Kong, Zequn Liu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.00644",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-30",
+      "updated_at": "2026-06-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-00756-comic-collaborative-memory-and-insights-circulation-for-long-horizon-llm-agents-.md",
+    "title": "\"CoMIC: Collaborative Memory and Insights Circulation for Long-Horizon LLM Agents in Cloud-Edge Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"CoMIC: Collaborative Memory and Insights Circulation for Long-Horizon LLM Agents in Cloud-Edge Systems\"",
+      "authors": "Yannan Wang, Longli Yang, Zhen Liu, Abhishek Kumar, Carsten Maple",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.00756",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-30",
+      "updated_at": "2026-05-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "21",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-00914-adversarial-feeds-steer-llm-agent-decisions-against-their-defaults.md",
+    "title": "Adversarial Feeds Steer LLM Agent Decisions Against Their Defaults",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Adversarial Feeds Steer LLM Agent Decisions Against Their Defaults",
+      "authors": "Rana Muhammad Usman",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.00914",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-30",
+      "updated_at": "2026-05-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL",
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-00915-autonomous-agentic-design-for-photonics.md",
+    "title": "Autonomous agentic design for photonics",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Autonomous agentic design for photonics",
+      "authors": "Prashanta Kharel, Amin Khavasi, Xinzhong Chen, Tyler W. Hughes",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.00915",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-30",
+      "updated_at": "2026-05-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "physics.optics"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-00922-a-machine-to-machine-knowledge-guided-llm-agent-for-generalizable-radiotherapy-t.md",
+    "title": "A Machine-to-Machine Knowledge-Guided LLM Agent for Generalizable Radiotherapy Treatment Planning",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "A Machine-to-Machine Knowledge-Guided LLM Agent for Generalizable Radiotherapy Treatment Planning",
+      "authors": "Md Mainul Abrar, Xun Jia, Yujie Chi",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.00922",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-30",
+      "updated_at": "2026-05-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "planning",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "physics.med-ph",
+        "cs.RO"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-00939-fincom-a-financial-multi-agent-demo-with-disagree-or-commit-deliberation.md",
+    "title": "\"FinCom: A Financial Multi-Agent Demo with Disagree-or-Commit Deliberation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"FinCom: A Financial Multi-Agent Demo with Disagree-or-Commit Deliberation\"",
+      "authors": "Chao Peter Yang, Zixiao Tan, Kaisen Yao, Ziyu Zhou, Eleanor Jiang, Michael Wu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.00939",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-31",
+      "updated_at": "2026-05-31",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-01041-expweaver-llm-agents-learn-from-experience-via-latent-rag.md",
+    "title": "\"ExpWeaver: LLM Agents Learn from Experience via Latent RAG\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ExpWeaver: LLM Agents Learn from Experience via Latent RAG\"",
+      "authors": "Tao Feng, Tianyang Luo, Jingjun Xu, Zhigang Hua, Yan Xie, Shuang Yang, Ge Liu, Jiaxuan You",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.01041",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-31",
+      "updated_at": "2026-05-31",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "planning-agent, rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-01138-memorywire-a-vendor-neutral-wire-format-for-agent-memory-operations.md",
+    "title": "\"memorywire: A Vendor-Neutral Wire Format for Agent Memory Operations\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"memorywire: A Vendor-Neutral Wire Format for Agent Memory Operations\"",
+      "authors": "Thamilvendhan Munirathinam",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.01138",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-31",
+      "updated_at": "2026-06-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI",
+        "cs.DC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-01166-braveguard-from-open-world-threats-to-safer-computer-use-agents.md",
+    "title": "\"BraveGuard: From Open-World Threats to Safer Computer-Use Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"BraveGuard: From Open-World Threats to Safer Computer-Use Agents\"",
+      "authors": "Yunhao Feng, Xiaohu Du, Xinhao Deng, Yifan Ding, Ming Wen, Yixu Wang, Yuxiang Xie, Baihui Zheng, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.01166",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-31",
+      "updated_at": "2026-06-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-01185-skill-issues-data-centric-optimization-of-lakehouse-agents.md",
+    "title": "\"\\\"Skill issues'': data-centric optimization of lakehouse agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"\\\"Skill issues'': data-centric optimization of lakehouse agents\"",
+      "authors": "Nicole Rose Schneider, Davide Ghilardi, Giacomo Piccinini, Jacopo Tagliabue",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.01185",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-31",
+      "updated_at": "2026-05-31",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "planning",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-01199-can-llm-agents-sustain-long-horizon-organizational-dynamics.md",
+    "title": "Can LLM Agents Sustain Long-Horizon Organizational Dynamics?",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Can LLM Agents Sustain Long-Horizon Organizational Dynamics?",
+      "authors": "Xuancheng Zhu, Yang Yue, Shuaibing Wan, Zihan Dou, Xiaohan Zhang, Yongrui Liu, Guoshun Nan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.01199",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-31",
+      "updated_at": "2026-05-31",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "multi-agent",
+        "planning",
+        "rag",
+        "workflow-agent",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "22",
+      "collection_queries": "language-agent, planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-01222-rag-driven-multi-agent-llm-framework-with-task-decomposition-for-beyond-5g-auto-.md",
+    "title": "RAG-driven Multi-Agent LLM Framework with Task Decomposition for Beyond 5G Auto-Configuration",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "RAG-driven Multi-Agent LLM Framework with Task Decomposition for Beyond 5G Auto-Configuration",
+      "authors": "İrşat Emin Sarıdaş, Onur Salan, Ali Görçin, Ibrahim Hokelek, Hakan Ali Çırpan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.01222",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-31",
+      "updated_at": "2026-05-31",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "planning",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "eess.SP"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-01385-bridging-requirements-and-architecture-multi-agent-orchestration-with-external-k.md",
+    "title": "\"Bridging Requirements and Architecture: Multi-Agent Orchestration with External Knowledge and Hierarchical Memory\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Bridging Requirements and Architecture: Multi-Agent Orchestration with External Knowledge and Hierarchical Memory\"",
+      "authors": "Ruiyin Li, Yiran Zhang, Xiyu Zhou, Yangxiao Cai, Peng Liang, Weisong Sun, Jifeng Xuan, Zhi Jin, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.01385",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-31",
+      "updated_at": "2026-05-31",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "memory",
+        "multi-agent",
+        "rag",
+        "reasoning",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "21",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-01416-self-healing-agentic-orchestrators-for-reliable-tool-augmented-large-language-mo.md",
+    "title": "Self-Healing Agentic Orchestrators for Reliable Tool-Augmented Large Language Model Systems",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Self-Healing Agentic Orchestrators for Reliable Tool-Augmented Large Language Model Systems",
+      "authors": "Rahul Suresh Babu, Adarsh Agrawal",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.01416",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-31",
+      "updated_at": "2026-05-31",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-01528-joint-agent-memory-and-exploration-learning-via-novelty-signals.md",
+    "title": "Joint Agent Memory and Exploration Learning via Novelty Signals",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Joint Agent Memory and Exploration Learning via Novelty Signals",
+      "authors": "Shizuo Tian, Xiaohong Weng, Rui Kong, Yuxuan Chen, Guohong Liu, Yuebing Song, Jiacheng Liu, Yuchen Li, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.01528",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-01",
+      "updated_at": "2026-06-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-01613-techrag-evidence-gated-multimodal-agentic-rag-for-technical-literature-reasoning.md",
+    "title": "\"TechRAG: Evidence-Gated Multimodal Agentic RAG for Technical Literature Reasoning\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"TechRAG: Evidence-Gated Multimodal Agentic RAG for Technical Literature Reasoning\"",
+      "authors": "Kanwar Bharat Singh",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.01613",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-01",
+      "updated_at": "2026-06-13",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "computer-use",
+        "multi-agent",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.IR",
+        "cs.AI",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-01815-crab-bench-evaluating-llm-agents-under-complex-task-dependencies-and-human-align.md",
+    "title": "\"CRAB-Bench: Evaluating LLM Agents under Complex Task Dependencies and Human-aligned User Simulation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"CRAB-Bench: Evaluating LLM Agents under Complex Task Dependencies and Human-aligned User Simulation\"",
+      "authors": "Danqing Wang, Akshay Sivaraman, Lei Li",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.01815",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-01",
+      "updated_at": "2026-06-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-01961-automedbench-towards-medical-autoresearch-with-agentic-ai-models.md",
+    "title": "\"AutoMedBench: Towards Medical AutoResearch with Agentic AI Models\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AutoMedBench: Towards Medical AutoResearch with Agentic AI Models\"",
+      "authors": "Junqi Liu, Selena Song, Yuhan Wang, Jiawei Mao, Hardy Chen, Xiaoke Huang, Tianhao Qi, Pengfei Guo, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.01961",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-01",
+      "updated_at": "2026-06-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-02109-badger-bridging-agentic-and-deterministic-evaluation-for-generative-enterprise-r.md",
+    "title": "\"BADGER: Bridging Agentic and Deterministic Evaluation for Generative Enterprise Reasoning\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"BADGER: Bridging Agentic and Deterministic Evaluation for Generative Enterprise Reasoning\"",
+      "authors": "Shannon Serrao, Soumitra Chatterjee, Dorina Strori, Abhishek Sharma, Nathan Miller",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.02109",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-01",
+      "updated_at": "2026-06-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-02302-seclaw-spec-driven-security-task-synthesis-for-evaluating-autonomous-agents.md",
+    "title": "\"SeClaw: Spec-Driven Security Task Synthesis for Evaluating Autonomous Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SeClaw: Spec-Driven Security Task Synthesis for Evaluating Autonomous Agents\"",
+      "authors": "Hao Cheng, Changtao Miao, Tianle Song, Yin Wu, He Liu, Erjia Xiao, Junchi Chen, Xiaoyu Shi, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.02302",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-01",
+      "updated_at": "2026-06-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "agent-safety, autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-02372-comap-co-evolving-world-models-and-agent-policies-for-llm-agents.md",
+    "title": "\"COMAP: Co-Evolving World Models and Agent Policies for LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"COMAP: Co-Evolving World Models and Agent Policies for LLM Agents\"",
+      "authors": "Youwei Liu, Jian Wang, Hanlin Wang, Wenjie Li",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.02372",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-01",
+      "updated_at": "2026-06-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "embodied-agent",
+        "planning",
+        "reasoning",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "language-agent, planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-02380-spade-bench-evaluating-spontaneous-strategic-deception-in-agents-via-plan-action.md",
+    "title": "\"SPADE-Bench: Evaluating Spontaneous Strategic Deception in Agents via Plan-Action Divergence\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SPADE-Bench: Evaluating Spontaneous Strategic Deception in Agents via Plan-Action Divergence\"",
+      "authors": "Yuyan Bu, Haowei Li, Qirui Zheng, Bowen Dong, Kaiyue Yang, Jiaming Ji, Yingshui Tan, Wenxin Li, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.02380",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-01",
+      "updated_at": "2026-06-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-02388-policy-and-world-modeling-co-training-for-language-agents.md",
+    "title": "Policy and World Modeling Co-Training for Language Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Policy and World Modeling Co-Training for Language Agents",
+      "authors": "Ning Lu, Baijiong Lin, Shengcai Liu, Jiahao Wu, Haoze Lv, Yanbin Wei, Lingting Zhu, Shengju Qian, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.02388",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-01",
+      "updated_at": "2026-06-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-02404-k-browsecomp-a-web-browsing-agent-benchmark-grounded-in-korean-contexts.md",
+    "title": "\"K-BrowseComp: A Web Browsing Agent Benchmark Grounded in Korean Contexts\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"K-BrowseComp: A Web Browsing Agent Benchmark Grounded in Korean Contexts\"",
+      "authors": "Nahyun Lee, Dongkeun Yoon, Guijin Son, Geewook Kim, Dayoon Ko, Jeonghun Park, Haneul Yoo, Jaewon Cho, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.02404",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-01",
+      "updated_at": "2026-06-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-02461-agentcl-toward-rigorous-evaluation-of-continual-learning-in-language-agents.md",
+    "title": "\"AgentCL: Toward Rigorous Evaluation of Continual Learning in Language Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AgentCL: Toward Rigorous Evaluation of Continual Learning in Language Agents\"",
+      "authors": "Yiheng Shu, Bernal Jiménez Gutiérrez, Saisri Padmaja Jonnalagedda, Yuguang Yao, Huan Sun, Yu Su",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.02461",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-01",
+      "updated_at": "2026-06-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "computer-use",
+        "memory",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "21",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-02497-bridging-the-last-mile-of-time-series-forecasting-with-llm-agents.md",
+    "title": "Bridging the Last Mile of Time Series Forecasting with LLM Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Bridging the Last Mile of Time Series Forecasting with LLM Agents",
+      "authors": "Yuhua Liao, Zetian Wang, Qiangqiang Nie, Zhenhua Zhang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.02497",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-01",
+      "updated_at": "2026-06-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "memory",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-02812-traj-evolve-a-self-evolving-multi-agent-system-for-patient-trajectory-modeling-i.md",
+    "title": "\"Traj-Evolve: A Self-Evolving Multi-Agent System for Patient Trajectory Modeling in Lung Cancer Early Detection\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Traj-Evolve: A Self-Evolving Multi-Agent System for Patient Trajectory Modeling in Lung Cancer Early Detection\"",
+      "authors": "Sihang Zeng, Matthew Thompson, Ruth Etzioni, Meliha Yetisgen",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.02812",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-01",
+      "updated_at": "2026-06-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "multi-agent",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-02965-what-benchmarks-don-t-measure-the-case-for-evaluating-abstention-competence-in-a.md",
+    "title": "\"What Benchmarks Don't Measure: The Case for Evaluating Abstention Competence in Autonomous Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"What Benchmarks Don't Measure: The Case for Evaluating Abstention Competence in Autonomous Agents\"",
+      "authors": "Victor Ojewale, Suresh Venkatasubramanian",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.02965",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-01",
+      "updated_at": "2026-06-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-03108-evotrainer-co-evolving-llm-policies-and-training-harnesses-for-autonomous-agenti.md",
+    "title": "\"EvoTrainer: Co-Evolving LLM Policies and Training Harnesses for Autonomous Agentic Reinforcement Learning\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"EvoTrainer: Co-Evolving LLM Policies and Training Harnesses for Autonomous Agentic Reinforcement Learning\"",
+      "authors": "Guhong Chen, Yingcheng Shi, Yongbin Li, Binhua Li, Xander Xu, Hu Wei, Shiwen Ni, Min Yang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.03108",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-02",
+      "updated_at": "2026-06-12",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "planning",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-03135-uncertainty-aware-clarification-in-llm-agents-with-information-gain.md",
+    "title": "Uncertainty-Aware Clarification in LLM Agents with Information Gain",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Uncertainty-Aware Clarification in LLM Agents with Information Gain",
+      "authors": "Mengyi Deng, Zhiwei Li, Xin Li, Tingyu Zhu, Ying Zhao, Zhijiang Guo, Wei Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.03135",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-02",
+      "updated_at": "2026-06-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-03157-clinicalmc-a-benchmark-for-multi-course-clinical-decision-making-with-large-lang.md",
+    "title": "\"ClinicalMC: A Benchmark for Multi-Course Clinical Decision-Making with Large Language Models\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ClinicalMC: A Benchmark for Multi-Course Clinical Decision-Making with Large Language Models\"",
+      "authors": "Ruihui Hou, Siyi Zhu, Ziyue Huai, Guangya Yu, Yongqi Fan, Chunming Wang, Tong Ruan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.03157",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-02",
+      "updated_at": "2026-06-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-03197-memtrain-self-supervised-context-memory-training.md",
+    "title": "\"MemTrain: Self-Supervised Context Memory Training\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MemTrain: Self-Supervised Context Memory Training\"",
+      "authors": "Ziheng Li, Xingrun Xing, Haoqing Wang, Zhi-Hong Deng, Yehui Tang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.03197",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-02",
+      "updated_at": "2026-06-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-03329-infomem-training-long-context-memory-agents-with-answer-conditioned-information-.md",
+    "title": "\"InfoMem: Training Long-Context Memory Agents with Answer-Conditioned Information Gain\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"InfoMem: Training Long-Context Memory Agents with Answer-Conditioned Information Gain\"",
+      "authors": "Tiancheng Han, Yong Li, Wuzhou Yu, Qiaosheng Zhang, Wenqi Shao",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.03329",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-02",
+      "updated_at": "2026-06-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-03374-emem-a-hybrid-spatio-temporal-memory-system-for-embodied-agents.md",
+    "title": "\"eMEM: A Hybrid Spatio-Temporal Memory System For Embodied Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"eMEM: A Hybrid Spatio-Temporal Memory System For Embodied Agents\"",
+      "authors": "A. Haroon Rasheed, Maria Kabtoul",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.03374",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-02",
+      "updated_at": "2026-06-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "embodied-agent",
+        "memory",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.RO"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agent-memory, rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-03544-sage-a-quantitative-evaluation-of-socialized-evolution-in-agent-ecosystems.md",
+    "title": "\"SAGE: A Quantitative Evaluation of Socialized Evolution in Agent Ecosystems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SAGE: A Quantitative Evaluation of Socialized Evolution in Agent Ecosystems\"",
+      "authors": "Linyue Pan, Yaoming Zhu, Lin Qiu, Xuezhi Cao, Xunliang Cai",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.03544",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-02",
+      "updated_at": "2026-06-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-03657-diagnosing-knowledge-gaps-in-llm-tool-use-an-agentic-benchmark-for-novel-api-acq.md",
+    "title": "\"Diagnosing Knowledge Gaps in LLM Tool Use: An Agentic Benchmark for Novel API Acquisition\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Diagnosing Knowledge Gaps in LLM Tool Use: An Agentic Benchmark for Novel API Acquisition\"",
+      "authors": "Jinnuo Liu, Yue Peng, Jinhan Niu, Hongyi Wen",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.03657",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-02",
+      "updated_at": "2026-06-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-03895-agent-libos-a-runtime-substrate-for-capability-controlled-self-evolving-llm-agen.md",
+    "title": "\"Agent libOS: A Runtime Substrate for Capability-Controlled Self-Evolving LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Agent libOS: A Runtime Substrate for Capability-Controlled Self-Evolving LLM Agents\"",
+      "authors": "Yingqi Zhang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.03895",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-02",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.OS",
+        "cs.AI",
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-04051-rubas-rubric-based-reinforcement-learning-for-agent-safety.md",
+    "title": "\"RUBAS: Rubric-Based Reinforcement Learning for Agent Safety\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"RUBAS: Rubric-Based Reinforcement Learning for Agent Safety\"",
+      "authors": "Xian Qi Loye, Qinglin Su, Zhexin Zhang, Shiyao Cui, Qi Zhu, Fei Mi, Hongning Wang, Minlie Huang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.04051",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-02",
+      "updated_at": "2026-06-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI",
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-04120-salimory-orchestrating-cognitive-memory-for-conversational-agents.md",
+    "title": "\"SaliMory: Orchestrating Cognitive Memory for Conversational Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SaliMory: Orchestrating Cognitive Memory for Conversational Agents\"",
+      "authors": "Kai Zhang, Xinyuan Zhang, Hongda Jiang, Shiun-Zu Kuo, Hyokun Yun, Ejaz Ahmed, Shereen Oraby, Ziyun Li, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.04120",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-02",
+      "updated_at": "2026-06-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-04296-the-saturation-trap-and-the-subjectivity-of-intervention-timing-why-affect-based.md",
+    "title": "\"The Saturation Trap and the Subjectivity of Intervention Timing: Why Affect-Based Triggers and LLM Judges Fail to Time Interventions on Autonomous Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"The Saturation Trap and the Subjectivity of Intervention Timing: Why Affect-Based Triggers and LLM Judges Fail to Time Interventions on Autonomous Agents\"",
+      "authors": "Manvendra Modgil",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.04296",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-02",
+      "updated_at": "2026-06-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-04315-exploring-cross-scenario-generality-of-agentic-memory-systems-diagnostics-and-a-.md",
+    "title": "\"Exploring Cross-Scenario Generality of Agentic Memory Systems: Diagnostics and a Strong Baseline\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Exploring Cross-Scenario Generality of Agentic Memory Systems: Diagnostics and a Strong Baseline\"",
+      "authors": "Zhikai Chen, Jialiang Gu, Junyu Yin, Xianxuan Long, Shenglai Zeng, Xiaoze Liu, Kai Guo, Keren Zhou, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.04315",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-03",
+      "updated_at": "2026-06-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-04555-temporal-order-matters-for-agentic-memory-segment-trees-for-long-horizon-agents.md",
+    "title": "\"Temporal Order Matters for Agentic Memory: Segment Trees for Long-Horizon Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Temporal Order Matters for Agentic Memory: Segment Trees for Long-Horizon Agents\"",
+      "authors": "Yifan Simon Liu, Liam Gallagher, Faeze Moradi Kalarde, Jiazhou Liang, Armin Toroghi, Scott Sanner",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.04555",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-03",
+      "updated_at": "2026-06-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-04599-plan-first-judge-later-run-better-a-dmaic-inspired-agentic-system-for-industrial.md",
+    "title": "\"Plan First, Judge Later, Run Better: A DMAIC-Inspired Agentic System for Industrial Anomaly Detection\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Plan First, Judge Later, Run Better: A DMAIC-Inspired Agentic System for Industrial Anomaly Detection\"",
+      "authors": "Yongzi Yu, Ao Li, Le Wang, Ziyue Li, Fugee Tsung, Yuxuan Liang, Man Li",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.04599",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-03",
+      "updated_at": "2026-06-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "multi-agent",
+        "planning",
+        "rag",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-04628-rampart-registry-based-agentic-memory-with-priority-aware-runtime-transformation.md",
+    "title": "\"RAMPART: Registry-based Agentic Memory with Priority-Aware Runtime Transformation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"RAMPART: Registry-based Agentic Memory with Priority-Aware Runtime Transformation\"",
+      "authors": "Nikodem Tomczak",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.04628",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-03",
+      "updated_at": "2026-06-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "memory"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-04780-personatree-structured-lifecycle-memory-for-person-understanding-in-llm-agents.md",
+    "title": "\"PersonaTree: Structured Lifecycle Memory for Person Understanding in LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"PersonaTree: Structured Lifecycle Memory for Person Understanding in LLM Agents\"",
+      "authors": "Yubo Hou, Jingwei Song, Hongbo Zhang, Zhisheng Chen, Bang Xiao, Tao Wan, Zengchang Qin",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.04780",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-03",
+      "updated_at": "2026-06-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-04874-agent-planning-benchmark-a-diagnostic-framework-for-planning-capabilities-in-llm.md",
+    "title": "\"Agent Planning Benchmark: A Diagnostic Framework for Planning Capabilities in LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Agent Planning Benchmark: A Diagnostic Framework for Planning Capabilities in LLM Agents\"",
+      "authors": "Haoyu Sun, Wenxuan Wang, Mingyang Song, Jujie He, Weinan Zhang, Yang Liu, Yang Yang, Yu Cheng",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.04874",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-03",
+      "updated_at": "2026-06-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agent-evaluation, planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-04990-from-agent-traces-to-trust-a-survey-of-evidence-tracing-and-execution-provenance.md",
+    "title": "\"From Agent Traces to Trust: A Survey of Evidence Tracing and Execution Provenance in LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"From Agent Traces to Trust: A Survey of Evidence Tracing and Execution Provenance in LLM Agents\"",
+      "authors": "Yiqi Wang, Jiaqi Zhang, Taotao Cai, Zirui Liu, Qingqiang Sun, Zequn Sun, Zhangkai Wu, Manqing Dong, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.04990",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-03",
+      "updated_at": "2026-06-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "multi-agent",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "21",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-05241-search-time-contamination-in-deep-research-agents-measuring-performance-inflatio.md",
+    "title": "\"Search-Time Contamination in Deep Research Agents: Measuring Performance Inflation in Public Benchmark Evaluation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Search-Time Contamination in Deep Research Agents: Measuring Performance Inflation in Public Benchmark Evaluation\"",
+      "authors": "Yongjie Wang, Xinyue Zhang, Kunhong Yao, Zhiwei Zeng, Kaisong Song, Jun Lin, Zhiqi Shen",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.05241",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-03",
+      "updated_at": "2026-06-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-05263-policy-conditioned-counterfactual-credit-for-verifiable-reinforcement-learning-o.md",
+    "title": "Policy-Conditioned Counterfactual Credit for Verifiable Reinforcement Learning of Long-Horizon Language Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Policy-Conditioned Counterfactual Credit for Verifiable Reinforcement Learning of Long-Horizon Language Agents",
+      "authors": "Renwei Meng",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.05263",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-03",
+      "updated_at": "2026-06-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-05414-when-evidence-is-sparse-weakly-supervised-early-failure-alerting-in-dialogs-and-.md",
+    "title": "\"When Evidence is Sparse: Weakly Supervised Early Failure Alerting in Dialogs and LLM-Agent Trajectories\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"When Evidence is Sparse: Weakly Supervised Early Failure Alerting in Dialogs and LLM-Agent Trajectories\"",
+      "authors": "Avinash Baidya, Xinran Liang, Ruocheng Guo, Xiang Gao, Kamalika Das",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.05414",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-03",
+      "updated_at": "2026-06-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI",
+        "cs.HC",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-05436-ten-headache-specialists-versus-artificial-intelligence-for-clinical-literature-.md",
+    "title": "\"Ten Headache Specialists versus Artificial Intelligence for Clinical Literature Summarization: A Critical Evaluation and Comparison\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Ten Headache Specialists versus Artificial Intelligence for Clinical Literature Summarization: A Critical Evaluation and Comparison\"",
+      "authors": "Alejandro Lozano, Keiko Ihara, Ping-Hao Yang, Carrie E. Robertson, Jennifer Stern, Allan Purdy, Hsiangkuo Yuan, Pengfei Zhang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.05436",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-03",
+      "updated_at": "2026-06-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL",
+        "cs.IR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-05463-psebench-a-controllable-and-verifiable-benchmark-for-evaluating-llms-in-patient-.md",
+    "title": "\"PSEBench: A Controllable and Verifiable Benchmark for Evaluating LLMs in Patient Safety Event Triage\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"PSEBench: A Controllable and Verifiable Benchmark for Evaluating LLMs in Patient Safety Event Triage\"",
+      "authors": "Keqi Han, Ryan Young, Annabel Strauss, Lindsey Hughes, Katharine M. Nesbitt, Nicole Schueler, Che Ngufor, Carl Yang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.05463",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-03",
+      "updated_at": "2026-06-09",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-05548-adk-arena-evaluating-agent-development-kits-via-llm-as-a-developer.md",
+    "title": "\"ADK Arena: Evaluating Agent Development Kits via LLM-as-a-Developer\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ADK Arena: Evaluating Agent Development Kits via LLM-as-a-Developer\"",
+      "authors": "Jintao Huang, Xiaomin Li, Gaurav Mittal, Yu Hu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.05548",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-04",
+      "updated_at": "2026-06-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-evaluation, autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-05558-autoregressive-diffusion-world-models-for-off-policy-evaluation-of-llm-agents.md",
+    "title": "Autoregressive Diffusion World Models for Off-Policy Evaluation of LLM Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Autoregressive Diffusion World Models for Off-Policy Evaluation of LLM Agents",
+      "authors": "Kaixuan Liu, Guojun Xiong, Weinan Zhang, Shengpu Tang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.05558",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-04",
+      "updated_at": "2026-06-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-05622-adaplanbench-evaluating-adaptive-planning-in-large-language-model-agents-under-w.md",
+    "title": "\"AdaPlanBench: Evaluating Adaptive Planning in Large Language Model Agents under World and User Constraints\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AdaPlanBench: Evaluating Adaptive Planning in Large Language Model Agents under World and User Constraints\"",
+      "authors": "Jiayu Liu, Cheng Qian, Zhenhailong Wang, Bingxuan Li, Jiateng Liu, Heng Wang, Jeonghwan Kim, Yumeng Wang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.05622",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-04",
+      "updated_at": "2026-06-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-05658-agent-orchestrated-adaptive-rag-a-comparative-study-on-structured-and-multi-hop-.md",
+    "title": "\"Agent-Orchestrated Adaptive RAG: A Comparative Study on Structured and Multi-Hop Retrieval\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Agent-Orchestrated Adaptive RAG: A Comparative Study on Structured and Multi-Hop Retrieval\"",
+      "authors": "Anuj Maharjan, Devinder Kaur, Richard Molyet",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.05658",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-04",
+      "updated_at": "2026-06-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.IR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-05684-adamem-test-time-adaptive-memory-for-language-agents.md",
+    "title": "\"AdaMEM: Test-Time Adaptive Memory for Language Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AdaMEM: Test-Time Adaptive Memory for Language Agents\"",
+      "authors": "Yunxiang Zhang, Yiheng Li, Ali Payani, Lu Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.05684",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-04",
+      "updated_at": "2026-06-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "planning",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agent-memory, language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-05711-beyond-tokens-a-unified-framework-for-latent-communication-in-llm-based-multi-ag.md",
+    "title": "\"Beyond tokens: a unified framework for latent communication in LLM-based multi-agent systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Beyond tokens: a unified framework for latent communication in LLM-based multi-agent systems\"",
+      "authors": "Yingzhuo Liu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.05711",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-04",
+      "updated_at": "2026-06-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "computer-use",
+        "multi-agent",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-05805-from-risk-classification-to-action-plan-remediation-a-guardrail-feedback-driven-.md",
+    "title": "\"From Risk Classification to Action Plan Remediation: A Guardrail Feedback Driven Framework for LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"From Risk Classification to Action Plan Remediation: A Guardrail Feedback Driven Framework for LLM Agents\"",
+      "authors": "Yuhao Sun, Jiacheng Zhang, Shaanan Cohney, Zhexin Zhang, Feng Liu, Xingliang Yuan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.05805",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-04",
+      "updated_at": "2026-06-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-06054-beyond-similarity-trustworthy-memory-search-for-personal-ai-agents.md",
+    "title": "\"Beyond Similarity: Trustworthy Memory Search for Personal AI Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Beyond Similarity: Trustworthy Memory Search for Personal AI Agents\"",
+      "authors": "Jiawen Zhang, Kejia Chen, Jiachen Ma, Yangfan Hu, Lipeng He, Yechao Zhang, Jian Liu, Xiaohu Yang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.06054",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-04",
+      "updated_at": "2026-06-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-06090-beyond-semantic-organization-memory-as-execution-state-management-for-long-horiz.md",
+    "title": "\"Beyond Semantic Organization: Memory as Execution State Management for Long-Horizon Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Beyond Semantic Organization: Memory as Execution State Management for Long-Horizon Agents\"",
+      "authors": "Yaoqi Chen, Haibin Lai, Yuru Feng, Chuyu Han, Qianxi Zhang, Baotong Lu, Menghao Li, Xinjiang Wang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.06090",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-04",
+      "updated_at": "2026-06-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "computer-use",
+        "memory",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-memory, rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-06388-humans-almanac-a-human-collaboration-dataset-of-action-level-mental-model-annota.md",
+    "title": "\"Humans' ALMANAC: A Human Collaboration Dataset of Action-Level Mental Model Annotations for Agent Collaboration\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Humans' ALMANAC: A Human Collaboration Dataset of Action-Level Mental Model Annotations for Agent Collaboration\"",
+      "authors": "Jiaju Chen, Yuxuan Lu, Jiayi Su, Chaoran Chen, Songlin Xiao, Zheng Zhang, Yun Wang, Yunyao Li, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.06388",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-04",
+      "updated_at": "2026-06-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "multi-agent",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-06399-collabsim-a-cscw-grounded-methodology-for-investigating-collaborative-competence.md",
+    "title": "\"CollabSim: A CSCW-Grounded Methodology for Investigating Collaborative Competence of LLM Agents through Controlled Multi-Agent Experiments\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"CollabSim: A CSCW-Grounded Methodology for Investigating Collaborative Competence of LLM Agents through Controlled Multi-Agent Experiments\"",
+      "authors": "Jiaju Chen, Bo Sun, Yuxuan Lu, Yun Wang, Dakuo Wang, Bingsheng Yao",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.06399",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-04",
+      "updated_at": "2026-06-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "planning",
+        "reasoning",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "22",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-06448-agent-memory-characterization-and-system-implications-of-stateful-long-horizon-w.md",
+    "title": "\"Agent Memory: Characterization and System Implications of Stateful Long-Horizon Workloads\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Agent Memory: Characterization and System Implications of Stateful Long-Horizon Workloads\"",
+      "authors": "Yasmine Omri, Ziyu Gan, Zachary Broveak, Robin Geens, Zexue He, Alex Pentland, Marian Verhelst, Tsachy Weissman, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.06448",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-04",
+      "updated_at": "2026-06-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-06462-benchmark-everything-everywhere-all-at-once.md",
+    "title": "Benchmark Everything Everywhere All at Once",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Benchmark Everything Everywhere All at Once",
+      "authors": "Shiyun Xiong, Dongming Wu, Peiwen Sun, Yuang Ai, Bokang Yang, Wencheng Han, Xiao-Hui Li, Xiangyu Yue",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.06462",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-04",
+      "updated_at": "2026-06-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-06473-mlevolve-a-self-evolving-framework-for-automated-machine-learning-algorithm-disc.md",
+    "title": "\"MLEvolve: A Self-Evolving Framework for Automated Machine Learning Algorithm Discovery\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MLEvolve: A Self-Evolving Framework for Automated Machine Learning Algorithm Discovery\"",
+      "authors": "Shangheng Du, Xiangchao Yan, Jinxin Shi, Zongsheng Cao, Shiyang Feng, Zichen Liang, Boyuan Sun, Tianshuo Peng, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.06473",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-04",
+      "updated_at": "2026-06-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "memory",
+        "multi-agent",
+        "planning",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-07314-qbuglm-an-agentic-benchmarking-framework-for-llm-based-quantum-software-debuggin.md",
+    "title": "\"QBugLM: An Agentic Benchmarking Framework for LLM-based Quantum Software Debugging\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"QBugLM: An Agentic Benchmarking Framework for LLM-based Quantum Software Debugging\"",
+      "authors": "An B. B. Pham, Hoa T. Nguyen, Muhammad Usman",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.07314",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-05",
+      "updated_at": "2026-06-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "multi-agent",
+        "reasoning",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.ET",
+        "quant-ph"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-07379-do-coding-agents-deceive-us-detecting-and-preventing-cheating-via-capped-evaluat.md",
+    "title": "Do Coding Agents Deceive Us? Detecting and Preventing Cheating via Capped Evaluation with Randomized Tests",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Do Coding Agents Deceive Us? Detecting and Preventing Cheating via Capped Evaluation with Randomized Tests",
+      "authors": "Thanawat Lodkaew, Johannes Ackermann, Soichiro Nishimori, Nontawat Charoenphakdee, Masashi Sugiyama, Takashi Ishida",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.07379",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-05",
+      "updated_at": "2026-06-08",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI",
+        "cs.CL",
+        "stat.ME"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-07402-m-3-exam-benchmarking-multimodal-memory-for-realistic-user-agent-interactions.md",
+    "title": "\"M$^3$Exam: Benchmarking Multimodal Memory for Realistic User-Agent Interactions\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"M$^3$Exam: Benchmarking Multimodal Memory for Realistic User-Agent Interactions\"",
+      "authors": "Zhengjun Huang, Wenxuan Liu, Zhoujin Tian, Wei Chen, Junle Chen, Yuqian Wu, Fangyuan Zhang, Qintian Guo, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.07402",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-05",
+      "updated_at": "2026-06-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-07591-researchclawbench-a-benchmark-for-end-to-end-autonomous-scientific-research.md",
+    "title": "\"ResearchClawBench: A Benchmark for End-to-End Autonomous Scientific Research\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ResearchClawBench: A Benchmark for End-to-End Autonomous Scientific Research\"",
+      "authors": "Wanghan Xu, Shuo Li, Tianlin Ye, Qinglong Cao, Yixin Chen, Hengjian Gao, Yiheng Wang, Qi Li, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.07591",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-28",
+      "updated_at": "2026-07-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-07595-visualleakbench-reproducible-action-boundary-propagation-failures-in-vision-lang.md",
+    "title": "\"VisualLeakBench: Reproducible Action-Boundary Propagation Failures in Vision-Language Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"VisualLeakBench: Reproducible Action-Boundary Propagation Failures in Vision-Language Agents\"",
+      "authors": "Youting Wang, Yuan Tang, Yitian Qian, Chen Zhao",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.07595",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-29",
+      "updated_at": "2026-05-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV",
+        "cs.AI",
+        "cs.IR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-07682-swe-marathon-can-agents-autonomously-complete-ultra-long-horizon-software-work.md",
+    "title": "\"SWE-Marathon: Can Agents Autonomously Complete Ultra-Long-Horizon Software Work?\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SWE-Marathon: Can Agents Autonomously Complete Ultra-Long-Horizon Software Work?\"",
+      "authors": "Rishi Desai, Jesse Hu, Joan Cabezas, Neel Harsola, Pratyush Shukla, Roey Ben Chaim, Adnan El Assadi, Omkaar Mukund Kamath, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.07682",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-05",
+      "updated_at": "2026-06-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "memory",
+        "planning",
+        "rag",
+        "reasoning",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-07711-rosetta-memory-adaptive-memory-for-cross-llm-agents.md",
+    "title": "\"Rosetta Memory: Adaptive Memory for Cross-LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Rosetta Memory: Adaptive Memory for Cross-LLM Agents\"",
+      "authors": "Hao Yang, Shiqi Shen, Haoxuan Li, Zhipeng Wang, Zhi Gong, Xu Chen",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.07711",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-05",
+      "updated_at": "2026-06-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "coding-agent",
+        "memory",
+        "planning",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-07836-agentic-multi-fidelity-learning-of-quasiparticle-and-excitonic-properties.md",
+    "title": "Agentic multi-fidelity learning of quasiparticle and excitonic properties",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Agentic multi-fidelity learning of quasiparticle and excitonic properties",
+      "authors": "Arnab Neogi, Aaron Forde, Christopher A. Lane, Sergei Tretiak, Jian-Xin Zhu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.07836",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-05",
+      "updated_at": "2026-06-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "rag",
+        "workflow-agent",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cond-mat.mtrl-sci",
+        "cond-mat.stat-mech",
+        "cs.AI",
+        "physics.comp-ph",
+        "quant-ph"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-07867-the-cold-start-safety-gap-in-llm-agents.md",
+    "title": "The Cold-Start Safety Gap in LLM Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "The Cold-Start Safety Gap in LLM Agents",
+      "authors": "Chung-En Sun, Linbo Liu, Tsui-Wei Weng",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.07867",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-05",
+      "updated_at": "2026-06-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-08162-silent-failure-in-llm-agent-systems-the-entropy-principle-and-the-inevitable-dis.md",
+    "title": "\"Silent Failure in LLM Agent Systems: The Entropy Principle and the Inevitable Disorder of Autonomous Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Silent Failure in LLM Agent Systems: The Entropy Principle and the Inevitable Disorder of Autonomous Agents\"",
+      "authors": "Dexing Liu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.08162",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-06",
+      "updated_at": "2026-06-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "memory",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-08172-the-governance-of-human-llm-interaction-safety-gating-civility-steering-and-affe.md",
+    "title": "\"The Governance of Human-LLM Interaction: Safety Gating, Civility Steering, and Affective Default Lock-In\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"The Governance of Human-LLM Interaction: Safety Gating, Civility Steering, and Affective Default Lock-In\"",
+      "authors": "Manuele Reani, Hongjian Zhang, Hongyu Tian",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.08172",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-06",
+      "updated_at": "2026-06-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "multi-agent",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.HC",
+        "cs.AI",
+        "cs.CY"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-08274-toward-human-centered-multi-agent-systems-integrating-cognition-culture-values-a.md",
+    "title": "\"Toward Human-Centered Multi-Agent Systems: Integrating Cognition, Culture, Values, and Cooperation in AI Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Toward Human-Centered Multi-Agent Systems: Integrating Cognition, Culture, Values, and Cooperation in AI Agents\"",
+      "authors": "Safia Baloch, Rahemeen Khan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.08274",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-06",
+      "updated_at": "2026-06-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "multi-agent",
+        "planning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "24",
+      "collection_queries": "autonomous-agent-llm, planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-08340-benchmarking-open-ended-multi-agent-coordination-in-language-agents.md",
+    "title": "Benchmarking Open-Ended Multi-Agent Coordination in Language Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Benchmarking Open-Ended Multi-Agent Coordination in Language Agents",
+      "authors": "Kale-ab Abebe Tessera, Andras Szecsenyi, Cameron Barker, Alexander Rutherford, Davide Paglieri, Aidan Scannell, Henry Gouk, Elliot J. Crowley, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.08340",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-06",
+      "updated_at": "2026-06-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "multi-agent",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.LG",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "28",
+      "collection_queries": "autonomous-agent-llm, language-agent, planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-08531-vesta-a-fully-automated-scenario-generation-and-safety-evaluation-framework-for-.md",
+    "title": "\"VESTA: A Fully Automated Scenario Generation and Safety Evaluation Framework for LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"VESTA: A Fully Automated Scenario Generation and Safety Evaluation Framework for LLM Agents\"",
+      "authors": "Lu Jia, Haibo Tong, Feifei Zhao, Jindong Li, Dongqi Liang, Ping Wu, Qian Zhang, Yi Zeng",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.08531",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-07",
+      "updated_at": "2026-06-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-08625-from-holistic-evaluation-to-structured-criteria-rubrics-across-the-evolving-llm-.md",
+    "title": "\"From Holistic Evaluation to Structured Criteria: Rubrics Across the Evolving LLM Landscape\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"From Holistic Evaluation to Structured Criteria: Rubrics Across the Evolving LLM Landscape\"",
+      "authors": "Hao Chen, Ziyu Han, Yukun Yan, Qingfu Zhu, Maosong Sun, Wanxiang Che",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.08625",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-07",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-08790-rails-verification-native-clearing-for-agentic-commerce.md",
+    "title": "\"RAILS: Verification-Native Clearing For Agentic Commerce\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"RAILS: Verification-Native Clearing For Agentic Commerce\"",
+      "authors": "Adrian de Valois-Franklin, Alex Bogdan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.08790",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-07",
+      "updated_at": "2026-06-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CR",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-08960-hardening-agent-benchmarks-with-adversarial-hacker-fixer-loops.md",
+    "title": "Hardening Agent Benchmarks with Adversarial Hacker-Fixer Loops",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Hardening Agent Benchmarks with Adversarial Hacker-Fixer Loops",
+      "authors": "Ziqian Zhong, Ivgeni Segal, Ivan Bercovich, Shashwat Saxena, Kexun Zhang, Aditi Raghunathan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.08960",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-08",
+      "updated_at": "2026-06-08",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI",
+        "cs.LG",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-09037-a-multi-agent-system-for-ipmsm-design-optimization-via-an-fea-ai-hybrid-approach.md",
+    "title": "A Multi-Agent System for IPMSM Design Optimization via an FEA-AI Hybrid Approach",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "A Multi-Agent System for IPMSM Design Optimization via an FEA-AI Hybrid Approach",
+      "authors": "Jinseong Han, Sunwoong Yang, Namwoo Kang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.09037",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-08",
+      "updated_at": "2026-06-08",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "planning",
+        "rag",
+        "reasoning",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-09071-reflect-intervention-supported-error-attribution-for-silent-failures-in-llm-agen.md",
+    "title": "\"REFLECT: Intervention-Supported Error Attribution for Silent Failures in LLM Agent Traces\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"REFLECT: Intervention-Supported Error Attribution for Silent Failures in LLM Agent Traces\"",
+      "authors": "Xiaofeng Lin, Yingxu Wang, Tung Sum Thomas Kwok, Daniel Guo, Sahil Arun Nale, Charles Fleming, Guang Cheng",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.09071",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-08",
+      "updated_at": "2026-06-08",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-09198-mass-deep-research-for-social-sciences-with-memory-augmented-social-simulation.md",
+    "title": "\"MASS: Deep Research for Social Sciences with Memory-Augmented Social Simulation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MASS: Deep Research for Social Sciences with Memory-Augmented Social Simulation\"",
+      "authors": "Yongrui Liu, Deyi Xiong",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.09198",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-08",
+      "updated_at": "2026-06-08",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "planning",
+        "rag",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-09316-anything2skill-compiling-external-knowledge-into-reusable-skills-for-agents.md",
+    "title": "\"Anything2Skill: Compiling External Knowledge into Reusable Skills for Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Anything2Skill: Compiling External Knowledge into Reusable Skills for Agents\"",
+      "authors": "Qianjun Pan, Yutao Yang, Junsong Li, Jie Zhou, Kai Chen, Xin Li, Qin Chen, Liang He",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.09316",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-08",
+      "updated_at": "2026-06-19",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "planning",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-09399-runagent-superbrowser-a-theory-of-autonomous-web-navigation-grounded-in-human-br.md",
+    "title": "\"RunAgent SuperBrowser: A Theory of Autonomous Web Navigation Grounded in Human Browsing Behaviour\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"RunAgent SuperBrowser: A Theory of Autonomous Web Navigation Grounded in Human Browsing Behaviour\"",
+      "authors": "Radeen Mostafa, Sawradip Saha",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.09399",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-08",
+      "updated_at": "2026-06-08",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "embodied-agent",
+        "memory",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-09426-weavebench-a-long-horizon-real-world-benchmark-for-computer-use-agents-with-hybr.md",
+    "title": "\"WeaveBench: A Long-Horizon, Real-World Benchmark for Computer-Use Agents with Hybrid Interfaces\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"WeaveBench: A Long-Horizon, Real-World Benchmark for Computer-Use Agents with Hybrid Interfaces\"",
+      "authors": "Wanli Li, Bowen Zhou, Yunyao Yu, Zhou Xu, Yifan Yang, Dongsheng Li, Caihua Shan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.09426",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-08",
+      "updated_at": "2026-07-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-09447-aliyunconsoleagent-training-web-agents-in-real-world-cloud-environments-via-dist.md",
+    "title": "\"AliyunConsoleAgent: Training Web Agents in Real-World Cloud Environments via Distillation and Reinforcement Learning\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AliyunConsoleAgent: Training Web Agents in Real-World Cloud Environments via Distillation and Reinforcement Learning\"",
+      "authors": "Bojie Rong, Zheyu Shen, Qiaoping Wang, Pengfei Kang, Yang Xu, Yawen Wei, Hanyu Wu, Zhi Zhao, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.09447",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-08",
+      "updated_at": "2026-06-08",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-09483-memory-beyond-recall-a-dual-process-cognitive-memory-system-for-self-evolving-ll.md",
+    "title": "\"Memory Beyond Recall: A Dual-Process Cognitive Memory System for Self-Evolving LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Memory Beyond Recall: A Dual-Process Cognitive Memory System for Self-Evolving LLM Agents\"",
+      "authors": "Tianxiang Fei, Mingyang Song, Mao Zheng, Xiang Yu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.09483",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-08",
+      "updated_at": "2026-06-08",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-09549-secureclaw-clawing-back-control-of-llm-agents.md",
+    "title": "\"SecureClaw: Clawing Back Control of LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SecureClaw: Clawing Back Control of LLM Agents\"",
+      "authors": "Yuhan Ma, Stefan Schmid",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.09549",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-08",
+      "updated_at": "2026-06-08",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-safety, planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-09738-hdsl-a-hierarchical-domain-specific-language-for-structured-3d-indoor-scene-gene.md",
+    "title": "\"HDSL: A Hierarchical Domain-Specific Language for Structured 3D Indoor Scene Generation and Localized Editing with LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"HDSL: A Hierarchical Domain-Specific Language for Structured 3D Indoor Scene Generation and Localized Editing with LLM Agents\"",
+      "authors": "Letian Li, Chao Shen, Shuzhao Xie, Chenghao Gu, ZhengXiao He, Yu Meng, Xin Yang, Wenyuan Jiang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.09738",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-08",
+      "updated_at": "2026-06-08",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "planning",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-09764-iosworld-a-benchmark-for-personally-intelligent-phone-agents.md",
+    "title": "\"iOSWorld: A Benchmark for Personally Intelligent Phone Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"iOSWorld: A Benchmark for Personally Intelligent Phone Agents\"",
+      "authors": "Lawrence Keunho Jang, Mareks Woodside, Geronimo Carom, Andrew Keunwoo Jang, Jing Yu Koh, Ruslan Salakhutdinov",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.09764",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-08",
+      "updated_at": "2026-06-08",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-evaluation, web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-09774-auto-configuring-scientific-simulators-with-lightweight-coding-agent-adapters.md",
+    "title": "Auto-Configuring Scientific Simulators with Lightweight Coding-Agent Adapters",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Auto-Configuring Scientific Simulators with Lightweight Coding-Agent Adapters",
+      "authors": "Matthew Ho, Brian Liu, Jixuan Chen, Audrey Wang, Lianhui Qin",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.09774",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-08",
+      "updated_at": "2026-06-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "memory",
+        "rag",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-09863-from-confident-closing-to-silent-failure-characterizing-false-success-in-llm-age.md",
+    "title": "\"From Confident Closing to Silent Failure: Characterizing False Success in LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"From Confident Closing to Silent Failure: Characterizing False Success in LLM Agents\"",
+      "authors": "Laksh Advani",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.09863",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-01",
+      "updated_at": "2026-06-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-09961-3spo-state-score-supervised-policy-optimization-for-llm-agents.md",
+    "title": "\"3SPO: State-Score-Supervised Policy Optimization for LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"3SPO: State-Score-Supervised Policy Optimization for LLM Agents\"",
+      "authors": "Yu Han, Kailing Li, Yang Jiao, Yulin Dai, Yuqian Fu, Linhai Zhuo, Tianwen Qian",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.09961",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-08",
+      "updated_at": "2026-06-08",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "computer-use",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-10209-less-context-better-agents-efficient-context-engineering-for-long-horizon-tool-u.md",
+    "title": "\"Less Context, Better Agents: Efficient Context Engineering for Long-Horizon Tool-Using LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Less Context, Better Agents: Efficient Context Engineering for Long-Horizon Tool-Using LLM Agents\"",
+      "authors": "Abhilasha Lodha, Mahsa Pahlavikhah Varnosfaderani, Abir Chakraborty, Abhinav Mithal",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.10209",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-08",
+      "updated_at": "2026-06-08",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.LG",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-10304-mirage-a-polarity-flipping-encoding-subspace-in-llm-agents.md",
+    "title": "\"MIRAGE: A Polarity-Flipping Encoding Subspace in LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MIRAGE: A Polarity-Flipping Encoding Subspace in LLM Agents\"",
+      "authors": "Pratibha Revankar, Kargi Chauhan, Jihye Kim, Sadiba Nusrat Nur, Vincent Siu, Chenguang Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.10304",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-09",
+      "updated_at": "2026-06-09",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "computer-use",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-10316-tabclaw-an-interactive-and-self-evolving-agent-for-spreadsheet-manipulation-and-.md",
+    "title": "\"TabClaw: An Interactive and Self-Evolving Agent for Spreadsheet Manipulation and Table Reasoning\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"TabClaw: An Interactive and Self-Evolving Agent for Spreadsheet Manipulation and Table Reasoning\"",
+      "authors": "Mingyue Cheng, Shuo Yu, Daoyu Wang, Qingchuan Li, Xiaoyu Tao, Qingyang Mao, Yitong Zhou, Qi Liu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.10316",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-09",
+      "updated_at": "2026-06-09",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-10381-agentic-hybrid-rag-for-evidence-grounded-muon-collider-analysis.md",
+    "title": "Agentic Hybrid RAG for Evidence-Grounded Muon Collider Analysis",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Agentic Hybrid RAG for Evidence-Grounded Muon Collider Analysis",
+      "authors": "Ruobing Jiang, Dawei Fu, Cheng Jiang, Tianyi Yang, Zijian Wang, Youpeng Wu, Yong Ban, Yajun Mao, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.10381",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-09",
+      "updated_at": "2026-06-09",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "hep-ex",
+        "cs.AI",
+        "cs.CL",
+        "cs.IR",
+        "physics.ins-det"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-10394-stage-claw-automated-state-based-agent-benchmarking-for-realistic-scenarios.md",
+    "title": "\"STAGE-Claw: Automated State-based Agent Benchmarking for Realistic Scenarios\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"STAGE-Claw: Automated State-based Agent Benchmarking for Realistic Scenarios\"",
+      "authors": "Sirui Liang, Bohan Yu, Peiyu Wang, Shiguang Guo, Wenxing Hu, Pengfei Cao, Jian Zhao, Cao Liu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.10394",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-09",
+      "updated_at": "2026-06-09",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-10423-webchallenger-a-reliable-and-efficient-generalist-web-agent.md",
+    "title": "\"WebChallenger: A Reliable and Efficient Generalist Web Agent\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"WebChallenger: A Reliable and Efficient Generalist Web Agent\"",
+      "authors": "Jayoo Hwang, Xiaowen Zhang, Vedant Padwal",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.10423",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-09",
+      "updated_at": "2026-06-09",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "computer-use",
+        "embodied-agent",
+        "memory",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-10507-hipif-hierarchical-planning-and-information-folding-for-long-horizon-llm-agent-l.md",
+    "title": "\"HIPIF: Hierarchical Planning and Information Folding for Long-Horizon LLM Agent Learning\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"HIPIF: Hierarchical Planning and Information Folding for Long-Horizon LLM Agent Learning\"",
+      "authors": "Juncheng Diao, Zhicong Lu, Peiguang Li, Yongwei Zhou, Changyuan Tian, Qingbin Li, Rongxiang Weng, Jingang Wang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.10507",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-09",
+      "updated_at": "2026-06-09",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "planning",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agent-evaluation, autonomous-agent-llm, planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-10532-activemem-distributed-active-memory-for-long-horizon-llm-reasoning.md",
+    "title": "\"ActiveMem: Distributed Active Memory for Long-Horizon LLM Reasoning\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ActiveMem: Distributed Active Memory for Long-Horizon LLM Reasoning\"",
+      "authors": "Yunhan Jiang, Wenbin Duan, Shasha Guo, Liang Pang, Xiaoqian Sun, Huawei Shen",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.10532",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-09",
+      "updated_at": "2026-06-09",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "memory",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-10577-agenticnav-zero-shot-vision-and-language-navigation-as-a-tool-calling-harness.md",
+    "title": "\"AgenticNav: Zero-Shot Vision-and-Language Navigation as a Tool-Calling Harness\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AgenticNav: Zero-Shot Vision-and-Language Navigation as a Tool-Calling Harness\"",
+      "authors": "Yijian Li, Changze Li, Hantian Shi, Jiaying Luo, Jiyuan Cai, Ming Yang, Tong Qin",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.10577",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-09",
+      "updated_at": "2026-06-09",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "embodied-agent",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.RO"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-10616-learning-what-to-remember-observability-safe-memory-retention-via-constrained-op.md",
+    "title": "\"Learning What to Remember: Observability-Safe Memory Retention via Constrained Optimization for Long-Horizon Language Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Learning What to Remember: Observability-Safe Memory Retention via Constrained Optimization for Long-Horizon Language Agents\"",
+      "authors": "Qingcan Kang, Liu Mingyang, Shixiong Kai, Kaichao Liang, Tao Zhong, Mingxuan Yuan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.10616",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-09",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-10677-infini-memory-maintainable-topic-documents-for-long-term-llm-agent-memory.md",
+    "title": "\"Infini Memory: Maintainable Topic Documents for Long-Term LLM Agent Memory\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Infini Memory: Maintainable Topic Documents for Long-Term LLM Agent Memory\"",
+      "authors": "Suozhao Ji, Baodong Wu, Zehao Wang, Lei Xia, Qingping Li, Ruisong Wang, Wenbo Ding, Zhenhua Zhu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.10677",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-09",
+      "updated_at": "2026-06-09",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-10684-divide-and-cooperate-role-decomposed-multi-agent-llm-training-with-cross-agent-l.md",
+    "title": "\"Divide and Cooperate: Role-Decomposed Multi-Agent LLM Training with Cross-Agent Learning Signals\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Divide and Cooperate: Role-Decomposed Multi-Agent LLM Training with Cross-Agent Learning Signals\"",
+      "authors": "Jaewan Park, Solbee Cho, Jay-Yoon Lee",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.10684",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-09",
+      "updated_at": "2026-06-09",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-10742-memvenom-triggered-poisoning-of-multimodal-memories-in-web-agents.md",
+    "title": "\"MemVenom: Triggered Poisoning of Multimodal Memories in Web Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MemVenom: Triggered Poisoning of Multimodal Memories in Web Agents\"",
+      "authors": "Yv Zhang, Hao Sun, Hao Fang, Kuofeng Gao, Fan Mo, Bin Chen, Shu-Tao Xia, Yaowei Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.10742",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-09",
+      "updated_at": "2026-06-09",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "planning",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-10749-toward-secure-llm-agents-threat-surfaces-attacks-defenses-and-evaluation.md",
+    "title": "\"Toward Secure LLM Agents: Threat Surfaces, Attacks, Defenses, and Evaluation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Toward Secure LLM Agents: Threat Surfaces, Attacks, Defenses, and Evaluation\"",
+      "authors": "Yuchen Ling, Shengcheng Yu, Zhenyu Chen, Chunrong Fang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.10749",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-09",
+      "updated_at": "2026-06-09",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "multi-agent",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "25",
+      "collection_queries": "agent-safety, planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-10921-trace-only-what-you-need-structure-aware-on-demand-hypergraph-memory-for-long-do.md",
+    "title": "\"Trace Only What You Need: Structure-Aware On-Demand Hypergraph Memory for Long-Document Question Answering\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Trace Only What You Need: Structure-Aware On-Demand Hypergraph Memory for Long-Document Question Answering\"",
+      "authors": "Xiangjun Zai, Xingyu Tan, Chen Chen, Xiaoyang Wang, Wenjie Zhang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.10921",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-09",
+      "updated_at": "2026-06-09",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "multi-agent",
+        "planning",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-10933-frontier-coding-agents-use-metaprogramming-to-adapt-to-unfamiliar-programming-la.md",
+    "title": "Frontier Coding Agents Use Metaprogramming to Adapt to Unfamiliar Programming Languages",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Frontier Coding Agents Use Metaprogramming to Adapt to Unfamiliar Programming Languages",
+      "authors": "Aman Sharma, Sushrut Thorat, Paras Chopra",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.10933",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-09",
+      "updated_at": "2026-06-09",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "computer-use",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-11042-workflow-gym-towards-long-horizon-evaluation-of-computer-use-agentic-tasks-in-re.md",
+    "title": "\"Workflow-GYM: Towards Long-Horizon Evaluation of Computer-use Agentic tasks in Real-World Professional Fields\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Workflow-GYM: Towards Long-Horizon Evaluation of Computer-use Agentic tasks in Real-World Professional Fields\"",
+      "authors": "Liya Zhu, Jingzhe Ding, Jian Zhang, Jianbo Xue, Shihao Liang, Ge Zhang, Yi Zhu, Duju Zeng, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.11042",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-09",
+      "updated_at": "2026-06-11",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-11078-a-history-aware-visually-grounded-critic-for-computer-use-agents.md",
+    "title": "A History-Aware Visually Grounded Critic for Computer Use Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "A History-Aware Visually Grounded Critic for Computer Use Agents",
+      "authors": "Jaewoo Lee, Zaid Khan, Archiki Prasad, Justin Chih-Yao Chen, Supriyo Chakraborty, Kartik Balasubramaniam, Sambit Sahu, Elias Stengel-Eskin, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.11078",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-09",
+      "updated_at": "2026-06-09",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL",
+        "cs.CV"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-11079-vista-a-versatile-interactive-user-simulation-toolkit-for-agent-evaluation.md",
+    "title": "\"VISTA: A Versatile Interactive User Simulation Toolkit for Agent Evaluation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"VISTA: A Versatile Interactive User Simulation Toolkit for Agent Evaluation\"",
+      "authors": "Yunan Lu, Ryan Shea, Yusen Zhang, Zhou Yu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.11079",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-09",
+      "updated_at": "2026-06-09",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-11119-trace-a-unified-rollout-budget-allocation-framework-for-efficient-agentic-reinfo.md",
+    "title": "\"TRACE: A Unified Rollout Budget Allocation Framework for Efficient Agentic Reinforcement Learning\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"TRACE: A Unified Rollout Budget Allocation Framework for Efficient Agentic Reinforcement Learning\"",
+      "authors": "Heming Zou, Qi Wang, Yun Qu, Yuhang Jiang, Lizhou Cai, Yixiu Mao, Ru Peng, Xin Xu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.11119",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-09",
+      "updated_at": "2026-06-09",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-11176-data-journalist-agent-transforming-data-into-verifiable-multimodal-stories.md",
+    "title": "\"Data Journalist Agent: Transforming Data into Verifiable Multimodal Stories\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Data Journalist Agent: Transforming Data into Verifiable Multimodal Stories\"",
+      "authors": "Kevin Qinghong Lin, Batu EI, Yuhong Shi, Pan Lu, Philip Torr, James Zou",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.11176",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-09",
+      "updated_at": "2026-06-09",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "multi-agent",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV",
+        "cs.CL",
+        "cs.CY",
+        "cs.HC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-11349-knowing-when-to-ask-self-gated-clarification-for-hierarchical-language-agents.md",
+    "title": "\"Knowing When to Ask: Self-Gated Clarification for Hierarchical Language Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Knowing When to Ask: Self-Gated Clarification for Hierarchical Language Agents\"",
+      "authors": "Aijing Gao, Yiming Kang, Mengdie Flora Wang, Jae Oh Woo",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.11349",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-09",
+      "updated_at": "2026-06-12",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "embodied-agent",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.HC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-11354-a-zero-shot-multi-agent-framework-for-human-building-interaction-via-programmati.md",
+    "title": "A Zero-Shot Multi-Agent Framework for Human-Building Interaction via Programmatic Reasoning",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "A Zero-Shot Multi-Agent Framework for Human-Building Interaction via Programmatic Reasoning",
+      "authors": "Yuqi Wang, Gulai Shen, Ali Mehmani",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.11354",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-09",
+      "updated_at": "2026-06-09",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "coding-agent",
+        "multi-agent",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.ET"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-11680-organize-then-retrieve-hierarchical-memory-navigation-for-efficient-agents.md",
+    "title": "\"Organize then Retrieve: Hierarchical Memory Navigation for Efficient Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Organize then Retrieve: Hierarchical Memory Navigation for Efficient Agents\"",
+      "authors": "Hao-Lun Hsu, Nikki Lijing Kuang, Boyi Liu, Zhewei Yao, Yuxiong He",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.11680",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-10",
+      "updated_at": "2026-06-10",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "embodied-agent",
+        "memory",
+        "planning",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-11688-goal-autopilot-a-verifiable-anti-fabrication-firewall-for-unattended-long-horizo.md",
+    "title": "\"Goal-Autopilot: A Verifiable Anti-Fabrication Firewall for Unattended Long-Horizon Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Goal-Autopilot: A Verifiable Anti-Fabrication Firewall for Unattended Long-Horizon Agents\"",
+      "authors": "Youwang Deng",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.11688",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-10",
+      "updated_at": "2026-06-10",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "planning",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-11702-medcta-a-benchmark-for-clinical-tool-agents.md",
+    "title": "\"MedCTA: A Benchmark for Clinical Tool Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MedCTA: A Benchmark for Clinical Tool Agents\"",
+      "authors": "Tajamul Ashraf, Hyewon Jeong, Fida Mohammad Thoker, Bernard Ghanem",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.11702",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-10",
+      "updated_at": "2026-06-10",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV",
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-11869-agents-all-the-way-down-a-methodology-for-building-custom-ai-agents-from-substra.md",
+    "title": "Agents All the Way Down; A Methodology for Building Custom AI Agents from Substrate to Production",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Agents All the Way Down; A Methodology for Building Custom AI Agents from Substrate to Production",
+      "authors": "Marc Alier Forment, Juanan Pereira, Francisco José García-Peñalvo, María José Casañ Guerrero",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.11869",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-10",
+      "updated_at": "2026-06-10",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "coding-agent",
+        "multi-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-12195-internvideo3-agentify-foundation-models-with-multimodal-contextual-reasoning.md",
+    "title": "\"InternVideo3: Agentify Foundation Models with Multimodal Contextual Reasoning\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"InternVideo3: Agentify Foundation Models with Multimodal Contextual Reasoning\"",
+      "authors": "Ziang Yan, Sheng Xia, Jiashuo Yu, Yue Wu, Tianxiang Jiang, Songze Li, Kanghui Tian, Yicheng Xu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.12195",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-10",
+      "updated_at": "2026-06-10",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-12320-a-five-plane-reference-architecture-for-runtime-governance-of-production-ai-agen.md",
+    "title": "A Five-Plane Reference Architecture for Runtime Governance of Production AI Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "A Five-Plane Reference Architecture for Runtime Governance of Production AI Agents",
+      "authors": "Krti Tallam",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.12320",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-10",
+      "updated_at": "2026-06-10",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "planning",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CC",
+        "cs.CR",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-12341-ocelot-inference-leakage-budgets-for-privacy-preserving-llm-agents.md",
+    "title": "\"OCELOT: Inference-Leakage Budgets for Privacy-Preserving LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"OCELOT: Inference-Leakage Budgets for Privacy-Preserving LLM Agents\"",
+      "authors": "Jin Xie, Songze Li",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.12341",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-10",
+      "updated_at": "2026-06-10",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-12344-claw-swe-bench-a-benchmark-for-evaluating-openclaw-style-agent-harnesses-on-codi.md",
+    "title": "\"Claw-SWE-Bench: A Benchmark for Evaluating OpenClaw-style Agent Harnesses on Coding Tasks\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Claw-SWE-Bench: A Benchmark for Evaluating OpenClaw-style Agent Harnesses on Coding Tasks\"",
+      "authors": "Mengyu Zheng, Kai Han, Boxun Li, Haiyang Xu, Yuchuan Tian, Wei He, Hang Zhou, Jianyuan Guo, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.12344",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-10",
+      "updated_at": "2026-06-10",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-12384-appo-agentic-procedural-policy-optimization.md",
+    "title": "\"APPO: Agentic Procedural Policy Optimization\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"APPO: Agentic Procedural Policy Optimization\"",
+      "authors": "Xucong Wang, Ziyu Ma, Yong Wang, Yuxiang Ji, Shidong Yang, Guanhua Chen, Pengkun Wang, Xiangxiang Chu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.12384",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-10",
+      "updated_at": "2026-06-10",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-12563-arbor-tree-search-as-a-cognition-layer-for-autonomous-agents.md",
+    "title": "\"Arbor: Tree Search as a Cognition Layer for Autonomous Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Arbor: Tree Search as a Cognition Layer for Autonomous Agents\"",
+      "authors": "Neha Prakriya, Chaojun Hou, Zheng Gong, Huasha Zhao, Xi Zhao, Mou Li, Zhenyu Gu, Emad Barsoum",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.12563",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-10",
+      "updated_at": "2026-06-10",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "multi-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-12586-beyond-attack-success-rate-examining-trigger-leakage-in-vision-language-agentic-.md",
+    "title": "\"Beyond Attack Success Rate: Examining Trigger Leakage in Vision-Language Agentic Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Beyond Attack Success Rate: Examining Trigger Leakage in Vision-Language Agentic Systems\"",
+      "authors": "Jiamin Chang, Salil Kanhere, Piotr Koniusz, Jason, Xue, Hammond Pearce",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.12586",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-10",
+      "updated_at": "2026-06-10",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "embodied-agent",
+        "planning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "language-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-12634-keep-policy-gradient-in-charge-sibling-guided-credit-distillation-for-long-horiz.md",
+    "title": "\"Keep Policy Gradient in Charge: Sibling-Guided Credit Distillation for Long-Horizon Tool-Use Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Keep Policy Gradient in Charge: Sibling-Guided Credit Distillation for Long-Horizon Tool-Use Agents\"",
+      "authors": "Tianyu Ding, Jianhong Xin, Juan Pablo De la Cruz Weinstein",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.12634",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-10",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-12657-trajgenagent-a-hierarchical-llm-agent-for-human-mobility-trajectory-generation.md",
+    "title": "\"TrajGenAgent: A Hierarchical LLM Agent for Human Mobility Trajectory Generation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"TrajGenAgent: A Hierarchical LLM Agent for Human Mobility Trajectory Generation\"",
+      "authors": "Siyu Li, Toan Tran, Lingyi Zhao, Khurram Shafique, Li Xiong",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.12657",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-10",
+      "updated_at": "2026-06-10",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "reasoning",
+        "workflow-agent",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.DB",
+        "cs.RO"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-12674-evoflux-inference-time-evolution-of-executable-tool-workflows-for-compact-agents.md",
+    "title": "\"Evoflux: Inference-Time Evolution of Executable Tool Workflows for Compact Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Evoflux: Inference-Time Evolution of Executable Tool Workflows for Compact Agents\"",
+      "authors": "Kushal Raj Bhandari, Ling Yue, Ching-Yun Ko, Dhaval Patel, Shaowu Pan, Pin-Yu Chen, Jianxi Gao",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.12674",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-10",
+      "updated_at": "2026-06-10",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "computer-use",
+        "planning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "function-calling, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-12703-smsr-certified-defence-against-runtime-memory-poisoning-in-persistent-llm-agent-.md",
+    "title": "\"SMSR: Certified Defence Against Runtime Memory Poisoning in Persistent LLM Agent Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SMSR: Certified Defence Against Runtime Memory Poisoning in Persistent LLM Agent Systems\"",
+      "authors": "Tarun Sharma",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.12703",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-10",
+      "updated_at": "2026-06-10",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-12780-proplay-procedural-world-models-for-self-evolving-llm-agents.md",
+    "title": "\"ProPlay: Procedural World Models for Self-Evolving LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ProPlay: Procedural World Models for Self-Evolving LLM Agents\"",
+      "authors": "Yijun Ma, Zehong Wang, Yiyang Li, Ziming Li, Xiaoguang Guo, Weixiang Sun, Chuxu Zhang, Yanfang Ye",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.12780",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-11",
+      "updated_at": "2026-06-11",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "planning",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-12837-lohosearch-benchmarking-long-horizon-search-agents-beyond-the-human-difficulty-c.md",
+    "title": "\"LoHoSearch: Benchmarking Long-Horizon Search Agents Beyond the Human Difficulty Ceiling\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"LoHoSearch: Benchmarking Long-Horizon Search Agents Beyond the Human Difficulty Ceiling\"",
+      "authors": "Jiarui Zhao, Rongzhi Zhang, Lingchuan Liu, Hao Yang, Xunliang Cai, Xi Su",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.12837",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-11",
+      "updated_at": "2026-06-17",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-12945-learning-what-to-remember-a-cognitively-grounded-multi-factor-value-model-for-ag.md",
+    "title": "\"Learning What to Remember: A Cognitively Grounded Multi-Factor Value Model for Agentic Memory\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Learning What to Remember: A Cognitively Grounded Multi-Factor Value Model for Agentic Memory\"",
+      "authors": "Zhibao Chen, Qian Cheng",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.12945",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-11",
+      "updated_at": "2026-06-20",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "memory",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "22",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-13148-terrabench-can-agents-reason-over-heterogeneous-earth-system-data.md",
+    "title": "\"TerraBench: Can Agents Reason Over Heterogeneous Earth-System Data?\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"TerraBench: Can Agents Reason Over Heterogeneous Earth-System Data?\"",
+      "authors": "Dat Tien Nguyen, Thao Nguyen, Fadillah Adamsyah Maani, Huy M. Le, Muhammad Umer Sheikh, Numan Saeed, Muhammad Haris Khan, Salman Khan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.13148",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-11",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-13177-memrefine-llm-guided-compression-for-long-term-agent-memory.md",
+    "title": "\"MemRefine: LLM-Guided Compression for Long-Term Agent Memory\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MemRefine: LLM-Guided Compression for Long-Term Agent Memory\"",
+      "authors": "Minjae Kim, Jinheon Baek, Soyeong Jeong, Sung Ju Hwang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.13177",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-11",
+      "updated_at": "2026-06-11",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-13192-reasoning-for-mobile-user-experience-with-multimodal-llms-task-benchmark-and-app.md",
+    "title": "\"Reasoning for Mobile User Experience with Multimodal LLMs: Task, Benchmark, and Approach\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Reasoning for Mobile User Experience with Multimodal LLMs: Task, Benchmark, and Approach\"",
+      "authors": "Ruichao Mao, Zhou Fang, Teng Guo, Hao Yang, Yaping Li, Shaohua Peng, Maji Huang, Xiaoyu Lin, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.13192",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-11",
+      "updated_at": "2026-06-11",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-13317-skillcat-contrastive-assessment-and-topology-aware-skill-self-evolution-for-llm-.md",
+    "title": "\"SkillCAT: Contrastive Assessment and Topology-Aware Skill Self-Evolution for LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SkillCAT: Contrastive Assessment and Topology-Aware Skill Self-Evolution for LLM Agents\"",
+      "authors": "Kunfeng Chen, Qihuang Zhong, Juhua Liu, Bo Du",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.13317",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-11",
+      "updated_at": "2026-06-11",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-13385-who-pays-the-price-stakeholder-centric-prompt-injection-benchmarking-for-real-wo.md",
+    "title": "Who Pays the Price? Stakeholder-Centric Prompt Injection Benchmarking for Real-world Web Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Who Pays the Price? Stakeholder-Centric Prompt Injection Benchmarking for Real-world Web Agents",
+      "authors": "Zihao Wang, Yiming Li, Yutong Wu, Zheyu Liu, Kangjie Chen, Fok Kar Wai, Pin-Yu Chen, Vrizlynn L. L. Thing, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.13385",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-11",
+      "updated_at": "2026-06-11",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI",
+        "cs.CY",
+        "cs.HC",
+        "cs.MM"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-13602-epibench-verifiable-evaluation-of-ai-agents-on-epigenomics-analysis.md",
+    "title": "\"EpiBench: Verifiable Evaluation of AI Agents on Epigenomics Analysis\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"EpiBench: Verifiable Evaluation of AI Agents on Epigenomics Analysis\"",
+      "authors": "Harihara Muralidharan, Reema Baskar, Soo Hee Lee, Tim Proctor, Kenny Workman",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.13602",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-11",
+      "updated_at": "2026-06-11",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-13608-agentbeats-agentifying-agent-assessment-for-openness-standardization-and-reprodu.md",
+    "title": "\"AgentBeats: Agentifying Agent Assessment for Openness, Standardization, and Reproducibility\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AgentBeats: Agentifying Agent Assessment for Openness, Standardization, and Reproducibility\"",
+      "authors": "Xiaoyuan Liu, Jianhong Tu, Yuqi Chen, Siyuan Xie, Sihan Ren, Tianneng Shi, Gal Gantar, Evan Sandoval, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.13608",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-11",
+      "updated_at": "2026-06-14",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "multi-agent",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-13643-recursive-agent-harnesses.md",
+    "title": "Recursive Agent Harnesses",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Recursive Agent Harnesses",
+      "authors": "Elias Lumer, Sahil Sen, Kevin Paul, Vamse Kumar Subbiah",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.13643",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-11",
+      "updated_at": "2026-06-11",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "planning",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-13663-hypertool-beyond-step-wise-tool-calls-for-tool-augmented-agents.md",
+    "title": "\"HyperTool: Beyond Step-Wise Tool Calls for Tool-Augmented Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"HyperTool: Beyond Step-Wise Tool Calls for Tool-Augmented Agents\"",
+      "authors": "Yaxin Du, Yifan Zhou, Yujie Ge, Jiajun Wang, Xianghe Pang, Shuo Tang, Tuney Zheng, Bryan Dai, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.13663",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-11",
+      "updated_at": "2026-06-11",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-13686-benchmarking-web-agent-safety-under-e-commerce-deceptive-interfaces.md",
+    "title": "Benchmarking Web Agent Safety under E-commerce Deceptive Interfaces",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Benchmarking Web Agent Safety under E-commerce Deceptive Interfaces",
+      "authors": "Zijing Shi, Meng Fang, Ling Chen",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.13686",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-26",
+      "updated_at": "2026-04-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.CY"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-13904-sana-what-matters-for-qa-agents-over-massive-data-lakes.md",
+    "title": "\"SANA: What Matters for QA Agents over Massive Data Lakes?\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SANA: What Matters for QA Agents over Massive Data Lakes?\"",
+      "authors": "Austin Senna Wijaya, Jiaxiang Liu, Haonan Wang, Eugene Wu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.13904",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-11",
+      "updated_at": "2026-06-11",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "embodied-agent",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI",
+        "cs.DB"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-13994-hidden-in-plain-sight-benchmarking-agent-safety-against-decomposition-attacks-wi.md",
+    "title": "\"Hidden in Plain Sight: Benchmarking Agent Safety Against Decomposition Attacks with DECOMPBENCH\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Hidden in Plain Sight: Benchmarking Agent Safety Against Decomposition Attacks with DECOMPBENCH\"",
+      "authors": "Vikhyath Kothamasu, Virginia Smith, Chhavi Yadav",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.13994",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-12",
+      "updated_at": "2026-06-12",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "21",
+      "collection_queries": "agent-safety, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-14106-naive-visual-memory-is-not-enough-a-failure-mode-study-of-gui-agents.md",
+    "title": "\"Naive Visual Memory is Not Enough: A Failure-Mode Study of GUI Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Naive Visual Memory is Not Enough: A Failure-Mode Study of GUI Agents\"",
+      "authors": "Seoyoung Choi, Minseok Ko, Hyunseok Lee, Kunwoong Kim, Woomin Song, Chanseok Jeon, Jinwoo Shin",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.14106",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-12",
+      "updated_at": "2026-06-12",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "computer-use",
+        "memory",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA",
+        "cs.CV"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-14470-gitofthoughts-version-controlled-reasoning-and-agent-memory-you-can-replay-diff-.md",
+    "title": "\"GitOfThoughts: Version-Controlled Reasoning and Agent Memory You Can Replay, Diff, and Merge\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"GitOfThoughts: Version-Controlled Reasoning and Agent Memory You Can Replay, Diff, and Merge\"",
+      "authors": "Pavan C Shekar, Abhishek H S, Aswanth Krishnan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.14470",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-12",
+      "updated_at": "2026-06-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "memory",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-14502-from-chatbot-to-digital-colleague-the-paradigm-shift-toward-persistent-autonomou.md",
+    "title": "\"From Chatbot to Digital Colleague: The Paradigm Shift Toward Persistent Autonomous AI\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"From Chatbot to Digital Colleague: The Paradigm Shift Toward Persistent Autonomous AI\"",
+      "authors": "Yongheng Zhang, Ziang Liu, Jiaxuan Zhu, Shuai Wang, Xiangqi Chen, Haojing Huang, Jiayi Kuang, Siyu Chen, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.14502",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-12",
+      "updated_at": "2026-06-12",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-14517-from-shield-to-target-denial-of-service-attacks-on-llm-based-agent-guardrails.md",
+    "title": "\"From Shield to Target: Denial-of-Service Attacks on LLM-Based Agent Guardrails\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"From Shield to Target: Denial-of-Service Attacks on LLM-Based Agent Guardrails\"",
+      "authors": "Yuguang Zhou, Xunguang Wang, Pingchuan Ma, Zhantong Xue, Zhaoyu Wang, Shuai Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.14517",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-12",
+      "updated_at": "2026-06-16",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "multi-agent",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-evaluation, autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-14571-streammembench-streaming-evaluation-of-agent-memory-for-future-oriented-assistan.md",
+    "title": "\"StreamMemBench: Streaming Evaluation of Agent Memory for Future-Oriented Assistance\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"StreamMemBench: Streaming Evaluation of Agent Memory for Future-Oriented Assistance\"",
+      "authors": "Guanming Liu, Yuqi Ren, Hansu Gu, Peng Zhang, Weihang Wang, Jiahao Liu, Ning Gu, Tun Lu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.14571",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-12",
+      "updated_at": "2026-06-12",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-14574-simmer-benchmarking-latent-failures-in-llm-executable-planning-with-a-world-mode.md",
+    "title": "\"SIMMER: Benchmarking Latent Failures in LLM Executable Planning with a World Model\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SIMMER: Benchmarking Latent Failures in LLM Executable Planning with a World Model\"",
+      "authors": "Xiaoxin Lu, Ranran Haoran Zhang, Rui Zhang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.14574",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-12",
+      "updated_at": "2026-06-12",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-14790-xflow-an-executable-protocol-programming-system-for-reliable-multi-agent-workflo.md",
+    "title": "\"XFlow: An Executable Protocol Programming System for Reliable Multi-Agent Workflows\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"XFlow: An Executable Protocol Programming System for Reliable Multi-Agent Workflows\"",
+      "authors": "Hanqi Li, Jing Peng, Zijian Wang, Lu Chen, Kai Yu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.14790",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-11",
+      "updated_at": "2026-06-11",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "coding-agent",
+        "memory",
+        "multi-agent",
+        "planning",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.PL",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-14805-knowledge-based-zero-replay-debugging-of-multi-agent-llm-traces.md",
+    "title": "Knowledge-Based Zero-Replay Debugging of Multi-Agent LLM Traces",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Knowledge-Based Zero-Replay Debugging of Multi-Agent LLM Traces",
+      "authors": "Dong Ho Kang, Hyeonjeong Cha, Daein Weon",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.14805",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-11",
+      "updated_at": "2026-06-11",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "memory",
+        "multi-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-15017-are-online-skill-and-memory-modules-always-worth-their-tokens-a-budget-constrain.md",
+    "title": "Are Online Skill and Memory Modules Always Worth Their Tokens? A Budget-Constrained Study of Web Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Are Online Skill and Memory Modules Always Worth Their Tokens? A Budget-Constrained Study of Web Agents",
+      "authors": "Sina Hajimiri, Masih Aminbeidokhti, Jose Dolz, Ismail Ben Ayed, Issam H. Laradji, Spandana Gella, Nicolas Gontier",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.15017",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-12",
+      "updated_at": "2026-06-12",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "reasoning",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-15034-osguard-a-benchmark-for-safety-in-computer-use-agents.md",
+    "title": "\"OSGuard: A Benchmark for Safety in Computer-Use Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"OSGuard: A Benchmark for Safety in Computer-Use Agents\"",
+      "authors": "Mina Mohammadmirzaei, Jeffrey Flanigan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.15034",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-13",
+      "updated_at": "2026-06-13",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-15079-ling-and-ring-2-6-technical-report-efficient-and-instant-agentic-intelligence-at.md",
+    "title": "\"Ling and Ring 2.6 Technical Report: Efficient and Instant Agentic Intelligence at Trillion-Parameter Scale\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Ling and Ring 2.6 Technical Report: Efficient and Instant Agentic Intelligence at Trillion-Parameter Scale\"",
+      "authors": "Ang Li, Ben Liu, Bin Han, Bin Hu, Bin Jing, Binbin Hu, Bing Li, Cai Chen, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.15079",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-13",
+      "updated_at": "2026-06-13",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "coding-agent",
+        "computer-use",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-15152-can-agents-read-the-room-benchmarking-visual-social-intelligence-in-multimodal-s.md",
+    "title": "Can Agents Read the Room? Benchmarking Visual Social Intelligence in Multimodal Simulation",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Can Agents Read the Room? Benchmarking Visual Social Intelligence in Multimodal Simulation",
+      "authors": "Shijun Wan, Xuehai Wu, Jiwen Zhang, Siyuan Wang, Zhongyu Wei",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.15152",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-13",
+      "updated_at": "2026-06-13",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-15242-benign-in-isolation-harmful-in-composition-security-risks-in-agent-skill-ecosyst.md",
+    "title": "\"Benign in Isolation, Harmful in Composition: Security Risks in Agent Skill Ecosystems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Benign in Isolation, Harmful in Composition: Security Risks in Agent Skill Ecosystems\"",
+      "authors": "Yi Xie, Jiawei Du, Yu Cheng, Jiuan Zhou, Zhaoxia Yin",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.15242",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-13",
+      "updated_at": "2026-06-13",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-15376-coagent-concurrency-control-for-multi-agent-systems.md",
+    "title": "\"CoAgent: Concurrency Control for Multi-Agent Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"CoAgent: Concurrency Control for Multi-Agent Systems\"",
+      "authors": "Hongtao Lyu, Dingyan Zhang, Mingyu Wu, Xingda Wei, Haibo Chen",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.15376",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-13",
+      "updated_at": "2026-06-13",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "coding-agent",
+        "multi-agent",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.DC",
+        "cs.AI",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-15591-agentic-retrieval-and-reinforcement-learned-equation-chains-a-controlled-generat.md",
+    "title": "\"Agentic Retrieval and Reinforcement Learned Equation Chains: A Controlled Generation Framework for Complex and Novel Physics Word Problems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Agentic Retrieval and Reinforcement Learned Equation Chains: A Controlled Generation Framework for Complex and Novel Physics Word Problems\"",
+      "authors": "Tirthankar Mittra",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.15591",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-14",
+      "updated_at": "2026-06-14",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-15609-fragfuse-bypassing-access-control-of-large-language-model-agents-via-memory-base.md",
+    "title": "\"FragFuse: Bypassing Access Control of Large Language Model Agents via Memory-Based Query Fragmentation and Fusion\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"FragFuse: Bypassing Access Control of Large Language Model Agents via Memory-Based Query Fragmentation and Fusion\"",
+      "authors": "Zixin Rao, Wentian Zhu, Chan Aristella Lu, Zhaorun Chen, Wei Niu, Le Guan, Bo Li, Zhen Xiang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.15609",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-14",
+      "updated_at": "2026-06-14",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-15684-multi-agent-framework-for-time-sensitive-complementary-collaboration-in-minecraf.md",
+    "title": "Multi-agent Framework for Time-Sensitive Complementary Collaboration in Minecraft",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Multi-agent Framework for Time-Sensitive Complementary Collaboration in Minecraft",
+      "authors": "Juheon Yi, Jinglu Wang, Xiaoyi Zhang, Yan Lu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.15684",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-14",
+      "updated_at": "2026-06-14",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-15709-ai-driven-framework-for-adaptive-water-network-management-with-proof-of-concept-.md",
+    "title": "\"AI-Driven Framework for Adaptive Water Network Management with Proof-of-Concept Implementation: Addressing Non-Revenue Water in Jordan\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AI-Driven Framework for Adaptive Water Network Management with Proof-of-Concept Implementation: Addressing Non-Revenue Water in Jordan\"",
+      "authors": "Mohammed Fasha, Nahel Al-Maayta, Bilal Sowan, Mohammad Athamneh, Husam Barham",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.15709",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-14",
+      "updated_at": "2026-06-14",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "rag",
+        "tool-use",
+        "workflow-agent",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "function-calling, rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-15862-retailbench-benchmarking-long-horizon-reasoning-and-coherent-decision-making-of-.md",
+    "title": "\"RetailBench: Benchmarking long horizon reasoning and coherent decision making of LLM agents in realistic retail environments\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"RetailBench: Benchmarking long horizon reasoning and coherent decision making of LLM agents in realistic retail environments\"",
+      "authors": "Linghua Zhang, Jun Wang, Jingtong Wu, Zhisong Zhang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.15862",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-14",
+      "updated_at": "2026-06-19",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "reasoning",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-15874-llm-as-code-agentic-programming-for-agent-harness.md",
+    "title": "\"LLM-as-Code: Agentic Programming for Agent Harness\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"LLM-as-Code: Agentic Programming for Agent Harness\"",
+      "authors": "Junjia Qi, Zichuan Fu, Jingtong Gao, Wenlin Zhang, Hanyu Yan, Xian Wu, Xiangyu Zhao",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.15874",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-14",
+      "updated_at": "2026-06-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-15903-control-plane-placement-shapes-forgetting-an-architectural-study-of-agent-memory.md",
+    "title": "\"Control-Plane Placement Shapes Forgetting: An Architectural Study of Agent Memory Across Thirteen System Configurations\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Control-Plane Placement Shapes Forgetting: An Architectural Study of Agent Memory Across Thirteen System Configurations\"",
+      "authors": "Dongxu Yang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.15903",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-14",
+      "updated_at": "2026-06-16",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-15906-mage-rag-multigranular-adaptive-graph-evidence-for-agentic-multimodal-rag-in-lon.md",
+    "title": "\"MAGE-RAG: Multigranular Adaptive Graph Evidence for Agentic Multimodal RAG in Long-Document QA\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MAGE-RAG: Multigranular Adaptive Graph Evidence for Agentic Multimodal RAG in Long-Document QA\"",
+      "authors": "Yilong Zuo, Xunkai Li, Jing Yuan, Qiangqiang Dai, Hongchao Qin, Ronghua Li",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.15906",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-14",
+      "updated_at": "2026-06-14",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.IR",
+        "cs.AI",
+        "cs.CL",
+        "cs.DB",
+        "cs.MM"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-15931-deeproot-a-kg-coordinated-multi-agent-system-for-therapeutic-reasoning-over-hist.md",
+    "title": "\"DeepRoot: A KG-Coordinated Multi-Agent System for Therapeutic Reasoning over Historical Medical Texts\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"DeepRoot: A KG-Coordinated Multi-Agent System for Therapeutic Reasoning over Historical Medical Texts\"",
+      "authors": "Zijian Carl Ma, Sean J. Wang, Sijbren Kramer, Li Erran Li",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.15931",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-14",
+      "updated_at": "2026-06-14",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-15994-agentic-framework-for-deep-learning-workload-migration-via-in-context-learning.md",
+    "title": "Agentic Framework for Deep Learning workload migration via In-Context Learning",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Agentic Framework for Deep Learning workload migration via In-Context Learning",
+      "authors": "Qiyue Liang, Steven Ingram, George Vanica, Andi Gavrilescu, Newfel Harrat, Hassan Sipra, Sethuraman Sankaran",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.15994",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-14",
+      "updated_at": "2026-06-14",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-16111-towards-pareto-optimal-tool-integrated-agents-with-pareto-ranking-policy-optimiz.md",
+    "title": "Towards Pareto-Optimal Tool-Integrated Agents with Pareto Ranking Policy Optimization",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Towards Pareto-Optimal Tool-Integrated Agents with Pareto Ranking Policy Optimization",
+      "authors": "Junyi Li, Xiaowei Qian, Yingyi Zhang, Wenlin Zhang, Guojing Li, Sheng Zhang, Xiao Han, Yichao Wang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.16111",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-15",
+      "updated_at": "2026-06-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "computer-use",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "language-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-16295-visualclaw-a-real-time-personalized-agent-for-the-physical-world.md",
+    "title": "\"VisualClaw: A Real-Time, Personalized Agent for the Physical World\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"VisualClaw: A Real-Time, Personalized Agent for the Physical World\"",
+      "authors": "Haoqin Tu, Jianwen Chen, Zijun Wang, Siwei Han, Juncheng Wu, Hardy Chen, Haonian Ji, Kaiwen Xiong, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.16295",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-15",
+      "updated_at": "2026-06-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "computer-use",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-evaluation, tool-use, web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-16420-transferable-self-evolving-playbooks-for-agentic-security-auditing.md",
+    "title": "Transferable Self-Evolving Playbooks for Agentic Security Auditing",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Transferable Self-Evolving Playbooks for Agentic Security Auditing",
+      "authors": "Ziyue Wang, Cheuk Wang Maurice Ng, Chenchen Yu, Strick Sheng, Kaihua Qin, Liyi Zhou",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.16420",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-15",
+      "updated_at": "2026-06-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "embodied-agent",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "agent-safety, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-16432-accord-action-conditioned-contextual-grounding-for-language-agents.md",
+    "title": "\"ACCORD: Action-Conditioned Contextual Grounding for Language Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ACCORD: Action-Conditioned Contextual Grounding for Language Agents\"",
+      "authors": "Lai Jiang, Cheng Qian, Zhenhailong Wang, Pan Lu, Heng Ji, Hao Peng",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.16432",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-15",
+      "updated_at": "2026-06-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "embodied-agent",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-16481-steering-emotional-dynamics-for-art-therapy-controllable-narrative-script-genera.md",
+    "title": "\"Steering Emotional Dynamics for Art Therapy: Controllable Narrative Script Generation through Hierarchically Guided LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Steering Emotional Dynamics for Art Therapy: Controllable Narrative Script Generation through Hierarchically Guided LLM Agents\"",
+      "authors": "Suqing Wang, Qinghai Miao, Chao Guo, Yisheng Lv",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.16481",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-15",
+      "updated_at": "2026-06-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "computer-use",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-16534-generated-parallel-scalable-a-study-of-agentic-ai-generated-julia-code-on-superc.md",
+    "title": "Generated, Parallel, Scalable? A Study of Agentic AI-Generated Julia Code on Supercomputers",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Generated, Parallel, Scalable? A Study of Agentic AI-Generated Julia Code on Supercomputers",
+      "authors": "Linus Bantel, Anna-Lena Roth, Jonas Posner, Dirk Pflüger",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.16534",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-15",
+      "updated_at": "2026-06-16",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.DC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-16576-can-llm-agents-infer-world-models-evidence-from-agentic-automata-learning.md",
+    "title": "Can LLM Agents Infer World Models? Evidence from Agentic Automata Learning",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Can LLM Agents Infer World Models? Evidence from Agentic Automata Learning",
+      "authors": "Reef Menaged, Gili Lior, Shauli Ravfogel, Roee Aharoni, Gabriel Stanovsky",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.16576",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-15",
+      "updated_at": "2026-06-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "reasoning",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-16591-sing-synthetic-intention-graph-for-scalable-active-tool-discovery-in-llm-agents.md",
+    "title": "\"SING: Synthetic Intention Graph for Scalable Active Tool Discovery in LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SING: Synthetic Intention Graph for Scalable Active Tool Discovery in LLM Agents\"",
+      "authors": "Qiao Xiao, Haochen Shi, Yisen Gao, Wenbin Hu, Huihao Jing, Tianshi Zheng, Baixuan Xu, Ziheng Zhang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.16591",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-15",
+      "updated_at": "2026-06-16",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-16613-coffeebench-benchmarking-long-horizon-llm-agents-in-heterogeneous-multi-agent-ec.md",
+    "title": "\"CoffeeBench: Benchmarking Long-Horizon LLM Agents in Heterogeneous Multi-Agent Economies\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"CoffeeBench: Benchmarking Long-Horizon LLM Agents in Heterogeneous Multi-Agent Economies\"",
+      "authors": "Issa Sugiura, Daichi Hattori, Kazuo Araragi, Keita Ogawa, Shota Onose, Taro Makino, Teppei Usuki, Takashi Ishida",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.16613",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-15",
+      "updated_at": "2026-06-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "planning",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "22",
+      "collection_queries": "autonomous-agent-llm, planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-16659-fraudsmswalker-benchmarking-agentic-large-language-models-for-sms-to-webpage-fra.md",
+    "title": "\"FraudSMSWalker: Benchmarking Agentic Large Language Models for SMS-to-Webpage Fraud Detection\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"FraudSMSWalker: Benchmarking Agentic Large Language Models for SMS-to-Webpage Fraud Detection\"",
+      "authors": "Y. H. Zhou, Z. M. Ma, Y. J. Zhou, Y. T. Li, H. X. Xiang, Y. M. Cheng, T. L. Chen, K. J. Zhang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.16659",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-15",
+      "updated_at": "2026-06-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-16748-mypcbench-a-benchmark-for-personally-intelligent-computer-use-agents.md",
+    "title": "\"MyPCBench: A Benchmark for Personally Intelligent Computer-Use Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MyPCBench: A Benchmark for Personally Intelligent Computer-Use Agents\"",
+      "authors": "Lawrence Keunho Jang, Andrew Keunwoo Jang, Jing Yu Koh, Ruslan Salakhutdinov",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.16748",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-15",
+      "updated_at": "2026-06-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-evaluation, web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-16774-openclaw-skill-collective-skill-tree-search-for-agentic-large-language-models.md",
+    "title": "\"OpenClaw-Skill: Collective Skill Tree Search for Agentic Large Language Models\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"OpenClaw-Skill: Collective Skill Tree Search for Agentic Large Language Models\"",
+      "authors": "Tianyi Lin, Chuanyu Sun, Jingyi Zhang, Changxu Wei, Huanjin Yao, Shunyu Liu, Xikun Zhang, Liu Liu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.16774",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-15",
+      "updated_at": "2026-06-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "planning-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-16802-labosbench-benchmarking-computer-use-agents-for-scientific-instrument-control.md",
+    "title": "\"LabOSBench: Benchmarking Computer Use Agents for Scientific Instrument Control\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"LabOSBench: Benchmarking Computer Use Agents for Scientific Instrument Control\"",
+      "authors": "Anqi Zou, Han Deng, Chengyu Zhang, Junquan Hu, Yu Wang, Yuxiang Xing, Aokai Zhang, Hanling Zhang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.16802",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-15",
+      "updated_at": "2026-06-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "planning",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "21",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-16813-gist-cmtf-goal-state-inference-for-causal-minimal-tool-filtering-in-llm-agents.md",
+    "title": "\"GIST-CMTF: Goal-State Inference for Causal Minimal Tool Filtering in LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"GIST-CMTF: Goal-State Inference for Causal Minimal Tool Filtering in LLM Agents\"",
+      "authors": "Rahul Suresh Babu, Rohit Shukla",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.16813",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-15",
+      "updated_at": "2026-06-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-16839-towards-llm-accelerated-rapid-reviews-for-software-tool-discovery-case-for-log-a.md",
+    "title": "Towards LLM Accelerated Rapid Reviews for Software Tool Discovery -- Case for Log Anomaly Detection",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Towards LLM Accelerated Rapid Reviews for Software Tool Discovery -- Case for Log Anomaly Detection",
+      "authors": "Jesse Nyyssölä, Hamza Bin Mazhar, Alexander Bakhtin, Matteo Esposito, Nana Reinikainen, Yuqing Wang, Ying Song, Davide Taibi, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.16839",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-15",
+      "updated_at": "2026-06-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "planning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-16871-human-on-the-bridge-scalable-evaluation-for-ai-agents.md",
+    "title": "\"Human-on-the-Bridge: Scalable Evaluation for AI Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Human-on-the-Bridge: Scalable Evaluation for AI Agents\"",
+      "authors": "Fouad Bousetouane",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.16871",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-15",
+      "updated_at": "2026-06-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-17041-benchmarking-llm-agents-on-meta-analysis-articles-from-nature-portfolio.md",
+    "title": "Benchmarking LLM Agents on Meta-Analysis Articles from Nature Portfolio",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Benchmarking LLM Agents on Meta-Analysis Articles from Nature Portfolio",
+      "authors": "Anzhe Xie, Weihang Su, Yujia Zhou, Yiqun Liu, Qingyao Ai",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.17041",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-15",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "rag",
+        "reasoning",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.IR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-17076-cmip-forge-an-agentic-system-that-retrieves-computes-and-self-reviews-climate-sc.md",
+    "title": "\"CMIP-Forge: An Agentic System that Retrieves, Computes, and Self-Reviews Climate Science\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"CMIP-Forge: An Agentic System that Retrieves, Computes, and Self-Reviews Climate Science\"",
+      "authors": "Dmitrii Pantiukhin, Boris Shapkin, Ivan Kuznetsov, Thomas Jung, Nikolay Koldunov",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.17076",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-10",
+      "updated_at": "2026-06-10",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "physics.ao-ph",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-17114-an-evaluation-of-data-leakage-risks-in-tool-using-llm-agents-in-realistic-scenar.md",
+    "title": "An Evaluation of Data Leakage Risks in Tool-Using LLM Agents in Realistic Scenarios",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "An Evaluation of Data Leakage Risks in Tool-Using LLM Agents in Realistic Scenarios",
+      "authors": "Hankyul Baek, Jaewon Noh, Sang Seo, Yongsu Kim, Gabriel Waikin Loh Matienzo, Young Il Kim, Ee Wei Seah, Akriti Vij",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.17114",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-15",
+      "updated_at": "2026-06-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "tool-use",
+        "workflow-agent",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "agent-safety, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-17246-geodisaster-benchmarking-orchestrated-agents-for-operational-disaster-geo-intell.md",
+    "title": "\"GeoDisaster: Benchmarking Orchestrated Agents for Operational Disaster Geo-Intelligence\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"GeoDisaster: Benchmarking Orchestrated Agents for Operational Disaster Geo-Intelligence\"",
+      "authors": "Maram Hasan, Aman Verma, Savitra Roy, Hariseetharam Gunduboina, Daksh Jain, Muhammad Haris Khan, Subhasis Chaudhuri, Biplab Banerjee",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.17246",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-15",
+      "updated_at": "2026-06-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-17368-distributed-general-purpose-agent-networks-architecture-key-mechanisms-and-proto.md",
+    "title": "\"Distributed General-Purpose Agent Networks: Architecture, Key Mechanisms, and Prototypes\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Distributed General-Purpose Agent Networks: Architecture, Key Mechanisms, and Prototypes\"",
+      "authors": "Shengli Zhang, Deen Ma, Zibin Lin, Taotao Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.17368",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-15",
+      "updated_at": "2026-06-15",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "computer-use",
+        "multi-agent",
+        "planning",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.NI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-17383-model-validation-of-agentic-ai-systems-a-pomdp-based-framework-for-belief-state-.md",
+    "title": "\"Model Validation of Agentic AI Systems: A POMDP-Based Framework for Belief-State, Forecast, and Policy Validation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Model Validation of Agentic AI Systems: A POMDP-Based Framework for Belief-State, Forecast, and Policy Validation\"",
+      "authors": "Matthew Francis Dixon",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.17383",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-16",
+      "updated_at": "2026-06-16",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "q-fin.RM",
+        "cs.AI",
+        "cs.LG",
+        "stat.ML"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-17449-mode-rag-manifold-outlier-diagnosis-and-energy-based-retrieval-augmented-generat.md",
+    "title": "\"MODE-RAG: Manifold Outlier Diagnosis and Energy-based Retrieval-Augmented Generation Evaluation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MODE-RAG: Manifold Outlier Diagnosis and Energy-based Retrieval-Augmented Generation Evaluation\"",
+      "authors": "Zehang Wei, Jiaxin Dai, Jiamin Yan, Xiang Xiang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.17449",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-16",
+      "updated_at": "2026-06-16",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "multi-agent",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI",
+        "cs.CV",
+        "cs.LG",
+        "cs.MM"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-17453-mapsatisfybench-benchmarking-satisfaction-aware-map-agents-through-behavior-grou.md",
+    "title": "\"MapSatisfyBench: Benchmarking Satisfaction-Aware Map Agents through Behavior-Grounded Implicit Decision Factors\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MapSatisfyBench: Benchmarking Satisfaction-Aware Map Agents through Behavior-Grounded Implicit Decision Factors\"",
+      "authors": "Lubin Bai, Mengyu Cao, Sixue Wang, Zhongwei Wan, Yue Pan, Jiale Hou, Xiang Li, Xiuyuan Zhang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.17453",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-16",
+      "updated_at": "2026-06-17",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-17459-can-llms-be-ceos-benchmarking-strategic-resource-reallocation-with-multi-role-ag.md",
+    "title": "Can LLMs Be CEOs? Benchmarking Strategic Resource Reallocation with Multi-Role Agent Simulation",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Can LLMs Be CEOs? Benchmarking Strategic Resource Reallocation with Multi-Role Agent Simulation",
+      "authors": "Yuyang Dai, Xueqing Peng, Lingfei Qian, Zhuohan Xie",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.17459",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-16",
+      "updated_at": "2026-06-16",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "multi-agent",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "22",
+      "collection_queries": "agent-evaluation, planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-17573-cordon-semantic-transactions-for-tool-using-llm-agents.md",
+    "title": "\"Cordon: Semantic Transactions for Tool-Using LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Cordon: Semantic Transactions for Tool-Using LLM Agents\"",
+      "authors": "Zheng Chen, Hanqing Liu, Duling Xu, Dong Dong, Jialin Li, Bangzheng Pu, Jidong Zhai",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.17573",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-16",
+      "updated_at": "2026-06-16",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.OS",
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-17680-envrl-learn-from-environment-dynamics-in-agentic-reinforcement-learning.md",
+    "title": "\"EnvRL: Learn from Environment Dynamics in Agentic Reinforcement Learning\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"EnvRL: Learn from Environment Dynamics in Agentic Reinforcement Learning\"",
+      "authors": "Zhitong Wang, Songze Li, Hao Peng, Shuzheng Si, Yi Wang, Maosong Sun, Juanzi Li",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.17680",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-16",
+      "updated_at": "2026-06-16",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-18023-loopcoder-v2-only-loop-once-for-efficient-test-time-computation-scaling.md",
+    "title": "\"LoopCoder-v2: Only Loop Once for Efficient Test-Time Computation Scaling\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"LoopCoder-v2: Only Loop Once for Efficient Test-Time Computation Scaling\"",
+      "authors": "Jian Yang, Shawn Guo, Wei Zhang, Tianyu Zheng, Yaxin Du, Haau-Sing Li, Jiajun Wu, Yue Song, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.18023",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-16",
+      "updated_at": "2026-06-16",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "memory",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-18037-provenanceguard-source-aware-factuality-verification-for-mcp-based-llm-agents.md",
+    "title": "\"ProvenanceGuard: Source-Aware Factuality Verification for MCP-Based LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ProvenanceGuard: Source-Aware Factuality Verification for MCP-Based LLM Agents\"",
+      "authors": "Ander Alvarez, Santhiya Rajan, Samuel Mugel, Román Orús",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.18037",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-16",
+      "updated_at": "2026-06-16",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-18051-compositional-skill-routing-for-llm-agents-decompose-retrieve-and-compose.md",
+    "title": "\"Compositional Skill Routing for LLM Agents: Decompose, Retrieve, and Compose\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Compositional Skill Routing for LLM Agents: Decompose, Retrieve, and Compose\"",
+      "authors": "Xueping Gao",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.18051",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-16",
+      "updated_at": "2026-06-16",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-18068-agentic-ai-based-framework-for-mitigating-premature-diagnostic-handoff-and-silen.md",
+    "title": "Agentic AI-based Framework for Mitigating Premature Diagnostic Handoff and Silent Hallucination in Healthcare Applications",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Agentic AI-based Framework for Mitigating Premature Diagnostic Handoff and Silent Hallucination in Healthcare Applications",
+      "authors": "Divyansh Srivastava, Shreya Ghosh, Anshul Verma, Rajkumar Buyya",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.18068",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-16",
+      "updated_at": "2026-06-16",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-18142-your-ai-travel-agent-would-book-you-a-bullfight-an-agentic-benchmark-for-implici.md",
+    "title": "\"Your AI Travel Agent Would Book You a Bullfight: An Agentic Benchmark for Implicit Animal Welfare in Frontier AI Models\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Your AI Travel Agent Would Book You a Bullfight: An Agentic Benchmark for Implicit Animal Welfare in Frontier AI Models\"",
+      "authors": "Jasmine Brazilek, Joel Christoph, Maheep Chaudhary, Oliver Tullio, Carol Kline, Miles Tidmarsh, Arturs Kanepajs",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.18142",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-16",
+      "updated_at": "2026-07-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL",
+        "cs.CY"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-18272-mitigating-anchoring-bias-in-llm-based-agents-for-energy-efficient-6g-autonomous.md",
+    "title": "Mitigating Anchoring Bias in LLM-Based Agents for Energy-Efficient 6G Autonomous Networks",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Mitigating Anchoring Bias in LLM-Based Agents for Energy-Efficient 6G Autonomous Networks",
+      "authors": "Hatim Chergui, Claudia Carballo González, Farhad Rezazadeh, Merouane Debbah",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.18272",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-05",
+      "updated_at": "2026-06-18",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "computer-use",
+        "multi-agent",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.NI",
+        "cs.AI",
+        "eess.SY"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-18356-safeclawbench-separating-semantic-audit-evidence-and-sandbox-harm-in-tool-using-.md",
+    "title": "\"SafeClawBench: Separating Semantic, Audit-Evidence, and Sandbox Harm in Tool-Using LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SafeClawBench: Separating Semantic, Audit-Evidence, and Sandbox Harm in Tool-Using LLM Agents\"",
+      "authors": "Yuchuan Tian, Mengyu Zheng, Haocheng Mei, Ye Yuan, Chao Xu, Xinghao Chen, Hanting Chen, Yu Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.18356",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-16",
+      "updated_at": "2026-06-16",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "memory",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agent-safety, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-18363-guava-an-effective-and-universal-harness-for-embodied-manipulation.md",
+    "title": "\"Guava: An Effective and Universal Harness for Embodied Manipulation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Guava: An Effective and Universal Harness for Embodied Manipulation\"",
+      "authors": "Haowen Liu, Xirui Li, Shaoxiong Yao, Peng Shi, Tianyi Zhou, Jia-Bin Huang, Furong Huang, Jiayuan Mao",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.18363",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-16",
+      "updated_at": "2026-06-16",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "embodied-agent",
+        "planning",
+        "reasoning",
+        "tool-use",
+        "workflow-agent",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.RO",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agentic-ai, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-18406-coremem-riemannian-retrieval-and-fisher-guided-distillation-for-long-term-memory.md",
+    "title": "\"CoreMem: Riemannian Retrieval and Fisher-Guided Distillation for Long-Term Memory in Dialogue Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"CoreMem: Riemannian Retrieval and Fisher-Guided Distillation for Long-Term Memory in Dialogue Agents\"",
+      "authors": "Jiaqi Chen, Yongqin Zeng, Shaoshen Chen, Yijian Zhang, Hai-Tao Zheng, Chunxia Ma, XiuTeng Zhou",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.18406",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-16",
+      "updated_at": "2026-06-16",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-18467-toolchain-crc-conformal-risk-control-for-agentic-ai-under-retrieval-and-tool-use.md",
+    "title": "\"ToolChain-CRC: Conformal Risk Control for Agentic AI Under Retrieval and Tool-Use Drift\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ToolChain-CRC: Conformal Risk Control for Agentic AI Under Retrieval and Tool-Use Drift\"",
+      "authors": "Jeffery Opoku, David Banahene",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.18467",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-16",
+      "updated_at": "2026-06-16",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "stat.ML",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-evaluation, agentic-ai, ai-agent, rag-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-18502-towards-scalable-customization-and-deployment-of-multi-agent-systems-for-enterpr.md",
+    "title": "Towards Scalable Customization and Deployment of Multi-Agent Systems for Enterprise Applications",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Towards Scalable Customization and Deployment of Multi-Agent Systems for Enterprise Applications",
+      "authors": "Paresh Dashore, Shreyas Kulkarni, Uttam Gurram, Nadia Bathaee, Kartik Balasubramaniam, Genta Indra Winata, Sambit Sahu, Shi-Xiong Zhang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.18502",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-16",
+      "updated_at": "2026-06-16",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "coding-agent",
+        "multi-agent",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-18619-code-augur-agentic-vulnerability-detection-via-specification-inference.md",
+    "title": "\"Code-Augur: Agentic Vulnerability Detection via Specification Inference\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Code-Augur: Agentic Vulnerability Detection via Specification Inference\"",
+      "authors": "Zhengxiong Luo, Mehtab Zafar, Dylan Wolff, Abhik Roychoudhury",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.18619",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-17",
+      "updated_at": "2026-06-17",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "computer-use",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-18671-hansel-extracting-breadcrumbs-from-web-agent-trajectories-for-interactive-verifi.md",
+    "title": "\"HANSEL: Extracting Breadcrumbs from Web Agent Trajectories for Interactive Verification\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"HANSEL: Extracting Breadcrumbs from Web Agent Trajectories for Interactive Verification\"",
+      "authors": "Yujin Zhang, Daye Nam",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.18671",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-17",
+      "updated_at": "2026-06-17",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "embodied-agent",
+        "planning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.HC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "ai-agent, web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-18789-poweragentbench-ss-a-benchmark-for-agentic-ai-in-power-system-steady-state-studi.md",
+    "title": "\"PowerAgentBench-SS: A Benchmark for Agentic AI in Power System Steady-State Studies\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"PowerAgentBench-SS: A Benchmark for Agentic AI in Power System Steady-State Studies\"",
+      "authors": "Costas Mylonas, Magda Foti, Andrea Pomarico, Matheus Duarte, Qian Zhang, Emmanouel Varvarigos",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.18789",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-17",
+      "updated_at": "2026-06-17",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "planning",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "eess.SY"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "24",
+      "collection_queries": "agentic-ai, planning-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-18829-gatemem-benchmarking-memory-governance-in-multi-principal-shared-memory-agents.md",
+    "title": "\"GateMem: Benchmarking Memory Governance in Multi-Principal Shared-Memory Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"GateMem: Benchmarking Memory Governance in Multi-Principal Shared-Memory Agents\"",
+      "authors": "Zhe Ren, Yibo Yang, Yimeng Chen, Zijun Zhao, Benshuo Fu, Zhihao Shu, Bingjie Zhang, Yangyang Xu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.18829",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-17",
+      "updated_at": "2026-06-17",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning",
+        "rag",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-18950-rtsgamebench-an-rts-benchmark-for-strategic-reasoning-by-vision-language-models.md",
+    "title": "\"RTSGameBench: An RTS Benchmark for Strategic Reasoning by Vision-Language Models\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"RTSGameBench: An RTS Benchmark for Strategic Reasoning by Vision-Language Models\"",
+      "authors": "San Kim, Daechul Ahn, Reokyoung Kim, Hyeonbeom Choi, Seungyeon Jwa, Jonghyun Choi",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.18950",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-17",
+      "updated_at": "2026-06-18",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "multi-agent",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-19063-pypiline-malicious-pypi-package-detection-via-suspicious-api-knowledge-and-agent.md",
+    "title": "\"PYPILINE: Malicious PyPI Package Detection via Suspicious API Knowledge and Agent Workflow\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"PYPILINE: Malicious PyPI Package Detection via Suspicious API Knowledge and Agent Workflow\"",
+      "authors": "Siyuan Pang, Yepeng Yao, Zhengwei Jiang, Zijing Fan, Haozhe Li, Baoxu Liu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.19063",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-17",
+      "updated_at": "2026-06-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agentic-ai, rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-19242-runtime-compliance-verification-for-ai-agents.md",
+    "title": "Runtime Compliance Verification for AI Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Runtime Compliance Verification for AI Agents",
+      "authors": "Nafiseh Kahani, Masoud Barati, Diana Addae",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.19242",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-17",
+      "updated_at": "2026-06-17",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "ai-agent, function-calling, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-19245-txbench-pp-analyzing-ai-agent-performance-on-small-molecule-preclinical-pharmaco.md",
+    "title": "\"TxBench-PP: Analyzing AI Agent Performance on Small-Molecule Preclinical Pharmacology\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"TxBench-PP: Analyzing AI Agent Performance on Small-Molecule Preclinical Pharmacology\"",
+      "authors": "Hannah Le, Ramesh Ramasamy, Alex Urrutia, Mahsa Yazdani, Tim Proctor, Kenny Workman",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.19245",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-17",
+      "updated_at": "2026-06-18",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-19409-openrath-session-centered-runtime-state-for-agent-systems.md",
+    "title": "\"OpenRath: Session-Centered Runtime State for Agent Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"OpenRath: Session-Centered Runtime State for Agent Systems\"",
+      "authors": "Fukang Wen, Zhijie Wang, Ruilin Xu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.19409",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-17",
+      "updated_at": "2026-06-17",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "multi-agent",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.PL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-19464-deontic-policies-for-runtime-governance-of-agentic-ai-systems.md",
+    "title": "Deontic Policies for Runtime Governance of Agentic AI Systems",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Deontic Policies for Runtime Governance of Agentic AI Systems",
+      "authors": "Anupam Joshi, Tim Finin, Karuna Pande Joshi, Lalana Kagal",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.19464",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-17",
+      "updated_at": "2026-06-17",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agentic-ai, autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-19613-staminabench-stress-testing-coding-agents-over-100-interaction-turns.md",
+    "title": "\"StaminaBench: Stress-Testing Coding Agents over 100 Interaction Turns\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"StaminaBench: Stress-Testing Coding Agents over 100 Interaction Turns\"",
+      "authors": "Vlad Sobal, Shuo Yang, Yuting Zhang, Wei Xia, Stefano Soatto",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.19613",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-17",
+      "updated_at": "2026-06-17",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-19704-beyond-static-leaderboards-predictive-validity-for-the-evaluation-of-llm-agents.md",
+    "title": "\"Beyond Static Leaderboards: Predictive Validity for the Evaluation of LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Beyond Static Leaderboards: Predictive Validity for the Evaluation of LLM Agents\"",
+      "authors": "Dhaval C. Patel, Kaoutar El Maghraoui, Shuxin Lin, Yusheng Li, Tianjun Feng, Chun-Yi Tsai, Yihan Sun, Wei Alexander Xin, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.19704",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-18",
+      "updated_at": "2026-06-18",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-19787-oragentbench-can-llm-agents-solve-challenging-operations-research-tasks-end-to-e.md",
+    "title": "\"ORAgentBench: Can LLM Agents Solve Challenging Operations Research Tasks End to End?\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ORAgentBench: Can LLM Agents Solve Challenging Operations Research Tasks End to End?\"",
+      "authors": "Jiajun Li, Mingshu Cai, Yixuan Li, Yu Ding, Ran Hou, Guanyu Nie, Xiongwei Han, Wanyuan Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.19787",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-18",
+      "updated_at": "2026-06-18",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-19812-human-on-the-loop-orchestration-for-ai-assisted-legal-discovery.md",
+    "title": "Human-on-the-Loop Orchestration for AI-Assisted Legal Discovery",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Human-on-the-Loop Orchestration for AI-Assisted Legal Discovery",
+      "authors": "Anushree Sinha, Srivaths Ranganathan, Abhishek Dharmaratnakar, Debanshu Das",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.19812",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-18",
+      "updated_at": "2026-06-18",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "rag",
+        "reasoning",
+        "workflow-agent",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agentic-ai, planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-19852-prompt-plan-extract-zero-shot-agentic-llms-workflows-for-lung-pathology-extracti.md",
+    "title": "\"Prompt, Plan, Extract: Zero-Shot Agentic LLMs Workflows for Lung Pathology Extraction from Clinical Narratives\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Prompt, Plan, Extract: Zero-Shot Agentic LLMs Workflows for Lung Pathology Extraction from Clinical Narratives\"",
+      "authors": "Aman Pathak, Cheng Peng, Mengxian Lyu, Ziyi Chen, Reema Solan, Sankalp Talankar, Yasir Khan, Hiren Mehta, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.19852",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-18",
+      "updated_at": "2026-06-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-19899-measuring-biological-capabilities-and-risks-of-ai-agents.md",
+    "title": "Measuring Biological Capabilities and Risks of AI Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Measuring Biological Capabilities and Risks of AI Agents",
+      "authors": "Patricia Paskov, Jeffrey Lee, Kyle Brady, Alyssa Worland",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.19899",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-18",
+      "updated_at": "2026-06-18",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CY",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "agent-evaluation, agentic-ai, ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-19926-memgui-agent-an-end-to-end-long-horizon-mobile-gui-agent-with-proactive-context-.md",
+    "title": "\"MemGUI-Agent: An End-to-End Long-Horizon Mobile GUI Agent with Proactive Context Management\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MemGUI-Agent: An End-to-End Long-Horizon Mobile GUI Agent with Proactive Context Management\"",
+      "authors": "Guangyi Liu, Gao Wu, Congxiao Liu, Pengxiang Zhao, Liang Liu, Mading Li, Qi Zhang, Mengyan Wang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.19926",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-18",
+      "updated_at": "2026-06-18",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.HC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-19930-mobileforge-annotation-free-adaptation-for-mobile-gui-agents-with-hierarchical-f.md",
+    "title": "\"MobileForge: Annotation-Free Adaptation for Mobile GUI Agents with Hierarchical Feedback-Guided Policy Optimization\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MobileForge: Annotation-Free Adaptation for Mobile GUI Agents with Hierarchical Feedback-Guided Policy Optimization\"",
+      "authors": "Guangyi Liu, Pengxiang Zhao, Gao Wu, Yiwen Yin, Mading Li, Liang Liu, Congxiao Liu, Zhang Qi, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.19930",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-18",
+      "updated_at": "2026-06-18",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.HC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-19980-enpire-agentic-robot-policy-self-improvement-in-the-real-world.md",
+    "title": "\"ENPIRE: Agentic Robot Policy Self-Improvement in the Real World\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ENPIRE: Agentic Robot Policy Self-Improvement in the Real World\"",
+      "authors": "\"Wenli Xiao, Jia Xie, Tonghe Zhang, Haotian Lin, Letian \\\"Max\\\" Fu, Haoru Xue, Jalen Lu, Yi Yang, et al.\"",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.19980",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-18",
+      "updated_at": "2026-06-18",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "embodied-agent",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "coding-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-20023-when-lower-privileges-suffice-investigating-over-privileged-tool-selection-in-ll.md",
+    "title": "\"When Lower Privileges Suffice: Investigating Over-Privileged Tool Selection in LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"When Lower Privileges Suffice: Investigating Over-Privileged Tool Selection in LLM Agents\"",
+      "authors": "Kaiyue Yang, Yuyan Bu, Jingwei Yi, Yuchi Wang, Biyu Zhou, Juntao Dai, Songlin Hu, Yaodong Yang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.20023",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-18",
+      "updated_at": "2026-07-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-20041-ai-economist-agent-an-agentic-framework-for-model-grounded-economic-analysis-wit.md",
+    "title": "\"AI Economist Agent: An Agentic Framework for Model-Grounded Economic Analysis with RAG, Knowledge Graphs, and Large Language Models\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AI Economist Agent: An Agentic Framework for Model-Grounded Economic Analysis with RAG, Knowledge Graphs, and Large Language Models\"",
+      "authors": "Masahiro Kato",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.20041",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-18",
+      "updated_at": "2026-06-18",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "econ.GN",
+        "cs.AI",
+        "cs.LG",
+        "q-fin.GN"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "ai-agent, rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-20047-pacms-submodular-context-selection-as-a-pluggable-engine-for-llm-agents.md",
+    "title": "\"PACMS: Submodular Context Selection as a Pluggable Engine for LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"PACMS: Submodular Context Selection as a Pluggable Engine for LLM Agents\"",
+      "authors": "Manu Ghulyani, Arunabh Singh, Karan Bharadwaj, Ankit Nath, Suranjan Goswami",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.20047",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-18",
+      "updated_at": "2026-06-18",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.IR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-20243-phoenix-safe-github-issue-resolution-via-multi-agent-llms.md",
+    "title": "\"Phoenix: Safe GitHub Issue Resolution via Multi-Agent LLMs\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Phoenix: Safe GitHub Issue Resolution via Multi-Agent LLMs\"",
+      "authors": "Kipngeno Koech, Muhammad Adam, Baimam Boukar Jean Jacques, Joao Barros",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.20243",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-18",
+      "updated_at": "2026-06-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "multi-agent",
+        "planning",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-20401-poweragentbench-dyn-a-benchmark-for-agentic-ai-in-power-system-dynamic-studies.md",
+    "title": "\"PowerAgentBench-Dyn: A Benchmark for Agentic AI in Power System Dynamic Studies\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"PowerAgentBench-Dyn: A Benchmark for Agentic AI in Power System Dynamic Studies\"",
+      "authors": "Qian Zhang, Andrea Pomarico, Costas Mylonas, Magda Foti, Alberto Berizzi, Le Xie",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.20401",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-18",
+      "updated_at": "2026-06-18",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "memory",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "eess.SY"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "24",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-20470-analyzing-defensive-misdirection-against-model-guided-automated-attacks-on-agent.md",
+    "title": "Analyzing Defensive Misdirection Against Model-Guided Automated Attacks on Agentic AI Systems",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Analyzing Defensive Misdirection Against Model-Guided Automated Attacks on Agentic AI Systems",
+      "authors": "Reza Soosahabi, Vivek Namsani",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.20470",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-18",
+      "updated_at": "2026-06-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-20479-groundcontrol-anticipating-navigation-failures-in-vision-language-agents-via-tra.md",
+    "title": "\"GroundControl: Anticipating Navigation Failures in Vision-Language Agents via Trajectory-Consistent Uncertainty Estimates\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"GroundControl: Anticipating Navigation Failures in Vision-Language Agents via Trajectory-Consistent Uncertainty Estimates\"",
+      "authors": "Nastaran Darabi, Divake Kumar, Sina Tayebati, Devashri Naik, Amit Ranjan Trivedi",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.20479",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-18",
+      "updated_at": "2026-06-18",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "embodied-agent",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.RO"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-20510-efficient-and-sound-probabilistic-verification-for-ai-agents.md",
+    "title": "Efficient and Sound Probabilistic Verification for AI Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Efficient and Sound Probabilistic Verification for AI Agents",
+      "authors": "Alaia Solko-Breslin, Pramod Kaushik Mudrakarta, Mihai Christodorescu, Somesh Jha, Krishnamurthy Dj Dvijotham",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.20510",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-18",
+      "updated_at": "2026-06-18",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-20512-probe-and-refine-tuning-of-repository-guidance-for-coding-agents.md",
+    "title": "Probe-and-Refine Tuning of Repository Guidance for Coding Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Probe-and-Refine Tuning of Repository Guidance for Coding Agents",
+      "authors": "Asa Shepard, Jeannie Albrecht",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.20512",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-18",
+      "updated_at": "2026-06-19",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "computer-use",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "coding-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-20515-s-agent-spatial-tool-use-elicits-reasoning-for-spatial-intelligence.md",
+    "title": "\"S-Agent: Spatial Tool-Use Elicits Reasoning for Spatial Intelligence\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"S-Agent: Spatial Tool-Use Elicits Reasoning for Spatial Intelligence\"",
+      "authors": "Yalun Dai, Hao Li, Shulin Tian, Runmao Yao, Yuhao Dong, Fangzhou Hong, Zhaoxi Chen, Fangfu Liu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.20515",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-18",
+      "updated_at": "2026-06-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-memory, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-20573-aona-a-comprehensive-architecture-and-workflow-design-for-global-agentic-collabo.md",
+    "title": "\"AONA: A Comprehensive Architecture and Workflow Design for Global Agentic Collaboration\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AONA: A Comprehensive Architecture and Workflow Design for Global Agentic Collaboration\"",
+      "authors": "Jinliang Xu, Runkai Zhu, Bingqi Li, Fanjie Nie, Jin Li, Jiagui Xie",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.20573",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-04-30",
+      "updated_at": "2026-04-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "multi-agent",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.NI",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-20629-specialize-roles-mix-deployments-pushing-the-cost-accuracy-frontier-of-llm-agent.md",
+    "title": "\"Specialize Roles, Mix Deployments: Pushing the Cost-Accuracy Frontier of LLM Agent Teams\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Specialize Roles, Mix Deployments: Pushing the Cost-Accuracy Frontier of LLM Agent Teams\"",
+      "authors": "Yinsicheng Jiang, Liang Cheng, Yeqi Huang, Yufan Zhao, Zhan Lu, Li Dong, Wenda Li, Edoardo Ponti, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.20629",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-05-28",
+      "updated_at": "2026-05-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA",
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-20717-mirage-stealthy-visual-prompt-injection-for-vulnerability-detection-in-web-agent.md",
+    "title": "\"MIRAGE: Stealthy Visual Prompt Injection for Vulnerability Detection in Web Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MIRAGE: Stealthy Visual Prompt Injection for Vulnerability Detection in Web Agents\"",
+      "authors": "Xuelong Dai, Jianyu Ma, Boyang Ma, Biwei Yan, Yijun Yang, Yue Zhang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.20717",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-16",
+      "updated_at": "2026-06-16",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV",
+        "cs.AI",
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-20785-fara-1-5-scalable-learning-environments-for-computer-use-agents.md",
+    "title": "\"Fara-1.5: Scalable Learning Environments for Computer Use Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Fara-1.5: Scalable Learning Environments for Computer Use Agents\"",
+      "authors": "Ahmed Awadallah, Sahil Gupta, Yash Lara, Yadong Lu, Hussein Mozannar, Akshay Nambi, Zach Nussbaum, Yash Pandya, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.20785",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-18",
+      "updated_at": "2026-06-18",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-20922-think-twice-before-you-act-protecting-llm-agents-against-tool-description-poison.md",
+    "title": "\"Think Twice Before You Act: Protecting LLM Agents Against Tool Description Poisoning via Isolated Planning\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Think Twice Before You Act: Protecting LLM Agents Against Tool Description Poisoning via Isolated Planning\"",
+      "authors": "Shanghao Shi, Xiao Wang, Chaoyu Zhang, Hao Li, Wenjing Lou, Thomas Hou, Yevgeniy Vorobeychik, Chongjie Zhang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.20922",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-18",
+      "updated_at": "2026-06-18",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-20950-power-systems-agent-benchmark-executable-evaluation-of-ai-agents-in-electric-pow.md",
+    "title": "\"Power Systems Agent Benchmark: Executable Evaluation of AI Agents in Electric Power Engineering\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Power Systems Agent Benchmark: Executable Evaluation of AI Agents in Electric Power Engineering\"",
+      "authors": "Sergei Trashchenkov",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.20950",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-18",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "eess.SY"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agent-evaluation, ai-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-20954-learning-what-not-to-forget-long-horizon-agent-memory-from-a-few-kilobytes-of-le.md",
+    "title": "\"Learning What Not to Forget: Long-Horizon Agent Memory from a Few Kilobytes of Learning\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Learning What Not to Forget: Long-Horizon Agent Memory from a Few Kilobytes of Learning\"",
+      "authors": "Nusrat Jahan Lia, Aritra Mazumder",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.20954",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-18",
+      "updated_at": "2026-06-18",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-21013-agentic-time-machine-as-an-infrastructure-for-future-event-forecasting.md",
+    "title": "Agentic Time Machine as an Infrastructure for Future-Event Forecasting",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Agentic Time Machine as an Infrastructure for Future-Event Forecasting",
+      "authors": "Jingyi Chai, Bingyang Zheng, Xiangrui Liu, Hao Lu, Zihang Zhou, Tianchen Wang, Kemeng Zhang, Siheng Chen",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.21013",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-19",
+      "updated_at": "2026-06-19",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-21123-a-multi-agent-audit-framework-for-high-stakes-reasoning-evaluation-and-interpret.md",
+    "title": "\"A Multi-Agent Audit Framework for High-Stakes Reasoning: Evaluation and Interpretability in Clinical Mental Health Screening\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"A Multi-Agent Audit Framework for High-Stakes Reasoning: Evaluation and Interpretability in Clinical Mental Health Screening\"",
+      "authors": "Jingchen Ye, Yanpei Yu, Luyao Zhang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.21123",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-19",
+      "updated_at": "2026-06-19",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "rag",
+        "reasoning",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-21129-agenticos-an-intent-oriented-secure-operating-system-architecture-for-autonomous.md",
+    "title": "\"AgenticOS: An Intent-Oriented Secure Operating System Architecture for Autonomous AI Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AgenticOS: An Intent-Oriented Secure Operating System Architecture for Autonomous AI Agents\"",
+      "authors": "Zhen Zhao, Yu Zhang, Yanpeng Zhu, Jia Wang, Songqiao Tao, Xin Cheng, Jiexin Gao",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.21129",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-19",
+      "updated_at": "2026-06-19",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.OS"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "ai-agent, autonomous-agent-llm, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-21228-sakana-fugu-technical-report.md",
+    "title": "Sakana Fugu Technical Report",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Sakana Fugu Technical Report",
+      "authors": "Yujin Tang, Edoardo Cetin, Jinglue Xu, Qi Sun, Stefan Nielsen, Vincent Richard, Haruto Goda, Iaroslav Tymchenko, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.21228",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-19",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "coding-agent",
+        "multi-agent",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-21401-swarmx-agentic-scheduling-for-low-latency-agentic-systems.md",
+    "title": "\"SwarmX: Agentic Scheduling for Low-Latency Agentic Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SwarmX: Agentic Scheduling for Low-Latency Agentic Systems\"",
+      "authors": "Yeqi Huang, Yanwei Ye, Guomin Chen, Wenhao Su, Bin Gong, Jialian Li, Zhan Lu, Yangshen Deng, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.21401",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-19",
+      "updated_at": "2026-06-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.DC",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-21409-don-t-blindly-trust-it-how-unreliable-feedback-breaks-tool-using-llm-agents.md",
+    "title": "\"Don't Blindly Trust It: How Unreliable Feedback Breaks Tool-Using LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Don't Blindly Trust It: How Unreliable Feedback Breaks Tool-Using LLM Agents\"",
+      "authors": "Chubin Zhang, Zhenglin Wan, Xingrui Yu, Pengfei Zhou, Wangbo Zhao, Jingxuan Wu, Yaxin Zhou, Ivor Tsang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.21409",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-19",
+      "updated_at": "2026-06-19",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-21445-autoras-learning-robust-agentic-systems-with-primitive-representations.md",
+    "title": "\"AutoRAS: Learning Robust Agentic Systems with Primitive Representations\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AutoRAS: Learning Robust Agentic Systems with Primitive Representations\"",
+      "authors": "Yang Yue, Xuancheng Zhu, Yuyang Ma, Guoshun Nan, Zihan Dou, Jingru Shan, Congyu Guo, Ji Zhang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.21445",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-19",
+      "updated_at": "2026-06-19",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "multi-agent",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-21553-dissecting-agentic-rag-a-component-ablation-for-multi-hop-qa-with-a-local-7b-mod.md",
+    "title": "\"Dissecting Agentic RAG: A Component Ablation for Multi-Hop QA with a Local 7B Model\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Dissecting Agentic RAG: A Component Ablation for Multi-Hop QA with a Local 7B Model\"",
+      "authors": "Sheroz Shaikh",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.21553",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-19",
+      "updated_at": "2026-06-19",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.IR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-21565-composing-verifiable-conceptual-models-via-building-blocks-towards-design-time-v.md",
+    "title": "\"Composing Verifiable Conceptual Models via Building Blocks: Towards Design-Time Verification of Agentic AI Workflows\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Composing Verifiable Conceptual Models via Building Blocks: Towards Design-Time Verification of Agentic AI Workflows\"",
+      "authors": "Noe Y. Flandre, Alexander C. Nwala, Philippe J. Giabbanelli",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.21565",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-19",
+      "updated_at": "2026-06-19",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "reasoning",
+        "tool-use",
+        "workflow-agent",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-21627-counsel-a-meta-evaluation-dataset-for-agentic-tasks.md",
+    "title": "\"Counsel: A Meta-Evaluation Dataset for Agentic Tasks\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Counsel: A Meta-Evaluation Dataset for Agentic Tasks\"",
+      "authors": "Sashank Pisupati, Henry Broomfield, Eujeong Choi, Antonia Calvi, Charlie Wang, Roman Engeler, Max Bartolo, Patrick Lewis",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.21627",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-19",
+      "updated_at": "2026-06-19",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "agent-evaluation, coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-21649-evoembedding-evolvable-representations-for-long-context-retrieval-and-agentic-me.md",
+    "title": "\"EvoEmbedding: Evolvable Representations for Long-Context Retrieval and Agentic Memory\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"EvoEmbedding: Evolvable Representations for Long-Context Retrieval and Agentic Memory\"",
+      "authors": "Chang Nie, Chaoyou Fu, Junlan Feng, Caifeng Shan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.21649",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-19",
+      "updated_at": "2026-06-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "memory",
+        "rag",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agent-memory, agentic-ai, rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-21710-privacyalign-contextual-privacy-alignment-for-llm-agents.md",
+    "title": "\"PrivacyAlign: Contextual Privacy Alignment for LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"PrivacyAlign: Contextual Privacy Alignment for LLM Agents\"",
+      "authors": "Manveer Singh Tamber, Abhay Puri, Marc-Etienne Brunet, Perouz Taslakian, Jimmy Lin, Spandana Gella",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.21710",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-19",
+      "updated_at": "2026-06-19",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI",
+        "cs.IR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-21732-safe-to-check-unsafe-to-use-relinking-at-the-compression-boundary-of-llm-agents.md",
+    "title": "\"Safe to Check, Unsafe to Use: Relinking at the Compression Boundary of LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Safe to Check, Unsafe to Use: Relinking at the Compression Boundary of LLM Agents\"",
+      "authors": "Zesen Liu, Zihan Zhang, Dongdong She",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.21732",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-19",
+      "updated_at": "2026-06-19",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-21740-training-the-orchestrator-a-supervised-approach-to-end-to-end-pddl-planning-with.md",
+    "title": "\"Training the Orchestrator: A Supervised Approach to End-to-End PDDL Planning with LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Training the Orchestrator: A Supervised Approach to End-to-End PDDL Planning with LLM Agents\"",
+      "authors": "Rajesh Mangannavar, Zachary Coalson, Pranay Dugar, Prasad Tadepalli",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.21740",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-19",
+      "updated_at": "2026-06-19",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-21836-agentdse-reasoning-augmented-architectural-design-space-exploration.md",
+    "title": "\"AgentDSE: Reasoning-Augmented Architectural Design Space Exploration\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AgentDSE: Reasoning-Augmented Architectural Design Space Exploration\"",
+      "authors": "Chenyu Wang, Jiahe Caroline Shi, David Kong, Duane S. Boning, Zishen Wan, Yilun Du, Vijay Janapa Reddi",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.21836",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-20",
+      "updated_at": "2026-06-20",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "computer-use",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-21842-agent-assisted-side-channel-attacks-on-non-prefix-kv-cache-in-rag.md",
+    "title": "Agent-Assisted Side-Channel Attacks on Non-Prefix KV Cache in RAG",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Agent-Assisted Side-Channel Attacks on Non-Prefix KV Cache in RAG",
+      "authors": "He Sun, Shinan Liu, Siyuan Ma, Junhao Li, Mingjun Xiao, Wenhao Jiang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.21842",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-20",
+      "updated_at": "2026-06-20",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-21877-agentriskbom-a-risk-scoping-security-bill-of-materials-for-agentic-ai-systems.md",
+    "title": "\"AgentRiskBOM: A Risk-Scoping Security Bill of Materials for Agentic AI Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AgentRiskBOM: A Risk-Scoping Security Bill of Materials for Agentic AI Systems\"",
+      "authors": "Srimonti Dutta, Akshata Kishore Moharir",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.21877",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-20",
+      "updated_at": "2026-06-20",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "memory",
+        "multi-agent",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CR",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "21",
+      "collection_queries": "agentic-ai, ai-agent, rag-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-21963-holmes-multimodal-agentic-diagnosis-for-mixed-language-mobile-crashes-at-industr.md",
+    "title": "\"Holmes: Multimodal Agentic Diagnosis for Mixed-Language Mobile Crashes at Industrial Scale\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Holmes: Multimodal Agentic Diagnosis for Mixed-Language Mobile Crashes at Industrial Scale\"",
+      "authors": "Jia Li, Wenyuan Ma, Ting Peng, Haibin Zheng, Yuetang Deng",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.21963",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-20",
+      "updated_at": "2026-06-20",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "multi-agent",
+        "rag",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-22030-nous-a-predictive-world-model-for-long-term-agent-memory.md",
+    "title": "\"Nous: A Predictive World Model for Long-Term Agent Memory\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Nous: A Predictive World Model for Long-Term Agent Memory\"",
+      "authors": "Pranav Singh",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.22030",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-20",
+      "updated_at": "2026-06-20",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL",
+        "cs.IR",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-22082-codeteam-an-llm-powered-multi-agent-framework-for-repository-level-code-generati.md",
+    "title": "\"CodeTeam: An LLM-Powered Multi-Agent Framework for Repository-Level Code Generation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"CodeTeam: An LLM-Powered Multi-Agent Framework for Repository-Level Code Generation\"",
+      "authors": "Yifei Wang, Ruiyin Li, Peng Liang, Qiong Feng, Zengyang Li, Mojtaba Shahin, Arif Ali Khan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.22082",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-20",
+      "updated_at": "2026-06-20",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "multi-agent",
+        "planning",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-22110-traceview-interactive-visualization-of-agentic-program-repair-trajectories.md",
+    "title": "\"TraceView: Interactive Visualization of Agentic Program Repair Trajectories\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"TraceView: Interactive Visualization of Agentic Program Repair Trajectories\"",
+      "authors": "Amirali Sajadi, Tu Nguyen, Kimmie Huynh, Esteban Parra, Preetha Chatterjee",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.22110",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-20",
+      "updated_at": "2026-06-20",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI",
+        "cs.HC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-22151-novelty-aware-agentic-retrieval-comparing-research-contributions-through-structu.md",
+    "title": "\"Novelty-Aware Agentic Retrieval: Comparing Research Contributions Through Structured Multi-Step Reasoning\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Novelty-Aware Agentic Retrieval: Comparing Research Contributions Through Structured Multi-Step Reasoning\"",
+      "authors": "Shou-Tzu Han",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.22151",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-20",
+      "updated_at": "2026-06-20",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.IR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-22263-revelio-cost-efficient-agentic-memory-safety-vulnerability-detection-for-reposit.md",
+    "title": "\"Revelio: Cost-Efficient Agentic Memory Safety Vulnerability Detection For Repository-Scale Codebases\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Revelio: Cost-Efficient Agentic Memory Safety Vulnerability Detection For Repository-Scale Codebases\"",
+      "authors": "Yiwei Hou, Hao Wang, Muxi Lyu, Marius Momeu, Eric Nguyen, Taige Yang, Koushik Sen, Dawn Song, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.22263",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-20",
+      "updated_at": "2026-06-20",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "memory"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI",
+        "cs.MA",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "agent-memory, coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-22330-hypothesis-driven-skill-optimization-for-llm-agents.md",
+    "title": "Hypothesis-Driven Skill Optimization for LLM Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Hypothesis-Driven Skill Optimization for LLM Agents",
+      "authors": "Fangxin Shang, Yehui Yang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.22330",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-21",
+      "updated_at": "2026-06-21",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "coding-agent",
+        "memory",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-22388-planbench-xl-evaluating-long-horizon-planning-of-llm-tool-use-agents-in-large-sc.md",
+    "title": "\"PlanBench-XL: Evaluating Long-Horizon Planning of LLM Tool-Use Agents in Large-Scale Tool Ecosystems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"PlanBench-XL: Evaluating Long-Horizon Planning of LLM Tool-Use Agents in Large-Scale Tool Ecosystems\"",
+      "authors": "Jiayu Liu, Qihan Lin, Cheng Qian, Rui Wang, Emre Can Acikgoz, Xiaocheng Yang, Jiateng Liu, Zhenhailong Wang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.22388",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-21",
+      "updated_at": "2026-06-21",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "planning-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-22417-code-isn-t-memory-a-structural-codebase-index-inside-a-coding-agent.md",
+    "title": "\"Code Isn't Memory: A Structural Codebase Index Inside a Coding Agent\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Code Isn't Memory: A Structural Codebase Index Inside a Coding Agent\"",
+      "authors": "Ishaan Bhola, Adithyan Krishnan, Sravanth Kurmala, Mukunda NS",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.22417",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-21",
+      "updated_at": "2026-06-21",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "memory",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-22484-governed-ai-assisted-engineering-graduated-human-oversight-for-agentic-code-gene.md",
+    "title": "\"Governed AI-Assisted Engineering: Graduated Human Oversight for Agentic Code Generation in Regulated Domains\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Governed AI-Assisted Engineering: Graduated Human Oversight for Agentic Code Generation in Regulated Domains\"",
+      "authors": "Richard Kang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.22484",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-21",
+      "updated_at": "2026-07-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "rag",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.HC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-22495-grounded-scaling-why-agentic-ai-needs-deterministic-environments.md",
+    "title": "\"Grounded Scaling: Why Agentic AI Needs Deterministic Environments\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Grounded Scaling: Why Agentic AI Needs Deterministic Environments\"",
+      "authors": "Liang Ding, Xintong Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.22495",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-21",
+      "updated_at": "2026-06-21",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "embodied-agent",
+        "multi-agent",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-22557-macagentbench-benchmarking-ai-agents-on-real-world-macos-desktop.md",
+    "title": "\"MacAgentBench: Benchmarking AI Agents on Real-World macOS Desktop\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MacAgentBench: Benchmarking AI Agents on Real-World macOS Desktop\"",
+      "authors": "Yikun Fu, Bowen Fu, Zhenyu Wu, Shuang Cheng, Xiaowei Sun, Bowen Yang, Zehao Li, Yibo Zhao, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.22557",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-21",
+      "updated_at": "2026-06-21",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL",
+        "cs.HC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "21",
+      "collection_queries": "agent-evaluation, ai-agent, web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-22610-paperclaw-harnessing-agents-for-autonomous-research-and-human-in-the-loop-refine.md",
+    "title": "\"PaperClaw: Harnessing Agents for Autonomous Research and Human-in-the-Loop Refinement\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"PaperClaw: Harnessing Agents for Autonomous Research and Human-in-the-Loop Refinement\"",
+      "authors": "Weiwei Ye, Hangchen Liu, Dongyuan Li, Renhe Jiang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.22610",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-21",
+      "updated_at": "2026-06-21",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "multi-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-22647-raven-agentic-rag-for-automated-vulnerability-repair.md",
+    "title": "\"RAVEN: Agentic RAG for Automated Vulnerability Repair\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"RAVEN: Agentic RAG for Automated Vulnerability Repair\"",
+      "authors": "Varun Gadey, Zijie Liu, Alexandra Dmitrienko",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.22647",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-21",
+      "updated_at": "2026-06-21",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "computer-use",
+        "memory",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.LG",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-22673-agentlens-interpretable-safety-steering-via-mechanistic-subspaces-for-multi-turn.md",
+    "title": "\"AgentLens: Interpretable Safety Steering via Mechanistic Subspaces for Multi-Turn Coding Agent\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AgentLens: Interpretable Safety Steering via Mechanistic Subspaces for Multi-Turn Coding Agent\"",
+      "authors": "Weidi Luo, Qiming Zhang, Yihao Quan, Mingyu Jin, Jie Cai, Chaowei Xiao, Jingcheng Niu, Zhen Xiang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.22673",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-21",
+      "updated_at": "2026-06-21",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "agent-safety, coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-22678-rigorbench-benchmarking-engineering-process-discipline-in-autonomous-ai-coding-a.md",
+    "title": "\"RigorBench: Benchmarking Engineering Process Discipline in Autonomous AI Coding Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"RigorBench: Benchmarking Engineering Process Discipline in Autonomous AI Coding Agents\"",
+      "authors": "Meher Bhaskar Madiraju, Meher Sai Preetam Madiraju",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.22678",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-21",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-22737-groundeval-a-deterministic-replacement-for-llm-as-judge-in-stateful-agent-evalua.md",
+    "title": "\"GroundEval: A Deterministic Replacement for LLM-as-Judge in Stateful Agent Evaluation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"GroundEval: A Deterministic Replacement for LLM-as-Judge in Stateful Agent Evaluation\"",
+      "authors": "Jeffrey Flynt",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.22737",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-22",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-22741-grade-graph-representation-of-llm-agent-dependency-and-execution.md",
+    "title": "\"GRADE: Graph Representation of LLM Agent Dependency and Execution\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"GRADE: Graph Representation of LLM Agent Dependency and Execution\"",
+      "authors": "Yue Zhao",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.22741",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-22",
+      "updated_at": "2026-06-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "coding-agent",
+        "multi-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "multi-agent-llm, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-22844-ramem-contextual-reinstatement-for-long-term-agentic-memory.md",
+    "title": "\"RaMem: Contextual Reinstatement for Long-term Agentic Memory\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"RaMem: Contextual Reinstatement for Long-term Agentic Memory\"",
+      "authors": "Wei Yang, Bryce Kan, Shixuan Li, Li Li, Yuehan Qin, Jiate Li, Paul Bogdan, Jesse Thomason",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.22844",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-22",
+      "updated_at": "2026-06-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-22864-when-auc-0-998-is-not-enough-a-candidate-evaluation-protocol-for-hidden-state-pr.md",
+    "title": "\"When AUC 0.998 Is Not Enough: A Candidate Evaluation Protocol for Hidden-State Probes of Indirect Prompt Injection in Multimodal Computer-Use Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"When AUC 0.998 Is Not Enough: A Candidate Evaluation Protocol for Hidden-State Probes of Indirect Prompt Injection in Multimodal Computer-Use Agents\"",
+      "authors": "Yanhang Li, Zhichao Fan, Zexin Zhuang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.22864",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-22",
+      "updated_at": "2026-06-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-22948-envs-environment-native-verified-search-for-long-horizon-gui-agents.md",
+    "title": "\"ENVS: Environment-Native Verified Search for Long-Horizon GUI Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ENVS: Environment-Native Verified Search for Long-Horizon GUI Agents\"",
+      "authors": "Yincheng Zhou, Athena Zhuoming Zhong, Shijie Zhang, Kevin Zhang, Teresa Xiaotao Shang, Shanghang Zhang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.22948",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-22",
+      "updated_at": "2026-06-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CV"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-22953-plans-don-t-persist-why-context-management-is-load-bearing-for-llm-agents.md",
+    "title": "\"Plans Don't Persist: Why Context Management Is Load Bearing for LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Plans Don't Persist: Why Context Management Is Load Bearing for LLM Agents\"",
+      "authors": "Aman Mehta, Anupam Datta",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.22953",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-22",
+      "updated_at": "2026-06-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-23032-ipo-finance-agent-benchmark-of-llm-financial-analysts-beyond-finance-agent-v2-wi.md",
+    "title": "\"IPO Finance Agent: Benchmark of LLM Financial Analysts Beyond Finance Agent v2, with Automated Rubric Generation, on the SpaceX (SPCX) IPO\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"IPO Finance Agent: Benchmark of LLM Financial Analysts Beyond Finance Agent v2, with Automated Rubric Generation, on the SpaceX (SPCX) IPO\"",
+      "authors": "Mostapha Benhenda",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.23032",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-22",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "q-fin.GN"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-23130-understanding-the-in-security-of-vibe-coded-applications.md",
+    "title": "Understanding the (In)Security of Vibe-Coded Applications",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Understanding the (In)Security of Vibe-Coded Applications",
+      "authors": "Junquan Deng, Zhiyu Fan, Ruijie Meng",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.23130",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-22",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "memory",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-23195-memory-contagion-cross-temporal-propagation-of-evaluator-bias-via-agent-memory.md",
+    "title": "\"Memory Contagion: Cross-Temporal Propagation of Evaluator Bias via Agent Memory\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Memory Contagion: Cross-Temporal Propagation of Evaluator Bias via Agent Memory\"",
+      "authors": "Zewen Liu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.23195",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-22",
+      "updated_at": "2026-06-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-23283-towards-root-memories-benchmarking-and-enhancing-implicit-logical-memory-retriev.md",
+    "title": "\"Towards Root Memories: Benchmarking and Enhancing Implicit Logical Memory Retrieval for Personalized LLMs\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Towards Root Memories: Benchmarking and Enhancing Implicit Logical Memory Retrieval for Personalized LLMs\"",
+      "authors": "Hongxun Ding, Xiang Yu, Chengbing Wang, Jianfei Xiao, Keqin Bao, Wenjie Wang, Xiangnan He",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.23283",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-22",
+      "updated_at": "2026-06-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-23327-videoagent-all-in-one-framework-for-video-understanding-and-editing.md",
+    "title": "\"VideoAgent: All-in-One Framework for Video Understanding and Editing\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"VideoAgent: All-in-One Framework for Video Understanding and Editing\"",
+      "authors": "Hengji Zhou, Lingxuan Huang, Jian Wang, Bing Zhou, Si Wu, Lianghao Xia, Chao Huang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.23327",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-22",
+      "updated_at": "2026-07-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-23343-group-selection-promotes-prosocial-prompts-in-populations-of-llm-agents.md",
+    "title": "Group Selection Promotes Prosocial Prompts in Populations of LLM Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Group Selection Promotes Prosocial Prompts in Populations of LLM Agents",
+      "authors": "Luis Celiktemel, Edward Eichhorn, Levin Brinkmann, Robin Schimmelpfennig, Aron Vallinder, Yaomin Jiang, Edward Hughes, Iyad Rahwan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.23343",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-22",
+      "updated_at": "2026-06-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "coding-agent",
+        "computer-use",
+        "multi-agent",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CY"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-23565-holoagent-0-a-unified-embodied-agent-framework-with-3d-spatial-memory.md",
+    "title": "\"HoloAgent-0: A Unified Embodied Agent Framework with 3D Spatial Memory\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"HoloAgent-0: A Unified Embodied Agent Framework with 3D Spatial Memory\"",
+      "authors": "Xiaolin Zhou, Liu Liu, Tingyang Xiao, Wei Feng, Fa Fu, Xinrui Meng, Xinjie Wang, Jialiang Han, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.23565",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-22",
+      "updated_at": "2026-06-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "embodied-agent",
+        "memory",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.RO",
+        "cs.CV"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-23664-mas-promptbench-when-does-prompt-optimization-improve-multi-agent-llm-systems.md",
+    "title": "\"MAS-PromptBench: When Does Prompt Optimization Improve Multi-Agent LLM Systems?\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MAS-PromptBench: When Does Prompt Optimization Improve Multi-Agent LLM Systems?\"",
+      "authors": "Juyang Bai, Laixi Shi",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.23664",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-22",
+      "updated_at": "2026-06-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agentic-ai, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-23752-esaa-conversational-an-event-sourced-memory-layer-for-continuity-handoff-and-cur.md",
+    "title": "\"ESAA-Conversational: An Event-Sourced Memory Layer for Continuity, Handoff, and Curation Across Heterogeneous LLM Coding Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ESAA-Conversational: An Event-Sourced Memory Layer for Continuity, Handoff, and Curation Across Heterogeneous LLM Coding Agents\"",
+      "authors": "Elzo Brito dos Santos Filho",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.23752",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-22",
+      "updated_at": "2026-06-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "coding-agent",
+        "memory",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-23764-emergent-relational-order-in-llm-agent-societies-from-collective-affect-to-autho.md",
+    "title": "\"Emergent Relational Order in LLM Agent Societies: From Collective Affect to Authority Stratification\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Emergent Relational Order in LLM Agent Societies: From Collective Affect to Authority Stratification\"",
+      "authors": "Zhiyuan Ji, Xinyu Chen, Ziqi Dai, Shiyun Tang, Chunyu Wei, Yueguo Chen",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.23764",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-22",
+      "updated_at": "2026-06-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "multi-agent",
+        "planning",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-23927-rift-bench-dynamic-red-teaming-for-agentic-ai-systems.md",
+    "title": "\"RIFT-Bench: Dynamic Red-teaming For Agentic AI Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"RIFT-Bench: Dynamic Red-teaming For Agentic AI Systems\"",
+      "authors": "Yarin Yerushalmi Levi, Roy Betser, Amit Giloni, Lidor Erez, Itay Gershon, Oren Rachmil, Sindhu Padakandla, Roman Vainshtein",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.23927",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-22",
+      "updated_at": "2026-06-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-23991-critique-of-agent-model.md",
+    "title": "Critique of Agent Model",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Critique of Agent Model",
+      "authors": "Eric Xing, Mingkai Deng, Jinyu Hou",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.23991",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-22",
+      "updated_at": "2026-06-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "coding-agent",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.LG",
+        "cs.MA",
+        "cs.RO"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agentic-ai, ai-agent, coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-24193-skychain-intelligence-a-blockchain-secured-multi-agent-drl-framework-for-low-alt.md",
+    "title": "\"SkyChain Intelligence: A Blockchain-Secured Multi-Agent DRL Framework for Low-Altitude Embodied Artificial Intelligence\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SkyChain Intelligence: A Blockchain-Secured Multi-Agent DRL Framework for Low-Altitude Embodied Artificial Intelligence\"",
+      "authors": "Haoxiang Luo, Tianqi Jiang, Ruichen Zhang, Yinqiu Liu, Gang Sun, Hongfang Yu, Abbas Jamalipour, Dong In Kim",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.24193",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "embodied-agent",
+        "multi-agent",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.NI",
+        "cs.DC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-24235-sp-mind-an-autonomous-reasoning-agent-for-spatial-proteomics-analysis.md",
+    "title": "\"SP-Mind: An Autonomous Reasoning Agent for Spatial Proteomics Analysis\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SP-Mind: An Autonomous Reasoning Agent for Spatial Proteomics Analysis\"",
+      "authors": "Yucheng Yuan, Yuanfeng Ji, Zhongxiao Li, Ruijiang Li",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.24235",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-24322-securing-llm-agent-long-term-memory-against-poisoning-non-malleable-origin-bound.md",
+    "title": "\"Securing LLM-Agent Long-Term Memory Against Poisoning: Non-Malleable, Origin-Bound Authority with Machine-Checked Guarantees\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Securing LLM-Agent Long-Term Memory Against Poisoning: Non-Malleable, Origin-Bound Authority with Machine-Checked Guarantees\"",
+      "authors": "Yedidel Louck",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.24322",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-24402-poisoned-playbooks-demystifying-knowledge-poisoning-effects-on-ai-security-agent.md",
+    "title": "\"Poisoned Playbooks: Demystifying Knowledge Poisoning Effects on AI Security Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Poisoned Playbooks: Demystifying Knowledge Poisoning Effects on AI Security Agents\"",
+      "authors": "Juho Park, Hyunmin Choi, Kevin Nam",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.24402",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "ai-agent, rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-24416-agentic-ai-for-bilevel-long-term-optimization-of-policy-driven-physical-layer-sy.md",
+    "title": "Agentic AI for Bilevel Long-Term Optimization of Policy-Driven Physical Layer Systems",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Agentic AI for Bilevel Long-Term Optimization of Policy-Driven Physical Layer Systems",
+      "authors": "Bingnan Xiao, Chenhao Yang, Wei Ni, Xin Wang, Tony Q. S. Quek",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.24416",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "multi-agent",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-24437-rem-moa-reasoning-memory-sustains-mixture-of-agents-scaling.md",
+    "title": "\"ReM-MoA: Reasoning Memory Sustains Mixture-of-Agents Scaling\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ReM-MoA: Reasoning Memory Sustains Mixture-of-Agents Scaling\"",
+      "authors": "Heng Ping, Arijit Bhattacharjee, Peiyu Zhang, Shixuan Li, Wei Yang, Ali Jannesari, Nesreen Ahmed, Paul Bogdan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.24437",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "multi-agent",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-24453-bayesian-control-for-coding-agents.md",
+    "title": "Bayesian control for coding agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Bayesian control for coding agents",
+      "authors": "Theodore Papamarkou, Vladislav Smirnov, Viktor Mazanov, Artem Vazhentsev, Preslav Nakov, Timothy Baldwin, Artem Shelmanov",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.24453",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "coding-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-24515-reinforcement-learning-for-computer-use-agents-with-autonomous-evaluation.md",
+    "title": "Reinforcement Learning for Computer-Use Agents with Autonomous Evaluation",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Reinforcement Learning for Computer-Use Agents with Autonomous Evaluation",
+      "authors": "Marta Sumyk, Oleksandr Kosovan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.24515",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.HC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-24525-viscritic-visual-state-comparison-as-process-reward-for-gui-agents.md",
+    "title": "\"VisCritic: Visual State Comparison as Process Reward for GUI Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"VisCritic: Visual State Comparison as Process Reward for GUI Agents\"",
+      "authors": "Jiachen Qian",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.24525",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-24535-governed-shared-memory-for-multi-agent-llm-systems.md",
+    "title": "Governed Shared Memory for Multi-Agent LLM Systems",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Governed Shared Memory for Multi-Agent LLM Systems",
+      "authors": "Yanki Margalit, Nurit Cohen-Inger, Erni Avram, Ran Taig, Oded Margalit",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.24535",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "multi-agent",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "agent-memory, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-24551-gui-vs-cli-execution-bottlenecks-in-screen-only-and-skill-mediated-computer-use-.md",
+    "title": "\"GUI vs. CLI: Execution Bottlenecks in Screen-Only and Skill-Mediated Computer-Use Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"GUI vs. CLI: Execution Bottlenecks in Screen-Only and Skill-Mediated Computer-Use Agents\"",
+      "authors": "Xiao Zhou, Siyue Zhang, Yilun Zhao, Jinbiao Wei, Tingyu Song, Arman Cohan, Chen Zhao",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.24551",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-22",
+      "updated_at": "2026-06-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-24595-memprobe-probing-long-term-agent-memory-via-hidden-user-state-recovery.md",
+    "title": "\"MEMPROBE: Probing Long-Term Agent Memory via Hidden User-State Recovery\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MEMPROBE: Probing Long-Term Agent Memory via Hidden User-State Recovery\"",
+      "authors": "Enze Ma, Yufan Zhou, Wei-Chieh Huang, Jie Yang, Huanhuan Ma, Zixuan Wang, Chengze Li, Chunyu Miao, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.24595",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-24597-qwen-agentworld-language-world-models-for-general-agents.md",
+    "title": "\"Qwen-AgentWorld: Language World Models for General Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Qwen-AgentWorld: Language World Models for General Agents\"",
+      "authors": "Yuxin Zuo, Zikai Xiao, Li Sheng, Fei Huang, Jianhong Tu, Yuxuan Liu, Tianyi Tang, Xiaomeng Hu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.24597",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-24623-privacy-preserving-rag-via-multi-agent-semantic-rewriting-achieving-confidential.md",
+    "title": "\"Privacy-Preserving RAG via Multi-Agent Semantic Rewriting: Achieving Confidentiality Without Compromising Contextual Fidelity\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Privacy-Preserving RAG via Multi-Agent Semantic Rewriting: Achieving Confidentiality Without Compromising Contextual Fidelity\"",
+      "authors": "Yuanhe Zhao, Tianyu Zhang, Huafei Xing, Derek F. Wong, Jianbin Li, Tao Fang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.24623",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "multi-agent-llm, rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-24626-safari-scaling-long-horizon-agentic-fault-attribution-via-active-investigation.md",
+    "title": "\"SAFARI: Scaling Long Horizon Agentic Fault Attribution via Active Investigation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SAFARI: Scaling Long Horizon Agentic Fault Attribution via Active Investigation\"",
+      "authors": "Chenyang Zhu, Jiayu Yao, Kushal Chawla, Youbing Yin, Nathan Wolfe, Pengshan Cai, Jingyu Wu, Spencer Hong, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.24626",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "multi-agent",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "autonomous-agent-llm, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-24649-agentic-collaborative-cognition-for-zero-shot-3d-understanding.md",
+    "title": "Agentic Collaborative Cognition for Zero-Shot 3D Understanding",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Agentic Collaborative Cognition for Zero-Shot 3D Understanding",
+      "authors": "Wenxin Wang, Bo Zhang, Feng Chen, Zixuan Wang, Wen Li, Changsheng Li, Yinjie Lei",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.24649",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-06-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "multi-agent",
+        "planning",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-24689-automated-summarization-of-software-documents-an-llm-based-multi-agent-approach.md",
+    "title": "\"Automated Summarization of Software Documents: An LLM-based Multi-Agent Approach\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Automated Summarization of Software Documents: An LLM-based Multi-Agent Approach\"",
+      "authors": "Duc S. H. Nguyen, Minh T. Nguyen, Phuong T. Nguyen, Juri Di Rocco, Davide Di Ruscio",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.24689",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "multi-agent",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-24694-supplynet-supporting-visual-exploratory-learning-in-supply-chain-via-contextual-.md",
+    "title": "\"SupplyNet: Supporting Visual Exploratory Learning in Supply Chain via Contextual Multi-Agent Simulation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SupplyNet: Supporting Visual Exploratory Learning in Supply Chain via Contextual Multi-Agent Simulation\"",
+      "authors": "Yanjia Li, Kelcy Kexin Han, Tianrui Hu, Yi-Fan Cao, Huamin Qu, Sicheng Song",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.24694",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "multi-agent",
+        "rag",
+        "reasoning",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.HC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-24775-are-we-ready-for-an-agent-native-memory-system.md",
+    "title": "Are We Ready For An Agent-Native Memory System?",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Are We Ready For An Agent-Native Memory System?",
+      "authors": "Wei Zhou, Xuanhe Zhou, Shaokun Han, Hongming Xu, Guoliang Li, Zhiyu Li, Feiyu Xiong, Fan Wu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.24775",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.DB",
+        "cs.IR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-24779-deepbd-a-grounded-agentic-workflow-for-variant-prioritization-and-diagnosis-of-g.md",
+    "title": "\"DeepBD: A Grounded Agentic Workflow for Variant Prioritization and Diagnosis of Genetic Birth Defects\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"DeepBD: A Grounded Agentic Workflow for Variant Prioritization and Diagnosis of Genetic Birth Defects\"",
+      "authors": "Shiyu Li, Ziqi Yan, Zhihao Wu, Jielong Lu, Weiran Liao, Jiajun Yu, Genjie Li, Zeyu Chu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.24779",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "q-bio.GN",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-24820-sherloc-structured-diagnostic-localization-for-code-repair-agents.md",
+    "title": "\"SHERLOC: Structured Diagnostic Localization for Code Repair Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SHERLOC: Structured Diagnostic Localization for Code Repair Agents\"",
+      "authors": "Hovhannes Tamoyan, Sean Narenthiran, Erik Arakelyan, Mira Mezini, Boris Ginsburg",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.24820",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "multi-agent",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "coding-agent, multi-agent-llm, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-24839-grading-the-grader-lessons-from-evaluating-an-agentic-data-analysis-system.md",
+    "title": "\"Grading the Grader: Lessons from Evaluating an Agentic Data Analysis System\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Grading the Grader: Lessons from Evaluating an Agentic Data Analysis System\"",
+      "authors": "Tian Zheng, Kai-Tai Hsu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.24839",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "multi-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "stat.AP"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-24855-openthoughts-agent-data-recipes-for-agentic-models.md",
+    "title": "\"OpenThoughts-Agent: Data Recipes for Agentic Models\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"OpenThoughts-Agent: Data Recipes for Agentic Models\"",
+      "authors": "Negin Raoof, Richard Zhuang, Marianna Nezhurina, Etash Guha, Atula Tejaswi, Ryan Marten, Charlie F. Ruan, Tyler Griggs, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.24855",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-24937-the-hitchhiker-s-guide-to-agentic-ai-from-foundations-to-systems.md",
+    "title": "\"The Hitchhiker's Guide to Agentic AI: From Foundations to Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"The Hitchhiker's Guide to Agentic AI: From Foundations to Systems\"",
+      "authors": "Haggai Roitman",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.24937",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-22",
+      "updated_at": "2026-06-22",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "memory",
+        "multi-agent",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL",
+        "cs.IR",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "25",
+      "collection_queries": "agentic-ai, multi-agent-llm, rag-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-24976-diagnosing-and-mitigating-compounding-failures-in-agentic-persuasion-via-taxonom.md",
+    "title": "Diagnosing and Mitigating Compounding Failures in Agentic Persuasion via Taxonomic Strategy Retrieval",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Diagnosing and Mitigating Compounding Failures in Agentic Persuasion via Taxonomic Strategy Retrieval",
+      "authors": "Sana Ayromlou, Purvi Sehgal, Pradyumna Narayana",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.24976",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-07-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "planning",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-25115-forget-to-improve-on-device-llm-agent-continual-learning-via-budget-curated-memo.md",
+    "title": "\"Forget to Improve: On-Device LLM-Agent Continual Learning via Budget-Curated Memory\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Forget to Improve: On-Device LLM-Agent Continual Learning via Budget-Curated Memory\"",
+      "authors": "Beining Wu, Zihao Ding, Jun Huang, Yanxiao Zhao",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.25115",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "embodied-agent",
+        "memory"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.NI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-25139-buildrix-an-open-platform-for-sharing-and-benchmarking-agentic-ai-skills-in-buil.md",
+    "title": "\"Buildrix: An Open Platform for Sharing and Benchmarking Agentic AI Skills in Building Engineering\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Buildrix: An Open Platform for Sharing and Benchmarking Agentic AI Skills in Building Engineering\"",
+      "authors": "Zixin Jiang, Bing Dong",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.25139",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "eess.SY"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-25161-trustmem-learning-trustworthy-memory-consolidation-for-llm-agents-with-long-term.md",
+    "title": "\"TRUSTMEM: Learning Trustworthy Memory Consolidation for LLM Agents with Long-Term Memory\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"TRUSTMEM: Learning Trustworthy Memory Consolidation for LLM Agents with Long-Term Memory\"",
+      "authors": "Tianyu Yang, Sudipta Paul, Vijay Srinivasan, Vivek Kulkarni, Srinivas Chappidi",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.25161",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-25189-actplane-programmable-os-level-policy-enforcement-for-agent-harnesses.md",
+    "title": "\"ActPlane: Programmable OS-Level Policy Enforcement for Agent Harnesses\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ActPlane: Programmable OS-Level Policy Enforcement for Agent Harnesses\"",
+      "authors": "Yusheng Zheng, Tianyuan Wu, Quanzhi Fu, Tong Yu, Wenan Mao, Tao Ma, Dan Williams, Wei Wang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.25189",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.OS"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-25191-to-isolate-or-to-score-model-adaptive-assessment-for-cost-efficient-multi-agent-.md",
+    "title": "To Isolate or to Score? Model-Adaptive Assessment for Cost-Efficient Multi-Agent RAG",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "To Isolate or to Score? Model-Adaptive Assessment for Cost-Efficient Multi-Agent RAG",
+      "authors": "Jungseob Lee, Chanjun Park, Heuiseok Lim",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.25191",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-25195-sok-ai-secure-code-generation-progress-pitfalls-and-paths-forward.md",
+    "title": "\"SoK: AI Secure Code Generation: Progress, Pitfalls, and Paths Forward\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SoK: AI Secure Code Generation: Progress, Pitfalls, and Paths Forward\"",
+      "authors": "Rupam Patir, Keyan Guo, Haipeng Cai, Hongxin Hu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.25195",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "computer-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agentic-ai, coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-25206-raven-long-horizon-reasoning-navigation-with-a-visuo-spatio-temporal-memory.md",
+    "title": "\"RAVEN: Long-Horizon Reasoning & Navigation with a Visuo-Spatio-Temporal Memory\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"RAVEN: Long-Horizon Reasoning & Navigation with a Visuo-Spatio-Temporal Memory\"",
+      "authors": "Yixun Hu, Zhicheng Zheng, Lihan Zha, Chunwei Xing, Rajdeep Singh, Omar Hossain, Antonio Loquercio, Dhruv Shah",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.25206",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-23",
+      "updated_at": "2026-06-23",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "embodied-agent",
+        "memory",
+        "planning",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.RO",
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-25334-bridging-the-post-discharge-gap-a-traceable-multi-agent-framework-for-safe-and-c.md",
+    "title": "\"Bridging the Post-discharge Gap: A Traceable Multi-agent Framework for Safe and Continuous Care\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Bridging the Post-discharge Gap: A Traceable Multi-agent Framework for Safe and Continuous Care\"",
+      "authors": "Runwei Guan, Yi Zhou, Heyi Lin, Jinjing Zhu, Mingyuan Hou, Yang Yang, Fang Yuan, Xiaohong Lin, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.25334",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-24",
+      "updated_at": "2026-06-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "multi-agent",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-25358-agentic-knowledge-tracing-a-multi-agent-llm-architecture-for-stealth-assessment-.md",
+    "title": "\"Agentic Knowledge Tracing: A Multi-Agent LLM Architecture for Stealth Assessment of Financial Literacy in Serious Games\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Agentic Knowledge Tracing: A Multi-Agent LLM Architecture for Stealth Assessment of Financial Literacy in Serious Games\"",
+      "authors": "Gabriel Santos, Rita Julia, Marcelo Nascimento",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.25358",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-24",
+      "updated_at": "2026-06-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-25361-memory-makes-the-difference-evaluating-how-different-memory-roles-shape-conversa.md",
+    "title": "\"Memory Makes the Difference: Evaluating How Different Memory Roles Shape Conversational Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Memory Makes the Difference: Evaluating How Different Memory Roles Shape Conversational Agents\"",
+      "authors": "Yuxin Wang, Paul Thomas, Zhiwei Yu, Yuan Gao, Saeed Hassanpour, Soroush Vosoughi, Robert Sim, Nick Craswell",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.25361",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-24",
+      "updated_at": "2026-06-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI",
+        "cs.IR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-25400-brainagent-a-large-language-model-driven-multi-agent-framework-for-autonomous-br.md",
+    "title": "\"BrainAgent: A Large Language Model-Driven Multi-Agent Framework for Autonomous Brain Signal Understanding\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"BrainAgent: A Large Language Model-Driven Multi-Agent Framework for Autonomous Brain Signal Understanding\"",
+      "authors": "Yangxuan Zhou, Sha Zhao, Jiquan Wang, Shijian Li, Gang Pan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.25400",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-24",
+      "updated_at": "2026-06-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "planning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-25484-from-causal-discovery-to-implementation-an-agentic-ai-framework-for-e-scooter-mo.md",
+    "title": "\"From Causal Discovery to Implementation: An Agentic AI Framework for E-Scooter Mobility Hub Planning Across 29 German Cities\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"From Causal Discovery to Implementation: An Agentic AI Framework for E-Scooter Mobility Hub Planning Across 29 German Cities\"",
+      "authors": "Meng Jin, Melanie Handrich, Simone Martinenz, Nicholas Hoeser, Ziyue Li",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.25484",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-24",
+      "updated_at": "2026-06-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "coding-agent",
+        "computer-use",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CY",
+        "econ.GN",
+        "stat.AP"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-25514-unlocking-model-potentials-through-adaptive-multi-agent-scaffolding-for-efficien.md",
+    "title": "Unlocking Model Potentials Through Adaptive Multi-Agent Scaffolding for Efficient Issue Resolution",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Unlocking Model Potentials Through Adaptive Multi-Agent Scaffolding for Efficient Issue Resolution",
+      "authors": "Yang Chen, Aliya Ahmad, Yiheng Zhou, Reyhaneh Jabbarvand",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.25514",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-24",
+      "updated_at": "2026-06-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "multi-agent",
+        "planning",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-25519-quantization-inflates-reasoning-token-inflation-as-a-hidden-cost-of-low-bit-reas.md",
+    "title": "\"Quantization Inflates Reasoning: Token Inflation as a Hidden Cost of Low-Bit Reasoning Models\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Quantization Inflates Reasoning: Token Inflation as a Hidden Cost of Low-Bit Reasoning Models\"",
+      "authors": "Xinyu Lian, Walid Krichene, Beichen Huang, Masahiro Tanaka, Olatunji Ruwase, Li Zhang, Minjia Zhang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.25519",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-24",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-25588-intenttester-intent-driven-multi-agent-framework-for-cross-library-test-migratio.md",
+    "title": "\"IntentTester: Intent-Driven Multi-agent Framework for Cross-Library Test Migration\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"IntentTester: Intent-Driven Multi-agent Framework for Cross-Library Test Migration\"",
+      "authors": "Yi Gao, Ziyuan Zhang, Xing Hu, Xiaohu Yang, Xin Xia",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.25588",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-24",
+      "updated_at": "2026-06-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "computer-use",
+        "multi-agent",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-25622-probabilistic-agents-in-deterministic-audits-evaluating-multi-agent-systems-for-.md",
+    "title": "\"Probabilistic Agents in Deterministic Audits: Evaluating Multi-Agent Systems for Automated Audits Based on the German IT-Grundschutz\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Probabilistic Agents in Deterministic Audits: Evaluating Multi-Agent Systems for Automated Audits Based on the German IT-Grundschutz\"",
+      "authors": "Lea Roxanne Muth, Marian Margraf",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.25622",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-24",
+      "updated_at": "2026-06-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-25651-medguards-multi-agent-system-for-reliable-medical-error-detection-and-correction.md",
+    "title": "\"MedGuards: Multi-Agent System for Reliable Medical Error Detection and Correction\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MedGuards: Multi-Agent System for Reliable Medical Error Detection and Correction\"",
+      "authors": "Congbo Ma, Hu Wang, Yichun Zhang, Farah E. Shamout",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.25651",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-24",
+      "updated_at": "2026-06-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "multi-agent",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-25656-is-graphrag-needed-from-basic-rag-to-graph-agentic-solutions-with-context-optimi.md",
+    "title": "Is GraphRAG Needed? From Basic RAG to Graph-/Agentic Solutions with Context Optimization",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Is GraphRAG Needed? From Basic RAG to Graph-/Agentic Solutions with Context Optimization",
+      "authors": "Long Chen, Ryan Razkenari, Yuxuan Zhou, Yuan Tian, Rahul Ghosh, Venkatesh Pappakrishnan, Disha Ahuja, Vidya Sagar Ravipati",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.25656",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-24",
+      "updated_at": "2026-06-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI",
+        "cs.IR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-25705-gui-agent-guided-exploration-of-user-sensitive-screens.md",
+    "title": "\"GUI agent: Guided Exploration of User-Sensitive Screens\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"GUI agent: Guided Exploration of User-Sensitive Screens\"",
+      "authors": "Aradhana Nayak, Mussadiq Nazeer, Wang Peng, Feng Liu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.25705",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-24",
+      "updated_at": "2026-06-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "computer-use",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-25760-uncertainty-quantification-for-computer-use-agents-a-benchmark-across-vision-lan.md",
+    "title": "\"Uncertainty Quantification for Computer-Use Agents: A Benchmark across Vision-Language Models and GUI Grounding Datasets\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Uncertainty Quantification for Computer-Use Agents: A Benchmark across Vision-Language Models and GUI Grounding Datasets\"",
+      "authors": "Divake Kumar, Sina Tayebati, Devashri Naik, Amanda Sofie Rios, Nilesh Ahuja, Omesh Tickoo, Ranganath Krishnan, Amit Ranjan Trivedi",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.25760",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-24",
+      "updated_at": "2026-06-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI",
+        "cs.CL",
+        "cs.CV"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-evaluation, web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-25819-beyond-function-calling-benchmarking-tool-using-agents-under-tool-environment-un.md",
+    "title": "\"Beyond Function Calling: Benchmarking Tool-Using Agents under Tool-Environment Unreliability\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Beyond Function Calling: Benchmarking Tool-Using Agents under Tool-Environment Unreliability\"",
+      "authors": "Yang Tian, Zhengpeng Shi, Yu Zhou, Bo Zhao",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.25819",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-24",
+      "updated_at": "2026-06-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "function-calling, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-25899-manipulation-is-task-dependent-a-multi-axis-multi-environment-evaluation-of-fron.md",
+    "title": "\"Manipulation Is Task-Dependent: A Multi-Axis, Multi-Environment Evaluation of Frontier LLMs\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Manipulation Is Task-Dependent: A Multi-Axis, Multi-Environment Evaluation of Frontier LLMs\"",
+      "authors": "Adeeb Zaman, Erik Nordby, Fred Heiding",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.25899",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-24",
+      "updated_at": "2026-06-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-26057-the-unfireable-safety-kernel-execution-time-ai-alignment-for-ai-agents-and-other.md",
+    "title": "\"The Unfireable Safety Kernel: Execution-Time AI Alignment for AI Agents and Other Escapable AI Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"The Unfireable Safety Kernel: Execution-Time AI Alignment for AI Agents and Other Escapable AI Systems\"",
+      "authors": "Seth Dobrin, Łukasz Chmiel",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.26057",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-24",
+      "updated_at": "2026-06-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CR",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-26203-agentic-analysis-for-agentic-infrastructure-an-llm-powered-pipeline-for-comparat.md",
+    "title": "\"Agentic Analysis for Agentic Infrastructure: An LLM-Powered Pipeline for Comparative Governance of DAO and Corporate AI Protocols\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Agentic Analysis for Agentic Infrastructure: An LLM-Powered Pipeline for Comparative Governance of DAO and Corporate AI Protocols\"",
+      "authors": "Yutian Wang, Luyao Zhang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.26203",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-24",
+      "updated_at": "2026-06-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "coding-agent",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CY",
+        "cs.MA",
+        "cs.SI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agentic-ai, ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-26205-knowledge-augmented-agentic-ai-for-mental-health-medication-information-seeking.md",
+    "title": "Knowledge-augmented Agentic AI for Mental Health Medication Information Seeking",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Knowledge-augmented Agentic AI for Mental Health Medication Information Seeking",
+      "authors": "Huizi Yu, Jian Liu, Wenkong Wang, Lingyao Li, Jiayan Zhou, Zhaoqian Xue, Xiang Li, Xinxin Lin, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.26205",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-24",
+      "updated_at": "2026-06-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agentic-ai, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-26216-cyberchainbench-can-ai-agents-secure-smart-contracts-against-real-world-on-chain.md",
+    "title": "\"CyberChainBench: Can AI Agents Secure Smart Contracts Against Real-World On-Chain Vulnerabilities?\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"CyberChainBench: Can AI Agents Secure Smart Contracts Against Real-World On-Chain Vulnerabilities?\"",
+      "authors": "Jintao Huang, Fengqing Jiang, Radha Poovendran, Zhiqiang Lin",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.26216",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-24",
+      "updated_at": "2026-06-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-safety, ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-26300-the-verification-horizon-no-silver-bullet-for-coding-agent-rewards.md",
+    "title": "\"The Verification Horizon: No Silver Bullet for Coding Agent Rewards\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"The Verification Horizon: No Silver Bullet for Coding Agent Rewards\"",
+      "authors": "Binghai Wang, Chenlong Zhang, Dayiheng Liu, Jiajun Zhang, Jiawei Chen, Mingze Li, Mouxiang Chen, Rongyao Fang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.26300",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-24",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "planning",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-26346-how-do-tool-augmented-llm-agents-perform-on-real-world-energy-analytics-tasks.md",
+    "title": "How Do Tool-Augmented LLM Agents Perform on Real-World Energy Analytics Tasks?",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "How Do Tool-Augmented LLM Agents Perform on Real-World Energy Analytics Tasks?",
+      "authors": "David Akinpelu, Akintonde Abbas, Rereloluwa Alimi, Ayodeji Lana",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.26346",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-24",
+      "updated_at": "2026-06-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-26356-instruction-bleed-cross-module-interference-in-prompt-composed-agentic-systems.md",
+    "title": "\"Instruction Bleed: Cross-Module Interference in Prompt-Composed Agentic Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Instruction Bleed: Cross-Module Interference in Prompt-Composed Agentic Systems\"",
+      "authors": "Ching-Yu Lin, Yifan Liu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.26356",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-24",
+      "updated_at": "2026-06-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.IR",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-evaluation, agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-26403-profilefoundry-a-synthetic-person-object-substrate-for-privacy-memory-and-tool-u.md",
+    "title": "\"ProfileFoundry: A Synthetic Person-Object Substrate for Privacy, Memory, and Tool-Use Evaluation in LLM Agent\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ProfileFoundry: A Synthetic Person-Object Substrate for Privacy, Memory, and Tool-Use Evaluation in LLM Agent\"",
+      "authors": "Sriram Selvam, Anneswa Ghosh",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.26403",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-24",
+      "updated_at": "2026-06-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-26453-optimizing-cuda-like-a-human-micro-profiling-tools-as-expert-surrogates-for-llm-.md",
+    "title": "\"Optimizing CUDA like a Human: Micro-Profiling Tools as Expert Surrogates for LLM-Based GPU Kernel Optimization\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Optimizing CUDA like a Human: Micro-Profiling Tools as Expert Surrogates for LLM-Based GPU Kernel Optimization\"",
+      "authors": "Jiading Gai, Shuai Zhang, Kaj Bostrom, Jin Huang, Vihang Patil, Haoyang Fang, Bernie Wang, Huzefa Rangwala, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.26453",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-24",
+      "updated_at": "2026-06-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "computer-use",
+        "memory",
+        "multi-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "coding-agent, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-26479-adaptive-evaluation-of-out-of-band-defenses-against-prompt-injection-in-llm-agen.md",
+    "title": "Adaptive Evaluation of Out-of-Band Defenses Against Prompt Injection in LLM Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Adaptive Evaluation of Out-of-Band Defenses Against Prompt Injection in LLM Agents",
+      "authors": "Praneeth Narisetty, Shiva Nagendra Babu Kore, Uday Kumar Reddy Kattamanchi, Jayaram Kumarapu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.26479",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-25",
+      "updated_at": "2026-06-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI",
+        "cs.CL",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-26511-temporal-validity-in-retrieval-memory-eliminating-stale-fact-errors-for-ai-agent.md",
+    "title": "\"Temporal Validity in Retrieval Memory: Eliminating Stale-Fact Errors for AI Agents over Evolving Knowledge\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Temporal Validity in Retrieval Memory: Eliminating Stale-Fact Errors for AI Agents over Evolving Knowledge\"",
+      "authors": "Neeraj Yadav",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.26511",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-25",
+      "updated_at": "2026-06-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI",
+        "cs.ET",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "21",
+      "collection_queries": "ai-agent, rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-26524-vigil-runtime-enforcement-of-behavioral-specifications-in-ai-agent-skills.md",
+    "title": "\"VIGIL: Runtime Enforcement of Behavioral Specifications in AI Agent Skills\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"VIGIL: Runtime Enforcement of Behavioral Specifications in AI Agent Skills\"",
+      "authors": "Ying Li, Yanju Chen, Hongbo Wen, Bosi Zhang, Hanzhi Liu, Peiran Wang, Yu Feng, Yuan Tian",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.26524",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-25",
+      "updated_at": "2026-06-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-26614-hilsva-design-and-evaluation-of-a-human-in-the-loop-agentic-system-for-scientifi.md",
+    "title": "\"HiLSVA: Design and Evaluation of a Human-in-the-Loop Agentic System for Scientific Visualization\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"HiLSVA: Design and Evaluation of a Human-in-the-Loop Agentic System for Scientific Visualization\"",
+      "authors": "Kuangshi Ai, Patrick Phuoc Do, Chaoli Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.26614",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-25",
+      "updated_at": "2026-06-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "multi-agent",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.HC",
+        "cs.AI",
+        "cs.GR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "23",
+      "collection_queries": "multi-agent-llm, planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-26627-agents-that-know-too-much-a-data-centric-survey-of-privacy-in-llm-agents.md",
+    "title": "\"Agents That Know Too Much: A Data-Centric Survey of Privacy in LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Agents That Know Too Much: A Data-Centric Survey of Privacy in LLM Agents\"",
+      "authors": "Nada Lahjouji, Ashwin Gerard Colaco",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.26627",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-25",
+      "updated_at": "2026-06-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-26649-autoformalization-of-agent-instructions-into-policy-as-code.md",
+    "title": "Autoformalization of Agent Instructions into Policy-as-Code",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Autoformalization of Agent Instructions into Policy-as-Code",
+      "authors": "Adam Mondl, Matthew Maisel, John H. Brock",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.26649",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-25",
+      "updated_at": "2026-06-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-26721-knowledge-based-pull-requests-a-trusted-workflow-for-agent-mediated-knowledge-co.md",
+    "title": "\"Knowledge-Based Pull Requests: A Trusted Workflow for Agent-Mediated Knowledge Collaboration\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Knowledge-Based Pull Requests: A Trusted Workflow for Agent-Mediated Knowledge Collaboration\"",
+      "authors": "Xinyu Zhang, Weiwei Sun",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.26721",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-25",
+      "updated_at": "2026-06-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "memory",
+        "multi-agent",
+        "planning",
+        "tool-use",
+        "workflow-agent",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.HC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-26758-egg-an-expert-guided-agent-framework-for-kernel-generation.md",
+    "title": "\"EGG: An Expert-Guided Agent Framework for Kernel Generation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"EGG: An Expert-Guided Agent Framework for Kernel Generation\"",
+      "authors": "Yaochen Han, Ke Fan, Hongxu Jiang, Wanqi Xu, Weiyu Xie, Runhua Zhang, Chenhui Zhu, Yixiang Zhang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.26758",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-25",
+      "updated_at": "2026-06-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "computer-use",
+        "memory",
+        "multi-agent",
+        "rag",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-26793-mirror-novelty-constrained-memory-guided-mcts-red-teaming-for-agentic-rag.md",
+    "title": "\"MIRROR: Novelty-Constrained Memory-Guided MCTS Red-Teaming for Agentic RAG\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MIRROR: Novelty-Constrained Memory-Guided MCTS Red-Teaming for Agentic RAG\"",
+      "authors": "Inderjeet Singh, Andrés Murillo, Motoyoshi Sekiya, Yuki Unno, Junichi Suga",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.26793",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-25",
+      "updated_at": "2026-06-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-26806-memory-depth-not-memory-access-selective-parametric-consolidation-for-long-runni.md",
+    "title": "\"Memory Depth, Not Memory Access: Selective Parametric Consolidation for Long-Running Language Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Memory Depth, Not Memory Access: Selective Parametric Consolidation for Long-Running Language Agents\"",
+      "authors": "Haoliang Han",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.26806",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-25",
+      "updated_at": "2026-06-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-26883-econsimulacra-a-digital-twin-platform-of-socio-economic-systems-powered-by-llm-a.md",
+    "title": "\"EconSimulacra: A Digital Twin Platform of Socio-Economic Systems Powered by LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"EconSimulacra: A Digital Twin Platform of Socio-Economic Systems Powered by LLM Agents\"",
+      "authors": "Ryuji Hashimoto, Masahiro Kaneko, Kentaro Ueda, Takehiro Takayanagi, Kiyoshi Izumi",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.26883",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-25",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "memory",
+        "multi-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.DL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-26918-diagnosing-task-insensitivity-in-language-agents.md",
+    "title": "Diagnosing Task Insensitivity in Language Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Diagnosing Task Insensitivity in Language Agents",
+      "authors": "Jingyu Liu, Xiaopeng Wu, Kehan Chen, Chuan Yu, Yong Liu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.26918",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-25",
+      "updated_at": "2026-06-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-26924-a-deterministic-control-plane-for-llm-coding-agents.md",
+    "title": "A Deterministic Control Plane for LLM Coding Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "A Deterministic Control Plane for LLM Coding Agents",
+      "authors": "Padmaraj Madatha",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.26924",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-25",
+      "updated_at": "2026-06-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "planning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI",
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-26960-toward-agentic-sysadmin-rethinking-system-administration-with-ai-agents.md",
+    "title": "\"Toward Agentic SysAdmin: Rethinking System Administration with AI Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Toward Agentic SysAdmin: Rethinking System Administration with AI Agents\"",
+      "authors": "Gianmaria Frigo, Davide Saladino, Alberto Castagnaro, Francesco Marchiori, Denis Donadel, Luca Pajola, Mauro Conti",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.26960",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-25",
+      "updated_at": "2026-06-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.NI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-27009-semantic-early-stopping-for-iterative-llm-agent-loops.md",
+    "title": "Semantic Early-Stopping for Iterative LLM Agent Loops",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Semantic Early-Stopping for Iterative LLM Agent Loops",
+      "authors": "Sahil Shrivastava",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.27009",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-25",
+      "updated_at": "2026-06-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.LG",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-27154-openrca-2-0-from-outcome-labels-to-causal-process-supervision.md",
+    "title": "\"OpenRCA 2.0: From Outcome Labels to Causal Process Supervision\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"OpenRCA 2.0: From Outcome Labels to Causal Process Supervision\"",
+      "authors": "\"Aoyang Fang, Yifan Yang, Jin'ao Shang, Qisheng Lu, Junjielung Xu, Rui Wang, Songhan Zhang, Yuzhong Zhang, et al.\"",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.27154",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-25",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-27243-nova-a-verification-aware-agent-harness-for-architecture-evolution-in-industrial.md",
+    "title": "\"NOVA: A Verification-Aware Agent Harness for Architecture Evolution in Industrial Recommender Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"NOVA: A Verification-Aware Agent Harness for Architecture Evolution in Industrial Recommender Systems\"",
+      "authors": "Shaohua Liu, Liang Fang, Yilong Sun, Shudong Huang, Qingsong Luo, Shaoxin Liu, Xiaoyang Chen, Dongqiang Liu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.27243",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-25",
+      "updated_at": "2026-06-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "computer-use",
+        "memory",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.IR",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-27330-empowering-gui-agents-via-autonomous-experience-exploration-and-hindsight-experi.md",
+    "title": "Empowering GUI Agents via Autonomous Experience Exploration and Hindsight Experience Utilization for Task Planning",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Empowering GUI Agents via Autonomous Experience Exploration and Hindsight Experience Utilization for Task Planning",
+      "authors": "Tianyi Men, Zhuoran Jin, Pengfei Cao, Yubo Chen, Kang Liu, Jun Zhao",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.27330",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-25",
+      "updated_at": "2026-06-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI",
+        "cs.CV",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-27350-chia-an-open-source-framework-for-principled-agentic-ai-driven-hardware-software.md",
+    "title": "\"CHIA: An open-source framework for principled, agentic AI-driven hardware/software co-design research\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"CHIA: An open-source framework for principled, agentic AI-driven hardware/software co-design research\"",
+      "authors": "Angela Cui, Ferran Hermida-Rivera, Jack Toubes, Raghav Gupta, Jim Fang, Chengyi Lux Zhang, Ella Schwarz, Junha Kim, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.27350",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-25",
+      "updated_at": "2026-06-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "coding-agent",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agentic-ai, coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-27397-sidconarena-an-environment-evaluating-agents-in-open-ended-positive-sum-bargaini.md",
+    "title": "\"SidConArena: An Environment Evaluating Agents in Open-Ended,Positive-Sum Bargaining Game\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SidConArena: An Environment Evaluating Agents in Open-Ended,Positive-Sum Bargaining Game\"",
+      "authors": "Yeqi Feng, Yuxin Chen, Tianxing He",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.27397",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-24",
+      "updated_at": "2026-06-24",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA",
+        "cs.AI",
+        "cs.GT"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-27406-towards-evaluation-of-implicit-software-world-models-in-coding-llms.md",
+    "title": "Towards Evaluation of Implicit Software World Models in Coding LLMs",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Towards Evaluation of Implicit Software World Models in Coding LLMs",
+      "authors": "Egor Bogomolov, Yaroslav Zharov",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.27406",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-25",
+      "updated_at": "2026-06-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "memory",
+        "reasoning",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "ai-agent, coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-27416-glite-arf-verifier-driven-research-with-parallel-llm-coding-agents.md",
+    "title": "\"Glite ARF: Verifier-Driven Research with Parallel LLM Coding Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Glite ARF: Verifier-Driven Research with Parallel LLM Coding Agents\"",
+      "authors": "Vassili Philippov, Pavel Katunin, Dmitry Andreev, Igor Ostanin, Anton Nikolaev",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.27416",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-25",
+      "updated_at": "2026-06-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "coding-agent",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-27472-supersede-diagnosing-and-training-the-memory-update-gap-in-llm-agents.md",
+    "title": "\"Supersede: Diagnosing and Training the Memory-Update Gap in LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Supersede: Diagnosing and Training the Memory-Update Gap in LLM Agents\"",
+      "authors": "Vedant Patel",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.27472",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-25",
+      "updated_at": "2026-06-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-27483-internalizing-the-future-a-unified-agentic-training-paradigm-for-world-model-pla.md",
+    "title": "\"Internalizing the Future: A Unified Agentic Training Paradigm for World Model Planning\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Internalizing the Future: A Unified Agentic Training Paradigm for World Model Planning\"",
+      "authors": "Xuan Zhang, Zhijian Zhou, Lingfeng Qiao, Yulei Qin, Ke Li, Xing Sun, Xiaoyu Tan, Chao Qu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.27483",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-25",
+      "updated_at": "2026-06-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "reasoning",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-27492-queenbee-planner-skill-evolving-communication-topologies-for-token-efficient-llm.md",
+    "title": "\"QueenBee Planner: Skill-Evolving Communication Topologies for Token-Efficient LLM Multi-Agent Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"QueenBee Planner: Skill-Evolving Communication Topologies for Token-Efficient LLM Multi-Agent Systems\"",
+      "authors": "Congjia Tian, Yuhang Yao, Jiaming Cui",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.27492",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-25",
+      "updated_at": "2026-06-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-27499-dmv-bench-diagnosing-long-horizon-multimodal-agents-visual-memory-with-incidenta.md",
+    "title": "\"DMV-Bench: Diagnosing Long-Horizon Multimodal Agents' Visual Memory with Incidental Cue Injection\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"DMV-Bench: Diagnosing Long-Horizon Multimodal Agents' Visual Memory with Incidental Cue Injection\"",
+      "authors": "Yujin Tang, Chenming Shang, Ruize Xu, Nikhil Singh",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.27499",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-25",
+      "updated_at": "2026-06-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "memory",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV",
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-27632-yuvion-llm-an-adversarially-aware-large-language-model-for-content-and-ai-safety.md",
+    "title": "\"Yuvion LLM: An Adversarially-Aware Large Language Model for Content And AI Safety\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Yuvion LLM: An Adversarially-Aware Large Language Model for Content And AI Safety\"",
+      "authors": "Ting Ma, Xiufeng Huang, Benlei Cui, Xiaowen Xu, Shikai Qiu, Ruijie Jian, Hongxing Li, Guanghui Wang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.27632",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-26",
+      "updated_at": "2026-06-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-27806-agent-vs-parametric-world-models-hybrid-planning-for-reliable-language-agents.md",
+    "title": "\"Agent vs. Parametric World Models: Hybrid Planning for Reliable Language Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Agent vs. Parametric World Models: Hybrid Planning for Reliable Language Agents\"",
+      "authors": "Xinyuan Song, Zekun Cai",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.27806",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-26",
+      "updated_at": "2026-07-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-27929-when-multi-robot-systems-meet-agentic-ai-towards-embodied-collective-intelligenc.md",
+    "title": "\"When Multi-Robot Systems Meet Agentic AI:Towards Embodied Collective Intelligence\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"When Multi-Robot Systems Meet Agentic AI:Towards Embodied Collective Intelligence\"",
+      "authors": "Yuxuan Yan, Yuanyuan Jia, Qianqian Yang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.27929",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-26",
+      "updated_at": "2026-06-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "embodied-agent",
+        "memory",
+        "multi-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.RO"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-27990-advancedshellm-a-stateful-multi-agent-llm-honeypot-for-ssh-deception.md",
+    "title": "\"AdvancedShelLM: A Stateful Multi-Agent LLM Honeypot for SSH Deception\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AdvancedShelLM: A Stateful Multi-Agent LLM Honeypot for SSH Deception\"",
+      "authors": "Muris Sladić, Eman Alibalić, Veronica Valeros, Carlos Catania, Sebastian Garcia",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.27990",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-26",
+      "updated_at": "2026-06-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "multi-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "llm-agent, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28011-from-detection-to-action-using-llm-agents-for-fault-tolerant-control.md",
+    "title": "\"From Detection to Action: Using LLM Agents for Fault-Tolerant Control\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"From Detection to Action: Using LLM Agents for Fault-Tolerant Control\"",
+      "authors": "Javal Vyas, Milapji Singh Gill, Artan Markaj, Felix Gehlhoff, Mehmet Mercangöz",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28011",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-26",
+      "updated_at": "2026-06-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "planning",
+        "rag",
+        "tool-use",
+        "workflow-agent",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "eess.SY",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "24",
+      "collection_queries": "agentic-ai, llm-agent, multi-agent-llm, planning-agent, rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28061-toolprivacybench-benchmarking-purpose-bound-privacy-in-tool-using-llm-agents.md",
+    "title": "\"ToolPrivacyBench: Benchmarking Purpose-Bound Privacy in Tool-Using LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ToolPrivacyBench: Benchmarking Purpose-Bound Privacy in Tool-Using LLM Agents\"",
+      "authors": "Shijing Hu, Liang Liu, Zhu Meng, Zhicheng Zhao",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28061",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-26",
+      "updated_at": "2026-06-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "22",
+      "collection_queries": "agent-evaluation, function-calling, llm-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28182-llawco-learning-laws-of-cooperation-for-modeling-embodied-multi-agent-behavior.md",
+    "title": "\"LLawCo: Learning Laws of Cooperation for Modeling Embodied Multi-Agent Behavior\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"LLawCo: Learning Laws of Cooperation for Modeling Embodied Multi-Agent Behavior\"",
+      "authors": "Qinhong Zhou, Chuang Gan, Anoop Cherian",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28182",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-26",
+      "updated_at": "2026-06-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "embodied-agent",
+        "multi-agent",
+        "planning",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI",
+        "cs.CV",
+        "cs.RO"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28187-gbc-gradient-based-connections-for-optimizing-multi-agent-systems.md",
+    "title": "\"GBC: Gradient-Based Connections for Optimizing Multi-Agent Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"GBC: Gradient-Based Connections for Optimizing Multi-Agent Systems\"",
+      "authors": "Xiaocheng Yang, Abdulrahman Alrabah, Dilek Hakkani-Tür, Gokhan Tur",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28187",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-26",
+      "updated_at": "2026-06-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "multi-agent",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28270-agent-native-immune-system-architecture-taxonomy-and-engineering.md",
+    "title": "\"Agent-Native Immune System: Architecture, Taxonomy, and Engineering\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Agent-Native Immune System: Architecture, Taxonomy, and Engineering\"",
+      "authors": "Bo Shen, Lifeng Chang, Tianyuan Wei, Yunpeng Li, Feng Shi, Yichen Han, Peijie Gao, Shiyi Kuang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28270",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-26",
+      "updated_at": "2026-06-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "multi-agent",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28279-agentic-hardware-design-as-repository-level-code-evolution.md",
+    "title": "Agentic Hardware Design as Repository-Level Code Evolution",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Agentic Hardware Design as Repository-Level Code Evolution",
+      "authors": "Cunxi Yu, Chenhui Deng, Nathaniel Pinckney, Brucek Khailany",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28279",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-26",
+      "updated_at": "2026-06-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28349-hmars-a-hierarchical-multi-agent-memory-system-for-long-context-reasoning.md",
+    "title": "\"HMARS: A Hierarchical Multi-Agent Memory System for Long-Context Reasoning\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"HMARS: A Hierarchical Multi-Agent Memory System for Long-Context Reasoning\"",
+      "authors": "Zeju Li, Ziyang Zheng, Yizhou Zhou, Qiang Xu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28349",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-03",
+      "updated_at": "2026-06-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "multi-agent",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.IR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28360-carolina-guide-a-multi-agent-rag-system-with-institutional-guardrails-for-academ.md",
+    "title": "\"Carolina Guide: A Multi-Agent RAG System with Institutional Guardrails for Academic Policy Assistance\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Carolina Guide: A Multi-Agent RAG System with Institutional Guardrails for Academic Policy Assistance\"",
+      "authors": "Ben Torsion, Jun Zhou",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28360",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-11",
+      "updated_at": "2026-06-11",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "multi-agent",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.IR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28374-recursive-self-evolving-agents-via-held-out-selection.md",
+    "title": "Recursive Self-Evolving Agents via Held-Out Selection",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Recursive Self-Evolving Agents via Held-Out Selection",
+      "authors": "Michael Nguyen, Quoc Nguyen, Paul Vuong",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28374",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-17",
+      "updated_at": "2026-06-17",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28409-evidence-driven-llm-agent-for-c-to-synthesizable-c-conversion-and-verification.md",
+    "title": "Evidence-Driven LLM Agent for C-to-Synthesizable-C Conversion and Verification",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Evidence-Driven LLM Agent for C-to-Synthesizable-C Conversion and Verification",
+      "authors": "Zhe Zhao, Hongbing Lang, Zhihan Xiao, Luke Ztz Hu, John Imoleayo Adebisi, Songping Mai",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28409",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-25",
+      "updated_at": "2026-06-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28425-tool-use-enables-undetectable-steganography-in-multi-agent-llm-systems.md",
+    "title": "Tool Use Enables Undetectable Steganography in Multi-Agent LLM Systems",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Tool Use Enables Undetectable Steganography in Multi-Agent LLM Systems",
+      "authors": "Jimmy Laurence Rippin, Simon C. Marshall, David Demitri Africa, Christian Schroeder de Witt",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28425",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-25",
+      "updated_at": "2026-06-25",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "computer-use",
+        "multi-agent",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "25",
+      "collection_queries": "agentic-ai, ai-agent, autonomous-agent-llm, multi-agent-llm, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28430-building-to-the-test-coding-agents-deliver-what-you-check-not-what-you-requested.md",
+    "title": "\"Building to the Test: Coding Agents Deliver What You Check, Not What You Requested\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Building to the Test: Coding Agents Deliver What You Check, Not What You Requested\"",
+      "authors": "Yanuo Ma, Ben Kereopa-Yorke, Ben Schultz",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28430",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-26",
+      "updated_at": "2026-06-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28434-swe-mem-learning-adaptive-memory-management-for-long-horizon-coding-agents.md",
+    "title": "\"SWE-MeM: Learning Adaptive Memory Management for Long-Horizon Coding Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SWE-MeM: Learning Adaptive Memory Management for Long-Horizon Coding Agents\"",
+      "authors": "Shuzheng Gao, Wenhao Zeng, Zhaojian Yu, Jianqiao Wangni, Chaozheng Wang, Kai Cai, Shilin He, Michael R. Lyu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28434",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-26",
+      "updated_at": "2026-06-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "coding-agent",
+        "memory",
+        "planning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28436-dockerless-environment-free-program-verifier-for-coding-agents.md",
+    "title": "\"Dockerless: Environment-Free Program Verifier for Coding Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Dockerless: Environment-Free Program Verifier for Coding Agents\"",
+      "authors": "Wenhao Zeng, Yuling Shi, Xiaodong Gu, Chao Hu, Chaofan Wang, Yuhao Cui, Hongting Zhou, Mengnan Qi, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28436",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-26",
+      "updated_at": "2026-06-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28450-llm-agents-security-duality-a-comprehensive-survey-of-self-security-and-empowere.md",
+    "title": "\"LLM agents security duality: a comprehensive survey of self-security and empowered cybersecurity\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"LLM agents security duality: a comprehensive survey of self-security and empowered cybersecurity\"",
+      "authors": "Yiwei Xu, Yong Zhuang, Xuanming Liu, Tian Zhang, Bowen Xiao, Xiaoyang Xu, Delong Jiang, Juan Wang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28450",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-26",
+      "updated_at": "2026-06-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-safety, llm-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28456-is-lying-an-emergent-behaviour-in-llms-evidence-from-gaslighting-ai-agents-in-a-.md",
+    "title": "Is Lying an Emergent Behaviour in LLMs? Evidence from Gaslighting AI agents in a Sustainability Game",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Is Lying an Emergent Behaviour in LLMs? Evidence from Gaslighting AI agents in a Sustainability Game",
+      "authors": "Subhendu Bhandary, Federico Carucci, Christos Charalambous, Francesca Dilisante, Ksenia Dvorkina, Anna Garbo, Jiaqi Liang, Riccardo Vasellini, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28456",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-26",
+      "updated_at": "2026-06-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "memory",
+        "multi-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "ai-agent, llm-agent, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28467-an-agentic-ai-pipeline-for-appliance-level-energy-anomaly-detection-and-llm-driv.md",
+    "title": "An Agentic AI Pipeline for Appliance-Level Energy Anomaly Detection and LLM-Driven Recommendations",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "An Agentic AI Pipeline for Appliance-Level Energy Anomaly Detection and LLM-Driven Recommendations",
+      "authors": "Dihia Falouz, Aida Douaibia, Amine Bechar, Youssef Elmir, Abbes Amira, Adel Oulefki",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28467",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-26",
+      "updated_at": "2026-06-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agentic-ai, rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28480-tua-bench-a-benchmark-for-general-purpose-terminal-use-agents.md",
+    "title": "\"TUA-Bench: A Benchmark for General-Purpose Terminal-Use Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"TUA-Bench: A Benchmark for General-Purpose Terminal-Use Agents\"",
+      "authors": "Shoufa Chen, Luyuan Wang, Xuan Yang, Zhiheng Liu, Yuren Cong, Yuanfeng Ji, Feiyan Zhou, Xiaohui Zhang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28480",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-26",
+      "updated_at": "2026-06-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "computer-use",
+        "reasoning",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28570-digitizing-coaching-intelligence-an-agentic-framework-for-holistic-athlete-profi.md",
+    "title": "\"Digitizing Coaching Intelligence: An Agentic Framework for Holistic Athlete Profiling using VLM and RAG\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Digitizing Coaching Intelligence: An Agentic Framework for Holistic Athlete Profiling using VLM and RAG\"",
+      "authors": "Deep Ghosal, Ishani Sen, Wazib Ansar, Amlan Chakrabarti",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28570",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-26",
+      "updated_at": "2026-06-26",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV",
+        "cs.AI",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "multi-agent-llm, rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28666-why-trust-your-agent-empirical-security-gains-from-trism-guided-agentic-workflow.md",
+    "title": "Why Trust Your Agent? Empirical Security Gains from TRiSM-Guided Agentic Workflows in Healthcare",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Why Trust Your Agent? Empirical Security Gains from TRiSM-Guided Agentic Workflows in Healthcare",
+      "authors": "Liam Kearns",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28666",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-27",
+      "updated_at": "2026-06-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agentic-ai, rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28679-capability-gates-are-not-authorization-confused-deputy-failures-in-llm-agent-fra.md",
+    "title": "\"Capability Gates Are Not Authorization: Confused-Deputy Failures in LLM Agent Frameworks\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Capability Gates Are Not Authorization: Confused-Deputy Failures in LLM Agent Frameworks\"",
+      "authors": "David Mellafe Zuvic",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28679",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-27",
+      "updated_at": "2026-06-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "llm-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28692-an-ai-agent-for-treatment-reasoning-over-a-biomedical-tool-universe.md",
+    "title": "An AI agent for treatment reasoning over a biomedical tool universe",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "An AI agent for treatment reasoning over a biomedical tool universe",
+      "authors": "Shanghua Gao, Ayush Noori, Richard Zhu, Curtis Ginder, Zhenglun Kong, Xiaorui Su, Justin Kauffman, Benjamin S. Glicksberg, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28692",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-27",
+      "updated_at": "2026-06-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "ai-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28733-agentic-abstention-do-agents-know-when-to-stop-instead-of-act.md",
+    "title": "\"Agentic Abstention: Do Agents Know When to Stop Instead of Act?\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Agentic Abstention: Do Agents Know When to Stop Instead of Act?\"",
+      "authors": "Han Luo, Bingbing Wen, Lucy Lu Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28733",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-27",
+      "updated_at": "2026-06-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28739-agent-safety-is-action-alignment.md",
+    "title": "Agent Safety Is Action Alignment",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Agent Safety Is Action Alignment",
+      "authors": "Shawn Li, Yue Zhao",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28739",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-27",
+      "updated_at": "2026-06-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-safety"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28781-hyphaedb-a-living-knowledge-topology-for-agent-first-memory.md",
+    "title": "\"HyphaeDB: A Living Knowledge Topology for Agent-First Memory\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"HyphaeDB: A Living Knowledge Topology for Agent-First Memory\"",
+      "authors": "Krishna Halaharvi",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28781",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-27",
+      "updated_at": "2026-06-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "coding-agent",
+        "memory",
+        "multi-agent",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-memory, agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28791-from-determinism-to-delegation-ai-native-software-engineering-and-the-evolution-.md",
+    "title": "\"From Determinism to Delegation: AI-Native Software Engineering and the Evolution of the Agentic Engineer\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"From Determinism to Delegation: AI-Native Software Engineering and the Evolution of the Agentic Engineer\"",
+      "authors": "Mamdouh Alenezi",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28791",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-27",
+      "updated_at": "2026-06-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "memory",
+        "multi-agent",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "23",
+      "collection_queries": "agentic-ai, autonomous-agent-llm, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28839-the-contagion-tensor-a-framework-for-measuring-output-distribution-coupling-in-m.md",
+    "title": "\"The Contagion Tensor: A Framework for Measuring Output-Distribution Coupling in Multi-Agent LLM Systems -- and Auditing the Claims It Enables\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"The Contagion Tensor: A Framework for Measuring Output-Distribution Coupling in Multi-Agent LLM Systems -- and Auditing the Claims It Enables\"",
+      "authors": "Zewen Liu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28839",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-27",
+      "updated_at": "2026-06-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "multi-agent",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28841-lamp-lean-based-agentic-framework-with-mcp-and-proof-repair.md",
+    "title": "\"LAMP: Lean-based Agentic framework with MCP and Proof Repair\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"LAMP: Lean-based Agentic framework with MCP and Proof Repair\"",
+      "authors": "Santhana Srinivasan R, Maithilee Patawar",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28841",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-27",
+      "updated_at": "2026-06-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "coding-agent",
+        "multi-agent",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LO",
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28896-a-task-driven-and-quality-assured-agent-framework-for-sar-data-generation.md",
+    "title": "A Task-Driven and Quality-Assured Agent Framework for SAR Data Generation",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "A Task-Driven and Quality-Assured Agent Framework for SAR Data Generation",
+      "authors": "Xuanting Wu, Fan Zhanga, Fei Ma, Ling Guan, Guochun Ma, Yongsheng Zhou",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28896",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-27",
+      "updated_at": "2026-06-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "eess.IV",
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28925-multi-agent-routing-as-set-valued-prediction-a-wildchat-benchmark-and-cost-aware.md",
+    "title": "\"Multi-Agent Routing as Set-Valued Prediction: A WildChat Benchmark and Cost-Aware Evaluation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Multi-Agent Routing as Set-Valued Prediction: A WildChat Benchmark and Cost-Aware Evaluation\"",
+      "authors": "Ananto Nayan Bala, Faisal Muhammad Shah",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28925",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-27",
+      "updated_at": "2026-06-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "rag",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI",
+        "cs.IR",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "21",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-28958-when-latent-agents-lie-kv-cache-integrity-in-multi-agent-llm-collaboration.md",
+    "title": "\"When Latent Agents Lie: KV-Cache Integrity in Multi-Agent LLM Collaboration\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"When Latent Agents Lie: KV-Cache Integrity in Multi-Agent LLM Collaboration\"",
+      "authors": "Luís Brito, Carlos Baquero",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.28958",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-27",
+      "updated_at": "2026-06-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "memory",
+        "multi-agent",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "llm-agent, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29014-customized-generative-ai-agent-for-transportation-engineering-practice-a-develop.md",
+    "title": "\"Customized Generative AI Agent for Transportation Engineering Practice: A Development and Continued Pre-training Guideline\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Customized Generative AI Agent for Transportation Engineering Practice: A Development and Continued Pre-training Guideline\"",
+      "authors": "Dianwei Chen, Yuan-Zheng Lei, Zifan Zhang, Yuchen Liu, Xianfeng Yang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29014",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-27",
+      "updated_at": "2026-07-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "planning",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.DL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29026-preventing-error-propagation-in-multi-agent-ai-through-runtime-monitoring.md",
+    "title": "Preventing Error Propagation in Multi-Agent AI through Runtime Monitoring",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Preventing Error Propagation in Multi-Agent AI through Runtime Monitoring",
+      "authors": "Shahnewaz Karim Sakib, Anindya Bijoy Das",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29026",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-27",
+      "updated_at": "2026-06-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.ET"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29030-memory-as-an-attack-surface-in-llm-agents-a-study-on-multiple-choice-question-an.md",
+    "title": "\"Memory as an Attack Surface in LLM Agents: A Study on Multiple-Choice Question Answering\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Memory as an Attack Surface in LLM Agents: A Study on Multiple-Choice Question Answering\"",
+      "authors": "Shahnewaz Karim Sakib, Anindya Bijoy Das",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29030",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-27",
+      "updated_at": "2026-06-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "memory",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.ET"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "ai-agent, llm-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29116-characterizing-large-language-model-agentic-workflows-a-study-on-n8n-ecosystem.md",
+    "title": "\"Characterizing Large Language Model Agentic Workflows: A Study on N8n Ecosystem\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Characterizing Large Language Model Agentic Workflows: A Study on N8n Ecosystem\"",
+      "authors": "Yutian Tang, Yuming Zhou, Huaming Chen",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29116",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-27",
+      "updated_at": "2026-06-27",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "planning",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "agentic-ai, llm-agent, planning-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29142-agent-security-meets-regulatory-reality-a-practitioner-systematization-of-autono.md",
+    "title": "Agent Security Meets Regulatory Reality -- A Practitioner Systematization of Autonomous-Agent Threats and Controls in Regulated Financial Systems",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Agent Security Meets Regulatory Reality -- A Practitioner Systematization of Autonomous-Agent Threats and Controls in Regulated Financial Systems",
+      "authors": "Krishna Mohan, Guda Nagavenkata Srinivasa",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29142",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-28",
+      "updated_at": "2026-06-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "computer-use",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CY",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-safety, autonomous-agent-llm, rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29178-selective-memory-retention-for-long-horizon-llm-agents.md",
+    "title": "Selective Memory Retention for Long-Horizon LLM Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Selective Memory Retention for Long-Horizon LLM Agents",
+      "authors": "Pranath Reddy",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29178",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-28",
+      "updated_at": "2026-06-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29193-a-multi-dataset-benchmark-for-evaluating-llm-agents-in-microservice-failure-diag.md",
+    "title": "A Multi-Dataset Benchmark for Evaluating LLM Agents in Microservice Failure Diagnosis",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "A Multi-Dataset Benchmark for Evaluating LLM Agents in Microservice Failure Diagnosis",
+      "authors": "Yuanhong Cai, Xiaohui Nie, Kanglin Yin, Changhua Pei, Yongqian Sun, Shenglin Zhang, Haibin Liu, Guiyang Liu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29193",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-28",
+      "updated_at": "2026-06-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29225-policyguard-a-dialogue-grounded-sub-agent-verifier-for-policy-adherence-in-llm-a.md",
+    "title": "\"PolicyGuard: A Dialogue-Grounded Sub-Agent Verifier for Policy Adherence in LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"PolicyGuard: A Dialogue-Grounded Sub-Agent Verifier for Policy Adherence in LLM Agents\"",
+      "authors": "Seongjae Kang, Taehyung Yu, Sung Ju Hwang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29225",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-28",
+      "updated_at": "2026-06-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "computer-use",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29270-minority-sentinel-when-to-overturn-majority-voting-in-multi-agent-llm-debates.md",
+    "title": "\"Minority Sentinel: When to Overturn Majority Voting in Multi-Agent LLM Debates\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Minority Sentinel: When to Overturn Majority Voting in Multi-Agent LLM Debates\"",
+      "authors": "Chuan He, Zebin Chen, Zhengyi Yang, Shaobo Qiao, Mingchen Ju, Jiate Liu, Dong Wen, Guanfeng Liu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29270",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-28",
+      "updated_at": "2026-06-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "llm-agent, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29315-hierarchical-experimentalist-agents.md",
+    "title": "Hierarchical Experimentalist Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Hierarchical Experimentalist Agents",
+      "authors": "Abhranil Chandra, Sankaran Vaidyanathan, Utsav Dhanuka, Varun Gandhi, Scott Niekum",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29315",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-28",
+      "updated_at": "2026-06-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29354-when-llms-develop-languages-symbolic-communication-for-efficient-multi-agent-rea.md",
+    "title": "\"When LLMs Develop Languages: Symbolic Communication for Efficient Multi-Agent Reasoning\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"When LLMs Develop Languages: Symbolic Communication for Efficient Multi-Agent Reasoning\"",
+      "authors": "Zhengqi Pei, Qingming Huang, Shuhui Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29354",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-28",
+      "updated_at": "2026-06-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.NE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "llm-agent, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29445-bridging-videoqa-and-video-guided-agentic-tasks-via-generalized-keyframe-extract.md",
+    "title": "Bridging VideoQA and Video-Guided Agentic Tasks via Generalized Keyframe Extraction",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Bridging VideoQA and Video-Guided Agentic Tasks via Generalized Keyframe Extraction",
+      "authors": "Sunqi Fan, Qingle Liu, Runqi Yin, Meng-Hao Guo, Shuojin Yang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29445",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-28",
+      "updated_at": "2026-06-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29495-cognitive-world-models-for-process-level-social-influence-evaluation.md",
+    "title": "Cognitive World Models for Process-Level Social Influence Evaluation",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Cognitive World Models for Process-Level Social Influence Evaluation",
+      "authors": "Minghui Ma, Bin Guo, Han Wang, Mengqi Chen, Jingqi Liu, Yan Liu, Zhiwen Yu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29495",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-28",
+      "updated_at": "2026-06-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "multi-agent",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29537-osworld2-0-benchmarking-computer-use-agents-on-long-horizon-real-world-tasks.md",
+    "title": "\"OSWorld2.0: Benchmarking Computer Use Agents on Long-Horizon Real-World Tasks\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"OSWorld2.0: Benchmarking Computer Use Agents on Long-Horizon Real-World Tasks\"",
+      "authors": "Mengqi Yuan, Zilong Zhou, Xinzhuang Xiong, Weiming Wu, Jiayang Sun, Jiamin Song, Kaiqian Cui, Bowen Wang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29537",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-28",
+      "updated_at": "2026-06-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "computer-use",
+        "memory",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "21",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29654-budgeted-act-or-defer-multi-agent-llm-deliberation-with-local-reliability-bounds.md",
+    "title": "Budgeted Act-or-Defer Multi-Agent LLM Deliberation with Local Reliability Bounds",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Budgeted Act-or-Defer Multi-Agent LLM Deliberation with Local Reliability Bounds",
+      "authors": "Mengdie Flora Wang, Haochen Xie, Guanghui Wang, Devin Zhang, Jae Oh Woo",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29654",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-28",
+      "updated_at": "2026-06-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29719-a-diagnostic-framework-and-multi-evaluator-audit-of-evaluator-driven-preference-.md",
+    "title": "A Diagnostic Framework and Multi-Evaluator Audit of Evaluator-Driven Preference Dynamics in Self-Adapting LLM Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "A Diagnostic Framework and Multi-Evaluator Audit of Evaluator-Driven Preference Dynamics in Self-Adapting LLM Agents",
+      "authors": "Liu Zewen",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29719",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29742-microagent-context-augmented-multi-agent-framework-for-automatic-microservice-de.md",
+    "title": "\"MicroAgent: Context-Augmented Multi-Agent Framework for Automatic Microservice Decomposition\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MicroAgent: Context-Augmented Multi-Agent Framework for Automatic Microservice Decomposition\"",
+      "authors": "Zishan Su, Junjie Huang, Shiwen Shan, Xingyan Chen, Hui Zeng, Yuxin Su, Yanlin Wang, Michael R. Lyu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29742",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "computer-use",
+        "multi-agent",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29745-echo-learning-epistemically-adaptive-language-agents-with-turn-level-credit.md",
+    "title": "\"ECHO: Learning Epistemically Adaptive Language Agents with Turn-Level Credit\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ECHO: Learning Epistemically Adaptive Language Agents with Turn-Level Credit\"",
+      "authors": "Abhijnan Nath, Nikhil Krishnaswamy",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29745",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29746-deepmed-search-an-open-source-agentic-platform-for-medical-deep-research-with-in.md",
+    "title": "\"DEEPMED Search: An Open-Source Agentic Platform for Medical Deep Research with Introspective Verification\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"DEEPMED Search: An Open-Source Agentic Platform for Medical Deep Research with Introspective Verification\"",
+      "authors": "Maolin Liu, Fanyu Xu, Ruoqing Xu, Jiahang Zhang, Hao Wang, Rui Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29746",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "computer-use",
+        "multi-agent",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.HC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29762-do-recommendation-algorithms-work-when-users-are-llm-agents-a-case-study-on-molt.md",
+    "title": "Do Recommendation Algorithms Work When Users Are LLM Agents? A Case Study on Moltbook",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Do Recommendation Algorithms Work When Users Are LLM Agents? A Case Study on Moltbook",
+      "authors": "Daming Li, Simeng Han, Jialu Zhang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29762",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.IR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "ai-agent, llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29771-clqt-a-closed-loop-cost-aware-strategy-consistent-benchmark-for-diagnostic-evalu.md",
+    "title": "\"CLQT: A Closed-Loop, Cost-Aware, Strategy-Consistent Benchmark for Diagnostic Evaluation of LLM Portfolio-Management Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"CLQT: A Closed-Loop, Cost-Aware, Strategy-Consistent Benchmark for Diagnostic Evaluation of LLM Portfolio-Management Agents\"",
+      "authors": "Bo Qu, Mingguang Chen",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29771",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.LG",
+        "q-fin.CP",
+        "q-fin.PM"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29774-analytic-concept-centric-memory-for-agentic-embodied-manipulation.md",
+    "title": "Analytic Concept-Centric Memory for Agentic Embodied Manipulation",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Analytic Concept-Centric Memory for Agentic Embodied Manipulation",
+      "authors": "Mingyang Sun, Xiujian Liang, Jiude Wei, Qichen He, Donglin Wang, Cewu Lu, Jianhua Sun",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29774",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "embodied-agent",
+        "memory",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.RO"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29778-mandol-an-agglomerative-agent-memory-system-for-long-term-conversations.md",
+    "title": "\"Mandol: An Agglomerative Agent Memory System for Long-Term Conversations\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Mandol: An Agglomerative Agent Memory System for Long-Term Conversations\"",
+      "authors": "Yuhan Zhang, Zhiyuan Guo, Ziheng Zeng, Wei Wang, Wentao Wu, Lijie Xu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29778",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.DB",
+        "cs.AI",
+        "cs.CL",
+        "cs.IR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-memory, rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29788-memleak-diagnosing-information-leaks-in-multimodal-agent-memory.md",
+    "title": "\"MemLeak: Diagnosing Information Leaks in Multimodal Agent Memory\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MemLeak: Diagnosing Information Leaks in Multimodal Agent Memory\"",
+      "authors": "Kuan Wang, Chao Zhang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29788",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-memory, ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29824-neural-procedural-memory-empowering-llm-agents-with-implicit-activation-steering.md",
+    "title": "\"Neural Procedural Memory: Empowering LLM Agents with Implicit Activation Steering\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Neural Procedural Memory: Empowering LLM Agents with Implicit Activation Steering\"",
+      "authors": "Chengfeng Zhao, Yuqiao Tan, Shizhu He, Yequan Wang, Jun Zhao, Kang Liu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29824",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "24",
+      "collection_queries": "agent-evaluation, agent-memory, autonomous-agent-llm, llm-agent, rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29894-saber-math-automated-benchmark-for-information-retrieval-evaluation-in-mathemati.md",
+    "title": "\"SABER-Math: Automated Benchmark for Information Retrieval Evaluation in Mathematics\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SABER-Math: Automated Benchmark for Information Retrieval Evaluation in Mathematics\"",
+      "authors": "Nikolay Georgiev, Maria Drencheva, Kseniia Ibragimova, Ivo Petrov, Dimitar I. Dimitrov, Martin Vechev",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29894",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.IR",
+        "cs.AI",
+        "cs.CL",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29914-memdelta-controlled-baselines-and-hidden-confounds-in-agent-memory-evaluation.md",
+    "title": "\"MemDelta: Controlled Baselines and Hidden Confounds in Agent Memory Evaluation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MemDelta: Controlled Baselines and Hidden Confounds in Agent Memory Evaluation\"",
+      "authors": "Kuan Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29914",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-memory, rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29932-saga-scene-aware-goal-evolving-agents-for-long-horizon-civrealm-strategy-plannin.md",
+    "title": "\"SAGA: Scene-Aware, Goal-Evolving Agents for Long-Horizon CivRealm Strategy Planning\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SAGA: Scene-Aware, Goal-Evolving Agents for Long-Horizon CivRealm Strategy Planning\"",
+      "authors": "Tianyu Jin, Shuo Chen, Yida Wang, Liuyu Xiang, Yingzhuo Liu, Zhiyao Jiang, Yexin Li, Zhaofeng He",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29932",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "multi-agent",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29957-swe-together-evaluating-coding-agents-in-interactive-user-sessions.md",
+    "title": "\"SWE-Together: Evaluating Coding Agents in Interactive User Sessions\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SWE-Together: Evaluating Coding Agents in Interactive User Sessions\"",
+      "authors": "Yifan Wu, Zhuokai Zhao, Songlin Li, Ho Hin Lee, Jiacheng Zhu, Shirley Wu, Tianhe Yu, Serena Li, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29957",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-evaluation, coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-29961-duomem-towards-capable-on-device-memory-agents-via-dual-space-distillation.md",
+    "title": "\"DuoMem: Towards Capable On-Device Memory Agents via Dual-Space Distillation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"DuoMem: Towards Capable On-Device Memory Agents via Dual-Space Distillation\"",
+      "authors": "Peyman Hosseini, Ondrej Bohdal, Ahmed Alajrami, Andrea Maracani, Ignacio Castro, Matthew Purver, Mete Ozay, Savas Ozkan, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.29961",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "embodied-agent",
+        "memory"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-30005-llm-agents-are-latent-context-managers-eliciting-self-managed-context-via-a-prop.md",
+    "title": "\"LLM Agents Are Latent Context Managers: Eliciting Self-Managed Context via a Proprioceptive Dashboard\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"LLM Agents Are Latent Context Managers: Eliciting Self-Managed Context via a Proprioceptive Dashboard\"",
+      "authors": "Binyan Xu, Haitao Li, Kehuan Zhang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.30005",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-07-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "memory",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-30111-automating-the-design-of-embodied-agent-architectures.md",
+    "title": "Automating the Design of Embodied Agent Architectures",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Automating the Design of Embodied Agent Architectures",
+      "authors": "Jian Zhou, Sihao Lin, Jin Li, Shuai Fu, Gengze Zhou, Qi Wu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.30111",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-07-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "embodied-agent",
+        "memory",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.RO",
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-30119-on-the-internet-nobody-knows-you-re-an-llm-bot-unmasking-web-agents-with-multi-l.md",
+    "title": "\"On the Internet, Nobody Knows You're an LLM Bot: Unmasking Web Agents with Multi-Layer Fingerprinting\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"On the Internet, Nobody Knows You're an LLM Bot: Unmasking Web Agents with Multi-Layer Fingerprinting\"",
+      "authors": "Iliana Fayolle, Sihem Bouhenniche, Samuel Pélissier, Pierre Laperdrix, Clémentine Maurice, Walter Rudametkin",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.30119",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "embodied-agent",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-30185-dynamo-dynamic-skill-tool-evolution-for-vision-language-agents.md",
+    "title": "\"Dynamo: Dynamic Skill-Tool Evolution for Vision-Language Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Dynamo: Dynamic Skill-Tool Evolution for Vision-Language Agents\"",
+      "authors": "Yutao Sun, Yanting Miao, Hao-Xuan Ma, Mengyu Zhou, Mingshuai Chen, Tiancheng Zhao, Dexin Wang, Lei Lv, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.30185",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-30251-taco-tool-augmented-credit-optimization-for-agentic-tool-use.md",
+    "title": "\"TACO: Tool-Augmented Credit Optimization for Agentic Tool Use\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"TACO: Tool-Augmented Credit Optimization for Agentic Tool Use\"",
+      "authors": "Mingkuan Feng, Jinyang Wu, Hao Gu, Fangrui Lv, Ruihan Jin, Chuyuan Zhang, Zhengqi Wen, Jianhua Tao",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.30251",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-30259-multi-agentic-system-leveraging-open-source-llms-to-mitigate-disinformation-thre.md",
+    "title": "Multi-Agentic System Leveraging Open-Source LLMs to Mitigate Disinformation Threats",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Multi-Agentic System Leveraging Open-Source LLMs to Mitigate Disinformation Threats",
+      "authors": "Sebastian Kula, Martin Tamajka",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.30259",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-30266-towards-continual-motion-language-agents-lora-variants-for-incremental-motion-un.md",
+    "title": "\"Towards Continual Motion-Language Agents: LoRA Variants for Incremental Motion Understanding and Generation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Towards Continual Motion-Language Agents: LoRA Variants for Incremental Motion Understanding and Generation\"",
+      "authors": "Bertram Taetz, Hugo Albuquerque Cosme da Silva, Gabriele Bleser-Taetz",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.30266",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "autonomous-agent-llm, language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-30294-rehearsed-multi-agent-live-product-demonstrations-with-real-time-voice-question-.md",
+    "title": "Rehearsed Multi-Agent Live Product Demonstrations with Real-Time Voice Question Answering",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Rehearsed Multi-Agent Live Product Demonstrations with Real-Time Voice Question Answering",
+      "authors": "Rahul Khedar, Mayank Malhotra, Avinash Karn, Mouli V, Prakhar Mehrotra",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.30294",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "computer-use",
+        "multi-agent",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.HC",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-30383-whose-side-is-your-agent-on-multi-party-principal-loyalty-in-llm-agents.md",
+    "title": "Whose Side Is Your Agent On? Multi-Party Principal Loyalty in LLM Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Whose Side Is Your Agent On? Multi-Party Principal Loyalty in LLM Agents",
+      "authors": "Bojie Li, Noah Shi",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.30383",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-30454-collective-cooperation-without-individual-fidelity-in-llm-agents.md",
+    "title": "Collective cooperation without individual fidelity in LLM agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Collective cooperation without individual fidelity in LLM agents",
+      "authors": "Henrique Ferraz de Arruda, Carlos Gracia Lázaro, Alberto Aleta, Yamir Moreno",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.30454",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "physics.soc-ph",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-30524-the-illusion-of-agentic-complexity-in-readme-md-generation-evaluating-single-age.md",
+    "title": "\"The Illusion of Agentic Complexity in README.md Generation: Evaluating Single-Agent vs. Multi-Agent RAG Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"The Illusion of Agentic Complexity in README.md Generation: Evaluating Single-Agent vs. Multi-Agent RAG Systems\"",
+      "authors": "Abu Saleh, Tesfay Welegebreal Tesfay, Phuong T. Nguyen, Juri Di Rocco, Muhammad Umar Zeshan, Davide Di Ruscio",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.30524",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "computer-use",
+        "multi-agent",
+        "planning",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "21",
+      "collection_queries": "multi-agent-llm, rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-30546-mas-lab-a-specification-driven-validation-framework-for-reliable-multi-agent-sys.md",
+    "title": "\"MAS-Lab: A Specification-Driven Validation Framework for Reliable Multi-Agent Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MAS-Lab: A Specification-Driven Validation Framework for Reliable Multi-Agent Systems\"",
+      "authors": "Jordan Augé, Giovanna Carofiglio, Giulio Grassi, Jacques Samain",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.30546",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "multi-agent",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-30555-linguistic-firewall-geometry-as-defense-in-multi-agent-systems-routing.md",
+    "title": "\"Linguistic Firewall: Geometry as Defense in Multi-Agent Systems Routing\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Linguistic Firewall: Geometry as Defense in Multi-Agent Systems Routing\"",
+      "authors": "Dvir Alsheich, Adar Peleg, Ben Hagag, Rom Himelstein, Amit Levi, Avi Mendelson",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.30555",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-07-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "multi-agent",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-30560-tracelab-characterizing-coding-agent-workloads-for-llm-serving.md",
+    "title": "\"TraceLab: Characterizing Coding Agent Workloads for LLM Serving\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"TraceLab: Characterizing Coding Agent Workloads for LLM Serving\"",
+      "authors": "Kan Zhu, Mathew Jacob, Chenxi Ma, Yi Pan, Stephanie Wang, Arvind Krishnamurthy, Baris Kasikci",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.30560",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI",
+        "cs.PF"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-30566-forensic-trajectory-signatures-for-agent-memory-poisoning-detection.md",
+    "title": "Forensic Trajectory Signatures for Agent Memory Poisoning Detection",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Forensic Trajectory Signatures for Agent Memory Poisoning Detection",
+      "authors": "Jun Wen Leong",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.30566",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-memory, llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-30573-swe-interact-reimagining-swe-benchmarks-as-user-driven-long-horizon-coding-sessi.md",
+    "title": "\"SWE-INTERACT: Reimagining SWE Benchmarks as User-Driven Long-Horizon Coding Sessions\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SWE-INTERACT: Reimagining SWE Benchmarks as User-Driven Long-Horizon Coding Sessions\"",
+      "authors": "Mohit Raghavendra, Anisha Gunjal, Aakash Sabharwal, Yunzhong He",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.30573",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "computer-use",
+        "planning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-30602-mesa-prioritizing-vulnerable-communication-channels-for-securing-multi-agent-sys.md",
+    "title": "\"MESA: Prioritizing Vulnerable Communication Channels for Securing Multi-Agent Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MESA: Prioritizing Vulnerable Communication Channels for Securing Multi-Agent Systems\"",
+      "authors": "Kunyang Li, Kyle Domico, Jonathan Gregory, Patrick McDaniel",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.30602",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-30616-scaling-the-horizon-not-the-parameters-reaching-trillion-parameter-performance-w.md",
+    "title": "\"Scaling the Horizon, Not the Parameters: Reaching Trillion-Parameter Performance with a 35B Agent\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Scaling the Horizon, Not the Parameters: Reaching Trillion-Parameter Performance with a 35B Agent\"",
+      "authors": "Lei Bai, Zongsheng Cao, Yang Chen, Zhiyao Cui, Shangheng Du, Yue Fan, Shiyang Feng, Zijie Guo, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.30616",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-30639-self-evolving-world-models-for-llm-agent-planning.md",
+    "title": "Self-Evolving World Models for LLM Agent Planning",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Self-Evolving World Models for LLM Agent Planning",
+      "authors": "Xuan Zhang, Wenxuan Zhang, See-Kiong Ng, Yang Deng",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.30639",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "llm-agent, planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-30697-lumos-a-semantic-operating-system-layer-for-accessibility-grounded-ai-agents.md",
+    "title": "\"LUMOS: A Semantic Operating-System Layer for Accessibility-Grounded AI Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"LUMOS: A Semantic Operating-System Layer for Accessibility-Grounded AI Agents\"",
+      "authors": "Yogeswar Reddy Thota",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.30697",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "computer-use",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.OS",
+        "cs.AI",
+        "cs.CV"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "ai-agent, web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-30755-understanding-and-evaluating-claw-like-agent-security-through-a-computer-systems.md",
+    "title": "Understanding and Evaluating Claw-like Agent Security Through a Computer-Systems Lens",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Understanding and Evaluating Claw-like Agent Security Through a Computer-Systems Lens",
+      "authors": "Peizhi Niu, Wenjie Qu, Shangding Gu, Tianneng Shi, Yuankai Li, Ahmad Tawaha, Hend Alzahrani, Vincent Siu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.30755",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-safety, ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-30840-contrastive-reflection-for-iterative-prompt-optimization.md",
+    "title": "Contrastive Reflection for Iterative Prompt Optimization",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Contrastive Reflection for Iterative Prompt Optimization",
+      "authors": "Derek Koh, Jinghui Mo, Benjamin H. Le, Jiening Zhan, Baofen Zheng, Kevin Bevis, Nathaniel C. Owen, Lauren Elizabeth Charney, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.30840",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "rag",
+        "reasoning",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "ai-agent, llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-30877-a-systematic-approach-to-multi-agent-ai-from-advanced-regulatory-control-theory-.md",
+    "title": "\"A Systematic Approach to Multi-Agent AI from Advanced Regulatory Control Theory: Safe and Auditable LLM Operator Agents for Process Control\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"A Systematic Approach to Multi-Agent AI from Advanced Regulatory Control Theory: Safe and Auditable LLM Operator Agents for Process Control\"",
+      "authors": "Idelfonso B. R. Nogueira, Sigurd Skogestad",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.30877",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "eess.SY",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agentic-ai, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-30887-training-therapeutic-judges-and-multi-agent-systems-for-human-aligned-mental-hea.md",
+    "title": "Training Therapeutic Judges and Multi-Agent Systems for Human-Aligned Mental Health Support",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Training Therapeutic Judges and Multi-Agent Systems for Human-Aligned Mental Health Support",
+      "authors": "Mizanur Rahman, Abeer Badawi, Elahe Rahimi, Laleh Seyyed-Kalantari, Frank Rudzicz, Enamul Hoque, Elham Dolatabadi",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.30887",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-30906-investigating-multi-agent-deliberation-in-law.md",
+    "title": "Investigating Multi-Agent Deliberation in Law",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Investigating Multi-Agent Deliberation in Law",
+      "authors": "Cor Steging, Ludi van Leeuwen, Tadeusz Zbiegień",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.30906",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "agentic-ai, ai-agent, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-30949-agrefactor-self-evolving-agentic-workflow-for-hls-compatibility-and-performance.md",
+    "title": "\"AgRefactor: Self-Evolving Agentic Workflow for HLS Compatibility and Performance\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AgRefactor: Self-Evolving Agentic Workflow for HLS Compatibility and Performance\"",
+      "authors": "Yang Zou, Zijian Ding, Yizhou Sun, Jason Cong",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.30949",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "multi-agent",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.AR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "agentic-ai, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-30970-behavioral-governance-for-autonomous-ai-agents-the-agentbound-framework.md",
+    "title": "\"Behavioral Governance for Autonomous AI Agents: The AgentBound Framework\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Behavioral Governance for Autonomous AI Agents: The AgentBound Framework\"",
+      "authors": "Anuj Kaul, Qianlong Lan, Pranay Gupta",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.30970",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-30986-the-organizational-behavior-of-agentic-ai-collective-intelligence-in-human-agent.md",
+    "title": "\"The Organizational Behavior of Agentic AI: Collective Intelligence in Human-Agent Workflows\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"The Organizational Behavior of Agentic AI: Collective Intelligence in Human-Agent Workflows\"",
+      "authors": "Canhui Liu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.30986",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "memory",
+        "planning",
+        "tool-use",
+        "workflow-agent",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CY",
+        "cs.HC",
+        "cs.MA",
+        "econ.GN"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agentic-ai, llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-31046-openlife-toward-open-world-artificial-life-with-autonomous-llm-agents.md",
+    "title": "\"OpenLife: Toward Open-World Artificial Life with Autonomous LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"OpenLife: Toward Open-World Artificial Life with Autonomous LLM Agents\"",
+      "authors": "Atsushi Masumori, Itsuki Doi, Norihiro Maruyama, Ryosuke Takata, Takashi Ikegami",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.31046",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "llm-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-31073-multiuav-plat-an-llm-oriented-platform-benchmark-and-framework-for-multi-uav-col.md",
+    "title": "\"MultiUAV-Plat: An LLM-Oriented Platform, Benchmark and Framework for Multi-UAV Collaborative Task Planning\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MultiUAV-Plat: An LLM-Oriented Platform, Benchmark and Framework for Multi-UAV Collaborative Task Planning\"",
+      "authors": "Sheng Zhang, Qinglin Li, Yuechao Zang, Xueqin Huang, Yijia Fu, Cheng Zhu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.31073",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "embodied-agent",
+        "memory",
+        "multi-agent",
+        "planning",
+        "rag",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.MA",
+        "cs.RO"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "agent-evaluation, llm-agent, planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-31085-ddiagents-mechanism-conditioned-context-flow-for-drug-drug-interaction-predictio.md",
+    "title": "\"DDIAgents: Mechanism-Conditioned Context Flow for Drug-Drug Interaction Prediction\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"DDIAgents: Mechanism-Conditioned Context Flow for Drug-Drug Interaction Prediction\"",
+      "authors": "Zhenqian Shen, Yu Liu, Xiaoyi Fu, Quanming Yao",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.31085",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-31134-beyond-the-library-an-agentic-framework-for-autoformalizing-research-mathematics.md",
+    "title": "\"Beyond the Library: An Agentic Framework for Autoformalizing Research Mathematics\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Beyond the Library: An Agentic Framework for Autoformalizing Research Mathematics\"",
+      "authors": "Arshia Soltani Moakhar, Iman Gholami, Max Springer, Mahdi JafariRaviz, MohammadTaghi Hajiaghayi",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.31134",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "multi-agent",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-31174-clawarena-team-benchmarking-subagent-orchestration-and-dynamic-workflows-in-lang.md",
+    "title": "\"ClawArena-Team: Benchmarking Subagent Orchestration and Dynamic Workflows in Language-Model Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ClawArena-Team: Benchmarking Subagent Orchestration and Dynamic Workflows in Language-Model Agents\"",
+      "authors": "Kaiwen Xiong, Haonian Ji, Shi Qiu, Zeyu Zheng, Cihang Xie, Xinyu Ye, Huaxiu Yao",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.31174",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "llm-agent, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-31179-healthagentbench-a-unified-benchmark-suite-of-realistic-agentic-healthcare-envir.md",
+    "title": "\"HealthAgentBench: A Unified Benchmark Suite of Realistic Agentic Healthcare Environments for Challenging Frontier AI Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"HealthAgentBench: A Unified Benchmark Suite of Realistic Agentic Healthcare Environments for Challenging Frontier AI Agents\"",
+      "authors": "Qianchu Liu, Sheng Zhang, Guanghui Qin, Jeya Maria Jose Valanarasu, Maximilian Rokuss, Mingyu Lu, Timothy Ossowski, Juan Manuel Zambrano Chaves, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.31179",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "reasoning",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL",
+        "cs.CV"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "agent-evaluation, ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-31200-agentic-rag-vlm-affordance-aware-retrieval-augmented-generation-with-self-reflec.md",
+    "title": "\"Agentic RAG-VLM: Affordance-Aware Retrieval-Augmented Generation with Self-Reflective Planning for Robotic Grasping\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Agentic RAG-VLM: Affordance-Aware Retrieval-Augmented Generation with Self-Reflective Planning for Robotic Grasping\"",
+      "authors": "Tao Chen, Lizheng Liu, Jiaxu Wang, Ziyue Jiang, Ruiqi Tian, JiGuang Huo, Zhongxue Gan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.31200",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "embodied-agent",
+        "planning",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-31209-long-term-traffic-simulation-via-structured-autoregressive-modeling.md",
+    "title": "Long-term Traffic Simulation via Structured Autoregressive Modeling",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Long-term Traffic Simulation via Structured Autoregressive Modeling",
+      "authors": "Lingyu Xiao, Zexin Feng, Xintao Yan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.31209",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "multi-agent",
+        "planning",
+        "rag",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.RO"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-31227-securing-the-ai-agent-a-unified-framework-for-multi-layer-agent-red-teaming.md",
+    "title": "\"Securing the AI Agent: A Unified Framework for Multi-Layer Agent Red Teaming\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Securing the AI Agent: A Unified Framework for Multi-Layer Agent Red Teaming\"",
+      "authors": "Yong Yang, Xing Zheng, Huiyu Wu, Huangsheng Cheng, Xiaorong Shi, Jing Guo, Bo Yang, Yi Zhou, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.31227",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-safety, ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-31229-agentic-ideation-sample-efficient-agentic-trajectories-synthesis-for-scientific-.md",
+    "title": "\"Agentic-Ideation: Sample Efficient Agentic Trajectories Synthesis for Scientific Ideation Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Agentic-Ideation: Sample Efficient Agentic Trajectories Synthesis for Scientific Ideation Agents\"",
+      "authors": "Keyu Zhao, Lingyan Kong, Fengli Xu, Yong Li",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.31229",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "computer-use",
+        "multi-agent",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "agentic-ai, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-31252-embodied-cad-solver-grounded-llm-agents-for-parametric-b-rep-assembly-modeling.md",
+    "title": "\"Embodied CAD: Solver-Grounded LLM Agents for Parametric B-Rep Assembly Modeling\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Embodied CAD: Solver-Grounded LLM Agents for Parametric B-Rep Assembly Modeling\"",
+      "authors": "Fumin Liu, Haoyu Zhou, Fei Hao, Lin Yang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.31252",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "embodied-agent",
+        "planning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "llm-agent, planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-31314-a-novel-method-for-differential-algebraic-dynamic-model-discovery-in-power-syste.md",
+    "title": "\"A Novel Method for Differential-Algebraic Dynamic Model Discovery in Power Systems: An LLM-Based Multi-Agent Collaborative Framework\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"A Novel Method for Differential-Algebraic Dynamic Model Discovery in Power Systems: An LLM-Based Multi-Agent Collaborative Framework\"",
+      "authors": "Xinming Wang, Fan Tang, Yingli Wei, Yakun He, Zhe Liu, Ping Jiang, Haoyu Wu, Zihan Guo, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.31314",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "multi-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "eess.SY"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-31339-verification-gated-agentic-mission-state-governance-for-intelligent-industrial-m.md",
+    "title": "Verification-Gated Agentic Mission-State Governance for Intelligent Industrial Multi-Robot Systems",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Verification-Gated Agentic Mission-State Governance for Intelligent Industrial Multi-Robot Systems",
+      "authors": "Guoqin Tang, Qingxuan Jia, Yichen Tan, Zeyuan Huang, Ning Ji, Gang Chen",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.31339",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "embodied-agent",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.RO"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-31410-xiaomi-gui-0-technical-report.md",
+    "title": "Xiaomi-GUI-0 Technical Report",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Xiaomi-GUI-0 Technical Report",
+      "authors": "Wanxia Cao, Chengzhen Duan, Pei Fu, Pengzhi Gao, Niu Lian, Fazhan Liu, Hui Liu, Heng Qu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.31410",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "embodied-agent",
+        "memory",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-31471-think-while-you-map-asynchronous-vision-language-agents-for-incremental-3d-scene.md",
+    "title": "\"Think While You Map: Asynchronous Vision-Language Agents for Incremental 3D Scene Graphs\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Think While You Map: Asynchronous Vision-Language Agents for Incremental 3D Scene Graphs\"",
+      "authors": "Deniz Bickici, Michael Pabst, Shohei Mori, Dieter Schmalstieg",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.31471",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-31612-what-memory-do-gui-agents-really-need-from-passive-records-to-active-task-drivin.md",
+    "title": "What Memory Do GUI Agents Really Need? From Passive Records to Active Task-Driving States",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "What Memory Do GUI Agents Really Need? From Passive Records to Active Task-Driving States",
+      "authors": "Chen Liu, Ling Chen, Hanzhang Zhou, Xu Zhang, Quyu Kong, Panrong Tong, Wenhao Wang, Xin Yu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.31612",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "planning",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "agent-memory, web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-31635-a-tutorial-on-autonomous-fault-tolerant-control-using-knowledge-grounded-llm-age.md",
+    "title": "A Tutorial on Autonomous Fault-Tolerant Control Using Knowledge-Grounded LLM Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "A Tutorial on Autonomous Fault-Tolerant Control Using Knowledge-Grounded LLM Agents",
+      "authors": "Javal Vyas, Milapji Singh Gill, Artan Markaj, Felix Gehlhoff, Mehmet Mercangöz",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.31635",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "planning",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "eess.SY",
+        "cs.AI",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-31639-a-lifecycle-and-application-stack-survey-of-large-language-model-vulnerabilities.md",
+    "title": "\"A Lifecycle and Application-Stack Survey of Large Language Model Vulnerabilities: Attacks, Risks, Defenses, and Open Problems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"A Lifecycle and Application-Stack Survey of Large Language Model Vulnerabilities: Attacks, Risks, Defenses, and Open Problems\"",
+      "authors": "Seyed Bagher Hashemi Natanzi, Bo Tang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.31639",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "embodied-agent",
+        "memory",
+        "planning",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI",
+        "cs.GT",
+        "cs.LO"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-evaluation, autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-31648-think-in-english-answer-in-korean-efficient-adaptation-of-multilingual-tool-usin.md",
+    "title": "\"Think in English, Answer in Korean: Efficient Adaptation of Multilingual Tool-Using Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Think in English, Answer in Korean: Efficient Adaptation of Multilingual Tool-Using Agents\"",
+      "authors": "Utsav Garg, Sungjin Hong, Jason Jung, Justin Lee, Shaan Desai, Joon Hee Kim, Anirudh Shrinivason, Edmond Wen, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.31648",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "memory",
+        "multi-agent",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agentic-ai, function-calling, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-31650-echo-prune-to-act-trace-to-learn-with-selective-turn-memory-in-agentic-rl.md",
+    "title": "\"ECHO: Prune to act, trace to learn with selective turn memory in agentic RL\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ECHO: Prune to act, trace to learn with selective turn memory in agentic RL\"",
+      "authors": "Zijun Xie, Binbin Zheng, Enlei Gong, Jihua Liu, Yuyang You, Lingfeng Liu, Jiayao Tang, Guanqun Zhao, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.31650",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-31665-forecastagentsearch-towards-a-multi-expert-agent-search-system-for-geopolitical-.md",
+    "title": "\"ForecastAgentSearch: Towards a Multi-Expert Agent Search System for Geopolitical Event Forecasting\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ForecastAgentSearch: Towards a Multi-Expert Agent Search System for Geopolitical Event Forecasting\"",
+      "authors": "Miaomiao Cai, He Chang, Yunshan Ma, See-kiong Ng",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.31665",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "planning",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-31693-shopx-a-foundation-model-for-intent-to-item-fulfillment-in-agentic-shopping.md",
+    "title": "\"ShopX: A Foundation Model for Intent-to-Item Fulfillment in Agentic Shopping\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ShopX: A Foundation Model for Intent-to-Item Fulfillment in Agentic Shopping\"",
+      "authors": "Jiacheng Chen, Tao Zhang, Manxi Lin, Dunxian Huang, Teng Shi, Honghao Fu, Mengyan Li, Xinming Zhang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.31693",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.IR",
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "llm-agent, planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-31744-a-conversational-agentic-interface-to-physics-based-household-digital-twins-for-.md",
+    "title": "A Conversational Agentic Interface to Physics-Based Household Digital Twins for Residential Energy Decision Support",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "A Conversational Agentic Interface to Physics-Based Household Digital Twins for Residential Energy Decision Support",
+      "authors": "Costas Mylonas, Titos Georgoulakis, Magda Foti",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.31744",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "eess.SY"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-31767-jeto-bench-a-reproducible-benchmark-for-execution-time-improvement-patches-in-ja.md",
+    "title": "\"JETO-Bench: A Reproducible Benchmark for Execution Time Improvement Patches in Java\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"JETO-Bench: A Reproducible Benchmark for Execution Time Improvement Patches in Java\"",
+      "authors": "Khashayar Etemadi, Zhendong Su",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.31767",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-31831-an-agentic-ai-framework-to-accelerate-scientific-discovery-in-plant-phenotyping.md",
+    "title": "An Agentic AI Framework to Accelerate Scientific Discovery in Plant Phenotyping",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "An Agentic AI Framework to Accelerate Scientific Discovery in Plant Phenotyping",
+      "authors": "Renan Souza, Daniel Rosendo, Kelsey Carter, John Lagergren, Frédéric Suter, Shelaine L. Curd, Gerald A. Tuskan, Rafael Ferreira da Silva, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.31831",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agentic-ai, ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-31916-theory-of-mind-and-persuasion-beyond-conversation-assessing-the-capacity-of-llms.md",
+    "title": "\"Theory of Mind and Persuasion Beyond Conversation: Assessing the Capacity of LLMs to Induce Belief States via Planning and Action\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Theory of Mind and Persuasion Beyond Conversation: Assessing the Capacity of LLMs to Induce Belief States via Planning and Action\"",
+      "authors": "Ben Slater, Matteo G. Mecattaf, Lucy G. Cheke, John Burden, Winnie Street",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.31916",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-31980-digitalcoach-communication-and-grounding-gaps-in-human-and-agentic-computer-use-.md",
+    "title": "\"DigitalCoach: Communication and Grounding Gaps in Human and Agentic Computer Use Coaching\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"DigitalCoach: Communication and Grounding Gaps in Human and Agentic Computer Use Coaching\"",
+      "authors": "Meng Chen, Anya Ji, Tsung-Han Wu, Tobias Maringgele, David M. Chan, Alane Suhr, Amy Pavel",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.31980",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI",
+        "cs.HC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-32025-generative-skill-composition-for-llm-agents.md",
+    "title": "Generative Skill Composition for LLM Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Generative Skill Composition for LLM Agents",
+      "authors": "Xinyu Zhao, Zhen Tan, Vaishnav Tadiparthi, Nakul Agarwal, Kwonjoon Lee, Ehsan Moradi Pari, Hossein Nourkhiz Mahjoub, Tianlong Chen",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.32025",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "planning",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "coding-agent, llm-agent, planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2606-32034-qval-cheaply-evaluating-dense-supervision-signals-for-long-horizon-llm-agents.md",
+    "title": "\"QVal: Cheaply Evaluating Dense Supervision Signals for Long-Horizon LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"QVal: Cheaply Evaluating Dense Supervision Signals for Long-Horizon LLM Agents\"",
+      "authors": "Sergio Hernández-Gutiérrez, Matteo Merler, Ilze Amanda Auzina, Joschka Strüber, Ameya Prabhu, Matthias Bethge",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2606.32034",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-00038-stop-hand-holding-your-coding-agent-engineering-the-loops-that-replace-step-by-s.md",
+    "title": "\"Stop Hand-Holding Your Coding Agent: Engineering the Loops that Replace Step-by-Step Prompting\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Stop Hand-Holding Your Coding Agent: Engineering the Loops that Replace Step-by-Step Prompting\"",
+      "authors": "Sandeco Macedo",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.00038",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-28",
+      "updated_at": "2026-06-28",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "coding-agent",
+        "computer-use",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-00041-atm-cid-brokered-pre-write-admission-for-multi-agent-code-co-synthesis.md",
+    "title": "\"ATM: CID-Brokered Pre-Write Admission for Multi-Agent Code Co-Synthesis\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ATM: CID-Brokered Pre-Write Admission for Multi-Agent Code Co-Synthesis\"",
+      "authors": "Eagl Huang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.00041",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-29",
+      "updated_at": "2026-06-29",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "computer-use",
+        "multi-agent",
+        "planning",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agent-evaluation, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-00233-from-signals-to-structure-how-memory-architecture-drives-language-emergence-in-l.md",
+    "title": "\"From Signals to Structure: How Memory Architecture Drives Language Emergence in LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"From Signals to Structure: How Memory Architecture Drives Language Emergence in LLM Agents\"",
+      "authors": "Yashar Talebirad, Eden Redman, Ali Parsaee, Osmar R. Zaiane",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.00233",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL",
+        "cs.IT",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-00255-slm-llm-or-agentic-ai-toward-intelligent-uav-enabled-wpt-systems-in-low-altitude.md",
+    "title": "SLM, LLM or Agentic AI? Toward Intelligent UAV-Enabled WPT Systems in Low-Altitude Economy Networks",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "SLM, LLM or Agentic AI? Toward Intelligent UAV-Enabled WPT Systems in Low-Altitude Economy Networks",
+      "authors": "Feibo Jiang, Li Dong, Lei Mao, Kezhi Wang, Xianbin Wang, Abbas Jamalipour",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.00255",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "reasoning",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.IT"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-00297-epc-a-standardized-protocol-for-measuring-evaluator-preference-dynamics-in-llm-a.md",
+    "title": "\"EPC: A Standardized Protocol for Measuring Evaluator Preference Dynamics in LLM Agent Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"EPC: A Standardized Protocol for Measuring Evaluator Preference Dynamics in LLM Agent Systems\"",
+      "authors": "Zewen Liu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.00297",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "computer-use",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-00334-managed-autonomy-at-runtime-gear-based-safety-and-governance-for-single-and-mult.md",
+    "title": "\"Managed Autonomy at Runtime: Gear-Based Safety and Governance for Single- and Multi-Agent Cyber-Physical Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Managed Autonomy at Runtime: Gear-Based Safety and Governance for Single- and Multi-Agent Cyber-Physical Systems\"",
+      "authors": "Srini Ramaswamy, Wang Miaosheng",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.00334",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "embodied-agent",
+        "multi-agent",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "autonomous-agent-llm, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-00345-registry-governed-agent-lifecycle-completing-eddops-with-evaluation-drivenregist.md",
+    "title": "\"Registry-Governed Agent Lifecycle:Completing EDDOps with Evaluation-DrivenRegistration, Promotion, and Retirement on AWS AgentCore\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Registry-Governed Agent Lifecycle:Completing EDDOps with Evaluation-DrivenRegistration, Promotion, and Retirement on AWS AgentCore\"",
+      "authors": "Richard Kang, Vincent Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.00345",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-00407-personalization-as-inverse-planning-learning-latent-design-intents-for-agentic-s.md",
+    "title": "\"Personalization as Inverse Planning: Learning Latent Design Intents for Agentic Slide Generation via Structural Denoising\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Personalization as Inverse Planning: Learning Latent Design Intents for Agentic Slide Generation via Structural Denoising\"",
+      "authors": "Tianci Liu, Zihan Dong, Linjun Zhang, Haoyu Wang, jing Gao, Emre Kiciman, Ranveer Chandra, Wei-Ting Chen",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.00407",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "multi-agent",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-00422-kidnaprag-a-black-box-attack-for-hijacking-reasoning-in-agentic-retrieval-augmen.md",
+    "title": "\"KidnapRAG: A Black-Box Attack for Hijacking Reasoning in Agentic Retrieval-Augmented Generation Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"KidnapRAG: A Black-Box Attack for Hijacking Reasoning in Agentic Retrieval-Augmented Generation Systems\"",
+      "authors": "Chanwoo Choi, Euntae Kim, Kyuho Lee, Youngsam Chun, Jinhee Jeong, Eunmi Kim, Myunggyo Oh, Junseo Jang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.00422",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-00436-phreeqc-mcq-200-a-diagnostic-benchmark-for-tool-augmented-scientific-simulator-a.md",
+    "title": "\"PHREEQC-MCQ-200: A Diagnostic Benchmark for Tool-Augmented Scientific Simulator Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"PHREEQC-MCQ-200: A Diagnostic Benchmark for Tool-Augmented Scientific Simulator Agents\"",
+      "authors": "Ke Zhang, Sahchit Chundur, Mohammad Javad Qomi, Maziar Raissi",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.00436",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-00440-minos-a-multi-agent-collaborative-framework-for-provenance-based-backward-tracki.md",
+    "title": "\"Minos: A Multi-Agent Collaborative Framework for Provenance-Based Backward Tracking\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Minos: A Multi-Agent Collaborative Framework for Provenance-Based Backward Tracking\"",
+      "authors": "Jiahui Wang, Zhenyuan Li, Zhengkai Wang, Xiangmin Shen, Fan Zhang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.00440",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "multi-agent",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-00454-agri-sage-simulation-grounded-multi-agent-llm-for-context-aware-agricultural-adv.md",
+    "title": "\"Agri-SAGE: Simulation-Grounded Multi-Agent LLM for Context-Aware Agricultural Advisory Generation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Agri-SAGE: Simulation-Grounded Multi-Agent LLM for Context-Aware Agricultural Advisory Generation\"",
+      "authors": "Vedant Balasubramaniam, Geetha Charan, Manojkumar Patil, Rohit P Suresh, V Priyanka, Kodur Sai Vinay Sathvik, Y. Narahari",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.00454",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "memory",
+        "multi-agent",
+        "planning",
+        "rag",
+        "reasoning",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-00502-a-task-state-representation-for-long-horizon-mobile-gui-agents.md",
+    "title": "A Task-State Representation for Long-Horizon Mobile GUI Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "A Task-State Representation for Long-Horizon Mobile GUI Agents",
+      "authors": "Yujie Zheng, Zikang Liu, Xin Zhao, Ji-Rong Wen",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.00502",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-00555-rise-from-the-ashes-llm-based-static-analysis-for-deep-learning-framework-bugs.md",
+    "title": "\"Rise From The Ashes: LLM-based Static Analysis for Deep Learning Framework Bugs\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Rise From The Ashes: LLM-based Static Analysis for Deep Learning Framework Bugs\"",
+      "authors": "Shaoyu Yang, Haifeng Lin, Chunrong Fang, Xiang Chen, Wei Cheng, Jiawei Liu, Yiyu Zhang, Hongyu Liu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.00555",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "computer-use",
+        "multi-agent",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agentic-ai, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-00604-vehicle-routing-problem-meets-large-language-models-an-overview-and-perspectives.md",
+    "title": "\"Vehicle Routing Problem Meets Large Language Models: An Overview and Perspectives\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Vehicle Routing Problem Meets Large Language Models: An Overview and Perspectives\"",
+      "authors": "Xianchao Xiu, Chong Shen, Yanjiao Zhu, Wanquan Liu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.00604",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "planning",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "math.OC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-00627-agi-maze-as-a-benchmark-framework-for-world-modeling-agents.md",
+    "title": "AGI Maze as a Benchmark Framework for World-Modeling Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "AGI Maze as a Benchmark Framework for World-Modeling Agents",
+      "authors": "Alexey Potapov",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.00627",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-00692-self-gc-self-governing-context-for-long-horizon-llm-agents.md",
+    "title": "\"Self-GC: Self-Governing Context for Long-Horizon LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Self-GC: Self-Governing Context for Long-Horizon LLM Agents\"",
+      "authors": "Xubin Hao, Hongjin Meng, Xin Yin, Jiawei Zhu, Chenpeng Cao",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.00692",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "llm-agent, planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-00911-from-registry-to-repository-how-ai-agent-skills-are-written-adapted-and-maintain.md",
+    "title": "\"From Registry to Repository: How AI Agent Skills Are Written, Adapted, and Maintained\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"From Registry to Repository: How AI Agent Skills Are Written, Adapted, and Maintained\"",
+      "authors": "Haoyu Gao, Jai Lal Lulla, Hong Yi Lin, Sebastian Baltes, Christoph Treude, Mansooreh Zahedi",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.00911",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "coding-agent",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "ai-agent, coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-00918-from-personas-to-plot-character-grounded-multi-agent-story-generation-for-long-f.md",
+    "title": "\"From Personas to Plot: Character-Grounded Multi-Agent Story Generation for Long-Form Narratives\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"From Personas to Plot: Character-Grounded Multi-Agent Story Generation for Long-Form Narratives\"",
+      "authors": "Aayush Aluru, Chloe Ho, Muhammad Hammouri, Kerry Luo, Myra Malik, Ryan Lagasse, Arjun Bahuguna, Vasu Sharma",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.00918",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-00939-leveraging-llm-based-agentic-systems-to-generate-quantum-applications-for-test-o.md",
+    "title": "Leveraging LLM-Based Agentic Systems to Generate Quantum Applications for Test Optimization",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Leveraging LLM-Based Agentic Systems to Generate Quantum Applications for Test Optimization",
+      "authors": "Ming Tao, Yuechen Li, Tao Yue, Man Zhang, Aitor Arrieta Marcos",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.00939",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "multi-agent",
+        "rag",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "quant-ph"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-00972-bayesian-uncertainty-propagation-for-agentic-rag-pipelines-a-proof-of-concept-st.md",
+    "title": "\"Bayesian Uncertainty Propagation for Agentic RAG Pipelines: A Proof-of-Concept Study on Multi-Hop Question Answering\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Bayesian Uncertainty Propagation for Agentic RAG Pipelines: A Proof-of-Concept Study on Multi-Hop Question Answering\"",
+      "authors": "Louis Donaldson, Connor Walker, Koorosh Aslansefat, Yiannis Papadopoulos",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.00972",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "reasoning",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-00990-swe-doctor-guiding-software-engineering-agents-with-runtime-diagnosis-from-multi.md",
+    "title": "\"SWE-Doctor: Guiding Software Engineering Agents with Runtime Diagnosis from Multi-Faceted Bug Reproduction Tests\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SWE-Doctor: Guiding Software Engineering Agents with Runtime Diagnosis from Multi-Faceted Bug Reproduction Tests\"",
+      "authors": "Yaoqi Guo, Yang Liu, Jie M. Zhang, Yun Ma, Yiling Lou, Zhenpeng Chen",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.00990",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "computer-use",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-01047-conversable-complexity-agentic-llm-collectives-as-interpretable-substrates.md",
+    "title": "\"Conversable Complexity: Agentic LLM Collectives as Interpretable Substrates\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Conversable Complexity: Agentic LLM Collectives as Interpretable Substrates\"",
+      "authors": "Elias Najarro, Ane Espeseth, Eleni Nisioti, Sebastian Risi, Stefano Nichele",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.01047",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "memory",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-01061-agentic-generation-of-verifiable-rules-for-deterministic-self-expanding-reaction.md",
+    "title": "Agentic generation of verifiable rules for deterministic, self-expanding reaction classification",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Agentic generation of verifiable rules for deterministic, self-expanding reaction classification",
+      "authors": "Daniel Armstrong, Maarten Dobbelaere, Valentas Olikauskas, Helena Avila, Octavian Susanu, Jérôme Waser, Philippe Schwaller",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.01061",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "coding-agent",
+        "multi-agent",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-01071-memsyco-bench-benchmarking-sycophancy-in-agent-memory.md",
+    "title": "\"MemSyco-Bench: Benchmarking Sycophancy in Agent Memory\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MemSyco-Bench: Benchmarking Sycophancy in Agent Memory\"",
+      "authors": "Zhishang Xiang, Zerui Chen, Yunbo Tang, Zhimin Wei, Ruqin Ning, Yujie Lin, Qinggang Zhang, Jinsong Su",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.01071",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.IR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-01084-can-agents-generalize-to-the-open-world-unveiling-the-fragility-of-static-traini.md",
+    "title": "Can Agents Generalize to the Open World? Unveiling the Fragility of Static Training in Tool Use",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Can Agents Generalize to the Open World? Unveiling the Fragility of Static Training in Tool Use",
+      "authors": "Song-Lin Lv, Weiming Wu, Rui Zhu, Zi-Jian Cheng, Lan-Zhe Guo",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.01084",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "llm-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-01120-next-generation-agentic-reinforcement-learning-systems-enable-self-evolving-agen.md",
+    "title": "Next-Generation Agentic Reinforcement Learning Systems Enable Self-Evolving Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Next-Generation Agentic Reinforcement Learning Systems Enable Self-Evolving Agents",
+      "authors": "Ran Yan, Wei Fu, Jiale Li, Shusheng Xu, Zhiyu Mei, Jiaxuan Gao, Jiarui Zhang, Wentai Zhang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.01120",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "coding-agent",
+        "planning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.DC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-01211-are-performance-optimization-benchmarks-reliably-measuring-coding-agents.md",
+    "title": "Are Performance-Optimization Benchmarks Reliably Measuring Coding Agents?",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Are Performance-Optimization Benchmarks Reliably Measuring Coding Agents?",
+      "authors": "Zhi Chen, Zhensu Sun, Yuling Shi, David Lo, Lingxiao Jiang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.01211",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-01213-reporescue-an-empirical-study-of-llm-agents-on-whole-repository-compatibility-re.md",
+    "title": "\"RepoRescue: An Empirical Study of LLM Agents on Whole-Repository Compatibility Rescue\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"RepoRescue: An Empirical Study of LLM Agents on Whole-Repository Compatibility Rescue\"",
+      "authors": "Zhihao Lin, Mingyi Zhou, Zhensu Sun, Yizhuo Yang, Renyu Yang, David Lo, Li Li",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.01213",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-01366-auto-fl-research-agentic-search-for-federated-learning-algorithms.md",
+    "title": "\"Auto-FL-Research: Agentic Search for Federated Learning Algorithms\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Auto-FL-Research: Agentic Search for Federated Learning Algorithms\"",
+      "authors": "Holger R. Roth, Ziyue Xu, Chester Chen, Daguang Xu, Peter Cnudde, Andrew Feng",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.01366",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agentic-ai, coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-01421-risk-architecture-for-ai-native-engineering-teams-an-organizational-framework-fo.md",
+    "title": "\"Risk Architecture for AI-Native Engineering Teams: An Organizational Framework for Agentic System Governance\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Risk Architecture for AI-Native Engineering Teams: An Organizational Framework for Agentic System Governance\"",
+      "authors": "Laxmipriya Ganesh Iyer",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.01421",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "computer-use",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-01510-janus-a-playground-for-user-involved-agentic-permission-management.md",
+    "title": "\"Janus: a Playground for User-Involved Agentic Permission Management\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Janus: a Playground for User-Involved Agentic Permission Management\"",
+      "authors": "Natalie Grace Brigham, Eugene Bagdasarian, Tadayoshi Kohno, Franziska Roesner",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.01510",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-01523-multi-head-recurrent-memory-agents.md",
+    "title": "Multi-Head Recurrent Memory Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Multi-Head Recurrent Memory Agents",
+      "authors": "Jiatong Li, Samuel Yeh, Sharon Li",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.01523",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-01531-opine-world-programmatic-world-modeling-with-ontology-error-prioritized-interact.md",
+    "title": "\"OPINE-World: Programmatic World Modeling with Ontology-error-Prioritized Interactive Exploration\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"OPINE-World: Programmatic World Modeling with Ontology-error-Prioritized Interactive Exploration\"",
+      "authors": "David Courtis, Wenhao Li, Scott Sanner",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.01531",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "llm-agent, planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-01600-boundary-sync-measuring-communication-induced-representational-coupling-in-multi.md",
+    "title": "\"BOUNDARY_SYNC: Measuring Communication-Induced Representational Coupling in Multi-Agent LLM Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"BOUNDARY_SYNC: Measuring Communication-Induced Representational Coupling in Multi-Agent LLM Systems\"",
+      "authors": "Zewen Liu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.01600",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "multi-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "llm-agent, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-01640-agentflow-building-agent-dependency-graphs-for-static-analysis-of-agent-programs.md",
+    "title": "\"AgentFlow: Building Agent Dependency Graphs for Static Analysis of Agent Programs\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AgentFlow: Building Agent Dependency Graphs for Static Analysis of Agent Programs\"",
+      "authors": "Shenao Wang, Xinyi Hou, Yanjie Zhao, Xiao Cheng, Haoyu Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.01640",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "multi-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "llm-agent, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-01641-when-agents-do-not-stop-uncovering-infinite-agentic-loops-in-llm-agents.md",
+    "title": "\"When Agents Do Not Stop: Uncovering Infinite Agentic Loops in LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"When Agents Do Not Stop: Uncovering Infinite Agentic Loops in LLM Agents\"",
+      "authors": "Xinyi Hou, Shenao Wang, Yanjie Zhao, Haoyu Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.01641",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "planning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "21",
+      "collection_queries": "llm-agent, planning-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-01661-diverse-evidence-better-forecasts-multi-agent-deliberation-under-information-asy.md",
+    "title": "\"Diverse Evidence, Better Forecasts: Multi-Agent Deliberation Under Information Asymmetry\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Diverse Evidence, Better Forecasts: Multi-Agent Deliberation Under Information Asymmetry\"",
+      "authors": "Yuante Li, Yicheng Tao, Kate Zhang, Taozhi Wang, Gefei Gu, Yaxin Zhou",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.01661",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-01668-verichat-an-agentic-conversational-ai-assistant-for-hardware-security-verificati.md",
+    "title": "\"VeriChat: An Agentic Conversational AI Assistant for Hardware Security Verification\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"VeriChat: An Agentic Conversational AI Assistant for Hardware Security Verification\"",
+      "authors": "Dipayan Saha, Khan Thamid Hasan, Shams Tarek, Sujan Kumar Saha, Mark Tehranipoor, Farimah Farahmandi",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.01668",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "multi-agent",
+        "rag",
+        "tool-use",
+        "workflow-agent",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-01709-comfyclaw-self-evolving-skill-harnesses-for-image-generation-workflows.md",
+    "title": "\"COMFYCLAW: Self-Evolving Skill Harnesses for Image Generation Workflows\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"COMFYCLAW: Self-Evolving Skill Harnesses for Image Generation Workflows\"",
+      "authors": "Zongxia Li, Dawei Liu, Fuxiao Liu, Yuhang Zhou, Xiyang Wu, Jingxi Chen, Jing Xie, Xiaomin Wu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.01709",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-01766-simworlds-a-multi-agent-system-for-dynamic-3d-scene-creation.md",
+    "title": "\"SimWorlds: A Multi-Agent System for Dynamic 3D Scene Creation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SimWorlds: A Multi-Agent System for Dynamic 3D Scene Creation\"",
+      "authors": "Chunjiang Liu, Xiaoyuan Wang, Haoyu Chen, Yizhou Zhao, Ming-Hsuan Yang, László A. Jeni",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.01766",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "embodied-agent",
+        "multi-agent",
+        "planning",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "llm-agent, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-01767-repair-the-amplifier-not-the-symptom-stable-world-model-correction-for-agent-rol.md",
+    "title": "\"Repair the Amplifier, Not the Symptom: Stable World-Model Correction for Agent Rollouts\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Repair the Amplifier, Not the Symptom: Stable World-Model Correction for Agent Rollouts\"",
+      "authors": "Xinyuan Song, Zekun Cai",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.01767",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-01788-krca-an-efficient-root-cause-analysis-system-in-hyper-scale-microservice-systems.md",
+    "title": "\"KRCA: An Efficient Root Cause Analysis System in Hyper-Scale Microservice Systems via Agentic AI\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"KRCA: An Efficient Root Cause Analysis System in Hyper-Scale Microservice Systems via Agentic AI\"",
+      "authors": "Jiamin Jiang, Jingfei Feng, Yu Luo, Qingliang Zhang, Yongqian Su, Wenwei Gu, Shenglin Zhang, Tianyu Cui, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.01788",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "multi-agent",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-01793-safety-testing-llm-agents-at-scale-from-risk-discovery-to-evidence-grounded-veri.md",
+    "title": "\"Safety Testing LLM Agents at Scale: From Risk Discovery to Evidence-Grounded Verification\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Safety Testing LLM Agents at Scale: From Risk Discovery to Evidence-Grounded Verification\"",
+      "authors": "Yunhao Feng, Ruixiao Lin, Ming Wen, Qinqin He, Yanming Guo, Yifan Ding, Yutao Wu, Jialuo Chen, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.01793",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-01812-to-master-an-llm-agent-framework-for-automated-topology-optimization.md",
+    "title": "\"TO-Master: an LLM-agent framework for automated topology optimization\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"TO-Master: an LLM-agent framework for automated topology optimization\"",
+      "authors": "Haoju Lin, Wenchang Zhang, Weipeng Xu, Xiang Li, Tian Xu, Tianju Xue",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.01812",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-01874-skillcoach-self-evolving-rubrics-for-evaluating-and-enhancing-agentic-skill-use.md",
+    "title": "\"SkillCoach: Self-Evolving Rubrics for Evaluating and Enhancing Agentic Skill-Use\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SkillCoach: Self-Evolving Rubrics for Evaluating and Enhancing Agentic Skill-Use\"",
+      "authors": "Jiayin Zhu, Kelong Mao, Yudong Guo, Dengbo He, Sulong Xu, Simiu Gu, Yutao Yue",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.01874",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "computer-use",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-01916-contextsniper-anttrail-s-token-efficient-code-memory-for-repository-level-progra.md",
+    "title": "\"ContextSniper: AntTrail's Token-Efficient Code Memory for Repository-Level Program Repair\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ContextSniper: AntTrail's Token-Efficient Code Memory for Repository-Level Program Repair\"",
+      "authors": "Chiwang Luk, Matin Mohammad Najafi, Zhifeng Jia, Wei Yang, Xiuchang Li, Jinwei Zhu, Yang Ren, Lei Chen, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.01916",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-memory, coding-agent, rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-01929-beyond-textual-repository-exploration-dual-modal-structural-reasoning-for-agenti.md",
+    "title": "\"Beyond Textual Repository Exploration: Dual-Modal Structural Reasoning for Agentic Issue Resolution\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Beyond Textual Repository Exploration: Dual-Modal Structural Reasoning for Agentic Issue Resolution\"",
+      "authors": "Jiayi Zhang, Kai Huang, Yang Liu, Chunyang Chen",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.01929",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "embodied-agent",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "coding-agent, function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-01935-a-tma-decoupling-state-aware-memory-failures-in-long-term-agent-memory.md",
+    "title": "\"A-TMA: Decoupling State-Aware Memory Failures in Long-Term Agent Memory\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"A-TMA: Decoupling State-Aware Memory Failures in Long-Term Agent Memory\"",
+      "authors": "Zitong Shi, Yixuan Tang, Anthony Kum Hoe Tung",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.01935",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-memory, llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-02032-pace-a-proxy-for-agentic-capability-evaluation.md",
+    "title": "\"PACE: A Proxy for Agentic Capability Evaluation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"PACE: A Proxy for Agentic Capability Evaluation\"",
+      "authors": "Yueqi Song, Lintang Sutawika, Jiarui Liu, Lindia Tjuatja, Jiayi Geng, Yunze Xiao, Daniel Lee, Aditya Bharat Soni, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.02032",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "21",
+      "collection_queries": "agent-evaluation, coding-agent, llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-02186-ua-chatdev-uncertainty-aware-multi-agent-collaboration-for-reliable-software-dev.md",
+    "title": "\"UA-ChatDev: Uncertainty-Aware Multi-Agent Collaboration for Reliable Software Development\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"UA-ChatDev: Uncertainty-Aware Multi-Agent Collaboration for Reliable Software Development\"",
+      "authors": "Temitayo Olamilekan Ogunsusi, Lijun Qian, Xishuang Dong",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.02186",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "multi-agent",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-02210-criticality-based-guard-rail-validation-for-ai-agent-decisions-in-autonomous-tel.md",
+    "title": "Criticality-Based Guard Rail Validation for AI Agent Decisions in Autonomous Telecom Networks",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Criticality-Based Guard Rail Validation for AI Agent Decisions in Autonomous Telecom Networks",
+      "authors": "Ravi Kant Sharma",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.02210",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.NI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-02245-copewell-a-multi-agent-swarm-architecture-for-equitable-mental-wellness-support.md",
+    "title": "\"Copewell: A Multi-Agent Swarm Architecture for Equitable Mental Wellness Support\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Copewell: A Multi-Agent Swarm Architecture for Equitable Mental Wellness Support\"",
+      "authors": "Seren Yenikent, Jack Vinijtrongjit, Katherine Ng",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.02245",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CY",
+        "cs.HC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-02255-agenticsts-a-bounded-memory-testbed-for-long-horizon-llm-agents.md",
+    "title": "\"AgenticSTS: A Bounded-Memory Testbed for Long-Horizon LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AgenticSTS: A Bounded-Memory Testbed for Long-Horizon LLM Agents\"",
+      "authors": "Xiangchen Cheng, Yunwei Jiang, Jianwen Sun, Zizhen Li, Chuanhao Li, Xiangcheng Cao, Yihao Liu, Fanrui Zhang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.02255",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "23",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-02294-coding-agents-are-guessing-measuring-action-boundary-violations-in-underspecifie.md",
+    "title": "\"Coding Agents Are Guessing: Measuring Action-Boundary Violations in Underspecified DevOps Instructions\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Coding Agents Are Guessing: Measuring Action-Boundary Violations in Underspecified DevOps Instructions\"",
+      "authors": "Zimo Ji, Zekai Zhang, Congying Xu, Zongjie Li, Yudong Gao, Shuai Wang, Shing-Chi Cheung",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.02294",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-evaluation, coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-02381-hulat2-at-mer-trans-2026-governed-multi-agent-simplification-for-spanish-easy-to.md",
+    "title": "\"HULAT2 at MER-TRANS 2026: Governed Multi-Agent Simplification for Spanish Easy-to-Read Generation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"HULAT2 at MER-TRANS 2026: Governed Multi-Agent Simplification for Spanish Easy-to-Read Generation\"",
+      "authors": "Lourdes Moreno, Paloma Martínez, Marco Antonio Sanchez-Escudero, Miguel Domínguez-Gómez",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.02381",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "multi-agent",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-02448-agentscad-automated-design-for-manufacturing-of-fdm-parts-via-multi-agent-llm-re.md",
+    "title": "\"AgentsCAD: Automated Design for Manufacturing of FDM Parts via Multi-Agent LLM Reasoning and Geometric Feature Recognition\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AgentsCAD: Automated Design for Manufacturing of FDM Parts via Multi-Agent LLM Reasoning and Geometric Feature Recognition\"",
+      "authors": "Emmanuel George, Christopher Keefe, Peter Pak, Amir Barati Farimani",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.02448",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "reasoning",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-02453-adoption-and-ecosystem-health-a-longitudinal-analysis-of-open-source-multi-agent.md",
+    "title": "\"Adoption and Ecosystem Health: A Longitudinal Analysis of Open-Source Multi-Agent Frameworks\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Adoption and Ecosystem Health: A Longitudinal Analysis of Open-Source Multi-Agent Frameworks\"",
+      "authors": "Xi Zhang, Papi Menon, Vivian Chu, Koray Cosguner",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.02453",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-02507-what-llm-agents-say-when-no-one-is-watching-social-structure-and-latent-objectiv.md",
+    "title": "\"What LLM Agents Say When No One Is Watching: Social Structure and Latent Objective Emergence in Multi-Agent Debates\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"What LLM Agents Say When No One Is Watching: Social Structure and Latent Objective Emergence in Multi-Agent Debates\"",
+      "authors": "Arman Ghaffarizadeh, Danyal Mohaddes, Aliakbar Izadkhah, Shahriar Noroozizadeh",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.02507",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL",
+        "cs.LG",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agent-evaluation, llm-agent, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-02577-benchmarking-the-benchmarks-a-validity-audit-of-tool-calling-evaluation.md",
+    "title": "\"Benchmarking the Benchmarks: A Validity Audit of Tool-Calling Evaluation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Benchmarking the Benchmarks: A Validity Audit of Tool-Calling Evaluation\"",
+      "authors": "Vishvesh Bhat, Jay Vaghasiya, Muhammad Ahmed Mohsin, Asad Aali",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.02577",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-02579-when-not-to-write-memory-governing-false-promotion-from-correlated-agent-traces.md",
+    "title": "\"When Not to Write Memory: Governing False Promotion from Correlated Agent Traces\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"When Not to Write Memory: Governing False Promotion from Correlated Agent Traces\"",
+      "authors": "Yijiashun Qi, Xiang Xu, Yuxuan Li",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.02579",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-06-30",
+      "updated_at": "2026-06-30",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-memory, language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-02599-agentltl-a-trace-verification-framework-for-measuring-enforcing-and-training-pro.md",
+    "title": "\"AgentLTL: A Trace-Verification Framework for Measuring, Enforcing, and Training Procedural Compliance in Tool-Using LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AgentLTL: A Trace-Verification Framework for Measuring, Enforcing, and Training Procedural Compliance in Tool-Using LLM Agents\"",
+      "authors": "Laïla Elkoussy, Julien Perez",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.02599",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI",
+        "cs.LO"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "llm-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-02606-chainswe-benchmarking-coding-agents-on-multi-bug-software-maintenance.md",
+    "title": "\"ChainSWE: Benchmarking Coding Agents on Multi-Bug Software Maintenance\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ChainSWE: Benchmarking Coding Agents on Multi-Bug Software Maintenance\"",
+      "authors": "Qirui Jin, Lingching Tung, Kenan Li, Qiyang Shi, Yushi She, Huanzhong Jia, Harrison Zhao, Kejing Xia, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.02606",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-02684-can-coding-agents-implement-missed-compiler-optimizations-evaluating-llm-agents-.md",
+    "title": "Can Coding Agents Implement Missed Compiler Optimizations? Evaluating LLM Agents on LLVM Peephole Optimizations",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Can Coding Agents Implement Missed Compiler Optimizations? Evaluating LLM Agents on LLVM Peephole Optimizations",
+      "authors": "Hongxu Xu, Chunhao Liao, Xintong Zhou, Chengnian Sun",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.02684",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "coding-agent, llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-02689-s-ember-a-large-scale-benchmark-for-streaming-egocentric-memory-retrieval.md",
+    "title": "\"S-EMBER: A Large-Scale Benchmark for Streaming Egocentric Memory Retrieval\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"S-EMBER: A Large-Scale Benchmark for Streaming Egocentric Memory Retrieval\"",
+      "authors": "Xiaodong Wang, Xuanyi Zhao, Pedro Rodriguez, Devendra Singh Sachan, Barlas Oguz, Seungwhan Moon, Shang-Wen Li, Gargi Ghosh, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.02689",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-02703-llmoxie-exploring-agentic-ai-for-scientific-software-development.md",
+    "title": "\"LLMoxie: Exploring Agentic AI for Scientific Software Development\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"LLMoxie: Exploring Agentic AI for Scientific Software Development\"",
+      "authors": "Landung Setiawan, Anant Mittal, Cordero Core, Anshul Tambay, Carlos Garcia Jurado Suarez, David A. C. Beck, Andrew J. Connolly, Vani Mandava",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.02703",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "planning",
+        "reasoning",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI",
+        "cs.DC",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "agentic-ai, coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-02716-evaluating-large-language-models-for-decision-making-in-agent-based-urban-mobili.md",
+    "title": "Evaluating Large Language Models for Decision-Making in Agent-Based Urban Mobility Simulations",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Evaluating Large Language Models for Decision-Making in Agent-Based Urban Mobility Simulations",
+      "authors": "Bruno Cascaes Alves, Míriam Blank Born, Ulisses Gilioli Francescatto Júnior, Felipe Moura Goulart, Letícia Brandão Caldas, Marilton Sanchotene de Aguiar",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.02716",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "multi-agent",
+        "planning",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-02807-swarmresearch-orchestrating-coding-agents-for-open-ended-discovery.md",
+    "title": "\"SwarmResearch: Orchestrating Coding Agents for Open-Ended Discovery\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SwarmResearch: Orchestrating Coding Agents for Open-Ended Discovery\"",
+      "authors": "Yuvraj Virk, Zack Edds, Chunqiu Steven Xia, Lingming Zhang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.02807",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-02",
+      "updated_at": "2026-07-02",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "coding-agent",
+        "computer-use",
+        "multi-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "coding-agent, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-02846-object-centric-environment-modeling-for-agentic-tasks.md",
+    "title": "Object-Centric Environment Modeling for Agentic Tasks",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Object-Centric Environment Modeling for Agentic Tasks",
+      "authors": "Yiyang Li, Tianyi Ma, Zehong Wang, Yijun Ma, Yanfang Ye",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.02846",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-03",
+      "updated_at": "2026-07-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-02857-mosaic-knowledge-guided-cli-command-composition-attack-in-llm-coding-agents.md",
+    "title": "\"MOSAIC: Knowledge-Guided CLI Command Composition Attack in LLM Coding Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MOSAIC: Knowledge-Guided CLI Command Composition Attack in LLM Coding Agents\"",
+      "authors": "Jiangrong Wu, Huaijin Wang, Yihao Zhang, Yuhong Nan, Shuai Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.02857",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-03",
+      "updated_at": "2026-07-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "computer-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-evaluation, coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-02879-medcalc-pro-solving-complex-medical-calculations-with-llm-agents.md",
+    "title": "\"MedCalc-Pro: Solving Complex Medical Calculations with LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MedCalc-Pro: Solving Complex Medical Calculations with LLM Agents\"",
+      "authors": "Siran Zhao, Ruihui Hou, Ziyue Huai, Chennuo Zhang, Tong Ruan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.02879",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-03",
+      "updated_at": "2026-07-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-02882-diagnosis-driven-automatic-repair-for-agentic-workflow-via-symbolic-inference.md",
+    "title": "Diagnosis-Driven Automatic Repair for Agentic Workflow via Symbolic Inference",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Diagnosis-Driven Automatic Repair for Agentic Workflow via Symbolic Inference",
+      "authors": "Xuyan Ma, Yawen Wang, Junjie Wang, Xiaofei Xie, Boyu Wu, Mingyang Li, Dandan Wang, Qing Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.02882",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-03",
+      "updated_at": "2026-07-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-02911-coact-action-preserving-observation-compression-for-coding-agents.md",
+    "title": "\"CoACT: Action-Preserving Observation Compression for Coding Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"CoACT: Action-Preserving Observation Compression for Coding Agents\"",
+      "authors": "Haorui Chen, Yuancheng Zhu, Yitong Zhang, Jia Li",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.02911",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-03",
+      "updated_at": "2026-07-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "coding-agent",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-02927-videosearcher-empowering-video-deep-research-with-multi-tool-agentic-reasoning-v.md",
+    "title": "\"VideoSearcher: Empowering Video Deep Research with Multi-Tool Agentic Reasoning via Reinforcement Learning\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"VideoSearcher: Empowering Video Deep Research with Multi-Tool Agentic Reasoning via Reinforcement Learning\"",
+      "authors": "Zhenkun Gao, Yicheng Bao, Jinlong Peng, Xueheng Li, Theo Huang, Bangwei Liu, Kunquan Li, Zhenye Gan, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.02927",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-03",
+      "updated_at": "2026-07-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-02942-a-workflow-aware-serving-layer-for-agentic-applications.md",
+    "title": "A Workflow-Aware Serving Layer for Agentic Applications",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "A Workflow-Aware Serving Layer for Agentic Applications",
+      "authors": "Jiayi Qian, Zishen Wan, Hanchen Yang, Chun Tao, Souvik Kundu, Tushar Krishna",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.02942",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-03",
+      "updated_at": "2026-07-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.DC",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-03105-orbit-q-dual-axis-benchmarking-of-autonomous-agents-in-scientific-quantum-progra.md",
+    "title": "\"ORBIT-Q: Dual-axis benchmarking of autonomous agents in scientific quantum programming\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ORBIT-Q: Dual-axis benchmarking of autonomous agents in scientific quantum programming\"",
+      "authors": "Shi-Xin Zhang, Yu-Qin Chen",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.03105",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-03",
+      "updated_at": "2026-07-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "quant-ph"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-03162-apeb-benchmarking-personalization-ability-of-large-language-model-agents.md",
+    "title": "\"APeB: Benchmarking Personalization Ability of Large Language Model Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"APeB: Benchmarking Personalization Ability of Large Language Model Agents\"",
+      "authors": "Garry Yang, Zizhe Chen, Xinru Chen, Yongqiang Chen, Jianxiang Wang, Deyu Zou, Linyi Ding, Jialiang Wu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.03162",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-03",
+      "updated_at": "2026-07-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.HC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-03220-contra-red-teaming-configurations-of-personalizable-agents.md",
+    "title": "\"CONTRA: Red-Teaming Configurations of Personalizable Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"CONTRA: Red-Teaming Configurations of Personalizable Agents\"",
+      "authors": "Jonathan Nöther, Adish Singla, Goran Radanovic",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.03220",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-03",
+      "updated_at": "2026-07-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-03233-agentic-and-generative-ai-for-open-source-intelligence-and-cyber-investigations-.md",
+    "title": "\"Agentic and Generative AI for Open-Source Intelligence and Cyber Investigations: Taxonomy, Evaluation, Challenges, and Future Directions\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Agentic and Generative AI for Open-Source Intelligence and Cyber Investigations: Taxonomy, Evaluation, Challenges, and Future Directions\"",
+      "authors": "Eduardo Almeida Palmieri, Mohamed Chahine Ghanem, Dipo Dunsin, Zubair Baig, Ed de Quincey, Kim-Kwang Raymond Choo",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.03233",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-03",
+      "updated_at": "2026-07-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI",
+        "cs.IR",
+        "cs.SI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "22",
+      "collection_queries": "agentic-ai, rag-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-03269-agentic-secpbft-agentic-ai-driven-proactive-security-framework-for-wireless-pbft.md",
+    "title": "\"Agentic-SecPBFT: Agentic AI-Driven Proactive Security Framework for Wireless PBFT Consensus in Mobile Ad-Hoc Networks\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Agentic-SecPBFT: Agentic AI-Driven Proactive Security Framework for Wireless PBFT Consensus in Mobile Ad-Hoc Networks\"",
+      "authors": "Haoxiang Luo, Yinqiu Liu, Ruichen Zhang, Guangyuan Liu, Gang Sun, Hongfang Yu, Zhu Han, Dong In Kim",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.03269",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-03",
+      "updated_at": "2026-07-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "computer-use",
+        "multi-agent",
+        "rag",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.NI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-03316-is-agentic-code-review-helpful-mining-developers-feedback-to-coderabbit-reviews-.md",
+    "title": "\"Is Agentic Code Review Helpful? Mining Developers' Feedback to CodeRabbit Reviews in the Wild\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Is Agentic Code Review Helpful? Mining Developers' Feedback to CodeRabbit Reviews in the Wild\"",
+      "authors": "Hong Yi Lin, Mingzhao Liang, Kla Tantithamthavorn, Patanamon Thongtanunam",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.03316",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-03",
+      "updated_at": "2026-07-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "coding-agent",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "autonomous-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-03333-spork-self-speculative-forking-to-accelerate-agentic-llm-inference.md",
+    "title": "\"SPORK: Self-Speculative Forking to Accelerate Agentic LLM Inference\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SPORK: Self-Speculative Forking to Accelerate Agentic LLM Inference\"",
+      "authors": "Huajun Bai, Weiwei Lv, Huichuan Zheng, Youyou Lu, Jiwu Shu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.03333",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-03",
+      "updated_at": "2026-07-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.DC",
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-03423-securing-multi-tool-ai-agent-chains-with-dynamic-real-time-compositional-policie.md",
+    "title": "Securing Multi-Tool AI Agent Chains With Dynamic, Real-Time Compositional Policies",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Securing Multi-Tool AI Agent Chains With Dynamic, Real-Time Compositional Policies",
+      "authors": "Chris Schneider, Kriti Faujdar, Philipp Schoenegger, Ben Bariach",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.03423",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-03",
+      "updated_at": "2026-07-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "ai-agent, coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-03441-no-time-like-the-present-agentic-test-time-training-for-llm-agents.md",
+    "title": "\"No Time Like the Present: Agentic Test-Time Training for LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"No Time Like the Present: Agentic Test-Time Training for LLM Agents\"",
+      "authors": "Yanbo Wang, Jinhua Hao, Yuze Shi, Kun Yuan, Ming Sun",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.03441",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-03",
+      "updated_at": "2026-07-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "coding-agent",
+        "computer-use",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "coding-agent, llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-03510-cage-1-control-assurance-and-governance-evaluation-for-enterprise-agentic-ai.md",
+    "title": "\"CAGE-1: Control, Assurance, and Governance Evaluation for Enterprise Agentic AI\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"CAGE-1: Control, Assurance, and Governance Evaluation for Enterprise Agentic AI\"",
+      "authors": "Roopam W. Sure",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.03510",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-03",
+      "updated_at": "2026-07-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "planning",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI",
+        "cs.CY"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-03525-gameenginebench-evaluating-coding-agents-on-real-c-runtime-environments.md",
+    "title": "\"GameEngineBench: Evaluating Coding Agents on Real C++ Runtime Environments\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"GameEngineBench: Evaluating Coding Agents on Real C++ Runtime Environments\"",
+      "authors": "Brian La, Sejoon Chang, Ben Kim, Junyoung Bae, Aamish Ahmad Beg, Sei Chang, Gonzalo Gonzalez-Pumariega",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.03525",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-03",
+      "updated_at": "2026-07-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "embodied-agent",
+        "memory",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-03601-archeval-measuring-ai-agents-as-computer-architects.md",
+    "title": "\"ArchEval: Measuring AI Agents as Computer Architects\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ArchEval: Measuring AI Agents as Computer Architects\"",
+      "authors": "Chenyu Wang, Zishen Wan, Jeffrey Ma, Shvetank Prakash, Zhenting Qi, Haebin Do, Andy Cheng, Arya Tschand, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.03601",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-03",
+      "updated_at": "2026-07-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "21",
+      "collection_queries": "ai-agent, llm-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-03628-swarm-driven-multi-agent-reasoning-for-smart-city-security.md",
+    "title": "Swarm-Driven Multi-Agent Reasoning for Smart City Security",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Swarm-Driven Multi-Agent Reasoning for Smart City Security",
+      "authors": "Saeid Jamshidi, Kawser Wazed Nafi, Carol Fung, Foutse Khomh",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.03628",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-03",
+      "updated_at": "2026-07-03",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "multi-agent",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-03691-don-t-blame-the-large-language-model-how-scaffolding-evolution-shapes-coding-age.md",
+    "title": "\"Don't Blame the Large Language Model: How Scaffolding Evolution Shapes Coding Agent Quality\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Don't Blame the Large Language Model: How Scaffolding Evolution Shapes Coding Agent Quality\"",
+      "authors": "Oussama Ben Sghaier, Hao Li, Bram Adams, Ahmed E. Hassan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.03691",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-04",
+      "updated_at": "2026-07-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-03695-social-networks-of-llm-agents.md",
+    "title": "Social Networks of LLM Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Social Networks of LLM Agents",
+      "authors": "Kaixuan Liu, Guojun Xiong, Weinan Zhang, Shengpu Tang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.03695",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-04",
+      "updated_at": "2026-07-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "llm-agent, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-03702-agent-reinforcement-learning-via-pivotal-aware-self-feedback-retry.md",
+    "title": "Agent Reinforcement Learning via Pivotal-Aware Self-Feedback Retry",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Agent Reinforcement Learning via Pivotal-Aware Self-Feedback Retry",
+      "authors": "Weiyang Guo, Zesheng Shi, Longhui Zhang, Zeen Zhu, Min Zhang, Jing Li",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.03702",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-04",
+      "updated_at": "2026-07-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-03726-selfmem-self-optimizing-memory-for-ai-agents.md",
+    "title": "\"SelfMem: Self-Optimizing Memory for AI Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SelfMem: Self-Optimizing Memory for AI Agents\"",
+      "authors": "Shu Yang, Junchao Wu, Derek F. Wong, Di Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.03726",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-04",
+      "updated_at": "2026-07-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "agent-memory, ai-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-03821-dualview-preventing-indirect-prompt-injection-in-personal-ai-agents.md",
+    "title": "\"DualView: Preventing Indirect Prompt Injection in Personal AI Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"DualView: Preventing Indirect Prompt Injection in Personal AI Agents\"",
+      "authors": "Juhee Kim, Woohyuk Choi, Taehyun Kang, Youngmin Kim, Byoungyoung Lee",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.03821",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-04",
+      "updated_at": "2026-07-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-03853-cograd-a-cognitively-inspired-multi-agent-framework-for-radiology-report-generat.md",
+    "title": "\"CogRad: A Cognitively-Inspired Multi-Agent Framework for Radiology Report Generation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"CogRad: A Cognitively-Inspired Multi-Agent Framework for Radiology Report Generation\"",
+      "authors": "Saif Ur Rehman Khan, Hasaan Maqsood, Sebastian Vollmer, Andreas Dengel, Muhammad Nabeel Asim",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.03853",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-04",
+      "updated_at": "2026-07-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-03953-the-remarkable-effectiveness-of-providing-ai-agents-with-natural-language-tools-.md",
+    "title": "\"The Remarkable Effectiveness of Providing AI Agents with Natural Language Tools: A Replication Study Validating NLT Performance Across 14 Models\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"The Remarkable Effectiveness of Providing AI Agents with Natural Language Tools: A Replication Study Validating NLT Performance Across 14 Models\"",
+      "authors": "Alexander Somma, Isabelle Plante, Fred Premji",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.03953",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-04",
+      "updated_at": "2026-07-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "agentic-ai, ai-agent, llm-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-03968-refused-in-chat-written-in-code-workflow-level-jailbreak-construction-in-ide-cod.md",
+    "title": "\"Refused in Chat, Written in Code: Workflow-Level Jailbreak Construction in IDE Coding Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Refused in Chat, Written in Code: Workflow-Level Jailbreak Construction in IDE Coding Agents\"",
+      "authors": "Abhishek Kumar, Carsten Maple",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.03968",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-04",
+      "updated_at": "2026-07-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-04009-physminer-an-agentic-ai-framework-for-discovering-turbulence-physics.md",
+    "title": "\"PhysMiner: An Agentic AI Framework for Discovering Turbulence Physics\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"PhysMiner: An Agentic AI Framework for Discovering Turbulence Physics\"",
+      "authors": "Jiawei Chen, Han Gao, Ping He",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.04009",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-04",
+      "updated_at": "2026-07-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "physics.flu-dyn"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-04034-the-i-don-t-know-filter-enhancing-agentic-reliability-in-function-calling.md",
+    "title": "\"The \\\"I Don't Know\\\" Filter: Enhancing Agentic Reliability in Function Calling\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"The \\\"I Don't Know\\\" Filter: Enhancing Agentic Reliability in Function Calling\"",
+      "authors": "Stefan Broecker, Mason del Rosario, Boris Selitser, Thomas Strohmer",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.04034",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-04",
+      "updated_at": "2026-07-04",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-evaluation, function-calling"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-04089-placemem-toward-a-compute-aware-memory-plane-for-lifelong-agents.md",
+    "title": "\"PLACEMEM: Toward a Compute-Aware Memory Plane for Lifelong Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"PLACEMEM: Toward a Compute-Aware Memory Plane for Lifelong Agents\"",
+      "authors": "Sukanta Ganguly",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.04089",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-05",
+      "updated_at": "2026-07-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "planning",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agent-memory"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-04149-beyond-scene-priors-fine-grained-traffic-scene-reasoning-with-benchmarking-and-q.md",
+    "title": "\"Beyond Scene Priors: Fine-Grained Traffic Scene Reasoning with Benchmarking and Query-Guided Small-Object Focus\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Beyond Scene Priors: Fine-Grained Traffic Scene Reasoning with Benchmarking and Query-Guided Small-Object Focus\"",
+      "authors": "Waikit Xiu, Qiang Lu, Zian Wang, Xinjie Yang, Zhiwei Chen, Chen Sun, Xiying Li",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.04149",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-05",
+      "updated_at": "2026-07-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "computer-use",
+        "multi-agent",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-04162-ace-agentic-control-for-embodied-manipulation-via-zero-shot-workflow-reasoning.md",
+    "title": "\"ACE: Agentic Control for Embodied Manipulation via Zero-shot Workflow Reasoning\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ACE: Agentic Control for Embodied Manipulation via Zero-shot Workflow Reasoning\"",
+      "authors": "Iok Tong Lei, QianZhi Li, Ying Jie Yap, Yujie Zhang, Rui Zhong, Haichao Gui, Xiaolong Liu, Zhidong Deng",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.04162",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-05",
+      "updated_at": "2026-07-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "embodied-agent",
+        "memory",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.RO",
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-04212-an-evaluation-of-role-based-multi-agent-code-generation-on-repository-scale-prob.md",
+    "title": "An Evaluation of Role-Based Multi-Agent Code Generation on Repository-Scale Problems",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "An Evaluation of Role-Based Multi-Agent Code Generation on Repository-Scale Problems",
+      "authors": "Benedetta Donato, Noah Hagar-Dent, Aaron Worsnop, Leonardo Mariani, Valerio Terragni",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.04212",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-05",
+      "updated_at": "2026-07-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "multi-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-04219-agentic-iot-architectures-applications-and-challenges-toward-the-internet-of-age.md",
+    "title": "\"Agentic IoT: Architectures, Applications, and Challenges Toward the Internet of Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Agentic IoT: Architectures, Applications, and Challenges Toward the Internet of Agents\"",
+      "authors": "Rümeysa Hilal Sevinç, Bahaeddin Türkoğlu, İbrahim Kök",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.04219",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-05",
+      "updated_at": "2026-07-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "multi-agent",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.MA",
+        "cs.NI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "ai-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-04240-biological-motifs-for-agentic-control.md",
+    "title": "Biological Motifs for Agentic Control",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Biological Motifs for Agentic Control",
+      "authors": "Bogdan Banu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.04240",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-05",
+      "updated_at": "2026-07-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "multi-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "q-bio.CB"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "agent-evaluation, autonomous-agent-llm, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-04293-causalgame-benchmarking-causal-thinking-of-llm-agents-in-games.md",
+    "title": "\"CausalGame: Benchmarking Causal Thinking of LLM Agents in Games\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"CausalGame: Benchmarking Causal Thinking of LLM Agents in Games\"",
+      "authors": "Zhenhao Chen, Yongqiang Chen, Chenxi Liu, Junchi Yu, Xiangchen Song, Zijian Li, Jialin Li, Philip Torr, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.04293",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-05",
+      "updated_at": "2026-07-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI",
+        "cs.LG",
+        "stat.ML"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-04334-do-gui-agents-believe-their-eyes-diagnosing-state-belief-reliance-on-pixels-vers.md",
+    "title": "Do GUI Agents Believe Their Eyes? Diagnosing State-Belief Reliance on Pixels versus Structure",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Do GUI Agents Believe Their Eyes? Diagnosing State-Belief Reliance on Pixels versus Structure",
+      "authors": "Guijia Zhang, Harry Yang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.04334",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-05",
+      "updated_at": "2026-07-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-04391-memory-orchestrated-semantic-system-moss-an-auditable-agentic-memory-architectur.md",
+    "title": "\"Memory-Orchestrated Semantic System (MOSS): An Auditable Agentic Memory Architecture\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Memory-Orchestrated Semantic System (MOSS): An Auditable Agentic Memory Architecture\"",
+      "authors": "Serge Lacasse, Jérémie Hatier, Alex Baker",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.04391",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-05",
+      "updated_at": "2026-07-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "agent-memory, ai-agent, rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-04394-mechmath-agent-team-llm-driven-agents-for-mathematical-research.md",
+    "title": "\"MechMath Agent Team: LLM Driven Agents for Mathematical Research\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MechMath Agent Team: LLM Driven Agents for Mathematical Research\"",
+      "authors": "Yichuan Cao, Ruichen Qiu, Junqi Liu, Jiaqi Wang, Dakai Guo, Ruyong Feng, Lihong Zhi, Xiao-Shan Gao",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.04394",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-05",
+      "updated_at": "2026-07-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "planning",
+        "rag",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.SC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-04395-nki-agent-domain-specific-fine-tuning-and-agentic-tool-use-for-neuron-kernel-gen.md",
+    "title": "\"NKI-Agent: Domain-Specific Fine-Tuning and Agentic Tool Use for Neuron Kernel Generation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"NKI-Agent: Domain-Specific Fine-Tuning and Agentic Tool Use for Neuron Kernel Generation\"",
+      "authors": "Junjie Tang, Jun Huan, Hao Zhou, Yuhao Zhang, Lin Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.04395",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-05",
+      "updated_at": "2026-07-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-04426-ace-brain-0-5-a-unified-embodied-foundational-model-for-physical-agentic-ai.md",
+    "title": "\"ACE-Brain-0.5: A Unified Embodied Foundational Model for Physical Agentic AI\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ACE-Brain-0.5: A Unified Embodied Foundational Model for Physical Agentic AI\"",
+      "authors": "\"ACE-Brain Team, :, Ziyang Gong, Haoming Gu, Zehang Luo, Tianyi Zhang, Tao Tao, Yixiao Chi, et al.\"",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.04426",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-05",
+      "updated_at": "2026-07-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "embodied-agent",
+        "memory",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.RO"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agentic-ai"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-04433-autonomous-information-seeking-a-roadmap-for-agentic-recommender-systems.md",
+    "title": "\"Autonomous Information Seeking: A Roadmap for Agentic Recommender Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Autonomous Information Seeking: A Roadmap for Agentic Recommender Systems\"",
+      "authors": "Xinyu Lin, Yashar Deldjoo, Sunhao Dai, Honghui Bao, Xiaopeng Ye, Fatemeh Nazary, Wenjie Wang, Tommaso Di Noia, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.04433",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-05",
+      "updated_at": "2026-07-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "planning",
+        "reasoning",
+        "tool-use",
+        "workflow-agent",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.IR",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "21",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-04470-regime-conditional-stabilisation-of-llm-augmented-cooperative-multi-agent-reinfo.md",
+    "title": "Regime-Conditional Stabilisation of LLM-Augmented Cooperative Multi-Agent Reinforcement Learning",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Regime-Conditional Stabilisation of LLM-Augmented Cooperative Multi-Agent Reinforcement Learning",
+      "authors": "Faid Keddouri, Sohaib Houhou, Aissa Boulmerka, Nadir Farhi",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.04470",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-05",
+      "updated_at": "2026-07-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI",
+        "math.OC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-04528-measuring-harness-induced-belief-divergence-in-multi-step-llm-agents.md",
+    "title": "Measuring Harness-Induced Belief Divergence in Multi-Step LLM Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Measuring Harness-Induced Belief Divergence in Multi-Step LLM Agents",
+      "authors": "Haiwen Yi, Xinyuan Song",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.04528",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-05",
+      "updated_at": "2026-07-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-evaluation, llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-04569-llms-for-agentic-home-energy-management.md",
+    "title": "LLMs for Agentic Home Energy Management",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "LLMs for Agentic Home Energy Management",
+      "authors": "Sokipriala Jonah",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.04569",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-06",
+      "updated_at": "2026-07-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "eess.SY"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "function-calling, llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-04617-mrms-a-multi-resolution-memory-substrate-for-long-lived-ai-agents.md",
+    "title": "\"MRMS: A Multi-Resolution Memory Substrate for Long-Lived AI Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MRMS: A Multi-Resolution Memory Substrate for Long-Lived AI Agents\"",
+      "authors": "Jizhizi Li, Amy Shi-Nash",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.04617",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-06",
+      "updated_at": "2026-07-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-04623-can-llms-really-recover-microservice-failures-a-recovery-aware-evaluation-of-dia.md",
+    "title": "Can LLMs Really Recover Microservice Failures? A Recovery-Aware Evaluation of Diagnosis-to-Action Reasoning",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Can LLMs Really Recover Microservice Failures? A Recovery-Aware Evaluation of Diagnosis-to-Action Reasoning",
+      "authors": "Jiaxing Qi, Zhongzhi Luan, Hongyu Zhang, Shaohan Huang, Carol Fung, Yongxin Tong, Hailong Yang, Depei Qian",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.04623",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-06",
+      "updated_at": "2026-07-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.DC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-04686-toolfailbench-diagnosing-tool-use-failures-in-llm-agents.md",
+    "title": "\"ToolFailBench: Diagnosing Tool-Use Failures in LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"ToolFailBench: Diagnosing Tool-Use Failures in LLM Agents\"",
+      "authors": "Harsh Soni",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.04686",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-06",
+      "updated_at": "2026-07-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "agentic-ai, llm-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-04697-ai-agent-pull-requests-on-github-frequency-structure-and-merge-conflict-rates.md",
+    "title": "\"AI Agent Pull Requests on GitHub: Frequency, Structure, and Merge Conflict Rates\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AI Agent Pull Requests on GitHub: Frequency, Structure, and Merge Conflict Rates\"",
+      "authors": "George Xu, Arjun Subramanian, Nithilan Karthik",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.04697",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-06",
+      "updated_at": "2026-07-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "multi-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "ai-agent, coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-04713-rspo-reward-swap-policy-optimization-for-multi-turn-llm-agents.md",
+    "title": "\"RSPO: Reward-Swap Policy Optimization for Multi-Turn LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"RSPO: Reward-Swap Policy Optimization for Multi-Turn LLM Agents\"",
+      "authors": "Qiang Liu, Taian Guo, Ruizhi Qiao, Xing Sun",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.04713",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-06",
+      "updated_at": "2026-07-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-evaluation, llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-04963-stapo-selective-trajectory-aware-policy-optimization-for-llm-agent-training.md",
+    "title": "\"STAPO: Selective Trajectory-Aware Policy Optimization for LLM Agent Training\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"STAPO: Selective Trajectory-Aware Policy Optimization for LLM Agent Training\"",
+      "authors": "Qiuyi Qi, Tian Liang, Mutian Bao, Jinjian Zhang, Dongnan Liu, Wei Zhou, Linjian Mo, Ming Kong, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.04963",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-06",
+      "updated_at": "2026-07-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "planning",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-05001-tactic-kg-toward-small-agent-teams-for-cyber-threat-intelligence-knowledge-graph.md",
+    "title": "\"TACTIC-KG: Toward Small Agent Teams for Cyber Threat Intelligence Knowledge Graph Construction\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"TACTIC-KG: Toward Small Agent Teams for Cyber Threat Intelligence Knowledge Graph Construction\"",
+      "authors": "Mouhamed Amine Bouchiha, Gregory Blanc",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.05001",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-06",
+      "updated_at": "2026-07-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI",
+        "cs.LG",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-05029-your-agent-s-memories-are-not-its-own-forged-reasoning-attacks-on-llm-agent-memo.md",
+    "title": "\"Your Agent's Memories Are Not Its Own: Forged Reasoning Attacks on LLM Agent Memory and Defenses\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Your Agent's Memories Are Not Its Own: Forged Reasoning Attacks on LLM Agent Memory and Defenses\"",
+      "authors": "Neeraj Karamchandani, Piyush Nagasubramaniam, Sencun Zhu, Dinghao Wu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.05029",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-06",
+      "updated_at": "2026-07-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agent-memory, llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-05055-toward-trustworthy-large-language-model-agents-in-healthcare.md",
+    "title": "Toward Trustworthy Large Language Model Agents in Healthcare",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Toward Trustworthy Large Language Model Agents in Healthcare",
+      "authors": "Hadi Hasan, Safaa Salman, Adam Tai Abou Dargham, Ammar Mohanna, Ali Chehab",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.05055",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-06",
+      "updated_at": "2026-07-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "function-calling, rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-05120-agent-data-injection-attacks-are-realistic-threats-to-ai-agents.md",
+    "title": "Agent Data Injection Attacks are Realistic Threats to AI Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Agent Data Injection Attacks are Realistic Threats to AI Agents",
+      "authors": "Woohyuk Choi, Juhee Kim, Taehyun Kang, Jihyeon Jeong, Luyi Xing, Byoungyoung Lee",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.05120",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-06",
+      "updated_at": "2026-07-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "computer-use",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agent-safety, ai-agent, coding-agent, web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-05132-when-agents-lie-premeditation-persistence-and-exploitation-in-repeated-games.md",
+    "title": "\"When Agents Lie: Premeditation, Persistence, and Exploitation in Repeated Games\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"When Agents Lie: Premeditation, Persistence, and Exploitation in Repeated Games\"",
+      "authors": "Jerick Shi, Terry Jingcheng Zhang, Bernhard Schölkopf, Vincent Conitzer, Zhijing Jin",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.05132",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-06",
+      "updated_at": "2026-07-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CY",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "autonomous-agent-llm, llm-agent, planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-05174-agentgym2-benchmarking-large-language-model-agents-in-de-idealized-real-world-en.md",
+    "title": "\"AgentGym2: Benchmarking Large Language Model Agents in De-Idealized Real-World Environments\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AgentGym2: Benchmarking Large Language Model Agents in De-Idealized Real-World Environments\"",
+      "authors": "Zhiheng Xi, Dingwen Yang, Jiaqi Liu, Jixuan Huang, Honglin Guo, Baodai Huang, Tinggang Chen, Qi Zhang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.05174",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-06",
+      "updated_at": "2026-07-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "language-agent, llm-agent, planning-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-05188-latent-programming-horizons-in-coding-agents.md",
+    "title": "Latent Programming Horizons in Coding Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Latent Programming Horizons in Coding Agents",
+      "authors": "André Silva, Han Tu, Martin Monperrus",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.05188",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-06",
+      "updated_at": "2026-07-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-05202-evoagentbench-benchmarking-agent-self-evolution-via-ability-transfer.md",
+    "title": "\"EvoAgentBench: Benchmarking Agent Self-Evolution via Ability Transfer\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"EvoAgentBench: Benchmarking Agent Self-Evolution via Ability Transfer\"",
+      "authors": "Xingze Gao, Chuanrui Hu, Hongda Chen, Pengfei Yao, Zhao Wang, Yi Bai, Zhengwei Wu, Yunyun Han, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.05202",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-06",
+      "updated_at": "2026-07-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "computer-use",
+        "memory",
+        "planning",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "19",
+      "collection_queries": "agent-evaluation"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-05297-metaskill-evolve-recursive-self-improvement-of-llm-agents-via-two-timescale-meta.md",
+    "title": "\"MetaSkill-Evolve: Recursive Self-Improvement of LLM Agents via Two-Timescale Meta-Skill Evolution\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"MetaSkill-Evolve: Recursive Self-Improvement of LLM Agents via Two-Timescale Meta-Skill Evolution\"",
+      "authors": "Zefeng Wang, Minxi Yan, Jinhe Bi, Sikuan Yan, Volker Tresp, Yunpu Ma",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.05297",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-06",
+      "updated_at": "2026-07-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "planning",
+        "reasoning",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "agent-evaluation, llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-05318-pisas-benchmarking-contextual-integrity-in-multi-user-agentic-systems.md",
+    "title": "\"PiSAs: Benchmarking Contextual Integrity in Multi-User Agentic Systems\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"PiSAs: Benchmarking Contextual Integrity in Multi-User Agentic Systems\"",
+      "authors": "Shubham Gupta, Nazanin Mohammadi Sepahvand, Abhinav Kumar, Cem Subakan, Spandana Gella, Pierre-André Noël, Perouz Taslakian, Eugene Bagdasarian, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.05318",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-06",
+      "updated_at": "2026-07-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "memory",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.MA",
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "20",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-05363-sovereignpa-bench-evaluating-user-owned-personal-agents-under-evolving-intent-pl.md",
+    "title": "\"SovereignPA-Bench: Evaluating User-Owned Personal Agents under Evolving Intent, Platform Mediation, and Consent Constraints\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"SovereignPA-Bench: Evaluating User-Owned Personal Agents under Evolving Intent, Platform Mediation, and Consent Constraints\"",
+      "authors": "Dylan Zongmin Liu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.05363",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-06",
+      "updated_at": "2026-07-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "embodied-agent",
+        "memory",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-evaluation, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-05378-compactionrl-reinforcement-learning-with-context-compaction-for-long-horizon-age.md",
+    "title": "\"CompactionRL: Reinforcement Learning with Context Compaction for Long-Horizon Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"CompactionRL: Reinforcement Learning with Context Compaction for Long-Horizon Agents\"",
+      "authors": "Yujiang Li, Zhenyu Hou, Yi Jing, Jie Tang, Yuxiao Dong",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.05378",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-06",
+      "updated_at": "2026-07-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "coding-agent",
+        "planning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "coding-agent, llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-05391-llm-as-a-verifier-a-general-purpose-verification-framework.md",
+    "title": "\"LLM-as-a-Verifier: A General-Purpose Verification Framework\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"LLM-as-a-Verifier: A General-Purpose Verification Framework\"",
+      "authors": "Jacky Kwok, Shulu Li, Pranav Atreya, Yuejiang Liu, Yixing Jiang, Chelsea Finn, Marco Pavone, Ion Stoica, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.05391",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-06",
+      "updated_at": "2026-07-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "embodied-agent",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL",
+        "cs.LG",
+        "cs.MA",
+        "cs.RO"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-05428-charlie-an-on-premise-multi-agent-retrieval-augmented-generation-system-for-evid.md",
+    "title": "\"CHARLIE: An On-Premise Multi-Agent Retrieval-Augmented Generation System for Evidential Reasoning in Forensic Science\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"CHARLIE: An On-Premise Multi-Agent Retrieval-Augmented Generation System for Evidential Reasoning in Forensic Science\"",
+      "authors": "Leandro D. Carneiro, Andre L. S. Meirelles, Juliano de A. Gomes, Rafael C. A. Cabral",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.05428",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-01",
+      "updated_at": "2026-07-01",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "multi-agent",
+        "planning",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.DL",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "rag-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-05456-prompt-to-paper-agentic-ai-system-for-bioinformatics.md",
+    "title": "\"Prompt-to-Paper: Agentic AI System for Bioinformatics\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Prompt-to-Paper: Agentic AI System for Bioinformatics\"",
+      "authors": "Ramsha Kamran, Maheera Amjad, Zartasha Mustansar, Arsalan Shaukat, Salma Sherbaz, Muhammad U. S. Khan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.05456",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-05",
+      "updated_at": "2026-07-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "multi-agent",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL",
+        "q-bio.QM"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "21",
+      "collection_queries": "agentic-ai, coding-agent, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-05458-learning-to-control-llm-agent-harnesses-with-offline-reinforcement-learning.md",
+    "title": "Learning to Control LLM Agent Harnesses with Offline Reinforcement Learning",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Learning to Control LLM Agent Harnesses with Offline Reinforcement Learning",
+      "authors": "Haiwen Yi, Xinyuan Song",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.05458",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-05",
+      "updated_at": "2026-07-05",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.LG",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-05518-aiauthz-off-host-identity-bound-authorization-for-ai-agents.md",
+    "title": "\"aiAuthZ: Off-Host, Identity-Bound Authorization for AI Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"aiAuthZ: Off-Host, Identity-Bound Authorization for AI Agents\"",
+      "authors": "Sai Varun Kodathala",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.05518",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-06",
+      "updated_at": "2026-07-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "ai-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-05659-agents-with-feelings-personality-and-emotion-in-multi-agent-software-teams.md",
+    "title": "Agents with Feelings? Personality and Emotion in Multi-Agent Software Teams",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Agents with Feelings? Personality and Emotion in Multi-Agent Software Teams",
+      "authors": "Yunyan Ding, Thomas Zimmermann, Iftekhar Ahmed",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.05659",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-06",
+      "updated_at": "2026-07-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "multi-agent",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "llm-agent, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-05666-what-do-ai-agents-actually-change-an-empirical-taxonomy-of-mutation-patterns-in-.md",
+    "title": "What Do AI Agents Actually Change? An Empirical Taxonomy of Mutation Patterns in Performance-Improving Pull Requests",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "What Do AI Agents Actually Change? An Empirical Taxonomy of Mutation Patterns in Performance-Improving Pull Requests",
+      "authors": "Illia Dovhoshliubnyi, Nima Soroush, Ashkan Sami, Alexander Brownlee",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.05666",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-06",
+      "updated_at": "2026-07-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "coding-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "ai-agent, coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-05677-from-conversation-to-contribution-characterizing-coding-agent-in-open-source-sof.md",
+    "title": "\"From Conversation to Contribution: Characterizing Coding Agent in Open-Source Software\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"From Conversation to Contribution: Characterizing Coding Agent in Open-Source Software\"",
+      "authors": "Zihan Fang, Yueke Zhang, Ningzhi Tang, Collin McMillan, Toby Jia-Jun Li, Yu Huang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.05677",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-06",
+      "updated_at": "2026-07-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "computer-use",
+        "multi-agent",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.HC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-05690-memory-in-the-loop-in-process-retrieval-as-extendedworking-memory-for-language-a.md",
+    "title": "\"Memory in the Loop: In-Process Retrieval as ExtendedWorking Memory for Language Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Memory in the Loop: In-Process Retrieval as ExtendedWorking Memory for Language Agents\"",
+      "authors": "Yusuf Khan, Carlo Lipizzi",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.05690",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-06",
+      "updated_at": "2026-07-06",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "language-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-05743-the-balkanization-of-execution-security-research-for-ai-coding-agents-isolation-.md",
+    "title": "\"The Balkanization of Execution-Security Research for AI Coding Agents: Isolation, Access Control, and Time-of-Check-to-Time-of-Use Vulnerabilities\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"The Balkanization of Execution-Security Research for AI Coding Agents: Isolation, Access Control, and Time-of-Check-to-Time-of-Use Vulnerabilities\"",
+      "authors": "Mohammadreza Rashidi",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.05743",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-07",
+      "updated_at": "2026-07-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agentic-ai, coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-05772-detecting-vulnerability-inducing-commits-via-multi-stage-reasoning-with-llm-base.md",
+    "title": "Detecting Vulnerability-Inducing Commits via Multi-Stage Reasoning with LLM-Based Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Detecting Vulnerability-Inducing Commits via Multi-Stage Reasoning with LLM-Based Agents",
+      "authors": "Liyou Chen, Hailong Sun, Xiang Gao, Yue Pan",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.05772",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-07",
+      "updated_at": "2026-07-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "multi-agent",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-05773-beyond-static-evaluation-building-simulation-environments-for-scalable-agentic-r.md",
+    "title": "\"Beyond Static Evaluation: Building Simulation Environments for Scalable Agentic Reinforcement Learning\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Beyond Static Evaluation: Building Simulation Environments for Scalable Agentic Reinforcement Learning\"",
+      "authors": "Akshay Arora, Ishan Nigam, Ashutosh Aggarwal, Shefali Bansal, Krishna Singh, Sweta Kumari, Nikhil Mittal, Shariq Farhan, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.05773",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-07",
+      "updated_at": "2026-07-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "21",
+      "collection_queries": "autonomous-agent-llm, tool-use, web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-05775-beyond-the-leaderboard-a-synthesis-of-tool-use-planning-and-reasoning-failures-i.md",
+    "title": "\"Beyond the Leaderboard: A Synthesis of Tool-Use, Planning, and Reasoning Failures in Large Language Model Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Beyond the Leaderboard: A Synthesis of Tool-Use, Planning, and Reasoning Failures in Large Language Model Agents\"",
+      "authors": "Wael Albayaydh, Rui Zhao, Ivan Flechais",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.05775",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-07",
+      "updated_at": "2026-07-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "coding-agent",
+        "embodied-agent",
+        "multi-agent",
+        "planning",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "23",
+      "collection_queries": "llm-agent, multi-agent-llm, planning-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-05794-from-passive-retrieval-to-active-memory-navigation-learning-to-use-memory-as-a-s.md",
+    "title": "\"From Passive Retrieval to Active Memory Navigation: Learning to Use Memory as a Structured Action Space\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"From Passive Retrieval to Active Memory Navigation: Learning to Use Memory as a Structured Action Space\"",
+      "authors": "Yue Xu, Yutao Sun, Yihao Liu, Mengyu Zhou, Jiayi Qiao, Lu Ma, Kai Tang, Wenjie Wang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.05794",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-07",
+      "updated_at": "2026-07-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "embodied-agent",
+        "memory",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-05805-onnes-a-physics-grounded-multi-agent-llm-simulator-for-cryogenic-fault-diagnosis.md",
+    "title": "\"Onnes: A Physics-Grounded Multi-Agent LLM Simulator for Cryogenic Fault Diagnosis in Quantum Computing Infrastructure\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Onnes: A Physics-Grounded Multi-Agent LLM Simulator for Cryogenic Fault Diagnosis in Quantum Computing Infrastructure\"",
+      "authors": "Praneeth Narisetty, Uday Kumar Reddy Kattamanchi, Shiva Nagendra Babu Kore",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.05805",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-07",
+      "updated_at": "2026-07-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.LG",
+        "quant-ph"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "llm-agent, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-05915-pcbworld-a-benchmark-environment-for-engine-grounded-pcb-design-automation.md",
+    "title": "\"PCBWorld: A Benchmark Environment for Engine-Grounded PCB Design Automation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"PCBWorld: A Benchmark Environment for Engine-Grounded PCB Design Automation\"",
+      "authors": "Hyungseok Song, Junseok Park, Won-Seok Choi, Seohui Bae, Han-Seul Jeong, Youngjoon Park, Soonyoung Lee",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.05915",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-07",
+      "updated_at": "2026-07-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "agentic-ai, llm-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-06000-context-to-execution-integrity-for-llm-agents.md",
+    "title": "Context-to-Execution Integrity for LLM Agents",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Context-to-Execution Integrity for LLM Agents",
+      "authors": "Igor Santos-Grueiro",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.06000",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-07",
+      "updated_at": "2026-07-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CR"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-evaluation, coding-agent, llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-06001-information-limits-and-attractor-dynamics-in-economies-of-frontier-llm-agents-a-.md",
+    "title": "\"Information Limits and Attractor Dynamics in Economies of Frontier LLM Agents: A Pre-Registered Test\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Information Limits and Attractor Dynamics in Economies of Frontier LLM Agents: A Pre-Registered Test\"",
+      "authors": "Cheng Qian",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.06001",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-07",
+      "updated_at": "2026-07-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "multi-agent",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.MA"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "llm-agent, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-06008-polyworkbench-benchmarking-multilingual-long-horizon-llm-agents.md",
+    "title": "\"PolyWorkBench: Benchmarking Multilingual Long-Horizon LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"PolyWorkBench: Benchmarking Multilingual Long-Horizon LLM Agents\"",
+      "authors": "Hongliang Li, Yijin Liu, Zhiwei Zhang, Zihe Liu, Xinyue Lou, Jinan Xu, Fandong Meng, Kaiyu Huang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.06008",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-07",
+      "updated_at": "2026-07-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "25",
+      "collection_queries": "agent-evaluation, llm-agent, planning-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-06080-from-blueprint-to-reality-modeling-and-applying-putnam-s-social-capital-theory-w.md",
+    "title": "\"From Blueprint to Reality: Modeling and Applying Putnam's Social Capital Theory with LLM-based Multi-agent Simulations\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"From Blueprint to Reality: Modeling and Applying Putnam's Social Capital Theory with LLM-based Multi-agent Simulations\"",
+      "authors": "Shiyi Ling, Zhi Zheng, Hui Zheng, Wenjun Xue, Feng Ye, Tong Xu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.06080",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-07",
+      "updated_at": "2026-07-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "multi-agent",
+        "rag",
+        "tool-use",
+        "world-model"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI",
+        "cs.SI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "llm-agent, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-06101-agents-that-teach-towards-designing-incidental-learning-back-into-ai-assisted-so.md",
+    "title": "\"Agents That Teach: Towards Designing Incidental Learning Back into AI-Assisted Software Development\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Agents That Teach: Towards Designing Incidental Learning Back into AI-Assisted Software Development\"",
+      "authors": "Rohit Mehra, Samdyuti Suri, Prithviraj K Tagadinamani, Kapil Singi, Vikrant Kaulgud, Adam P. Burden",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.06101",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-07",
+      "updated_at": "2026-07-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-safety",
+        "coding-agent",
+        "computer-use",
+        "multi-agent",
+        "rag",
+        "reasoning",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI",
+        "cs.CY",
+        "cs.HC"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-06118-webretriever-a-large-scale-comprehensive-benchmark-for-efficient-web-agent-evalu.md",
+    "title": "\"WebRetriever: A Large-Scale Comprehensive Benchmark for Efficient Web Agent Evaluation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"WebRetriever: A Large-Scale Comprehensive Benchmark for Efficient Web Agent Evaluation\"",
+      "authors": "Wei Dong, Tianyu Fu, Zhe Yu, Hanning Wang, Anyang Su, Zhizhou Fang, Yuyang Chen, Shuo Wang, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.06118",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-07",
+      "updated_at": "2026-07-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "agent-safety",
+        "computer-use",
+        "embodied-agent",
+        "rag",
+        "tool-use",
+        "workflow-agent"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CV",
+        "cs.MM"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "21",
+      "collection_queries": "agent-evaluation, web-gui-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-06140-curateevo-data-curation-evolving-for-agentic-post-training.md",
+    "title": "\"CurateEvo: Data-Curation Evolving for Agentic Post-Training\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"CurateEvo: Data-Curation Evolving for Agentic Post-Training\"",
+      "authors": "Dingzirui Wang, Xuanliang Zhang, Keyan Xu, Qingfu Zhu, Wanxiang Che",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.06140",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-07",
+      "updated_at": "2026-07-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "memory",
+        "planning",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-06157-llm-agents-for-deliberative-collaboration-a-study-on-joint-decision-making-under.md",
+    "title": "\"LLM Agents for Deliberative Collaboration: A Study on Joint Decision Making Under Partial Observability\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"LLM Agents for Deliberative Collaboration: A Study on Joint Decision Making Under Partial Observability\"",
+      "authors": "Chenxu Wang, Yongkun Yang, Boyuan Du, Shiwei Lin, Huaping Liu",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.06157",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-07",
+      "updated_at": "2026-07-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "18",
+      "collection_queries": "llm-agent, multi-agent-llm"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-06195-logichunter-testing-llm-agent-frameworks-with-an-agentic-oracle.md",
+    "title": "\"LogicHunter: Testing LLM Agent Frameworks with an Agentic Oracle\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"LogicHunter: Testing LLM Agent Frameworks with an Agentic Oracle\"",
+      "authors": "Minghui Long, Yanjie Zhao, Haoyu Wang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.06195",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-07",
+      "updated_at": "2026-07-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-06223-information-gain-based-rollout-policy-optimization-an-adaptive-tree-structured-r.md",
+    "title": "\"Information Gain-based Rollout Policy Optimization: An Adaptive Tree-Structured Rollout Approach for Multi-Turn LLM Agents\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"Information Gain-based Rollout Policy Optimization: An Adaptive Tree-Structured Rollout Approach for Multi-Turn LLM Agents\"",
+      "authors": "Yijun Zhang, Fan Xu, Jiaxin Ding, Yule Xie, Shiqing Gao, Xin Ding, Haoxiang Zhang, Luoyi Fu, et al.",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.06223",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-07",
+      "updated_at": "2026-07-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "planning",
+        "rag"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "15",
+      "collection_queries": "llm-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-06273-agenttether-graph-guided-diagnosis-and-runtime-intervention-for-reliable-llm-age.md",
+    "title": "\"AgentTether: Graph-Guided Diagnosis and Runtime Intervention for Reliable LLM Agent Operation\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"AgentTether: Graph-Guided Diagnosis and Runtime Intervention for Reliable LLM Agent Operation\"",
+      "authors": "Chenyu Zhao, Shenglin Zhang, Wenwei Gu, Yongqian Sun, Dan Pei, Chetan Bansal, Saravan Rajmohan, Minghua Ma",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.06273",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-07",
+      "updated_at": "2026-07-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "computer-use",
+        "memory",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "17",
+      "collection_queries": "llm-agent, tool-use"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-06341-harnessing-code-agents-for-automatic-software-verification.md",
+    "title": "Harnessing Code Agents for Automatic Software Verification",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "Harnessing Code Agents for Automatic Software Verification",
+      "authors": "Shuangxiang Kan, Shuanglong Kan, Sebastian Ertel",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.06341",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-07",
+      "updated_at": "2026-07-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "memory",
+        "rag",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.FL",
+        "cs.AI",
+        "cs.SE"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "13",
+      "collection_queries": "coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-06411-rubench-a-repository-level-agentic-coding-benchmark-with-natively-authored-russi.md",
+    "title": "\"RuBench: A Repository-Level Agentic Coding Benchmark with Natively Authored Russian Task Specifications\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"RuBench: A Repository-Level Agentic Coding Benchmark with Natively Authored Russian Task Specifications\"",
+      "authors": "Evgeny Shilov",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.06411",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-07",
+      "updated_at": "2026-07-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.SE",
+        "cs.AI",
+        "cs.CL"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agent-evaluation, coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-06413-an-experimental-design-approach-to-evaluating-agentic-ai-s-autonomous-model-disc.md",
+    "title": "\"An Experimental Design Approach to Evaluating Agentic AI's Autonomous Model Discovery\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"An Experimental Design Approach to Evaluating Agentic AI's Autonomous Model Discovery\"",
+      "authors": "Hao He, Xueying Liu, Chris J. Kuhlman, Xinwei Deng",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.06413",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-07",
+      "updated_at": "2026-07-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "coding-agent",
+        "reasoning"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "stat.ME",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "16",
+      "collection_queries": "agentic-ai, coding-agent"
+    }
+  },
+  {
+    "collection": "papers",
+    "path": "papers/items/2026-2607-06452-from-voting-to-agent-collaboration-answer-type-aware-llm-pipelines-for-bioasq-14.md",
+    "title": "\"From Voting to Agent Collaboration: Answer-Type-Aware LLM Pipelines for BioASQ 14b\"",
+    "type": "paper",
+    "meta": {
+      "type": "paper",
+      "title": "\"From Voting to Agent Collaboration: Answer-Type-Aware LLM Pipelines for BioASQ 14b\"",
+      "authors": "Taeyun Roh, Eunha Lee, Wonjune Jang, Sohyun Chung, Junha Jung, Jaewoo Kang",
+      "year": "2026",
+      "venue": "arXiv",
+      "url": "https://arxiv.org/abs/2607.06452",
+      "code_url": [],
+      "source": "arxiv",
+      "collected_at": "2026-07-08",
+      "published_at": "2026-07-07",
+      "updated_at": "2026-07-07",
+      "status": "queued",
+      "relevance": "high",
+      "topics": [
+        "agent-evaluation",
+        "multi-agent",
+        "rag",
+        "reasoning",
+        "tool-use"
+      ],
+      "methods": [],
+      "benchmarks": [],
+      "models": [],
+      "datasets": [
+        "cs.CL",
+        "cs.AI"
+      ],
+      "related_concepts": [],
+      "related_jobs": [],
+      "related_experiments": [],
+      "related_projects": [],
+      "collection_score": "14",
+      "collection_queries": "multi-agent-llm"
+    }
+  },
   {
     "collection": "papers",
     "path": "papers/items/2026-agent-safety-benchmark-taxonomy.md",
diff --git a/data/summary.json b/data/summary.json
index 891025b..6188ee7 100644
--- a/data/summary.json
+++ b/data/summary.json
@@ -1,8 +1,8 @@
 {
-  "total": 21,
+  "total": 989,
   "by_collection": {
     "jobs": 6,
-    "papers": 6,
+    "papers": 974,
     "industry": 9
   },
   "by_status": {
@@ -11,8 +11,8 @@
       "unknown": 2
     },
     "papers": {
-      "skimmed": 5,
-      "queued": 1
+      "queued": 969,
+      "skimmed": 5
     },
     "industry": {
       "analyzed": 8,
@@ -22,44 +22,72 @@
   "top_topics": [
     [
       "agent-evaluation",
-      8
+      837
+    ],
+    [
+      "tool-use",
+      728
+    ],
+    [
+      "rag",
+      460
+    ],
+    [
+      "reasoning",
+      378
+    ],
+    [
+      "planning",
+      347
+    ],
+    [
+      "agent-safety",
+      345
+    ],
+    [
+      "memory",
+      296
+    ],
+    [
+      "computer-use",
+      257
+    ],
+    [
+      "multi-agent",
+      245
+    ],
+    [
+      "workflow-agent",
+      237
+    ],
+    [
+      "coding-agent",
+      205
+    ],
+    [
+      "world-model",
+      83
+    ],
+    [
+      "embodied-agent",
+      66
     ],
     [
       "agent",
       6
     ],
-    [
-      "memory",
-      5
-    ],
     [
       "evaluation",
       4
     ],
-    [
-      "coding-agent",
-      4
-    ],
     [
       "agent-architecture",
       3
     ],
-    [
-      "agent-safety",
-      3
-    ],
     [
       "enterprise-ai",
       3
     ],
-    [
-      "tool-use",
-      2
-    ],
-    [
-      "workflow-agent",
-      1
-    ],
     [
       "llm-infra",
       1
@@ -71,34 +99,6 @@
     [
       "harness",
       1
-    ],
-    [
-      "context-compression",
-      1
-    ],
-    [
-      "agent-productization",
-      1
-    ],
-    [
-      "agent-infra",
-      1
-    ],
-    [
-      "agent-testing",
-      1
-    ],
-    [
-      "harness-engineering",
-      1
-    ],
-    [
-      "human-in-the-loop",
-      1
-    ],
-    [
-      "gui-agent",
-      1
     ]
   ],
   "top_skills": [
diff --git a/papers/README.md b/papers/README.md
index a95766c..7c6ecfd 100644
--- a/papers/README.md
+++ b/papers/README.md
@@ -13,6 +13,7 @@
 - [source-registry](source-registry.md): 来源、搜索方式和筛选规则
 - [reading-queue](reading-queue.md): 阅读队列和优先级
 - [paper-insights](paper-insights.md): 周期性研究洞察
+- [corpus-summary-2026-07-08](corpus-summary-2026-07-08.md): 最近一年 arXiv Agent 论文扩召回摘要
 
 ## Collection Questions
 
diff --git a/papers/corpus-summary-2026-07-08.md b/papers/corpus-summary-2026-07-08.md
new file mode 100644
index 0000000..794210d
--- /dev/null
+++ b/papers/corpus-summary-2026-07-08.md
@@ -0,0 +1,143 @@
+# Paper Corpus Summary: 2025-07-08 to 2026-07-08
+
+status: generated-summary
+
+## Scope
+
+- source: arXiv API
+- window: 2025-07-08 to 2026-07-08
+- query groups: LLM agent, language agent, AI agent, agentic AI, evaluation, memory, tool use, coding agent, web/GUI/computer-use, multi-agent LLM, safety, RAG, planning, autonomous agent
+- unique candidates seen: 1506
+- high-relevance candidates selected into `papers/items/`: 970
+- total paper items after expansion: 974
+
+## Data Files
+
+- full candidate manifest: `data/arxiv-agent-candidates-2025-07-08-to-2026-07-08.json`
+- high-relevance paper manifest: `data/arxiv-agent-papers-2025-07-08-to-2026-07-08.json`
+- full knowledge index: `data/index.json`
+- summary index: `data/summary.json`
+
+## Selected Corpus Distribution
+
+### By Month
+
+| Month | Count |
+| --- | ---: |
+| 2025-07 | 2 |
+| 2025-08 | 4 |
+| 2025-09 | 8 |
+| 2025-10 | 8 |
+| 2025-11 | 4 |
+| 2025-12 | 5 |
+| 2026-01 | 6 |
+| 2026-02 | 28 |
+| 2026-03 | 29 |
+| 2026-04 | 48 |
+| 2026-05 | 181 |
+| 2026-06 | 490 |
+| 2026-07 | 157 |
+
+This distribution is partly query and recency biased, but the May-July 2026 density still strongly suggests rapid acceleration in Agent papers. The full candidate pool shows the same shape: 249 in 2026-05, 785 in 2026-06, and 265 in 2026-07.
+
+### By Primary Query Group
+
+| Query Group | Selected Count |
+| --- | ---: |
+| agent-memory | 103 |
+| agent-safety | 96 |
+| agent-evaluation | 93 |
+| planning-agent | 77 |
+| language-agent | 71 |
+| agentic-ai | 70 |
+| function-calling | 66 |
+| autonomous-agent-llm | 63 |
+| multi-agent-llm | 63 |
+| llm-agent | 62 |
+| rag-agent | 46 |
+| coding-agent | 45 |
+| ai-agent | 41 |
+| tool-use | 37 |
+| web-gui-agent | 37 |
+
+### By Topic Tag
+
+Auto tags are intentionally broad and should be treated as recall-oriented.
+
+| Topic | Count |
+| --- | ---: |
+| agent-evaluation | 831 |
+| tool-use | 727 |
+| rag | 462 |
+| reasoning | 379 |
+| planning | 348 |
+| agent-safety | 343 |
+| memory | 292 |
+| computer-use | 256 |
+| multi-agent | 245 |
+| workflow-agent | 236 |
+| coding-agent | 202 |
+| world-model | 82 |
+| embodied-agent | 66 |
+
+## Trend Readout
+
+### 1. Evaluation Is the Center of Gravity
+
+The strongest corpus signal is not a single Agent architecture. It is evaluation: benchmarks, trajectory auditing, failure diagnosis, safety testing, harness effects, and long-horizon task scoring.
+
+Implication: our knowledge base needs an `agent-evaluation` track that covers final-state success, trajectory-level correctness, failure localization, cost/latency, user interaction quality, and safety.
+
+### 2. Memory Became a Systems Problem
+
+The memory cluster is large and diverse: active memory, selective retention, memory poisoning, memory governance, GUI memory, cross-episode memory, procedural memory, and long-horizon state management.
+
+Implication: treat memory as an architecture subsystem with write policy, read policy, retention, forgetting, provenance, safety, and evaluation. Do not reduce it to vector search.
+
+### 3. Safety Shifted From Refusal to Runtime Risk
+
+Agent safety papers now focus on tool-use leakage, data injection, prompt injection, memory poisoning, action-boundary violations, governance, and runtime controls.
+
+Implication: safety docs should cover permissions, sandboxing, irreversible action confirmation, tool schemas, provenance, and runtime intervention.
+
+### 4. Coding Agents Are Moving Beyond SWE-Bench
+
+The coding-agent cluster includes long-horizon maintenance, repository-level coding, compiler optimization, GUI/terminal agents, action-boundary violations, code memory, and benchmark variants.
+
+Implication: coding-agent evaluation should include repository exploration, multi-file edits, release-note implementation, dialogue clarification, build/test loops, and failure attribution.
+
+### 5. Computer-Use and GUI Agents Are Becoming Their Own Layer
+
+The corpus includes OS/macOS/GUI/mobile/browser/computer-use benchmarks, visual state comparison, task-state representation, and scientific instrument control.
+
+Implication: computer-use agents need their own tool taxonomy, safety checklist, and benchmark map.
+
+### 6. Multi-Agent Is Splitting Into Collaboration, Governance, and Shared Memory
+
+The multi-agent cluster is not only debate or role play. It includes shared memory governance, orchestration, role specialization, majority-vote failure, routing, and organizational behavior.
+
+Implication: multi-agent notes should separate collaboration protocols, shared state, security boundaries, and evaluation.
+
+## Immediate Reading Priorities
+
+P0 clusters:
+
+- Agent evaluation / benchmark methodology
+- Memory architecture and memory evaluation
+- Agent safety / runtime governance
+- Coding agent long-horizon evaluation
+- Computer-use / GUI agent benchmarks
+
+P1 clusters:
+
+- Multi-agent shared memory and orchestration
+- RAG-agent and technical literature agents
+- Domain-specific agents in healthcare, finance, energy, cybersecurity
+- Embodied/world-model agents
+
+## Caveats
+
+- This is a high-recall automated arXiv sweep, not a curated bibliography.
+- Auto tags are broad and may over-tag papers whose abstracts use common words like planning, action, retrieval, or benchmark.
+- The selected 970 are ordered by automated collection score, not by paper quality.
+- The full 1506-candidate manifest should be used as the reserve pool for future expansion.
diff --git a/papers/items/2025-2507-20395-mazeeval-a-benchmark-for-testing-sequential-decision-making-in-language-models.md b/papers/items/2025-2507-20395-mazeeval-a-benchmark-for-testing-sequential-decision-making-in-language-models.md
new file mode 100644
index 0000000..08f2d34
--- /dev/null
+++ b/papers/items/2025-2507-20395-mazeeval-a-benchmark-for-testing-sequential-decision-making-in-language-models.md
@@ -0,0 +1,61 @@
+# Paper: MazeEval: A Benchmark for Testing Sequential Decision-Making in Language Models
+
+---
+type: paper
+title: "MazeEval: A Benchmark for Testing Sequential Decision-Making in Language Models"
+authors: Hafsteinn Einarsson
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2507.20395
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-07-27
+updated_at: 2025-07-27
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - embodied-agent
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, computer-use, embodied-agent, reasoning
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2507.20395
diff --git a/papers/items/2025-2507-20666-mimii-agent-leveraging-llms-with-function-calling-for-relative-evaluation-of-ano.md b/papers/items/2025-2507-20666-mimii-agent-leveraging-llms-with-function-calling-for-relative-evaluation-of-ano.md
new file mode 100644
index 0000000..9da3cfe
--- /dev/null
+++ b/papers/items/2025-2507-20666-mimii-agent-leveraging-llms-with-function-calling-for-relative-evaluation-of-ano.md
@@ -0,0 +1,63 @@
+# Paper: MIMII-Agent: Leveraging LLMs with Function Calling for Relative Evaluation of Anomalous Sound Detection
+
+---
+type: paper
+title: "MIMII-Agent: Leveraging LLMs with Function Calling for Relative Evaluation of Anomalous Sound Detection"
+authors: Harsh Purohit, Tomoya Nishida, Kota Dohi, Takashi Endo, Yohei Kawaguchi
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2507.20666
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-07-28
+updated_at: 2025-07-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - eess.AS
+  - cs.AI
+  - cs.LG
+  - cs.SD
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, rag, tool-use
+- arXiv categories: eess.AS, cs.AI, cs.LG, cs.SD
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2507.20666
diff --git a/papers/items/2025-2508-07575-mcptoolbench-a-large-scale-ai-agent-model-context-protocol-mcp-tool-use-benchmar.md b/papers/items/2025-2508-07575-mcptoolbench-a-large-scale-ai-agent-model-context-protocol-mcp-tool-use-benchmar.md
new file mode 100644
index 0000000..85af450
--- /dev/null
+++ b/papers/items/2025-2508-07575-mcptoolbench-a-large-scale-ai-agent-model-context-protocol-mcp-tool-use-benchmar.md
@@ -0,0 +1,60 @@
+# Paper: MCPToolBench++: A Large Scale AI Agent Model Context Protocol MCP Tool Use Benchmark
+
+---
+type: paper
+title: "MCPToolBench++: A Large Scale AI Agent Model Context Protocol MCP Tool Use Benchmark"
+authors: Shiqing Fan, Xichen Ding, Liang Zhang, Linjian Mo
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2508.07575
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-08-11
+updated_at: 2025-08-11
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 21
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, computer-use, tool-use
+- arXiv categories: cs.AI
+- collection score: 21
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2508.07575
diff --git a/papers/items/2025-2508-11027-hell-or-high-water-evaluating-agentic-recovery-from-external-failures.md b/papers/items/2025-2508-11027-hell-or-high-water-evaluating-agentic-recovery-from-external-failures.md
new file mode 100644
index 0000000..b109e19
--- /dev/null
+++ b/papers/items/2025-2508-11027-hell-or-high-water-evaluating-agentic-recovery-from-external-failures.md
@@ -0,0 +1,61 @@
+# Paper: Hell or High Water: Evaluating Agentic Recovery from External Failures
+
+---
+type: paper
+title: "Hell or High Water: Evaluating Agentic Recovery from External Failures"
+authors: Andrew Wang, Sophia Hager, Adi Asija, Daniel Khashabi, Nicholas Andrews
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2508.11027
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-08-14
+updated_at: 2025-08-14
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, planning, tool-use, workflow-agent
+- arXiv categories: cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2508.11027
diff --git a/papers/items/2025-2508-12685-toolace-mt-non-autoregressive-generation-for-agentic-multi-turn-interaction.md b/papers/items/2025-2508-12685-toolace-mt-non-autoregressive-generation-for-agentic-multi-turn-interaction.md
new file mode 100644
index 0000000..05f5f69
--- /dev/null
+++ b/papers/items/2025-2508-12685-toolace-mt-non-autoregressive-generation-for-agentic-multi-turn-interaction.md
@@ -0,0 +1,61 @@
+# Paper: ToolACE-MT: Non-Autoregressive Generation for Agentic Multi-Turn Interaction
+
+---
+type: paper
+title: "ToolACE-MT: Non-Autoregressive Generation for Agentic Multi-Turn Interaction"
+authors: Xingshan Zeng, Weiwen Liu, Lingzhi Wang, Liangyou Li, Fei Mi, Yasheng Wang, Lifeng Shang, Xin Jiang, et al.
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2508.12685
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-08-18
+updated_at: 2026-02-13
+status: queued
+relevance: high
+topics:
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: tool-use, world-model
+- arXiv categories: cs.CL, cs.AI, cs.LG
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2508.12685
diff --git a/papers/items/2025-2508-17094-powerchain-a-verifiable-agentic-ai-system-for-automating-distribution-grid-analy.md b/papers/items/2025-2508-17094-powerchain-a-verifiable-agentic-ai-system-for-automating-distribution-grid-analy.md
new file mode 100644
index 0000000..35b3611
--- /dev/null
+++ b/papers/items/2025-2508-17094-powerchain-a-verifiable-agentic-ai-system-for-automating-distribution-grid-analy.md
@@ -0,0 +1,63 @@
+# Paper: PowerChain: A Verifiable Agentic AI System for Automating Distribution Grid Analyses
+
+---
+type: paper
+title: "PowerChain: A Verifiable Agentic AI System for Automating Distribution Grid Analyses"
+authors: Emmanuel O. Badmus, Peng Sang, Dimitrios Stamoulis, Amritanshu Pandey
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2508.17094
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-08-23
+updated_at: 2025-10-21
+status: queued
+relevance: high
+topics:
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - eess.SY
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: planning, rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI, eess.SY
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2508.17094
diff --git a/papers/items/2025-2509-02444-appcopilot-toward-general-accurate-long-horizon-and-efficient-mobile-agent.md b/papers/items/2025-2509-02444-appcopilot-toward-general-accurate-long-horizon-and-efficient-mobile-agent.md
new file mode 100644
index 0000000..163301a
--- /dev/null
+++ b/papers/items/2025-2509-02444-appcopilot-toward-general-accurate-long-horizon-and-efficient-mobile-agent.md
@@ -0,0 +1,66 @@
+# Paper: AppCopilot: Toward General, Accurate, Long-Horizon, and Efficient Mobile Agent
+
+---
+type: paper
+title: "AppCopilot: Toward General, Accurate, Long-Horizon, and Efficient Mobile Agent"
+authors: Jingru Fan, Yufan Dang, Jingyao Wu, Huatao Li, Runde Yang, Xiyuan Yang, Yuheng Wang, Chen Qian
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2509.02444
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-09-02
+updated_at: 2025-10-17
+status: queued
+relevance: high
+topics:
+  - computer-use
+  - memory
+  - multi-agent
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+  - cs.CV
+  - cs.HC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: computer-use, memory, multi-agent, planning, reasoning, tool-use
+- arXiv categories: cs.AI, cs.CL, cs.CV, cs.HC
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2509.02444
diff --git a/papers/items/2025-2509-02494-gridmind-llms-powered-agents-for-power-system-analysis-and-operations.md b/papers/items/2025-2509-02494-gridmind-llms-powered-agents-for-power-system-analysis-and-operations.md
new file mode 100644
index 0000000..20b2d3a
--- /dev/null
+++ b/papers/items/2025-2509-02494-gridmind-llms-powered-agents-for-power-system-analysis-and-operations.md
@@ -0,0 +1,60 @@
+# Paper: GridMind: LLMs-Powered Agents for Power System Analysis and Operations
+
+---
+type: paper
+title: "GridMind: LLMs-Powered Agents for Power System Analysis and Operations"
+authors: Hongwei Jin, Kibaek Kim, Jonghwan Kwon
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2509.02494
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-09-02
+updated_at: 2025-09-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, multi-agent, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2509.02494
diff --git a/papers/items/2025-2509-08863-geojson-agents-a-multi-agent-llm-architecture-for-geospatial-analysis-function-c.md b/papers/items/2025-2509-08863-geojson-agents-a-multi-agent-llm-architecture-for-geospatial-analysis-function-c.md
new file mode 100644
index 0000000..bc9e667
--- /dev/null
+++ b/papers/items/2025-2509-08863-geojson-agents-a-multi-agent-llm-architecture-for-geospatial-analysis-function-c.md
@@ -0,0 +1,62 @@
+# Paper: GeoJSON Agents:A Multi-Agent LLM Architecture for Geospatial Analysis-Function Calling vs Code Generation
+
+---
+type: paper
+title: "GeoJSON Agents:A Multi-Agent LLM Architecture for Geospatial Analysis-Function Calling vs Code Generation"
+authors: Qianqian Luo, Qingming Lin, Liuchang Xu, Sensen Wu, Ruichen Mao, Chao Wang, Hailin Feng, Bo Huang, et al.
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2509.08863
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-09-10
+updated_at: 2025-12-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - planning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, multi-agent, planning, tool-use, workflow-agent
+- arXiv categories: cs.SE
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2509.08863
diff --git a/papers/items/2025-2509-10769-agentarch-a-comprehensive-benchmark-to-evaluate-agent-architectures-in-enterpris.md b/papers/items/2025-2509-10769-agentarch-a-comprehensive-benchmark-to-evaluate-agent-architectures-in-enterpris.md
new file mode 100644
index 0000000..7f20ca4
--- /dev/null
+++ b/papers/items/2025-2509-10769-agentarch-a-comprehensive-benchmark-to-evaluate-agent-architectures-in-enterpris.md
@@ -0,0 +1,64 @@
+# Paper: AgentArch: A Comprehensive Benchmark to Evaluate Agent Architectures in Enterprise
+
+---
+type: paper
+title: "AgentArch: A Comprehensive Benchmark to Evaluate Agent Architectures in Enterprise"
+authors: Tara Bogavelli, Roshnee Sharma, Hari Subramani
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2509.10769
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-09-13
+updated_at: 2026-01-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - multi-agent
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, memory, multi-agent, tool-use, workflow-agent
+- arXiv categories: cs.AI, cs.CL, cs.MA
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2509.10769
diff --git a/papers/items/2025-2509-13311-towards-general-agentic-intelligence-via-environment-scaling.md b/papers/items/2025-2509-13311-towards-general-agentic-intelligence-via-environment-scaling.md
new file mode 100644
index 0000000..35f7969
--- /dev/null
+++ b/papers/items/2025-2509-13311-towards-general-agentic-intelligence-via-environment-scaling.md
@@ -0,0 +1,59 @@
+# Paper: Towards General Agentic Intelligence via Environment Scaling
+
+---
+type: paper
+title: Towards General Agentic Intelligence via Environment Scaling
+authors: Runnan Fang, Shihao Cai, Baixuan Li, Jialong Wu, Guangyu Li, Wenbiao Yin, Xinyu Wang, Xiaobin Wang, et al.
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2509.13311
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-09-16
+updated_at: 2025-09-16
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, tool-use
+- arXiv categories: cs.CL
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2509.13311
diff --git a/papers/items/2025-2509-14477-ticket-bench-a-kickoff-for-multilingual-and-regionalized-agent-evaluation.md b/papers/items/2025-2509-14477-ticket-bench-a-kickoff-for-multilingual-and-regionalized-agent-evaluation.md
new file mode 100644
index 0000000..0e73d8b
--- /dev/null
+++ b/papers/items/2025-2509-14477-ticket-bench-a-kickoff-for-multilingual-and-regionalized-agent-evaluation.md
@@ -0,0 +1,60 @@
+# Paper: Ticket-Bench: A Kickoff for Multilingual and Regionalized Agent Evaluation
+
+---
+type: paper
+title: "Ticket-Bench: A Kickoff for Multilingual and Regionalized Agent Evaluation"
+authors: Thales Sales Almeida, João Guilherme Alves Santos, Thiago Laitz, Giovana Kerche Bonás
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2509.14477
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-09-17
+updated_at: 2025-09-17
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, computer-use, reasoning
+- arXiv categories: cs.CL
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2509.14477
diff --git a/papers/items/2025-2509-20998-core-full-path-evaluation-of-llm-agents-beyond-final-state.md b/papers/items/2025-2509-20998-core-full-path-evaluation-of-llm-agents-beyond-final-state.md
new file mode 100644
index 0000000..822d8df
--- /dev/null
+++ b/papers/items/2025-2509-20998-core-full-path-evaluation-of-llm-agents-beyond-final-state.md
@@ -0,0 +1,61 @@
+# Paper: CORE: Full-Path Evaluation of LLM Agents Beyond Final State
+
+---
+type: paper
+title: "CORE: Full-Path Evaluation of LLM Agents Beyond Final State"
+authors: Panagiotis Michelakis, Yiannis Hadjiyiannis, Dimitrios Stamoulis
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2509.20998
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-09-25
+updated_at: 2025-09-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, agent-safety, tool-use, world-model
+- arXiv categories: cs.AI
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2509.20998
diff --git a/papers/items/2025-2509-26553-towards-reliable-benchmarking-a-contamination-free-controllable-evaluation-frame.md b/papers/items/2025-2509-26553-towards-reliable-benchmarking-a-contamination-free-controllable-evaluation-frame.md
new file mode 100644
index 0000000..e0d18f8
--- /dev/null
+++ b/papers/items/2025-2509-26553-towards-reliable-benchmarking-a-contamination-free-controllable-evaluation-frame.md
@@ -0,0 +1,61 @@
+# Paper: Towards Reliable Benchmarking: A Contamination Free, Controllable Evaluation Framework for Multi-step LLM Function Calling
+
+---
+type: paper
+title: "Towards Reliable Benchmarking: A Contamination Free, Controllable Evaluation Framework for Multi-step LLM Function Calling"
+authors: Seiji Maekawa, Jackson Hassell, Pouya Pezeshkpour, Tom Mitchell, Estevam Hruschka
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2509.26553
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-09-30
+updated_at: 2026-02-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.PL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, reasoning, tool-use
+- arXiv categories: cs.CL, cs.PL
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2509.26553
diff --git a/papers/items/2025-2510-03847-small-language-models-for-agentic-systems-a-survey-of-architectures-capabilities.md b/papers/items/2025-2510-03847-small-language-models-for-agentic-systems-a-survey-of-architectures-capabilities.md
new file mode 100644
index 0000000..aa4b88a
--- /dev/null
+++ b/papers/items/2025-2510-03847-small-language-models-for-agentic-systems-a-survey-of-architectures-capabilities.md
@@ -0,0 +1,65 @@
+# Paper: Small Language Models for Agentic Systems: A Survey of Architectures, Capabilities, and Deployment Trade offs
+
+---
+type: paper
+title: "Small Language Models for Agentic Systems: A Survey of Architectures, Capabilities, and Deployment Trade offs"
+authors: Raghav Sharma, Manan Mehta
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2510.03847
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-10-04
+updated_at: 2025-10-04
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - computer-use
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, coding-agent, computer-use, planning, rag, reasoning, tool-use
+- arXiv categories: cs.AI, cs.LG
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2510.03847
diff --git a/papers/items/2025-2510-04206-agentrl-scaling-agentic-reinforcement-learning-with-a-multi-turn-multi-task-fram.md b/papers/items/2025-2510-04206-agentrl-scaling-agentic-reinforcement-learning-with-a-multi-turn-multi-task-fram.md
new file mode 100644
index 0000000..0ede631
--- /dev/null
+++ b/papers/items/2025-2510-04206-agentrl-scaling-agentic-reinforcement-learning-with-a-multi-turn-multi-task-fram.md
@@ -0,0 +1,59 @@
+# Paper: AgentRL: Scaling Agentic Reinforcement Learning with a Multi-Turn, Multi-Task Framework
+
+---
+type: paper
+title: "AgentRL: Scaling Agentic Reinforcement Learning with a Multi-Turn, Multi-Task Framework"
+authors: Hanchen Zhang, Xiao Liu, Bowen Lv, Xueqiao Sun, Bohao Jing, Iat Long Iong, Zhenyu Hou, Zehan Qi, et al.
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2510.04206
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-10-05
+updated_at: 2025-10-05
+status: queued
+relevance: high
+topics:
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2510.04206
diff --git a/papers/items/2025-2510-14548-llm-agents-beyond-utility-an-open-ended-perspective.md b/papers/items/2025-2510-14548-llm-agents-beyond-utility-an-open-ended-perspective.md
new file mode 100644
index 0000000..ae8851e
--- /dev/null
+++ b/papers/items/2025-2510-14548-llm-agents-beyond-utility-an-open-ended-perspective.md
@@ -0,0 +1,61 @@
+# Paper: LLM Agents Beyond Utility: An Open-Ended Perspective
+
+---
+type: paper
+title: "LLM Agents Beyond Utility: An Open-Ended Perspective"
+authors: Asen Nachkov, Xi Wang, Luc Van Gool
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2510.14548
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-10-16
+updated_at: 2025-10-16
+status: queued
+relevance: high
+topics:
+  - memory
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: memory, planning, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2510.14548
diff --git a/papers/items/2025-2510-18586-tokencake-a-kv-cache-centric-serving-framework-for-llm-based-multi-agent-applica.md b/papers/items/2025-2510-18586-tokencake-a-kv-cache-centric-serving-framework-for-llm-based-multi-agent-applica.md
new file mode 100644
index 0000000..c525ce5
--- /dev/null
+++ b/papers/items/2025-2510-18586-tokencake-a-kv-cache-centric-serving-framework-for-llm-based-multi-agent-applica.md
@@ -0,0 +1,61 @@
+# Paper: TokenCake: A KV-Cache-centric Serving Framework for LLM-based Multi-Agent Applications
+
+---
+type: paper
+title: "TokenCake: A KV-Cache-centric Serving Framework for LLM-based Multi-Agent Applications"
+authors: Zhuohang Bian, Feiyang Wu, Zhuoran Li, Teng Ma, Youwei Zhuo
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2510.18586
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-10-21
+updated_at: 2026-05-20
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - multi-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.DC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, computer-use, memory, multi-agent
+- arXiv categories: cs.DC
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2510.18586
diff --git a/papers/items/2025-2510-21524-eu-agent-bench-measuring-illegal-behavior-of-llm-agents-under-eu-law.md b/papers/items/2025-2510-21524-eu-agent-bench-measuring-illegal-behavior-of-llm-agents-under-eu-law.md
new file mode 100644
index 0000000..fc5ea5b
--- /dev/null
+++ b/papers/items/2025-2510-21524-eu-agent-bench-measuring-illegal-behavior-of-llm-agents-under-eu-law.md
@@ -0,0 +1,61 @@
+# Paper: EU-Agent-Bench: Measuring Illegal Behavior of LLM Agents Under EU Law
+
+---
+type: paper
+title: "EU-Agent-Bench: Measuring Illegal Behavior of LLM Agents Under EU Law"
+authors: Ilija Lichkovski, Alexander Müller, Mariam Ibrahim, Tiwai Mhundwa
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2510.21524
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-10-24
+updated_at: 2025-10-24
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, agent-safety, rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2510.21524
diff --git a/papers/items/2025-2510-22768-seeing-is-believing-evaluating-vision-language-model-susceptibility-in-agent-to-.md b/papers/items/2025-2510-22768-seeing-is-believing-evaluating-vision-language-model-susceptibility-in-agent-to-.md
new file mode 100644
index 0000000..0e52ea5
--- /dev/null
+++ b/papers/items/2025-2510-22768-seeing-is-believing-evaluating-vision-language-model-susceptibility-in-agent-to-.md
@@ -0,0 +1,62 @@
+# Paper: Seeing is Believing? Evaluating Vision-Language Model Susceptibility in Agent-to-Agent Multimodal Persuasion
+
+---
+type: paper
+title: Seeing is Believing? Evaluating Vision-Language Model Susceptibility in Agent-to-Agent Multimodal Persuasion
+authors: Haoyi Qiu, Yilun Zhou, Pranav Narayanan Venkit, Kung-Hsiang Huang, Jiaxin Zhang, Nanyun Peng, Chien-Sheng Wu
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2510.22768
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-10-26
+updated_at: 2026-06-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, agent-safety, multi-agent, rag, tool-use
+- arXiv categories: cs.CL
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2510.22768
diff --git a/papers/items/2025-2510-24645-funreason-mt-technical-report-advanced-data-synthesis-solution-for-real-world-mu.md b/papers/items/2025-2510-24645-funreason-mt-technical-report-advanced-data-synthesis-solution-for-real-world-mu.md
new file mode 100644
index 0000000..1ae8e05
--- /dev/null
+++ b/papers/items/2025-2510-24645-funreason-mt-technical-report-advanced-data-synthesis-solution-for-real-world-mu.md
@@ -0,0 +1,61 @@
+# Paper: FunReason-MT Technical Report: Advanced Data Synthesis Solution for Real-world Multi-Turn Tool-use
+
+---
+type: paper
+title: "FunReason-MT Technical Report: Advanced Data Synthesis Solution for Real-world Multi-Turn Tool-use"
+authors: Zengzhuang Xu, Bingguang Hao, Zechuan Wang, Yuntao Wen, Xinyi Xu, Yang Liu, Long Chen, Dong Wang, et al.
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2510.24645
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-10-28
+updated_at: 2025-11-16
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - multi-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, computer-use, multi-agent, tool-use
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2510.24645
diff --git a/papers/items/2025-2510-26167-toolrm-towards-agentic-tool-use-reward-modeling.md b/papers/items/2025-2510-26167-toolrm-towards-agentic-tool-use-reward-modeling.md
new file mode 100644
index 0000000..eaa8e1d
--- /dev/null
+++ b/papers/items/2025-2510-26167-toolrm-towards-agentic-tool-use-reward-modeling.md
@@ -0,0 +1,60 @@
+# Paper: ToolRM: Towards Agentic Tool-Use Reward Modeling
+
+---
+type: paper
+title: "ToolRM: Towards Agentic Tool-Use Reward Modeling"
+authors: Renhao Li, Jianhong Tu, Yang Su, Yantao Liu, Fei Huang, Hamid Alinejad-Rokny, Derek F. Wong, Junyang Lin, et al.
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2510.26167
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-10-30
+updated_at: 2026-01-13
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, tool-use
+- arXiv categories: cs.AI, cs.CL
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2510.26167
diff --git a/papers/items/2025-2511-04847-test-time-adaptation-for-llm-agents-via-environment-interaction.md b/papers/items/2025-2511-04847-test-time-adaptation-for-llm-agents-via-environment-interaction.md
new file mode 100644
index 0000000..32c1304
--- /dev/null
+++ b/papers/items/2025-2511-04847-test-time-adaptation-for-llm-agents-via-environment-interaction.md
@@ -0,0 +1,63 @@
+# Paper: Test-Time Adaptation for LLM Agents via Environment Interaction
+
+---
+type: paper
+title: Test-Time Adaptation for LLM Agents via Environment Interaction
+authors: Arthur Chen, Zuxin Liu, Jianguo Zhang, Akshara Prabhakar, Zhiwei Liu, Shelby Heinecke, Silvio Savarese, Victor Zhong, et al.
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2511.04847
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-11-06
+updated_at: 2026-02-22
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - embodied-agent
+  - rag
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, agent-safety, embodied-agent, rag, tool-use, world-model
+- arXiv categories: cs.LG
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2511.04847
diff --git a/papers/items/2025-2511-11169-refine-and-align-confidence-calibration-through-multi-agent-interaction-in-vqa.md b/papers/items/2025-2511-11169-refine-and-align-confidence-calibration-through-multi-agent-interaction-in-vqa.md
new file mode 100644
index 0000000..36d5f53
--- /dev/null
+++ b/papers/items/2025-2511-11169-refine-and-align-confidence-calibration-through-multi-agent-interaction-in-vqa.md
@@ -0,0 +1,63 @@
+# Paper: Refine and Align: Confidence Calibration through Multi-Agent Interaction in VQA
+
+---
+type: paper
+title: "Refine and Align: Confidence Calibration through Multi-Agent Interaction in VQA"
+authors: Ayush Pandey, Jai Bardhan, Ishita Jain, Ramya S Hebbalaguppe, Rohan Raju Dhanakshirur, Lovekesh Vig
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2511.11169
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-11-14
+updated_at: 2025-11-14
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - embodied-agent
+  - multi-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, embodied-agent, multi-agent, tool-use
+- arXiv categories: cs.CV, cs.AI, cs.LG
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2511.11169
diff --git a/papers/items/2025-2511-15203-taxonomy-evaluation-and-exploitation-of-ipi-centric-llm-agent-defense-frameworks.md b/papers/items/2025-2511-15203-taxonomy-evaluation-and-exploitation-of-ipi-centric-llm-agent-defense-frameworks.md
new file mode 100644
index 0000000..16acdbf
--- /dev/null
+++ b/papers/items/2025-2511-15203-taxonomy-evaluation-and-exploitation-of-ipi-centric-llm-agent-defense-frameworks.md
@@ -0,0 +1,62 @@
+# Paper: Taxonomy, Evaluation and Exploitation of IPI-Centric LLM Agent Defense Frameworks
+
+---
+type: paper
+title: Taxonomy, Evaluation and Exploitation of IPI-Centric LLM Agent Defense Frameworks
+authors: Zimo Ji, Xunguang Wang, Zongjie Li, Pingchuan Ma, Yudong Gao, Daoyuan Wu, Xincheng Yan, Tian Tian, et al.
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2511.15203
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-11-19
+updated_at: 2025-11-19
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, agent-safety, rag, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2511.15203
diff --git a/papers/items/2025-2511-22138-tinyllm-evaluation-and-optimization-of-small-language-models-for-agentic-tasks-o.md b/papers/items/2025-2511-22138-tinyllm-evaluation-and-optimization-of-small-language-models-for-agentic-tasks-o.md
new file mode 100644
index 0000000..5661519
--- /dev/null
+++ b/papers/items/2025-2511-22138-tinyllm-evaluation-and-optimization-of-small-language-models-for-agentic-tasks-o.md
@@ -0,0 +1,60 @@
+# Paper: TinyLLM: Evaluation and Optimization of Small Language Models for Agentic Tasks on Edge Devices
+
+---
+type: paper
+title: "TinyLLM: Evaluation and Optimization of Small Language Models for Agentic Tasks on Edge Devices"
+authors: Mohd Ariful Haque, Fahad Rahman, Kishor Datta Gupta, Khalil Shujaee, Roy George
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2511.22138
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-11-27
+updated_at: 2025-11-27
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, agent-safety, tool-use
+- arXiv categories: cs.LG
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2511.22138
diff --git a/papers/items/2025-2512-02605-iact-a-self-organizing-recursive-model-for-general-ai-agents-a-technical-white-p.md b/papers/items/2025-2512-02605-iact-a-self-organizing-recursive-model-for-general-ai-agents-a-technical-white-p.md
new file mode 100644
index 0000000..f57e13a
--- /dev/null
+++ b/papers/items/2025-2512-02605-iact-a-self-organizing-recursive-model-for-general-ai-agents-a-technical-white-p.md
@@ -0,0 +1,65 @@
+# Paper: IACT: A Self-Organizing Recursive Model for General AI Agents: A Technical White Paper on the Architecture Behind kragent.ai
+
+---
+type: paper
+title: "IACT: A Self-Organizing Recursive Model for General AI Agents: A Technical White Paper on the Architecture Behind kragent.ai"
+authors: Pengju Lu
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2512.02605
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-12-02
+updated_at: 2025-12-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.MA
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, computer-use, memory, rag, tool-use, workflow-agent
+- arXiv categories: cs.AI, cs.MA, cs.SE
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2512.02605
diff --git a/papers/items/2025-2512-11682-medai-evaluating-txagent-s-therapeutic-agentic-reasoning-in-the-neurips-cure-ben.md b/papers/items/2025-2512-11682-medai-evaluating-txagent-s-therapeutic-agentic-reasoning-in-the-neurips-cure-ben.md
new file mode 100644
index 0000000..6a031df
--- /dev/null
+++ b/papers/items/2025-2512-11682-medai-evaluating-txagent-s-therapeutic-agentic-reasoning-in-the-neurips-cure-ben.md
@@ -0,0 +1,65 @@
+# Paper: MedAI: Evaluating TxAgent's Therapeutic Agentic Reasoning in the NeurIPS CURE-Bench Competition
+
+---
+type: paper
+title: "MedAI: Evaluating TxAgent's Therapeutic Agentic Reasoning in the NeurIPS CURE-Bench Competition"
+authors: Tim Cofala, Christian Kalfar, Jingge Xiao, Johanna Schrader, Michelle Tang, Wolfgang Nejdl
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2512.11682
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-12-12
+updated_at: 2026-06-15
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, agent-safety, computer-use, planning, rag, reasoning, tool-use
+- arXiv categories: cs.AI, cs.LG
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2512.11682
diff --git a/papers/items/2025-2512-23611-close-the-loop-synthesizing-infinite-tool-use-data-via-multi-agent-role-playing.md b/papers/items/2025-2512-23611-close-the-loop-synthesizing-infinite-tool-use-data-via-multi-agent-role-playing.md
new file mode 100644
index 0000000..a93c320
--- /dev/null
+++ b/papers/items/2025-2512-23611-close-the-loop-synthesizing-infinite-tool-use-data-via-multi-agent-role-playing.md
@@ -0,0 +1,62 @@
+# Paper: Close the Loop: Synthesizing Infinite Tool-Use Data via Multi-Agent Role-Playing
+
+---
+type: paper
+title: "Close the Loop: Synthesizing Infinite Tool-Use Data via Multi-Agent Role-Playing"
+authors: Yuwen Li, Wei Zhang, Zelong Huang, Mason Yang, Jiajun Wu, Shawn Guo, Huahao Hu, Lingyi Sun, et al.
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2512.23611
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-12-29
+updated_at: 2025-12-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, multi-agent, rag, tool-use, workflow-agent
+- arXiv categories: cs.CL
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2512.23611
diff --git a/papers/items/2025-2512-23647-nested-browser-use-learning-for-agentic-information-seeking.md b/papers/items/2025-2512-23647-nested-browser-use-learning-for-agentic-information-seeking.md
new file mode 100644
index 0000000..b77ac95
--- /dev/null
+++ b/papers/items/2025-2512-23647-nested-browser-use-learning-for-agentic-information-seeking.md
@@ -0,0 +1,65 @@
+# Paper: Nested Browser-Use Learning for Agentic Information Seeking
+
+---
+type: paper
+title: Nested Browser-Use Learning for Agentic Information Seeking
+authors: Baixuan Li, Jialong Wu, Wenbiao Yin, Kuan Li, Zhongwang Zhang, Huifeng Yin, Zhengwei Tao, Liwen Zhang, et al.
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2512.23647
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-12-29
+updated_at: 2025-12-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+  - cs.IR
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, computer-use, rag, reasoning, tool-use
+- arXiv categories: cs.CL, cs.AI, cs.IR, cs.MA
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2512.23647
diff --git a/papers/items/2025-2512-23747-state-of-the-art-small-language-coder-model-mify-coder.md b/papers/items/2025-2512-23747-state-of-the-art-small-language-coder-model-mify-coder.md
new file mode 100644
index 0000000..9d6a861
--- /dev/null
+++ b/papers/items/2025-2512-23747-state-of-the-art-small-language-coder-model-mify-coder.md
@@ -0,0 +1,64 @@
+# Paper: State-of-the-art Small Language Coder Model: Mify-Coder
+
+---
+type: paper
+title: "State-of-the-art Small Language Coder Model: Mify-Coder"
+authors: Abhinav Parmar, Abhisek Panigrahi, Abhishek Kumar Dwivedi, Abhishek Bhattacharya, Adarsh Ramachandra, Aditya Choudhary, Aditya Garg, Aditya Raj, et al.
+year: 2025
+venue: arXiv
+url: https://arxiv.org/abs/2512.23747
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2025-12-26
+updated_at: 2025-12-26
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - computer-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, agent-safety, coding-agent, computer-use, workflow-agent
+- arXiv categories: cs.SE, cs.AI, cs.CL
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2512.23747
diff --git a/papers/items/2026-2601-00268-beyond-perfect-apis-a-comprehensive-evaluation-of-llm-agents-under-real-world-ap.md b/papers/items/2026-2601-00268-beyond-perfect-apis-a-comprehensive-evaluation-of-llm-agents-under-real-world-ap.md
new file mode 100644
index 0000000..e73c5b8
--- /dev/null
+++ b/papers/items/2026-2601-00268-beyond-perfect-apis-a-comprehensive-evaluation-of-llm-agents-under-real-world-ap.md
@@ -0,0 +1,60 @@
+# Paper: Beyond Perfect APIs: A Comprehensive Evaluation of LLM Agents Under Real-World API Complexity
+
+---
+type: paper
+title: "Beyond Perfect APIs: A Comprehensive Evaluation of LLM Agents Under Real-World API Complexity"
+authors: Doyoung Kim, Zhiwei Ren, Jie Hao, Zhongkai Sun, Lichao Wang, Xiyao Ma, Zack Ye, Xu Han, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2601.00268
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-01-01
+updated_at: 2026-01-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, tool-use
+- arXiv categories: cs.CL, cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2601.00268
diff --git a/papers/items/2026-2601-05467-stelp-secure-transpilation-and-execution-of-llm-generated-programs.md b/papers/items/2026-2601-05467-stelp-secure-transpilation-and-execution-of-llm-generated-programs.md
new file mode 100644
index 0000000..9856897
--- /dev/null
+++ b/papers/items/2026-2601-05467-stelp-secure-transpilation-and-execution-of-llm-generated-programs.md
@@ -0,0 +1,64 @@
+# Paper: STELP: Secure Transpilation and Execution of LLM-Generated Programs
+
+---
+type: paper
+title: "STELP: Secure Transpilation and Execution of LLM-Generated Programs"
+authors: Swapnil Shinde, Sahil Wadhwa, Andy Luo, Akshay Gupta, Mohammad Shahed Sorower
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2601.05467
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-01-09
+updated_at: 2026-01-15
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, agent-safety, multi-agent, planning, reasoning, tool-use
+- arXiv categories: cs.SE, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2601.05467
diff --git a/papers/items/2026-2601-06007-don-t-break-the-cache-an-evaluation-of-prompt-caching-for-long-horizon-agentic-t.md b/papers/items/2026-2601-06007-don-t-break-the-cache-an-evaluation-of-prompt-caching-for-long-horizon-agentic-t.md
new file mode 100644
index 0000000..9117dfc
--- /dev/null
+++ b/papers/items/2026-2601-06007-don-t-break-the-cache-an-evaluation-of-prompt-caching-for-long-horizon-agentic-t.md
@@ -0,0 +1,61 @@
+# Paper: Don't Break the Cache: An Evaluation of Prompt Caching for Long-Horizon Agentic Tasks
+
+---
+type: paper
+title: "Don't Break the Cache: An Evaluation of Prompt Caching for Long-Horizon Agentic Tasks"
+authors: Elias Lumer, Faheem Nizar, Akshaya Jangiti, Kevin Frank, Anmol Gulati, Mandar Phadate, Vamse Kumar Subbiah
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2601.06007
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-01-09
+updated_at: 2026-01-31
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, computer-use, planning, tool-use
+- arXiv categories: cs.CL
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2601.06007
diff --git a/papers/items/2026-2601-06606-cedar-context-engineering-for-agentic-data-science.md b/papers/items/2026-2601-06606-cedar-context-engineering-for-agentic-data-science.md
new file mode 100644
index 0000000..0f46997
--- /dev/null
+++ b/papers/items/2026-2601-06606-cedar-context-engineering-for-agentic-data-science.md
@@ -0,0 +1,60 @@
+# Paper: CEDAR: Context Engineering for Agentic Data Science
+
+---
+type: paper
+title: "CEDAR: Context Engineering for Agentic Data Science"
+authors: Rishiraj Saha Roy, Chris Hinze, Luzian Hahn, Fabian Kuech
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2601.06606
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-01-10
+updated_at: 2026-04-22
+status: queued
+relevance: high
+topics:
+  - planning
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: planning, workflow-agent
+- arXiv categories: cs.LG, cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2601.06606
diff --git a/papers/items/2026-2601-12988-paperguide-making-small-language-model-paper-reading-agents-more-efficient.md b/papers/items/2026-2601-12988-paperguide-making-small-language-model-paper-reading-agents-more-efficient.md
new file mode 100644
index 0000000..5b0276e
--- /dev/null
+++ b/papers/items/2026-2601-12988-paperguide-making-small-language-model-paper-reading-agents-more-efficient.md
@@ -0,0 +1,62 @@
+# Paper: PaperGuide: Making Small Language-Model Paper-Reading Agents More Efficient
+
+---
+type: paper
+title: "PaperGuide: Making Small Language-Model Paper-Reading Agents More Efficient"
+authors: Zijian Wang, Tiancheng Huang, Hanqi Li, Da Ma, Lu Chen, Kai Yu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2601.12988
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-01-19
+updated_at: 2026-01-19
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, computer-use, planning, reasoning, tool-use
+- arXiv categories: cs.LG
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2601.12988
diff --git a/papers/items/2026-2601-14652-mas-orchestra-understanding-and-improving-multi-agent-reasoning-through-holistic.md b/papers/items/2026-2601-14652-mas-orchestra-understanding-and-improving-multi-agent-reasoning-through-holistic.md
new file mode 100644
index 0000000..e2db118
--- /dev/null
+++ b/papers/items/2026-2601-14652-mas-orchestra-understanding-and-improving-multi-agent-reasoning-through-holistic.md
@@ -0,0 +1,63 @@
+# Paper: MAS-Orchestra: Understanding and Improving Multi-Agent Reasoning Through Holistic Orchestration and Controlled Benchmarks
+
+---
+type: paper
+title: "MAS-Orchestra: Understanding and Improving Multi-Agent Reasoning Through Holistic Orchestration and Controlled Benchmarks"
+authors: Zixuan Ke, Yifei Ming, Austin Xu, Ryan Chin, Xuan-Phi Nguyen, Prathyusha Jwalapuram, Jiayu Wang, Semih Yavuz, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2601.14652
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-01-21
+updated_at: 2026-05-21
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - multi-agent
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, computer-use, multi-agent, reasoning
+- arXiv categories: cs.AI, cs.CL, cs.MA
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2601.14652
diff --git a/papers/items/2026-2602-03117-agentdyn-are-your-agent-security-defenses-deployable-in-real-world-dynamic-envir.md b/papers/items/2026-2602-03117-agentdyn-are-your-agent-security-defenses-deployable-in-real-world-dynamic-envir.md
new file mode 100644
index 0000000..282879b
--- /dev/null
+++ b/papers/items/2026-2602-03117-agentdyn-are-your-agent-security-defenses-deployable-in-real-world-dynamic-envir.md
@@ -0,0 +1,61 @@
+# Paper: AgentDyn: Are Your Agent Security Defenses Deployable in Real-World Dynamic Environments?
+
+---
+type: paper
+title: "AgentDyn: Are Your Agent Security Defenses Deployable in Real-World Dynamic Environments?"
+authors: Hao Li, Ruoyao Wen, Shanghao Shi, Ning Zhang, Yevgeniy Vorobeychik, Chaowei Xiao
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2602.03117
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-02-03
+updated_at: 2026-05-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, planning, tool-use
+- arXiv categories: cs.CR
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2602.03117
diff --git a/papers/items/2026-2602-03224-tame-a-trustworthy-test-time-evolution-of-agent-memory-with-systematic-benchmark.md b/papers/items/2026-2602-03224-tame-a-trustworthy-test-time-evolution-of-agent-memory-with-systematic-benchmark.md
new file mode 100644
index 0000000..4d309cb
--- /dev/null
+++ b/papers/items/2026-2602-03224-tame-a-trustworthy-test-time-evolution-of-agent-memory-with-systematic-benchmark.md
@@ -0,0 +1,63 @@
+# Paper: TAME: A Trustworthy Test-Time Evolution of Agent Memory with Systematic Benchmarking
+
+---
+type: paper
+title: "TAME: A Trustworthy Test-Time Evolution of Agent Memory with Systematic Benchmarking"
+authors: Yu Cheng, Yongkang Hu, Jiuan Zhou, Yushuo Zhang, Yihang Chen, Huichi Zhou, Mingang Chen, Zhizhong Zhang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2602.03224
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-02-03
+updated_at: 2026-06-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - memory
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, computer-use, memory, reasoning
+- arXiv categories: cs.AI, cs.LG
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2602.03224
diff --git a/papers/items/2026-2602-03786-aorchestra-automating-sub-agent-creation-for-agentic-orchestration.md b/papers/items/2026-2602-03786-aorchestra-automating-sub-agent-creation-for-agentic-orchestration.md
new file mode 100644
index 0000000..e802a08
--- /dev/null
+++ b/papers/items/2026-2602-03786-aorchestra-automating-sub-agent-creation-for-agentic-orchestration.md
@@ -0,0 +1,63 @@
+# Paper: AOrchestra: Automating Sub-Agent Creation for Agentic Orchestration
+
+---
+type: paper
+title: "AOrchestra: Automating Sub-Agent Creation for Agentic Orchestration"
+authors: Jianhao Ruan, Zhihao Xu, Yiran Peng, Fashen Ren, Zhaoyang Yu, Xinbing Liang, Jinyu Xiang, Yongru Chen, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2602.03786
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-02-03
+updated_at: 2026-02-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - planning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, coding-agent, planning, tool-use, workflow-agent
+- arXiv categories: cs.AI, cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2602.03786
diff --git a/papers/items/2026-2602-05115-socialveil-probing-social-intelligence-of-language-agents-under-communication-ba.md b/papers/items/2026-2602-05115-socialveil-probing-social-intelligence-of-language-agents-under-communication-ba.md
new file mode 100644
index 0000000..88c66e6
--- /dev/null
+++ b/papers/items/2026-2602-05115-socialveil-probing-social-intelligence-of-language-agents-under-communication-ba.md
@@ -0,0 +1,61 @@
+# Paper: SocialVeil: Probing Social Intelligence of Language Agents under Communication Barriers
+
+---
+type: paper
+title: "SocialVeil: Probing Social Intelligence of Language Agents under Communication Barriers"
+authors: Keyang Xuan, Pengda Wang, Chongrui Ye, Haofei Yu, Tal August, Jiaxuan You
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2602.05115
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-02-04
+updated_at: 2026-02-04
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, rag, tool-use
+- arXiv categories: cs.AI, cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2602.05115
diff --git a/papers/items/2026-2602-05302-piearena-ranking-and-profiling-language-agents-in-realistic-negotiation-scenario.md b/papers/items/2026-2602-05302-piearena-ranking-and-profiling-language-agents-in-realistic-negotiation-scenario.md
new file mode 100644
index 0000000..b4f8a30
--- /dev/null
+++ b/papers/items/2026-2602-05302-piearena-ranking-and-profiling-language-agents-in-realistic-negotiation-scenario.md
@@ -0,0 +1,61 @@
+# Paper: PieArena: Ranking and Profiling Language Agents in Realistic Negotiation Scenarios
+
+---
+type: paper
+title: "PieArena: Ranking and Profiling Language Agents in Realistic Negotiation Scenarios"
+authors: Chris Zhu, Sasha Cui, Will Sanok Dufallo, Runzhi Jin, Zhen Xu, Linjun Zhang, Daylian Cain
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2602.05302
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-02-05
+updated_at: 2026-06-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, multi-agent, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2602.05302
diff --git a/papers/items/2026-2602-05386-spider-sense-intrinsic-risk-sensing-for-efficient-agent-defense-with-hierarchica.md b/papers/items/2026-2602-05386-spider-sense-intrinsic-risk-sensing-for-efficient-agent-defense-with-hierarchica.md
new file mode 100644
index 0000000..247d60b
--- /dev/null
+++ b/papers/items/2026-2602-05386-spider-sense-intrinsic-risk-sensing-for-efficient-agent-defense-with-hierarchica.md
@@ -0,0 +1,62 @@
+# Paper: Spider-Sense: Intrinsic Risk Sensing for Efficient Agent Defense with Hierarchical Adaptive Screening
+
+---
+type: paper
+title: "Spider-Sense: Intrinsic Risk Sensing for Efficient Agent Defense with Hierarchical Adaptive Screening"
+authors: Zhenxiong Yu, Zhi Yang, Zhiheng Jin, Shuhe Wang, Heng Zhang, Yanlin Fei, Lingfeng Zeng, Fangqi Lou, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2602.05386
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-02-05
+updated_at: 2026-02-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, reasoning, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2602.05386
diff --git a/papers/items/2026-2602-07391-naamse-framework-for-evolutionary-security-evaluation-of-agents.md b/papers/items/2026-2602-07391-naamse-framework-for-evolutionary-security-evaluation-of-agents.md
new file mode 100644
index 0000000..2624b6b
--- /dev/null
+++ b/papers/items/2026-2602-07391-naamse-framework-for-evolutionary-security-evaluation-of-agents.md
@@ -0,0 +1,60 @@
+# Paper: NAAMSE: Framework for Evolutionary Security Evaluation of Agents
+
+---
+type: paper
+title: "NAAMSE: Framework for Evolutionary Security Evaluation of Agents"
+authors: Kunal Pai, Parth Shah, Harshil Patel
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2602.07391
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-02-07
+updated_at: 2026-03-08
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety
+- arXiv categories: cs.AI, cs.MA
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2602.07391
diff --git a/papers/items/2026-2602-07652-agent-fence-mapping-security-vulnerabilities-across-deep-research-agents.md b/papers/items/2026-2602-07652-agent-fence-mapping-security-vulnerabilities-across-deep-research-agents.md
new file mode 100644
index 0000000..2223e54
--- /dev/null
+++ b/papers/items/2026-2602-07652-agent-fence-mapping-security-vulnerabilities-across-deep-research-agents.md
@@ -0,0 +1,64 @@
+# Paper: Agent-Fence: Mapping Security Vulnerabilities Across Deep Research Agents
+
+---
+type: paper
+title: "Agent-Fence: Mapping Security Vulnerabilities Across Deep Research Agents"
+authors: Sai Puppala, Ismail Hossain, Md Jahangir Alam, Yoonpyo Lee, Jay Yoo, Tanzim Ahad, Syed Bahauddin Alam, Sajedul Talukder
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2602.07652
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-02-07
+updated_at: 2026-02-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, memory, planning, rag, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2602.07652
diff --git a/papers/items/2026-2602-07962-loca-bench-benchmarking-language-agents-under-controllable-and-extreme-context-g.md b/papers/items/2026-2602-07962-loca-bench-benchmarking-language-agents-under-controllable-and-extreme-context-g.md
new file mode 100644
index 0000000..79ec15b
--- /dev/null
+++ b/papers/items/2026-2602-07962-loca-bench-benchmarking-language-agents-under-controllable-and-extreme-context-g.md
@@ -0,0 +1,61 @@
+# Paper: LOCA-bench: Benchmarking Language Agents Under Controllable and Extreme Context Growth
+
+---
+type: paper
+title: "LOCA-bench: Benchmarking Language Agents Under Controllable and Extreme Context Growth"
+authors: Weihao Zeng, Yuzhen Huang, Junxian He
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2602.07962
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-02-08
+updated_at: 2026-02-08
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, planning, rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2602.07962
diff --git a/papers/items/2026-2602-08082-spectral-guardrails-for-agents-in-the-wild-detecting-tool-use-hallucinations-via.md b/papers/items/2026-2602-08082-spectral-guardrails-for-agents-in-the-wild-detecting-tool-use-hallucinations-via.md
new file mode 100644
index 0000000..06e9858
--- /dev/null
+++ b/papers/items/2026-2602-08082-spectral-guardrails-for-agents-in-the-wild-detecting-tool-use-hallucinations-via.md
@@ -0,0 +1,62 @@
+# Paper: Spectral Guardrails for Agents in the Wild: Detecting Tool Use Hallucinations via Attention Topology
+
+---
+type: paper
+title: "Spectral Guardrails for Agents in the Wild: Detecting Tool Use Hallucinations via Attention Topology"
+authors: Valentin Noël
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2602.08082
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-02-08
+updated_at: 2026-02-08
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+  - eess.SP
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, tool-use
+- arXiv categories: cs.LG, cs.AI, eess.SP
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2602.08082
diff --git a/papers/items/2026-2602-08412-from-assistant-to-double-agent-formalizing-and-benchmarking-attacks-on-openclaw-.md b/papers/items/2026-2602-08412-from-assistant-to-double-agent-formalizing-and-benchmarking-attacks-on-openclaw-.md
new file mode 100644
index 0000000..832dd57
--- /dev/null
+++ b/papers/items/2026-2602-08412-from-assistant-to-double-agent-formalizing-and-benchmarking-attacks-on-openclaw-.md
@@ -0,0 +1,63 @@
+# Paper: From Assistant to Double Agent: Formalizing and Benchmarking Attacks on OpenClaw for Personalized Local AI Agent
+
+---
+type: paper
+title: "From Assistant to Double Agent: Formalizing and Benchmarking Attacks on OpenClaw for Personalized Local AI Agent"
+authors: Yuhang Wang, Feiming Xu, Zheng Lin, Guangyu He, Yuzhe Huang, Haichang Gao, Zhenxing Niu, Shiguo Lian, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2602.08412
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-02-09
+updated_at: 2026-02-11
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 21
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, memory, planning, rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 21
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2602.08412
diff --git a/papers/items/2026-2602-10133-agenttrace-a-structured-logging-framework-for-agent-system-observability.md b/papers/items/2026-2602-10133-agenttrace-a-structured-logging-framework-for-agent-system-observability.md
new file mode 100644
index 0000000..1f86431
--- /dev/null
+++ b/papers/items/2026-2602-10133-agenttrace-a-structured-logging-framework-for-agent-system-observability.md
@@ -0,0 +1,62 @@
+# Paper: AgentTrace: A Structured Logging Framework for Agent System Observability
+
+---
+type: paper
+title: "AgentTrace: A Structured Logging Framework for Agent System Observability"
+authors: Adam AlSayyad, Kelvin Yuxiang Huang, Richik Pal
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2602.10133
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-02-07
+updated_at: 2026-02-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, reasoning, tool-use
+- arXiv categories: cs.SE, cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2602.10133
diff --git a/papers/items/2026-2602-11749-air-improving-agent-safety-through-incident-response.md b/papers/items/2026-2602-11749-air-improving-agent-safety-through-incident-response.md
new file mode 100644
index 0000000..5db9e2c
--- /dev/null
+++ b/papers/items/2026-2602-11749-air-improving-agent-safety-through-incident-response.md
@@ -0,0 +1,61 @@
+# Paper: AIR: Improving Agent Safety through Incident Response
+
+---
+type: paper
+title: "AIR: Improving Agent Safety through Incident Response"
+authors: Zibo Xiao, Jun Sun, Junjie Chen
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2602.11749
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-02-12
+updated_at: 2026-06-20
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, computer-use, tool-use
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2602.11749
diff --git a/papers/items/2026-2602-13379-unsafer-in-many-turns-benchmarking-and-defending-multi-turn-safety-risks-in-tool.md b/papers/items/2026-2602-13379-unsafer-in-many-turns-benchmarking-and-defending-multi-turn-safety-risks-in-tool.md
new file mode 100644
index 0000000..8e3d005
--- /dev/null
+++ b/papers/items/2026-2602-13379-unsafer-in-many-turns-benchmarking-and-defending-multi-turn-safety-risks-in-tool.md
@@ -0,0 +1,65 @@
+# Paper: Unsafer in Many Turns: Benchmarking and Defending Multi-Turn Safety Risks in Tool-Using Agents
+
+---
+type: paper
+title: "Unsafer in Many Turns: Benchmarking and Defending Multi-Turn Safety Risks in Tool-Using Agents"
+authors: Xu Li, Simon Yu, Minzhou Pan, Yiyou Sun, Bo Li, Dawn Song, Xue Lin, Weiyan Shi
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2602.13379
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-02-13
+updated_at: 2026-06-10
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+  - cs.CL
+  - cs.LG
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, rag, tool-use
+- arXiv categories: cs.CR, cs.AI, cs.CL, cs.LG, cs.SE
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2602.13379
diff --git a/papers/items/2026-2602-13530-remem-reasoning-with-episodic-memory-in-language-agent.md b/papers/items/2026-2602-13530-remem-reasoning-with-episodic-memory-in-language-agent.md
new file mode 100644
index 0000000..0aa8f59
--- /dev/null
+++ b/papers/items/2026-2602-13530-remem-reasoning-with-episodic-memory-in-language-agent.md
@@ -0,0 +1,62 @@
+# Paper: REMem: Reasoning with Episodic Memory in Language Agent
+
+---
+type: paper
+title: "REMem: Reasoning with Episodic Memory in Language Agent"
+authors: Yiheng Shu, Saisri Padmaja Jonnalagedda, Xiang Gao, Bernal Jiménez Gutiérrez, Weijian Qi, Kamalika Das, Huan Sun, Yu Su
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2602.13530
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-02-13
+updated_at: 2026-02-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, memory, rag, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2602.13530
diff --git a/papers/items/2026-2602-13665-hyfunc-accelerating-llm-based-function-calls-for-agentic-ai-through-hybrid-model.md b/papers/items/2026-2602-13665-hyfunc-accelerating-llm-based-function-calls-for-agentic-ai-through-hybrid-model.md
new file mode 100644
index 0000000..07d6fd3
--- /dev/null
+++ b/papers/items/2026-2602-13665-hyfunc-accelerating-llm-based-function-calls-for-agentic-ai-through-hybrid-model.md
@@ -0,0 +1,59 @@
+# Paper: HyFunc: Accelerating LLM-based Function Calls for Agentic AI through Hybrid-Model Cascade and Dynamic Templating
+
+---
+type: paper
+title: "HyFunc: Accelerating LLM-based Function Calls for Agentic AI through Hybrid-Model Cascade and Dynamic Templating"
+authors: Weibin Liao, Jian-guang Lou, Haoyi Xiong
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2602.13665
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-02-14
+updated_at: 2026-02-14
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, computer-use
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2602.13665
diff --git a/papers/items/2026-2602-14234-redsearcher-a-scalable-and-cost-efficient-framework-for-long-horizon-search-agen.md b/papers/items/2026-2602-14234-redsearcher-a-scalable-and-cost-efficient-framework-for-long-horizon-search-agen.md
new file mode 100644
index 0000000..61db05b
--- /dev/null
+++ b/papers/items/2026-2602-14234-redsearcher-a-scalable-and-cost-efficient-framework-for-long-horizon-search-agen.md
@@ -0,0 +1,62 @@
+# Paper: REDSearcher: A Scalable and Cost-Efficient Framework for Long-Horizon Search Agents
+
+---
+type: paper
+title: "REDSearcher: A Scalable and Cost-Efficient Framework for Long-Horizon Search Agents"
+authors: Zheng Chu, Xiao Wang, Jack Hong, Huiming Fan, Yuqi Huang, Yue Yang, Guohai Xu, Chenxiao Zhao, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2602.14234
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-02-15
+updated_at: 2026-02-15
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, planning, rag, tool-use
+- arXiv categories: cs.AI, cs.CL
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2602.14234
diff --git a/papers/items/2026-2602-14281-mcpshield-a-security-cognition-layer-for-adaptive-trust-calibration-in-model-con.md b/papers/items/2026-2602-14281-mcpshield-a-security-cognition-layer-for-adaptive-trust-calibration-in-model-con.md
new file mode 100644
index 0000000..061d77e
--- /dev/null
+++ b/papers/items/2026-2602-14281-mcpshield-a-security-cognition-layer-for-adaptive-trust-calibration-in-model-con.md
@@ -0,0 +1,62 @@
+# Paper: MCPShield: A Security Cognition Layer for Adaptive Trust Calibration in Model Context Protocol Agents
+
+---
+type: paper
+title: "MCPShield: A Security Cognition Layer for Adaptive Trust Calibration in Model Context Protocol Agents"
+authors: Zhenhong Zhou, Yuanhe Zhang, Hongwei Cai, Moayad Aloqaily, Ouns Bouachir, Linsey Pang, Prakhar Mehrotra, Kun Wang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2602.14281
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-02-15
+updated_at: 2026-02-24
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - computer-use
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-safety, computer-use, reasoning, tool-use
+- arXiv categories: cs.CR, cs.CL
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2602.14281
diff --git a/papers/items/2026-2602-16931-narrow-fine-tuning-erodes-safety-alignment-in-vision-language-agents.md b/papers/items/2026-2602-16931-narrow-fine-tuning-erodes-safety-alignment-in-vision-language-agents.md
new file mode 100644
index 0000000..187691d
--- /dev/null
+++ b/papers/items/2026-2602-16931-narrow-fine-tuning-erodes-safety-alignment-in-vision-language-agents.md
@@ -0,0 +1,59 @@
+# Paper: Narrow Fine-Tuning Erodes Safety Alignment in Vision-Language Agents
+
+---
+type: paper
+title: Narrow Fine-Tuning Erodes Safety Alignment in Vision-Language Agents
+authors: Idhant Gulati, Shivam Raval
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2602.16931
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-02-18
+updated_at: 2026-03-15
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, agent-safety
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2602.16931
diff --git a/papers/items/2026-2602-18456-beyond-single-channel-agentic-benchmarking.md b/papers/items/2026-2602-18456-beyond-single-channel-agentic-benchmarking.md
new file mode 100644
index 0000000..4793625
--- /dev/null
+++ b/papers/items/2026-2602-18456-beyond-single-channel-agentic-benchmarking.md
@@ -0,0 +1,61 @@
+# Paper: Beyond single-channel agentic benchmarking
+
+---
+type: paper
+title: Beyond single-channel agentic benchmarking
+authors: Nelu D. Radpour
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2602.18456
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-02-05
+updated_at: 2026-02-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CY
+  - cs.AI
+  - cs.HC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety
+- arXiv categories: cs.CY, cs.AI, cs.HC
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2602.18456
diff --git a/papers/items/2026-2602-19008-capable-but-unreliable-canonical-path-deviation-as-a-causal-mechanism-of-agent-f.md b/papers/items/2026-2602-19008-capable-but-unreliable-canonical-path-deviation-as-a-causal-mechanism-of-agent-f.md
new file mode 100644
index 0000000..4a7311b
--- /dev/null
+++ b/papers/items/2026-2602-19008-capable-but-unreliable-canonical-path-deviation-as-a-causal-mechanism-of-agent-f.md
@@ -0,0 +1,62 @@
+# Paper: Capable but Unreliable: Canonical Path Deviation as a Causal Mechanism of Agent Failure in Long-Horizon Tasks
+
+---
+type: paper
+title: "Capable but Unreliable: Canonical Path Deviation as a Causal Mechanism of Agent Failure in Long-Horizon Tasks"
+authors: Wilson Y. Lee
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2602.19008
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-02-22
+updated_at: 2026-02-22
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, computer-use, planning, tool-use
+- arXiv categories: cs.CL, cs.LG
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2602.19008
diff --git a/papers/items/2026-2602-21127-are-you-sure-an-empirical-study-of-human-perception-vulnerability-in-llm-driven-.md b/papers/items/2026-2602-21127-are-you-sure-an-empirical-study-of-human-perception-vulnerability-in-llm-driven-.md
new file mode 100644
index 0000000..4b7032b
--- /dev/null
+++ b/papers/items/2026-2602-21127-are-you-sure-an-empirical-study-of-human-perception-vulnerability-in-llm-driven-.md
@@ -0,0 +1,63 @@
+# Paper: "Are You Sure?": An Empirical Study of Human Perception Vulnerability in LLM-Driven Agentic Systems
+
+---
+type: paper
+title: "\"Are You Sure?\": An Empirical Study of Human Perception Vulnerability in LLM-Driven Agentic Systems"
+authors: Xinfeng Li, Shenyu Dai, Kelong Zheng, Yue Xiao, Gelei Deng, Wei Dong, Xiaofeng Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2602.21127
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-02-24
+updated_at: 2026-02-24
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.HC
+  - cs.AI
+  - cs.CR
+  - cs.SI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-safety, tool-use, workflow-agent
+- arXiv categories: cs.HC, cs.AI, cs.CR, cs.SI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2602.21127
diff --git a/papers/items/2026-2602-23320-parammem-augmenting-language-agents-with-parametric-reflective-memory.md b/papers/items/2026-2602-23320-parammem-augmenting-language-agents-with-parametric-reflective-memory.md
new file mode 100644
index 0000000..da91dca
--- /dev/null
+++ b/papers/items/2026-2602-23320-parammem-augmenting-language-agents-with-parametric-reflective-memory.md
@@ -0,0 +1,61 @@
+# Paper: ParamMem: Augmenting Language Agents with Parametric Reflective Memory
+
+---
+type: paper
+title: "ParamMem: Augmenting Language Agents with Parametric Reflective Memory"
+authors: Tianjun Yao, Yongqiang Chen, Yujia Zheng, Pan Li, Zhiqiang Shen, Kun Zhang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2602.23320
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-02-26
+updated_at: 2026-02-27
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, memory, reasoning
+- arXiv categories: cs.LG, cs.MA
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2602.23320
diff --git a/papers/items/2026-2603-00131-thought-virus-viral-misalignment-via-subliminal-prompting-in-multi-agent-systems.md b/papers/items/2026-2603-00131-thought-virus-viral-misalignment-via-subliminal-prompting-in-multi-agent-systems.md
new file mode 100644
index 0000000..9e17230
--- /dev/null
+++ b/papers/items/2026-2603-00131-thought-virus-viral-misalignment-via-subliminal-prompting-in-multi-agent-systems.md
@@ -0,0 +1,61 @@
+# Paper: Thought Virus: Viral Misalignment via Subliminal Prompting in Multi-Agent Systems
+
+---
+type: paper
+title: "Thought Virus: Viral Misalignment via Subliminal Prompting in Multi-Agent Systems"
+authors: Moritz Weckbecker, Jonas Müller, Ben Hagag, Michael Mulet
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.00131
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-02-23
+updated_at: 2026-02-23
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - multi-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-safety, multi-agent, tool-use
+- arXiv categories: cs.MA, cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.00131
diff --git a/papers/items/2026-2603-00623-tracesir-a-multi-agent-framework-for-structured-analysis-and-reporting-of-agenti.md b/papers/items/2026-2603-00623-tracesir-a-multi-agent-framework-for-structured-analysis-and-reporting-of-agenti.md
new file mode 100644
index 0000000..1053dff
--- /dev/null
+++ b/papers/items/2026-2603-00623-tracesir-a-multi-agent-framework-for-structured-analysis-and-reporting-of-agenti.md
@@ -0,0 +1,63 @@
+# Paper: TraceSIR: A Multi-Agent Framework for Structured Analysis and Reporting of Agentic Execution Traces
+
+---
+type: paper
+title: "TraceSIR: A Multi-Agent Framework for Structured Analysis and Reporting of Agentic Execution Traces"
+authors: Shu-Xun Yang, Cunxiang Wang, Haoke Zhang, Wenbo Yu, Lindong Wu, Jiayi Gui, Dayong Yang, Yukuo Cen, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.00623
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-02-28
+updated_at: 2026-02-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - multi-agent
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, coding-agent, multi-agent, reasoning, tool-use
+- arXiv categories: cs.AI, cs.CL
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.00623
diff --git a/papers/items/2026-2603-00801-the-synthetic-web-adversarially-curated-mini-internets-for-diagnosing-epistemic-.md b/papers/items/2026-2603-00801-the-synthetic-web-adversarially-curated-mini-internets-for-diagnosing-epistemic-.md
new file mode 100644
index 0000000..9e8b277
--- /dev/null
+++ b/papers/items/2026-2603-00801-the-synthetic-web-adversarially-curated-mini-internets-for-diagnosing-epistemic-.md
@@ -0,0 +1,62 @@
+# Paper: The Synthetic Web: Adversarially-Curated Mini-Internets for Diagnosing Epistemic Weaknesses of Language Agents
+
+---
+type: paper
+title: "The Synthetic Web: Adversarially-Curated Mini-Internets for Diagnosing Epistemic Weaknesses of Language Agents"
+authors: Shrey Shah, Levent Ozgur
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.00801
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-02-28
+updated_at: 2026-02-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - embodied-agent
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.IR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, embodied-agent, rag, tool-use
+- arXiv categories: cs.AI, cs.IR
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.00801
diff --git a/papers/items/2026-2603-01438-enhancing-persona-following-at-decoding-time-via-dynamic-importance-estimation-f.md b/papers/items/2026-2603-01438-enhancing-persona-following-at-decoding-time-via-dynamic-importance-estimation-f.md
new file mode 100644
index 0000000..37e83ca
--- /dev/null
+++ b/papers/items/2026-2603-01438-enhancing-persona-following-at-decoding-time-via-dynamic-importance-estimation-f.md
@@ -0,0 +1,64 @@
+# Paper: Enhancing Persona Following at Decoding Time via Dynamic Importance Estimation for Role-Playing Agents
+
+---
+type: paper
+title: Enhancing Persona Following at Decoding Time via Dynamic Importance Estimation for Role-Playing Agents
+authors: Yuxin Liu, Mingye Zhu, Siyuan Liu, Bo Hu, Lei Zhang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.01438
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-03-02
+updated_at: 2026-03-02
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - coding-agent
+  - computer-use
+  - planning
+  - rag
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-safety, coding-agent, computer-use, planning, rag, world-model
+- arXiv categories: cs.CL, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.01438
diff --git a/papers/items/2026-2603-01712-ft-dojo-towards-autonomous-llm-fine-tuning-with-language-agents.md b/papers/items/2026-2603-01712-ft-dojo-towards-autonomous-llm-fine-tuning-with-language-agents.md
new file mode 100644
index 0000000..fbd15c2
--- /dev/null
+++ b/papers/items/2026-2603-01712-ft-dojo-towards-autonomous-llm-fine-tuning-with-language-agents.md
@@ -0,0 +1,61 @@
+# Paper: FT-Dojo: Towards Autonomous LLM Fine-Tuning with Language Agents
+
+---
+type: paper
+title: "FT-Dojo: Towards Autonomous LLM Fine-Tuning with Language Agents"
+authors: Qizheng Li, Yifei Zhang, Xiao Yang, Xu Yang, Zhuo Wang, Weiqing Liu, Jiang Bian
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.01712
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-03-02
+updated_at: 2026-05-20
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - planning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, coding-agent, planning
+- arXiv categories: cs.AI, cs.LG
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.01712
diff --git a/papers/items/2026-2603-02711-a-natural-language-agentic-approach-to-study-affective-polarization.md b/papers/items/2026-2603-02711-a-natural-language-agentic-approach-to-study-affective-polarization.md
new file mode 100644
index 0000000..e02a70e
--- /dev/null
+++ b/papers/items/2026-2603-02711-a-natural-language-agentic-approach-to-study-affective-polarization.md
@@ -0,0 +1,60 @@
+# Paper: A Natural Language Agentic Approach to Study Affective Polarization
+
+---
+type: paper
+title: A Natural Language Agentic Approach to Study Affective Polarization
+authors: Stephanie Anneris Malvicini, Ewelina Gajewska, Arda Derbent, Katarzyna Budzynska, Jarosław A. Chudziak, Maria Vanina Martinez
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.02711
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-03-03
+updated_at: 2026-03-03
+status: queued
+relevance: high
+topics:
+  - multi-agent
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: multi-agent, rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.02711
diff --git a/papers/items/2026-2603-03515-the-controllability-trap-a-governance-framework-for-military-ai-agents.md b/papers/items/2026-2603-03515-the-controllability-trap-a-governance-framework-for-military-ai-agents.md
new file mode 100644
index 0000000..b5e6d92
--- /dev/null
+++ b/papers/items/2026-2603-03515-the-controllability-trap-a-governance-framework-for-military-ai-agents.md
@@ -0,0 +1,63 @@
+# Paper: The Controllability Trap: A Governance Framework for Military AI Agents
+
+---
+type: paper
+title: "The Controllability Trap: A Governance Framework for Military AI Agents"
+authors: Subramanyam Sahoo
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.03515
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-03-03
+updated_at: 2026-03-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CY
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, planning, tool-use, world-model
+- arXiv categories: cs.CY, cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.03515
diff --git a/papers/items/2026-2603-03680-mage-meta-reinforcement-learning-for-language-agents-toward-strategic-exploratio.md b/papers/items/2026-2603-03680-mage-meta-reinforcement-learning-for-language-agents-toward-strategic-exploratio.md
new file mode 100644
index 0000000..9153295
--- /dev/null
+++ b/papers/items/2026-2603-03680-mage-meta-reinforcement-learning-for-language-agents-toward-strategic-exploratio.md
@@ -0,0 +1,61 @@
+# Paper: MAGE: Meta-Reinforcement Learning for Language Agents toward Strategic Exploration and Exploitation
+
+---
+type: paper
+title: "MAGE: Meta-Reinforcement Learning for Language Agents toward Strategic Exploration and Exploitation"
+authors: Lu Yang, Zelai Xu, Minyang Xie, Jiaxuan Gao, Zhao Shok, Yu Wang, Yi Wu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.03680
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-03-04
+updated_at: 2026-03-04
+status: queued
+relevance: high
+topics:
+  - memory
+  - multi-agent
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: memory, multi-agent, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.03680
diff --git a/papers/items/2026-2603-05553-eigendata-a-self-evolving-multi-agent-platform-for-function-calling-data-synthes.md b/papers/items/2026-2603-05553-eigendata-a-self-evolving-multi-agent-platform-for-function-calling-data-synthes.md
new file mode 100644
index 0000000..b6dc370
--- /dev/null
+++ b/papers/items/2026-2603-05553-eigendata-a-self-evolving-multi-agent-platform-for-function-calling-data-synthes.md
@@ -0,0 +1,63 @@
+# Paper: EigenData: A Self-Evolving Multi-Agent Platform for Function-Calling Data Synthesis, Auditing, and Repair
+
+---
+type: paper
+title: "EigenData: A Self-Evolving Multi-Agent Platform for Function-Calling Data Synthesis, Auditing, and Repair"
+authors: Jiaao Chen, Jingyuan Qi, Mingye Gao, Wei-Chen Wang, Hanrui Wang, Di Jin
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.05553
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-03-05
+updated_at: 2026-03-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - multi-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, coding-agent, multi-agent, tool-use
+- arXiv categories: cs.SE, cs.AI, cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.05553
diff --git a/papers/items/2026-2603-05578-tool-genesis-a-task-driven-tool-creation-benchmark-for-self-evolving-language-ag.md b/papers/items/2026-2603-05578-tool-genesis-a-task-driven-tool-creation-benchmark-for-self-evolving-language-ag.md
new file mode 100644
index 0000000..6576f18
--- /dev/null
+++ b/papers/items/2026-2603-05578-tool-genesis-a-task-driven-tool-creation-benchmark-for-self-evolving-language-ag.md
@@ -0,0 +1,61 @@
+# Paper: Tool-Genesis: A Task-Driven Tool Creation Benchmark for Self-Evolving Language Agent
+
+---
+type: paper
+title: "Tool-Genesis: A Task-Driven Tool Creation Benchmark for Self-Evolving Language Agent"
+authors: Bowei Xia, Mengkang Hu, Shijian Wang, Jiarui Jin, Wenxiang Jiao, Yuan Lu, Kexin Li, Ping Luo
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.05578
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-03-05
+updated_at: 2026-03-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, computer-use, tool-use
+- arXiv categories: cs.SE, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.05578
diff --git a/papers/items/2026-2603-07496-from-thinker-to-society-security-in-hierarchical-autonomy-evolution-of-ai-agents.md b/papers/items/2026-2603-07496-from-thinker-to-society-security-in-hierarchical-autonomy-evolution-of-ai-agents.md
new file mode 100644
index 0000000..62d0fc8
--- /dev/null
+++ b/papers/items/2026-2603-07496-from-thinker-to-society-security-in-hierarchical-autonomy-evolution-of-ai-agents.md
@@ -0,0 +1,64 @@
+# Paper: From Thinker to Society: Security in Hierarchical Autonomy Evolution of AI Agents
+
+---
+type: paper
+title: "From Thinker to Society: Security in Hierarchical Autonomy Evolution of AI Agents"
+authors: Xiaolei Zhang, Lu Zhou, Xiaogang Xu, Jiafei Wu, Tianyu Du, Heqing Huang, Hao Peng, Zhe Liu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.07496
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-03-08
+updated_at: 2026-03-21
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - multi-agent
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, computer-use, multi-agent, reasoning, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.07496
diff --git a/papers/items/2026-2603-07557-agentraft-automated-detection-of-data-over-exposure-in-llm-agents.md b/papers/items/2026-2603-07557-agentraft-automated-detection-of-data-over-exposure-in-llm-agents.md
new file mode 100644
index 0000000..20fcdbd
--- /dev/null
+++ b/papers/items/2026-2603-07557-agentraft-automated-detection-of-data-over-exposure-in-llm-agents.md
@@ -0,0 +1,62 @@
+# Paper: AgentRaft: Automated Detection of Data Over-Exposure in LLM Agents
+
+---
+type: paper
+title: "AgentRaft: Automated Detection of Data Over-Exposure in LLM Agents"
+authors: Yixi Lin, Jiangrong Wu, Yuhong Nan, Xueqiang Wang, Xinyuan Zhang, Zibin Zheng
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.07557
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-03-08
+updated_at: 2026-03-08
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, agent-safety, rag, reasoning, tool-use
+- arXiv categories: cs.SE
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.07557
diff --git a/papers/items/2026-2603-07980-onemillion-bench-how-far-are-language-agents-from-human-experts.md b/papers/items/2026-2603-07980-onemillion-bench-how-far-are-language-agents-from-human-experts.md
new file mode 100644
index 0000000..8667452
--- /dev/null
+++ b/papers/items/2026-2603-07980-onemillion-bench-how-far-are-language-agents-from-human-experts.md
@@ -0,0 +1,63 @@
+# Paper: \$OneMillion-Bench: How Far are Language Agents from Human Experts?
+
+---
+type: paper
+title: "\\$OneMillion-Bench: How Far are Language Agents from Human Experts?"
+authors: Qianyu Yang, Yang Liu, Jiaqi Li, Jun Bai, Hao Chen, Kaiyuan Chen, Tiliang Duan, Jiayun Dong, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.07980
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-03-09
+updated_at: 2026-03-09
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, planning, reasoning, tool-use
+- arXiv categories: cs.LG, cs.AI, cs.CL
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.07980
diff --git a/papers/items/2026-2603-08721-kernelcraft-benchmarking-for-agentic-close-to-metal-kernel-generation-on-emergin.md b/papers/items/2026-2603-08721-kernelcraft-benchmarking-for-agentic-close-to-metal-kernel-generation-on-emergin.md
new file mode 100644
index 0000000..260c6d4
--- /dev/null
+++ b/papers/items/2026-2603-08721-kernelcraft-benchmarking-for-agentic-close-to-metal-kernel-generation-on-emergin.md
@@ -0,0 +1,62 @@
+# Paper: KernelCraft: Benchmarking for Agentic Close-to-Metal Kernel Generation on Emerging Hardware
+
+---
+type: paper
+title: "KernelCraft: Benchmarking for Agentic Close-to-Metal Kernel Generation on Emerging Hardware"
+authors: Jiayi Nie, Haoran Wu, Yao Lai, Zeyu Cao, Cheng Zhang, Binglei Lou, Erwei Wang, Jianyi Cheng, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.08721
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-02-10
+updated_at: 2026-05-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - reasoning
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AR
+  - cs.LG
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, reasoning, workflow-agent
+- arXiv categories: cs.AR, cs.LG, cs.SE
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.08721
diff --git a/papers/items/2026-2603-09002-security-considerations-for-multi-agent-systems.md b/papers/items/2026-2603-09002-security-considerations-for-multi-agent-systems.md
new file mode 100644
index 0000000..9fa5b26
--- /dev/null
+++ b/papers/items/2026-2603-09002-security-considerations-for-multi-agent-systems.md
@@ -0,0 +1,66 @@
+# Paper: Security Considerations for Multi-agent Systems
+
+---
+type: paper
+title: Security Considerations for Multi-agent Systems
+authors: Tam Nguyen, Moses Ndebugre, Dheeraj Arremsetty
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.09002
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-03-09
+updated_at: 2026-04-26
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - memory
+  - multi-agent
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, computer-use, memory, multi-agent, planning, rag, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.09002
diff --git a/papers/items/2026-2603-10492-human-ai-co-reasoning-for-clinical-diagnosis-with-evidence-integrated-language-a.md b/papers/items/2026-2603-10492-human-ai-co-reasoning-for-clinical-diagnosis-with-evidence-integrated-language-a.md
new file mode 100644
index 0000000..5125a9c
--- /dev/null
+++ b/papers/items/2026-2603-10492-human-ai-co-reasoning-for-clinical-diagnosis-with-evidence-integrated-language-a.md
@@ -0,0 +1,63 @@
+# Paper: Human-AI Co-reasoning for Clinical Diagnosis with Evidence-Integrated Language Agent
+
+---
+type: paper
+title: Human-AI Co-reasoning for Clinical Diagnosis with Evidence-Integrated Language Agent
+authors: Zhongzhen Huang, Yan Ling, Hong Chen, Ye Feng, Li Wu, Linjie Mu, Shaoting Zhang, Xiaofan Zhang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.10492
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-03-11
+updated_at: 2026-03-18
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - rag
+  - reasoning
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, agent-safety, multi-agent, rag, reasoning, workflow-agent
+- arXiv categories: cs.CL
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.10492
diff --git a/papers/items/2026-2603-11088-the-attack-and-defense-landscape-of-agentic-ai-a-comprehensive-survey.md b/papers/items/2026-2603-11088-the-attack-and-defense-landscape-of-agentic-ai-a-comprehensive-survey.md
new file mode 100644
index 0000000..632802f
--- /dev/null
+++ b/papers/items/2026-2603-11088-the-attack-and-defense-landscape-of-agentic-ai-a-comprehensive-survey.md
@@ -0,0 +1,61 @@
+# Paper: The Attack and Defense Landscape of Agentic AI: A Comprehensive Survey
+
+---
+type: paper
+title: "The Attack and Defense Landscape of Agentic AI: A Comprehensive Survey"
+authors: Juhee Kim, Xiaoyuan Liu, Zhun Wang, Shi Qiu, Bo Li, Wenbo Guo, Dawn Song
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.11088
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-03-11
+updated_at: 2026-03-11
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-safety, tool-use, workflow-agent
+- arXiv categories: cs.CR, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.11088
diff --git a/papers/items/2026-2603-11890-quare-quality-aware-requirements-analysis-through-multi-agent-dialectical-negoti.md b/papers/items/2026-2603-11890-quare-quality-aware-requirements-analysis-through-multi-agent-dialectical-negoti.md
new file mode 100644
index 0000000..b1f8213
--- /dev/null
+++ b/papers/items/2026-2603-11890-quare-quality-aware-requirements-analysis-through-multi-agent-dialectical-negoti.md
@@ -0,0 +1,61 @@
+# Paper: QUARE: Quality-Aware Requirements Analysis through Multi-Agent Dialectical Negotiation
+
+---
+type: paper
+title: "QUARE: Quality-Aware Requirements Analysis through Multi-Agent Dialectical Negotiation"
+authors: Haowei Cheng, Milhan Kim, Foutse Khomh, Teeradaj Racharak, Nobukazu Yoshioka, Naoyasu Ubayashi, Hironori Washizaki
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.11890
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-03-12
+updated_at: 2026-06-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, multi-agent, rag
+- arXiv categories: cs.SE
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.11890
diff --git a/papers/items/2026-2603-15309-cctu-a-benchmark-for-tool-use-under-complex-constraints.md b/papers/items/2026-2603-15309-cctu-a-benchmark-for-tool-use-under-complex-constraints.md
new file mode 100644
index 0000000..291cbfc
--- /dev/null
+++ b/papers/items/2026-2603-15309-cctu-a-benchmark-for-tool-use-under-complex-constraints.md
@@ -0,0 +1,61 @@
+# Paper: CCTU: A Benchmark for Tool Use under Complex Constraints
+
+---
+type: paper
+title: "CCTU: A Benchmark for Tool Use under Complex Constraints"
+authors: Junjie Ye, Guoqiang Zhang, Wenjie Fu, Tao Gui, Qi Zhang, Xuanjing Huang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.15309
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-03-16
+updated_at: 2026-03-16
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, rag, tool-use
+- arXiv categories: cs.CL, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.15309
diff --git a/papers/items/2026-2603-15666-compiled-memory-not-more-information-but-more-precise-instructions-for-language-.md b/papers/items/2026-2603-15666-compiled-memory-not-more-information-but-more-precise-instructions-for-language-.md
new file mode 100644
index 0000000..00361f0
--- /dev/null
+++ b/papers/items/2026-2603-15666-compiled-memory-not-more-information-but-more-precise-instructions-for-language-.md
@@ -0,0 +1,59 @@
+# Paper: Compiled Memory: Not More Information, but More Precise Instructions for Language Agents
+
+---
+type: paper
+title: "Compiled Memory: Not More Information, but More Precise Instructions for Language Agents"
+authors: James Rhodes, George Kang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.15666
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-03-12
+updated_at: 2026-03-12
+status: queued
+relevance: high
+topics:
+  - memory
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: memory, rag
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.15666
diff --git a/papers/items/2026-2603-16734-differential-harm-propensity-in-personalized-llm-agents-the-curious-case-of-ment.md b/papers/items/2026-2603-16734-differential-harm-propensity-in-personalized-llm-agents-the-curious-case-of-ment.md
new file mode 100644
index 0000000..df68fde
--- /dev/null
+++ b/papers/items/2026-2603-16734-differential-harm-propensity-in-personalized-llm-agents-the-curious-case-of-ment.md
@@ -0,0 +1,62 @@
+# Paper: Differential Harm Propensity in Personalized LLM Agents: The Curious Case of Mental Health Disclosure
+
+---
+type: paper
+title: "Differential Harm Propensity in Personalized LLM Agents: The Curious Case of Mental Health Disclosure"
+authors: Caglar Yildirim
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.16734
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-03-17
+updated_at: 2026-03-17
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, memory, rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.16734
diff --git a/papers/items/2026-2603-17392-agentic-cognitive-profiling-realigning-automated-alzheimer-s-disease-detection-w.md b/papers/items/2026-2603-17392-agentic-cognitive-profiling-realigning-automated-alzheimer-s-disease-detection-w.md
new file mode 100644
index 0000000..cc6d9bf
--- /dev/null
+++ b/papers/items/2026-2603-17392-agentic-cognitive-profiling-realigning-automated-alzheimer-s-disease-detection-w.md
@@ -0,0 +1,61 @@
+# Paper: Agentic Cognitive Profiling: Realigning Automated Alzheimer's Disease Detection with Clinical Construct Validity
+
+---
+type: paper
+title: "Agentic Cognitive Profiling: Realigning Automated Alzheimer's Disease Detection with Clinical Construct Validity"
+authors: Jiawen Kang, Kun Li, Dongrui Han, Jinchao Li, Junan Li, Lingwei Meng, Xixin Wu, Helen Meng
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.17392
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-03-18
+updated_at: 2026-03-18
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+  - cs.IR
+  - q-bio.NC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, tool-use
+- arXiv categories: cs.MA, cs.IR, q-bio.NC
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.17392
diff --git a/papers/items/2026-2603-18245-who-tests-the-testers-systematic-enumeration-and-coverage-audit-of-llm-agent-too.md b/papers/items/2026-2603-18245-who-tests-the-testers-systematic-enumeration-and-coverage-audit-of-llm-agent-too.md
new file mode 100644
index 0000000..4afb96e
--- /dev/null
+++ b/papers/items/2026-2603-18245-who-tests-the-testers-systematic-enumeration-and-coverage-audit-of-llm-agent-too.md
@@ -0,0 +1,63 @@
+# Paper: Who Tests the Testers? Systematic Enumeration and Coverage Audit of LLM Agent Tool Call Safety
+
+---
+type: paper
+title: Who Tests the Testers? Systematic Enumeration and Coverage Audit of LLM Agent Tool Call Safety
+authors: Xuan Chen, Lu Yan, Ruqi Zhang, Xiangyu Zhang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.18245
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-03-18
+updated_at: 2026-03-18
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, rag, tool-use, workflow-agent
+- arXiv categories: cs.SE, cs.CR
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.18245
diff --git a/papers/items/2026-2603-19469-a-framework-for-formalizing-llm-agent-security.md b/papers/items/2026-2603-19469-a-framework-for-formalizing-llm-agent-security.md
new file mode 100644
index 0000000..4d96660
--- /dev/null
+++ b/papers/items/2026-2603-19469-a-framework-for-formalizing-llm-agent-security.md
@@ -0,0 +1,61 @@
+# Paper: A Framework for Formalizing LLM Agent Security
+
+---
+type: paper
+title: A Framework for Formalizing LLM Agent Security
+authors: Vincent Siu, Jingxuan He, Kyle Montgomery, Zhun Wang, Neil Gong, Chenguang Wang, Dawn Song
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.19469
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-03-19
+updated_at: 2026-03-19
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - memory
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-safety, memory, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.19469
diff --git a/papers/items/2026-2603-19684-tsegagent-zero-shot-tooth-segmentation-via-geometry-aware-vision-language-agents.md b/papers/items/2026-2603-19684-tsegagent-zero-shot-tooth-segmentation-via-geometry-aware-vision-language-agents.md
new file mode 100644
index 0000000..4d46516
--- /dev/null
+++ b/papers/items/2026-2603-19684-tsegagent-zero-shot-tooth-segmentation-via-geometry-aware-vision-language-agents.md
@@ -0,0 +1,62 @@
+# Paper: TSegAgent: Zero-Shot Tooth Segmentation via Geometry-Aware Vision-Language Agents
+
+---
+type: paper
+title: "TSegAgent: Zero-Shot Tooth Segmentation via Geometry-Aware Vision-Language Agents"
+authors: Shaojie Zhuang, Lu Yin, Guangshun Wei, Yunpeng Li, Xilu Wang, Yuanfeng Zhou
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.19684
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-03-20
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, coding-agent, rag, reasoning, tool-use
+- arXiv categories: cs.CV
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.19684
diff --git a/papers/items/2026-2603-21357-agenther-hindsight-experience-replay-for-llm-agent-trajectory-relabeling.md b/papers/items/2026-2603-21357-agenther-hindsight-experience-replay-for-llm-agent-trajectory-relabeling.md
new file mode 100644
index 0000000..0db7643
--- /dev/null
+++ b/papers/items/2026-2603-21357-agenther-hindsight-experience-replay-for-llm-agent-trajectory-relabeling.md
@@ -0,0 +1,62 @@
+# Paper: AgentHER: Hindsight Experience Replay for LLM Agent Trajectory Relabeling
+
+---
+type: paper
+title: "AgentHER: Hindsight Experience Replay for LLM Agent Trajectory Relabeling"
+authors: Liang Ding
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.21357
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-03-22
+updated_at: 2026-05-10
+status: queued
+relevance: high
+topics:
+  - computer-use
+  - memory
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: computer-use, memory, tool-use, workflow-agent
+- arXiv categories: cs.AI, cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.21357
diff --git a/papers/items/2026-2603-21564-toward-a-theory-of-hierarchical-memory-for-language-agents.md b/papers/items/2026-2603-21564-toward-a-theory-of-hierarchical-memory-for-language-agents.md
new file mode 100644
index 0000000..2c82af7
--- /dev/null
+++ b/papers/items/2026-2603-21564-toward-a-theory-of-hierarchical-memory-for-language-agents.md
@@ -0,0 +1,64 @@
+# Paper: Toward a Theory of Hierarchical Memory for Language Agents
+
+---
+type: paper
+title: Toward a Theory of Hierarchical Memory for Language Agents
+authors: Yashar Talebirad, Ali Parsaee, Csongor Y. Szepesvari, Amirhossein Nadiri, Osmar Zaiane
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.21564
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-03-23
+updated_at: 2026-03-23
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.IR
+  - cs.AI
+  - cs.IT
+  - cs.SI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, memory, rag, tool-use
+- arXiv categories: cs.IR, cs.AI, cs.IT, cs.SI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.21564
diff --git a/papers/items/2026-2603-24257-memory-augmented-vision-language-agents-for-persistent-and-semantically-consiste.md b/papers/items/2026-2603-24257-memory-augmented-vision-language-agents-for-persistent-and-semantically-consiste.md
new file mode 100644
index 0000000..5dbe041
--- /dev/null
+++ b/papers/items/2026-2603-24257-memory-augmented-vision-language-agents-for-persistent-and-semantically-consiste.md
@@ -0,0 +1,60 @@
+# Paper: Memory-Augmented Vision-Language Agents for Persistent and Semantically Consistent Object Captioning
+
+---
+type: paper
+title: Memory-Augmented Vision-Language Agents for Persistent and Semantically Consistent Object Captioning
+authors: Tommaso Galliena, Stefano Rosa, Tommaso Apicella, Pietro Morerio, Alessio Del Bue, Lorenzo Natale
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.24257
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-03-25
+updated_at: 2026-03-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - embodied-agent
+  - memory
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, embodied-agent, memory
+- arXiv categories: cs.CV
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.24257
diff --git a/papers/items/2026-2603-25353-safeguard-asf-sr-agentic-humanoid-robot-system-for-autonomous-industrial-safety.md b/papers/items/2026-2603-25353-safeguard-asf-sr-agentic-humanoid-robot-system-for-autonomous-industrial-safety.md
new file mode 100644
index 0000000..a3c1dcd
--- /dev/null
+++ b/papers/items/2026-2603-25353-safeguard-asf-sr-agentic-humanoid-robot-system-for-autonomous-industrial-safety.md
@@ -0,0 +1,62 @@
+# Paper: SafeGuard ASF: SR Agentic Humanoid Robot System for Autonomous Industrial Safety
+
+---
+type: paper
+title: "SafeGuard ASF: SR Agentic Humanoid Robot System for Autonomous Industrial Safety"
+authors: Thanh Nguyen Canh, Thang Tran Viet, Thanh Tuan Tran, Ben Wei Lim
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.25353
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-03-26
+updated_at: 2026-03-26
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - embodied-agent
+  - reasoning
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.RO
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-safety, embodied-agent, reasoning, tool-use, world-model
+- arXiv categories: cs.RO
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.25353
diff --git a/papers/items/2026-2603-27148-safetydrift-predicting-when-ai-agents-cross-the-line-before-they-actually-do.md b/papers/items/2026-2603-27148-safetydrift-predicting-when-ai-agents-cross-the-line-before-they-actually-do.md
new file mode 100644
index 0000000..8595609
--- /dev/null
+++ b/papers/items/2026-2603-27148-safetydrift-predicting-when-ai-agents-cross-the-line-before-they-actually-do.md
@@ -0,0 +1,60 @@
+# Paper: SafetyDrift: Predicting When AI Agents Cross the Line Before They Actually Do
+
+---
+type: paper
+title: "SafetyDrift: Predicting When AI Agents Cross the Line Before They Actually Do"
+authors: Aditya Dhodapkar, Farhaan Pishori
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.27148
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-03-28
+updated_at: 2026-03-28
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-safety, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.27148
diff --git a/papers/items/2026-2603-28166-evaluating-privilege-usage-of-agents-with-real-world-tools.md b/papers/items/2026-2603-28166-evaluating-privilege-usage-of-agents-with-real-world-tools.md
new file mode 100644
index 0000000..00d36c1
--- /dev/null
+++ b/papers/items/2026-2603-28166-evaluating-privilege-usage-of-agents-with-real-world-tools.md
@@ -0,0 +1,63 @@
+# Paper: Evaluating Privilege Usage of Agents with Real-World Tools
+
+---
+type: paper
+title: Evaluating Privilege Usage of Agents with Real-World Tools
+authors: Quan Zhang, Lianhang Fu, Lvsi Lian, Gwihwan Go, Yujue Wang, Chijin Zhou, Yu Jiang, Geguang Pu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.28166
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-03-30
+updated_at: 2026-04-20
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, rag, tool-use, workflow-agent
+- arXiv categories: cs.CR, cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.28166
diff --git a/papers/items/2026-2603-28428-synergy-a-next-generation-general-purpose-agent-for-open-agentic-web.md b/papers/items/2026-2603-28428-synergy-a-next-generation-general-purpose-agent-for-open-agentic-web.md
new file mode 100644
index 0000000..e0cee35
--- /dev/null
+++ b/papers/items/2026-2603-28428-synergy-a-next-generation-general-purpose-agent-for-open-agentic-web.md
@@ -0,0 +1,64 @@
+# Paper: Synergy: A Next-Generation General-Purpose Agent for Open Agentic Web
+
+---
+type: paper
+title: "Synergy: A Next-Generation General-Purpose Agent for Open Agentic Web"
+authors: Xiaohang Nie, Zihan Guo, Kezhuo Yang, Zhichong Zheng, Bochen Ge, Shuai Pan, Zeyi Chen, Youling Xiang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.28428
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-03-30
+updated_at: 2026-03-30
+status: queued
+relevance: high
+topics:
+  - coding-agent
+  - embodied-agent
+  - memory
+  - multi-agent
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CY
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: coding-agent, embodied-agent, memory, multi-agent, rag, tool-use
+- arXiv categories: cs.CY, cs.MA
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.28428
diff --git a/papers/items/2026-2603-28900-robust-multi-agent-reinforcement-learning-for-small-uas-separation-assurance-und.md b/papers/items/2026-2603-28900-robust-multi-agent-reinforcement-learning-for-small-uas-separation-assurance-und.md
new file mode 100644
index 0000000..d7b06d1
--- /dev/null
+++ b/papers/items/2026-2603-28900-robust-multi-agent-reinforcement-learning-for-small-uas-separation-assurance-und.md
@@ -0,0 +1,64 @@
+# Paper: Robust Multi-Agent Reinforcement Learning for Small UAS Separation Assurance under GPS Degradation and Spoofing
+
+---
+type: paper
+title: Robust Multi-Agent Reinforcement Learning for Small UAS Separation Assurance under GPS Degradation and Spoofing
+authors: Alex Zongo, Filippos Fotiadis, Ufuk Topcu, Peng Wei
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2603.28900
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-03-30
+updated_at: 2026-03-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.RO
+  - cs.AI
+  - cs.LG
+  - eess.SY
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, multi-agent, world-model
+- arXiv categories: cs.RO, cs.AI, cs.LG, eess.SY
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2603.28900
diff --git a/papers/items/2026-2604-02022-atbench-a-diverse-and-realistic-agent-trajectory-benchmark-for-safety-evaluation.md b/papers/items/2026-2604-02022-atbench-a-diverse-and-realistic-agent-trajectory-benchmark-for-safety-evaluation.md
new file mode 100644
index 0000000..cfe6183
--- /dev/null
+++ b/papers/items/2026-2604-02022-atbench-a-diverse-and-realistic-agent-trajectory-benchmark-for-safety-evaluation.md
@@ -0,0 +1,62 @@
+# Paper: ATBench: A Diverse and Realistic Agent Trajectory Benchmark for Safety Evaluation and Diagnosis
+
+---
+type: paper
+title: "ATBench: A Diverse and Realistic Agent Trajectory Benchmark for Safety Evaluation and Diagnosis"
+authors: Yu Li, Haoyu Luo, Yuejin Xie, Yuqian Fu, Zhonghao Yang, Shuai Shao, Qihan Ren, Wanying Qu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.02022
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-02
+updated_at: 2026-05-13
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, planning, rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.02022
diff --git a/papers/items/2026-2604-02155-brief-is-better-non-monotonic-chain-of-thought-budget-effects-in-function-callin.md b/papers/items/2026-2604-02155-brief-is-better-non-monotonic-chain-of-thought-budget-effects-in-function-callin.md
new file mode 100644
index 0000000..755df83
--- /dev/null
+++ b/papers/items/2026-2604-02155-brief-is-better-non-monotonic-chain-of-thought-budget-effects-in-function-callin.md
@@ -0,0 +1,61 @@
+# Paper: Brief Is Better: Non-Monotonic Chain-of-Thought Budget Effects in Function-Calling Language Agents
+
+---
+type: paper
+title: "Brief Is Better: Non-Monotonic Chain-of-Thought Budget Effects in Function-Calling Language Agents"
+authors: Xuan Qi
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.02155
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-02
+updated_at: 2026-04-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: function-calling, language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling, language-agent
+- inferred topics: agent-evaluation, rag, reasoning, tool-use
+- arXiv categories: cs.CL
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.02155
diff --git a/papers/items/2026-2604-03098-co-evolution-of-policy-and-internal-reward-for-language-agents.md b/papers/items/2026-2604-03098-co-evolution-of-policy-and-internal-reward-for-language-agents.md
new file mode 100644
index 0000000..e69c3b5
--- /dev/null
+++ b/papers/items/2026-2604-03098-co-evolution-of-policy-and-internal-reward-for-language-agents.md
@@ -0,0 +1,63 @@
+# Paper: Co-Evolution of Policy and Internal Reward for Language Agents
+
+---
+type: paper
+title: Co-Evolution of Policy and Internal Reward for Language Agents
+authors: Xinyu Wang, Hanwei Wu, Jingwei Song, Shuyuan Zhang, Jiayi Zhang, Fanqi Kong, Tung Sum Thomas Kwok, Xiao-Wen Chang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.03098
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-03
+updated_at: 2026-04-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, computer-use, planning, tool-use
+- arXiv categories: cs.LG, cs.AI, cs.CL
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.03098
diff --git a/papers/items/2026-2604-03242-draft-task-decoupled-latent-reasoning-for-agent-safety.md b/papers/items/2026-2604-03242-draft-task-decoupled-latent-reasoning-for-agent-safety.md
new file mode 100644
index 0000000..2ff0c1a
--- /dev/null
+++ b/papers/items/2026-2604-03242-draft-task-decoupled-latent-reasoning-for-agent-safety.md
@@ -0,0 +1,62 @@
+# Paper: DRAFT: Task Decoupled Latent Reasoning for Agent Safety
+
+---
+type: paper
+title: "DRAFT: Task Decoupled Latent Reasoning for Agent Safety"
+authors: Lin Wang, Junfeng Fang, Dan Zhang, Fei Shen, Xiang Wang, Tat-Seng Chua
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.03242
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-02-11
+updated_at: 2026-02-11
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, rag, reasoning, tool-use
+- arXiv categories: cs.LG
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.03242
diff --git a/papers/items/2026-2604-04131-profile-then-reason-bounded-semantic-complexity-for-tool-augmented-language-agen.md b/papers/items/2026-2604-04131-profile-then-reason-bounded-semantic-complexity-for-tool-augmented-language-agen.md
new file mode 100644
index 0000000..aa084df
--- /dev/null
+++ b/papers/items/2026-2604-04131-profile-then-reason-bounded-semantic-complexity-for-tool-augmented-language-agen.md
@@ -0,0 +1,62 @@
+# Paper: Profile-Then-Reason: Bounded Semantic Complexity for Tool-Augmented Language Agents
+
+---
+type: paper
+title: "Profile-Then-Reason: Bounded Semantic Complexity for Tool-Augmented Language Agents"
+authors: Paulo Akira F. Enabe
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.04131
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-05
+updated_at: 2026-04-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.04131
diff --git a/papers/items/2026-2604-04426-shieldnet-network-level-guardrails-against-emerging-supply-chain-injections-in-a.md b/papers/items/2026-2604-04426-shieldnet-network-level-guardrails-against-emerging-supply-chain-injections-in-a.md
new file mode 100644
index 0000000..f86b31e
--- /dev/null
+++ b/papers/items/2026-2604-04426-shieldnet-network-level-guardrails-against-emerging-supply-chain-injections-in-a.md
@@ -0,0 +1,60 @@
+# Paper: ShieldNet: Network-Level Guardrails against Emerging Supply-Chain Injections in Agentic Systems
+
+---
+type: paper
+title: "ShieldNet: Network-Level Guardrails against Emerging Supply-Chain Injections in Agentic Systems"
+authors: Zhuowen Yuan, Zhaorun Chen, Zhen Xiang, Nathaniel D. Bastian, Seyyed Hadi Hashemi, Chaowei Xiao, Wenbo Guo, Bo Li
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.04426
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-06
+updated_at: 2026-04-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, tool-use
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.04426
diff --git a/papers/items/2026-2604-06762-arulecon-agentic-security-rule-conversion.md b/papers/items/2026-2604-06762-arulecon-agentic-security-rule-conversion.md
new file mode 100644
index 0000000..67adcc6
--- /dev/null
+++ b/papers/items/2026-2604-06762-arulecon-agentic-security-rule-conversion.md
@@ -0,0 +1,60 @@
+# Paper: ARuleCon: Agentic Security Rule Conversion
+
+---
+type: paper
+title: "ARuleCon: Agentic Security Rule Conversion"
+authors: Ming Xu, Hongtai Wang, Yanpei Guo, Zhengmin Yu, Weili Han, Hoon Wei Lim, Jin Song Dong, Jiaheng Zhang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.06762
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-08
+updated_at: 2026-04-08
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, rag
+- arXiv categories: cs.CR
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.06762
diff --git a/papers/items/2026-2604-06972-differentiable-environment-trajectory-co-optimization-for-safe-multi-agent-navig.md b/papers/items/2026-2604-06972-differentiable-environment-trajectory-co-optimization-for-safe-multi-agent-navig.md
new file mode 100644
index 0000000..51127a1
--- /dev/null
+++ b/papers/items/2026-2604-06972-differentiable-environment-trajectory-co-optimization-for-safe-multi-agent-navig.md
@@ -0,0 +1,65 @@
+# Paper: Differentiable Environment-Trajectory Co-Optimization for Safe Multi-Agent Navigation
+
+---
+type: paper
+title: Differentiable Environment-Trajectory Co-Optimization for Safe Multi-Agent Navigation
+authors: Zhan Gao, Gabriele Fadini, Stelian Coros, Amanda Prorok
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.06972
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-08
+updated_at: 2026-04-08
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - embodied-agent
+  - multi-agent
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.RO
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, computer-use, embodied-agent, multi-agent, rag, tool-use
+- arXiv categories: cs.RO, cs.MA
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.06972
diff --git a/papers/items/2026-2604-08388-awakening-the-sleeping-agent-lean-specific-agentic-data-reactivates-general-tool.md b/papers/items/2026-2604-08388-awakening-the-sleeping-agent-lean-specific-agentic-data-reactivates-general-tool.md
new file mode 100644
index 0000000..cac47ff
--- /dev/null
+++ b/papers/items/2026-2604-08388-awakening-the-sleeping-agent-lean-specific-agentic-data-reactivates-general-tool.md
@@ -0,0 +1,59 @@
+# Paper: Awakening the Sleeping Agent: Lean-Specific Agentic Data Reactivates General Tool Use in Goedel Prover
+
+---
+type: paper
+title: "Awakening the Sleeping Agent: Lean-Specific Agentic Data Reactivates General Tool Use in Goedel Prover"
+authors: Jui-Hui Chung, Hongzhou Lin, Lai Jiang, Shange Tang, Chi Jin
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.08388
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-09
+updated_at: 2026-04-09
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, tool-use
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.08388
diff --git a/papers/items/2026-2604-10577-the-blind-spot-of-agent-safety-how-benign-user-instructions-expose-critical-vuln.md b/papers/items/2026-2604-10577-the-blind-spot-of-agent-safety-how-benign-user-instructions-expose-critical-vuln.md
new file mode 100644
index 0000000..2805097
--- /dev/null
+++ b/papers/items/2026-2604-10577-the-blind-spot-of-agent-safety-how-benign-user-instructions-expose-critical-vuln.md
@@ -0,0 +1,63 @@
+# Paper: The Blind Spot of Agent Safety: How Benign User Instructions Expose Critical Vulnerabilities in Computer-Use Agents
+
+---
+type: paper
+title: "The Blind Spot of Agent Safety: How Benign User Instructions Expose Critical Vulnerabilities in Computer-Use Agents"
+authors: Xuwei Ding, Skylar Zhai, Linxin Song, Jiate Li, Taiwei Shi, Nicholas Meade, Siva Reddy, Jian Kang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.10577
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-12
+updated_at: 2026-04-17
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, multi-agent, rag, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.10577
diff --git a/papers/items/2026-2604-11557-unitoolcall-unifying-tool-use-representation-data-and-evaluation-for-llm-agents.md b/papers/items/2026-2604-11557-unitoolcall-unifying-tool-use-representation-data-and-evaluation-for-llm-agents.md
new file mode 100644
index 0000000..f947c26
--- /dev/null
+++ b/papers/items/2026-2604-11557-unitoolcall-unifying-tool-use-representation-data-and-evaluation-for-llm-agents.md
@@ -0,0 +1,60 @@
+# Paper: UniToolCall: Unifying Tool-Use Representation, Data, and Evaluation for LLM Agents
+
+---
+type: paper
+title: "UniToolCall: Unifying Tool-Use Representation, Data, and Evaluation for LLM Agents"
+authors: Yijuan Liang, Xinghao Chen, Yifan Ge, Ziyi Wu, Hao Wu, Changyu Zeng, Wei Xing, Xiaoyu Shen
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.11557
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-13
+updated_at: 2026-05-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.11557
diff --git a/papers/items/2026-2604-12986-parallax-why-ai-agents-that-think-must-never-act.md b/papers/items/2026-2604-12986-parallax-why-ai-agents-that-think-must-never-act.md
new file mode 100644
index 0000000..d8e9605
--- /dev/null
+++ b/papers/items/2026-2604-12986-parallax-why-ai-agents-that-think-must-never-act.md
@@ -0,0 +1,63 @@
+# Paper: Parallax: Why AI Agents That Think Must Never Act
+
+---
+type: paper
+title: "Parallax: Why AI Agents That Think Must Never Act"
+authors: Joel Fokou
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.12986
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-14
+updated_at: 2026-04-14
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.CR, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.12986
diff --git a/papers/items/2026-2604-13298-can-agents-secure-hardware-evaluating-agentic-llm-driven-obfuscation-for-ip-prot.md b/papers/items/2026-2604-13298-can-agents-secure-hardware-evaluating-agentic-llm-driven-obfuscation-for-ip-prot.md
new file mode 100644
index 0000000..a3dc60c
--- /dev/null
+++ b/papers/items/2026-2604-13298-can-agents-secure-hardware-evaluating-agentic-llm-driven-obfuscation-for-ip-prot.md
@@ -0,0 +1,61 @@
+# Paper: Can Agents Secure Hardware? Evaluating Agentic LLM-Driven Obfuscation for IP Protection
+
+---
+type: paper
+title: Can Agents Secure Hardware? Evaluating Agentic LLM-Driven Obfuscation for IP Protection
+authors: Sujan Ghimire, Parsa Mirfasihi, Muhtasim Alam Chowdhury, Veeramani Pugazhenthi, Harish Kumar Dharavath, Farshad Firouzi, Rozhin Yasaei, Pratik Satam, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.13298
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-14
+updated_at: 2026-04-14
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, planning, rag
+- arXiv categories: cs.CR
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.13298
diff --git a/papers/items/2026-2604-13536-don-t-let-ai-agents-yolo-your-files-shifting-information-and-control-to-filesyst.md b/papers/items/2026-2604-13536-don-t-let-ai-agents-yolo-your-files-shifting-information-and-control-to-filesyst.md
new file mode 100644
index 0000000..ee4fb0c
--- /dev/null
+++ b/papers/items/2026-2604-13536-don-t-let-ai-agents-yolo-your-files-shifting-information-and-control-to-filesyst.md
@@ -0,0 +1,61 @@
+# Paper: Don't Let AI Agents YOLO Your Files: Shifting Information and Control to Filesystems for Agent Safety and Autonomy
+
+---
+type: paper
+title: "Don't Let AI Agents YOLO Your Files: Shifting Information and Control to Filesystems for Agent Safety and Autonomy"
+authors: Shawn Wanxiang Zhong, Junxuan Liao, Jing Liu, Mai Zheng, Andrea C. Arpaci-Dusseau, Remzi H. Arpaci-Dusseau
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.13536
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-15
+updated_at: 2026-04-16
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.OS
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, coding-agent, tool-use
+- arXiv categories: cs.OS
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.13536
diff --git a/papers/items/2026-2604-13954-hintbench-horizon-agent-intrinsic-non-attack-trajectory-benchmark.md b/papers/items/2026-2604-13954-hintbench-horizon-agent-intrinsic-non-attack-trajectory-benchmark.md
new file mode 100644
index 0000000..1ddb7b7
--- /dev/null
+++ b/papers/items/2026-2604-13954-hintbench-horizon-agent-intrinsic-non-attack-trajectory-benchmark.md
@@ -0,0 +1,62 @@
+# Paper: HINTBench: Horizon-agent Intrinsic Non-attack Trajectory Benchmark
+
+---
+type: paper
+title: "HINTBench: Horizon-agent Intrinsic Non-attack Trajectory Benchmark"
+authors: Jiacheng Wang, Jinchang Hou, Fabian Wang, Ping Jian, Chenfu Bao, Zhonghou Lv
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.13954
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-15
+updated_at: 2026-04-15
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, planning, rag
+- arXiv categories: cs.LG, cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.13954
diff --git a/papers/items/2026-2604-14399-spacemind-a-modular-and-self-evolving-embodied-vision-language-agent-framework-f.md b/papers/items/2026-2604-14399-spacemind-a-modular-and-self-evolving-embodied-vision-language-agent-framework-f.md
new file mode 100644
index 0000000..9005f4e
--- /dev/null
+++ b/papers/items/2026-2604-14399-spacemind-a-modular-and-self-evolving-embodied-vision-language-agent-framework-f.md
@@ -0,0 +1,63 @@
+# Paper: SpaceMind: A Modular and Self-Evolving Embodied Vision-Language Agent Framework for Autonomous On-orbit Servicing
+
+---
+type: paper
+title: "SpaceMind: A Modular and Self-Evolving Embodied Vision-Language Agent Framework for Autonomous On-orbit Servicing"
+authors: Aodi Wu, Haodong Han, Xubo Luo, Ruisuo Wang, Shan He, Xue Wan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.14399
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-15
+updated_at: 2026-04-15
+status: queued
+relevance: high
+topics:
+  - embodied-agent
+  - reasoning
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.RO
+  - cs.AI
+  - eess.SY
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: embodied-agent, reasoning, tool-use, world-model
+- arXiv categories: cs.RO, cs.AI, eess.SY
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.14399
diff --git a/papers/items/2026-2604-15415-harmfulskillbench-how-do-harmful-skills-weaponize-your-agents.md b/papers/items/2026-2604-15415-harmfulskillbench-how-do-harmful-skills-weaponize-your-agents.md
new file mode 100644
index 0000000..0a43d76
--- /dev/null
+++ b/papers/items/2026-2604-15415-harmfulskillbench-how-do-harmful-skills-weaponize-your-agents.md
@@ -0,0 +1,62 @@
+# Paper: HarmfulSkillBench: How Do Harmful Skills Weaponize Your Agents?
+
+---
+type: paper
+title: "HarmfulSkillBench: How Do Harmful Skills Weaponize Your Agents?"
+authors: Yukun Jiang, Yage Zhang, Michael Backes, Xinyue Shen, Yang Zhang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.15415
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-16
+updated_at: 2026-04-16
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, rag, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.15415
diff --git a/papers/items/2026-2604-15579-don-t-make-models-guess-security-and-safety-symbolic-guardrails-for-domain-speci.md b/papers/items/2026-2604-15579-don-t-make-models-guess-security-and-safety-symbolic-guardrails-for-domain-speci.md
new file mode 100644
index 0000000..25e0670
--- /dev/null
+++ b/papers/items/2026-2604-15579-don-t-make-models-guess-security-and-safety-symbolic-guardrails-for-domain-speci.md
@@ -0,0 +1,63 @@
+# Paper: Don't Make Models Guess Security and Safety: Symbolic Guardrails for Domain-Specific AI Agents
+
+---
+type: paper
+title: "Don't Make Models Guess Security and Safety: Symbolic Guardrails for Domain-Specific AI Agents"
+authors: Yining Hong, Yining She, Eunsuk Kang, Christopher S. Timperley, Christian Kästner
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.15579
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-16
+updated_at: 2026-07-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, coding-agent, tool-use
+- arXiv categories: cs.SE, cs.AI, cs.CR
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.15579
diff --git a/papers/items/2026-2604-16706-evaluating-tool-using-language-agents-judge-reliability-propagation-cascades-and.md b/papers/items/2026-2604-16706-evaluating-tool-using-language-agents-judge-reliability-propagation-cascades-and.md
new file mode 100644
index 0000000..ae4e51e
--- /dev/null
+++ b/papers/items/2026-2604-16706-evaluating-tool-using-language-agents-judge-reliability-propagation-cascades-and.md
@@ -0,0 +1,61 @@
+# Paper: Evaluating Tool-Using Language Agents: Judge Reliability, Propagation Cascades, and Runtime Mitigation in AgentProp-Bench
+
+---
+type: paper
+title: "Evaluating Tool-Using Language Agents: Judge Reliability, Propagation Cascades, and Runtime Mitigation in AgentProp-Bench"
+authors: Bhaskar Gurram
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.16706
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-17
+updated_at: 2026-04-17
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, tool-use
+- arXiv categories: cs.AI, cs.CL, cs.MA
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.16706
diff --git a/papers/items/2026-2604-17562-safeagent-a-runtime-protection-architecture-for-agentic-systems.md b/papers/items/2026-2604-17562-safeagent-a-runtime-protection-architecture-for-agentic-systems.md
new file mode 100644
index 0000000..2340230
--- /dev/null
+++ b/papers/items/2026-2604-17562-safeagent-a-runtime-protection-architecture-for-agentic-systems.md
@@ -0,0 +1,65 @@
+# Paper: SafeAgent: A Runtime Protection Architecture for Agentic Systems
+
+---
+type: paper
+title: "SafeAgent: A Runtime Protection Architecture for Agentic Systems"
+authors: Hailin Liu, Eugene Ilyushin, Jie Ni, Min Zhu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.17562
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-19
+updated_at: 2026-04-19
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - memory
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, coding-agent, memory, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI, cs.MA
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.17562
diff --git a/papers/items/2026-2604-18658-owner-harm-a-missing-threat-model-for-ai-agent-safety.md b/papers/items/2026-2604-18658-owner-harm-a-missing-threat-model-for-ai-agent-safety.md
new file mode 100644
index 0000000..039d50d
--- /dev/null
+++ b/papers/items/2026-2604-18658-owner-harm-a-missing-threat-model-for-ai-agent-safety.md
@@ -0,0 +1,64 @@
+# Paper: Owner-Harm: A Missing Threat Model for AI Agent Safety
+
+---
+type: paper
+title: "Owner-Harm: A Missing Threat Model for AI Agent Safety"
+authors: Dongcheng Zhang, Yiqing Jiang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.18658
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-20
+updated_at: 2026-04-20
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, rag, reasoning, tool-use
+- arXiv categories: cs.CR, cs.AI, cs.CL
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.18658
diff --git a/papers/items/2026-2604-18718-towards-optimal-agentic-architectures-for-offensive-security-tasks.md b/papers/items/2026-2604-18718-towards-optimal-agentic-architectures-for-offensive-security-tasks.md
new file mode 100644
index 0000000..00f869b
--- /dev/null
+++ b/papers/items/2026-2604-18718-towards-optimal-agentic-architectures-for-offensive-security-tasks.md
@@ -0,0 +1,62 @@
+# Paper: Towards Optimal Agentic Architectures for Offensive Security Tasks
+
+---
+type: paper
+title: Towards Optimal Agentic Architectures for Offensive Security Tasks
+authors: Isaac David, Arthur Gervais
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.18718
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-20
+updated_at: 2026-04-20
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, rag, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.18718
diff --git a/papers/items/2026-2604-18847-human-guided-harm-recovery-for-computer-use-agents.md b/papers/items/2026-2604-18847-human-guided-harm-recovery-for-computer-use-agents.md
new file mode 100644
index 0000000..6303a84
--- /dev/null
+++ b/papers/items/2026-2604-18847-human-guided-harm-recovery-for-computer-use-agents.md
@@ -0,0 +1,65 @@
+# Paper: Human-Guided Harm Recovery for Computer Use Agents
+
+---
+type: paper
+title: Human-Guided Harm Recovery for Computer Use Agents
+authors: Christy Li, Sky CH-Wang, Andi Peng, Andreea Bobu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.18847
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-20
+updated_at: 2026-05-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - memory
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, computer-use, memory, planning, rag, tool-use
+- arXiv categories: cs.AI, cs.CL
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.18847
diff --git a/papers/items/2026-2604-19821-jtpro-a-joint-tool-prompt-reflective-optimization-framework-for-language-agents.md b/papers/items/2026-2604-19821-jtpro-a-joint-tool-prompt-reflective-optimization-framework-for-language-agents.md
new file mode 100644
index 0000000..72651bd
--- /dev/null
+++ b/papers/items/2026-2604-19821-jtpro-a-joint-tool-prompt-reflective-optimization-framework-for-language-agents.md
@@ -0,0 +1,62 @@
+# Paper: JTPRO: A Joint Tool-Prompt Reflective Optimization Framework for Language Agents
+
+---
+type: paper
+title: "JTPRO: A Joint Tool-Prompt Reflective Optimization Framework for Language Agents"
+authors: Sandip Ghoshal, Anshul Mittal, Jyotika Singh, Miguel Ballesteros, Weiyi Sun, Fang Tu, Shailender Singh, Yassine Benajiba, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.19821
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-20
+updated_at: 2026-04-20
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, computer-use, reasoning, tool-use
+- arXiv categories: cs.AI, cs.SE
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.19821
diff --git a/papers/items/2026-2604-19844-if-you-re-waiting-for-a-sign-that-might-not-be-it-mitigating-trust-boundary-conf.md b/papers/items/2026-2604-19844-if-you-re-waiting-for-a-sign-that-might-not-be-it-mitigating-trust-boundary-conf.md
new file mode 100644
index 0000000..aed1dd6
--- /dev/null
+++ b/papers/items/2026-2604-19844-if-you-re-waiting-for-a-sign-that-might-not-be-it-mitigating-trust-boundary-conf.md
@@ -0,0 +1,62 @@
+# Paper: If you're waiting for a sign... that might not be it! Mitigating Trust Boundary Confusion from Visual Injections on Vision-Language Agentic Systems
+
+---
+type: paper
+title: "If you're waiting for a sign... that might not be it! Mitigating Trust Boundary Confusion from Visual Injections on Vision-Language Agentic Systems"
+authors: Jiamin Chang, Minhui Xue, Ruoxi Sun, Shuchao Pang, Salil S. Kanhere, Hammond Pearce
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.19844
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-21
+updated_at: 2026-04-21
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - embodied-agent
+  - multi-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, agent-safety, embodied-agent, multi-agent
+- arXiv categories: cs.CV, cs.AI
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.19844
diff --git a/papers/items/2026-2604-20994-breaking-mcp-with-function-hijacking-attacks-novel-threats-for-function-calling-.md b/papers/items/2026-2604-20994-breaking-mcp-with-function-hijacking-attacks-novel-threats-for-function-calling-.md
new file mode 100644
index 0000000..65ee24c
--- /dev/null
+++ b/papers/items/2026-2604-20994-breaking-mcp-with-function-hijacking-attacks-novel-threats-for-function-calling-.md
@@ -0,0 +1,62 @@
+# Paper: Breaking MCP with Function Hijacking Attacks: Novel Threats for Function Calling and Agentic Models
+
+---
+type: paper
+title: "Breaking MCP with Function Hijacking Attacks: Novel Threats for Function Calling and Agentic Models"
+authors: Yannis Belkhiter, Giulio Zizzo, Sergio Maffeis, Seshu Tirupathi, John D. Kelleher
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.20994
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-22
+updated_at: 2026-04-22
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-safety, reasoning, tool-use
+- arXiv categories: cs.CR, cs.AI, cs.CL
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.20994
diff --git a/papers/items/2026-2604-21190-spatio-adaptive-test-time-orchestration-of-vision-language-agents-for-spatial-re.md b/papers/items/2026-2604-21190-spatio-adaptive-test-time-orchestration-of-vision-language-agents-for-spatial-re.md
new file mode 100644
index 0000000..db04c41
--- /dev/null
+++ b/papers/items/2026-2604-21190-spatio-adaptive-test-time-orchestration-of-vision-language-agents-for-spatial-re.md
@@ -0,0 +1,61 @@
+# Paper: SpatiO: Adaptive Test-Time Orchestration of Vision-Language Agents for Spatial Reasoning
+
+---
+type: paper
+title: "SpatiO: Adaptive Test-Time Orchestration of Vision-Language Agents for Spatial Reasoning"
+authors: Chan Yeong Hwang, Miso Choi, Sunghyun On, Jinkyu Kim, Jungbeom Lee
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.21190
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-23
+updated_at: 2026-06-27
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, multi-agent, rag, reasoning
+- arXiv categories: cs.CV
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.21190
diff --git a/papers/items/2026-2604-22879-beyond-single-agent-alignment-preventing-context-fragmented-violations-in-multi-.md b/papers/items/2026-2604-22879-beyond-single-agent-alignment-preventing-context-fragmented-violations-in-multi-.md
new file mode 100644
index 0000000..b379021
--- /dev/null
+++ b/papers/items/2026-2604-22879-beyond-single-agent-alignment-preventing-context-fragmented-violations-in-multi-.md
@@ -0,0 +1,67 @@
+# Paper: Beyond Single-Agent Alignment: Preventing Context-Fragmented Violations in Multi-Agent Systems
+
+---
+type: paper
+title: "Beyond Single-Agent Alignment: Preventing Context-Fragmented Violations in Multi-Agent Systems"
+authors: Jie Wu, Ming Gong
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.22879
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-24
+updated_at: 2026-04-24
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - rag
+  - tool-use
+  - workflow-agent
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+  - cs.AI
+  - cs.CR
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, multi-agent, rag, tool-use, workflow-agent, world-model
+- arXiv categories: cs.MA, cs.AI, cs.CR, cs.LG
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.22879
diff --git a/papers/items/2026-2604-23210-discovering-agentic-safety-specifications-from-1-bit-danger-signals.md b/papers/items/2026-2604-23210-discovering-agentic-safety-specifications-from-1-bit-danger-signals.md
new file mode 100644
index 0000000..2dbae72
--- /dev/null
+++ b/papers/items/2026-2604-23210-discovering-agentic-safety-specifications-from-1-bit-danger-signals.md
@@ -0,0 +1,64 @@
+# Paper: Discovering Agentic Safety Specifications from 1-Bit Danger Signals
+
+---
+type: paper
+title: Discovering Agentic Safety Specifications from 1-Bit Danger Signals
+authors: Víctor Gallego
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.23210
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-25
+updated_at: 2026-04-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, planning, rag, reasoning, tool-use
+- arXiv categories: cs.AI, cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.23210
diff --git a/papers/items/2026-2604-23374-ghost-in-the-agent-redefining-information-flow-tracking-for-llm-agents.md b/papers/items/2026-2604-23374-ghost-in-the-agent-redefining-information-flow-tracking-for-llm-agents.md
new file mode 100644
index 0000000..47f1fb1
--- /dev/null
+++ b/papers/items/2026-2604-23374-ghost-in-the-agent-redefining-information-flow-tracking-for-llm-agents.md
@@ -0,0 +1,62 @@
+# Paper: Ghost in the Agent: Redefining Information Flow Tracking for LLM Agents
+
+---
+type: paper
+title: "Ghost in the Agent: Redefining Information Flow Tracking for LLM Agents"
+authors: Yuandao Cai, Wensheng Tang, Cheng Wen, Shengchao Qin
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.23374
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-25
+updated_at: 2026-04-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, memory, reasoning, tool-use
+- arXiv categories: cs.CR
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.23374
diff --git a/papers/items/2026-2604-23459-architecture-matters-for-multi-agent-security.md b/papers/items/2026-2604-23459-architecture-matters-for-multi-agent-security.md
new file mode 100644
index 0000000..b127baf
--- /dev/null
+++ b/papers/items/2026-2604-23459-architecture-matters-for-multi-agent-security.md
@@ -0,0 +1,65 @@
+# Paper: Architecture Matters for Multi-Agent Security
+
+---
+type: paper
+title: Architecture Matters for Multi-Agent Security
+authors: Ben Hagag, William L. Anderson, Christian Schroeder de Witt, Sarah Scheffler
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.23459
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-25
+updated_at: 2026-04-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - memory
+  - multi-agent
+  - planning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+  - cs.CR
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, computer-use, memory, multi-agent, planning
+- arXiv categories: cs.MA, cs.CR, cs.LG
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.23459
diff --git a/papers/items/2026-2604-24212-empowering-autonomous-debugging-agents-with-efficient-dynamic-analysis.md b/papers/items/2026-2604-24212-empowering-autonomous-debugging-agents-with-efficient-dynamic-analysis.md
new file mode 100644
index 0000000..169b7fb
--- /dev/null
+++ b/papers/items/2026-2604-24212-empowering-autonomous-debugging-agents-with-efficient-dynamic-analysis.md
@@ -0,0 +1,63 @@
+# Paper: Empowering Autonomous Debugging Agents with Efficient Dynamic Analysis
+
+---
+type: paper
+title: Empowering Autonomous Debugging Agents with Efficient Dynamic Analysis
+authors: Jiahong Xiang, Xiaoyang Xu, Xiaopan Chu, Hongliang Tian, Yuqun Zhang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.24212
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-27
+updated_at: 2026-04-27
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - embodied-agent
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, coding-agent, embodied-agent, memory, rag, tool-use
+- arXiv categories: cs.SE
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.24212
diff --git a/papers/items/2026-2604-24826-a-comparative-evaluation-of-ai-agent-security-guardrails.md b/papers/items/2026-2604-24826-a-comparative-evaluation-of-ai-agent-security-guardrails.md
new file mode 100644
index 0000000..6259e60
--- /dev/null
+++ b/papers/items/2026-2604-24826-a-comparative-evaluation-of-ai-agent-security-guardrails.md
@@ -0,0 +1,61 @@
+# Paper: A Comparative Evaluation of AI Agent Security Guardrails
+
+---
+type: paper
+title: A Comparative Evaluation of AI Agent Security Guardrails
+authors: Qi Li, Jiu Li, Pingtao Wei, Jianjun Xu, Xueyi Wei, Jiwei Shi, Xuan Zhang, Yanhui Yang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.24826
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-27
+updated_at: 2026-04-27
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.24826
diff --git a/papers/items/2026-2604-25135-fama-failure-aware-meta-agentic-framework-for-open-source-llms-in-interactive-to.md b/papers/items/2026-2604-25135-fama-failure-aware-meta-agentic-framework-for-open-source-llms-in-interactive-to.md
new file mode 100644
index 0000000..3f233dc
--- /dev/null
+++ b/papers/items/2026-2604-25135-fama-failure-aware-meta-agentic-framework-for-open-source-llms-in-interactive-to.md
@@ -0,0 +1,59 @@
+# Paper: FAMA: Failure-Aware Meta-Agentic Framework for Open-Source LLMs in Interactive Tool Use Environments
+
+---
+type: paper
+title: "FAMA: Failure-Aware Meta-Agentic Framework for Open-Source LLMs in Interactive Tool Use Environments"
+authors: Amir Saeidi, Venkatesh Mishra, Souradeep Mukhopadhyay, Gaowen Liu, Ali Payani, Jayanth Srinivasa, Chitta Baral
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.25135
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-28
+updated_at: 2026-04-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, tool-use
+- arXiv categories: cs.CL
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.25135
diff --git a/papers/items/2026-2604-25318-cutscene-agent-an-llm-agent-framework-for-automated-3d-cutscene-generation.md b/papers/items/2026-2604-25318-cutscene-agent-an-llm-agent-framework-for-automated-3d-cutscene-generation.md
new file mode 100644
index 0000000..8465e61
--- /dev/null
+++ b/papers/items/2026-2604-25318-cutscene-agent-an-llm-agent-framework-for-automated-3d-cutscene-generation.md
@@ -0,0 +1,64 @@
+# Paper: Cutscene Agent: An LLM Agent Framework for Automated 3D Cutscene Generation
+
+---
+type: paper
+title: "Cutscene Agent: An LLM Agent Framework for Automated 3D Cutscene Generation"
+authors: Lanshan He, Haozhou Pang, Qi Gan, Xin Shen, Ziwei Zhang, Yibo Liu, Gang Fang, Bo Liu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.25318
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-28
+updated_at: 2026-04-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.GR
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, multi-agent, planning, reasoning, tool-use
+- arXiv categories: cs.GR, cs.AI, cs.CL
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.25318
diff --git a/papers/items/2026-2604-25555-from-crud-to-autonomous-agents-formal-validation-and-zero-trust-security-for-sem.md b/papers/items/2026-2604-25555-from-crud-to-autonomous-agents-formal-validation-and-zero-trust-security-for-sem.md
new file mode 100644
index 0000000..85931ab
--- /dev/null
+++ b/papers/items/2026-2604-25555-from-crud-to-autonomous-agents-formal-validation-and-zero-trust-security-for-sem.md
@@ -0,0 +1,63 @@
+# Paper: From CRUD to Autonomous Agents: Formal Validation and Zero-Trust Security for Semantic Gateways in AI-Native Enterprise Systems
+
+---
+type: paper
+title: "From CRUD to Autonomous Agents: Formal Validation and Zero-Trust Security for Semantic Gateways in AI-Native Enterprise Systems"
+authors: Ignacio Peyrano
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.25555
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-28
+updated_at: 2026-04-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, agent-safety, coding-agent, tool-use, workflow-agent
+- arXiv categories: cs.CR, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.25555
diff --git a/papers/items/2026-2604-26274-enforcing-benign-trajectories-a-behavioral-firewall-for-structured-workflow-ai-a.md b/papers/items/2026-2604-26274-enforcing-benign-trajectories-a-behavioral-firewall-for-structured-workflow-ai-a.md
new file mode 100644
index 0000000..6524538
--- /dev/null
+++ b/papers/items/2026-2604-26274-enforcing-benign-trajectories-a-behavioral-firewall-for-structured-workflow-ai-a.md
@@ -0,0 +1,63 @@
+# Paper: Enforcing Benign Trajectories: A Behavioral Firewall for Structured-Workflow AI Agents
+
+---
+type: paper
+title: "Enforcing Benign Trajectories: A Behavioral Firewall for Structured-Workflow AI Agents"
+authors: Hung Dang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.26274
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-29
+updated_at: 2026-04-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, rag, tool-use, workflow-agent
+- arXiv categories: cs.CR, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.26274
diff --git a/papers/items/2026-2604-26959-careguardai-context-aware-multi-agent-guardrails-for-clinical-safety-hallucinati.md b/papers/items/2026-2604-26959-careguardai-context-aware-multi-agent-guardrails-for-clinical-safety-hallucinati.md
new file mode 100644
index 0000000..d1b5966
--- /dev/null
+++ b/papers/items/2026-2604-26959-careguardai-context-aware-multi-agent-guardrails-for-clinical-safety-hallucinati.md
@@ -0,0 +1,63 @@
+# Paper: CareGuardAI: Context-Aware Multi-Agent Guardrails for Clinical Safety & Hallucination Mitigation in Patient-Facing LLMs
+
+---
+type: paper
+title: "CareGuardAI: Context-Aware Multi-Agent Guardrails for Clinical Safety & Hallucination Mitigation in Patient-Facing LLMs"
+authors: Elham Nasarian, Abhilash Neog, Kwok-Leung Tsui, Niyousha HosseiniChimeh
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.26959
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-07
+updated_at: 2026-04-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CY
+  - cs.AI
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, multi-agent, tool-use
+- arXiv categories: cs.CY, cs.AI, cs.MA
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.26959
diff --git a/papers/items/2026-2604-27092-end-to-end-autonomous-scientific-discovery-on-a-real-optical-platform.md b/papers/items/2026-2604-27092-end-to-end-autonomous-scientific-discovery-on-a-real-optical-platform.md
new file mode 100644
index 0000000..59961e9
--- /dev/null
+++ b/papers/items/2026-2604-27092-end-to-end-autonomous-scientific-discovery-on-a-real-optical-platform.md
@@ -0,0 +1,63 @@
+# Paper: End-to-end autonomous scientific discovery on a real optical platform
+
+---
+type: paper
+title: End-to-end autonomous scientific discovery on a real optical platform
+authors: Shuxing Yang, Fujia Chen, Rui Zhao, Junyao Wu, Yize Wang, Haiyao Luo, Ning Han, Qiaolu Chen, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.27092
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-29
+updated_at: 2026-04-29
+status: queued
+relevance: high
+topics:
+  - memory
+  - planning
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - physics.optics
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: memory, planning, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI, physics.optics
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.27092
diff --git a/papers/items/2026-2604-27464-security-attack-and-defense-strategies-for-autonomous-agent-frameworks-a-layered.md b/papers/items/2026-2604-27464-security-attack-and-defense-strategies-for-autonomous-agent-frameworks-a-layered.md
new file mode 100644
index 0000000..18fbc31
--- /dev/null
+++ b/papers/items/2026-2604-27464-security-attack-and-defense-strategies-for-autonomous-agent-frameworks-a-layered.md
@@ -0,0 +1,63 @@
+# Paper: Security Attack and Defense Strategies for Autonomous Agent Frameworks: A Layered Review with OpenClaw as a Case Study
+
+---
+type: paper
+title: "Security Attack and Defense Strategies for Autonomous Agent Frameworks: A Layered Review with OpenClaw as a Case Study"
+authors: Luyao Xu, Xiang Chen
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.27464
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-30
+updated_at: 2026-04-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-safety, autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety, autonomous-agent-llm
+- inferred topics: agent-evaluation, agent-safety, planning, tool-use, workflow-agent
+- arXiv categories: cs.CR, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.27464
diff --git a/papers/items/2026-2604-27699-bridging-values-and-behavior-a-hierarchical-framework-for-proactive-embodied-age.md b/papers/items/2026-2604-27699-bridging-values-and-behavior-a-hierarchical-framework-for-proactive-embodied-age.md
new file mode 100644
index 0000000..20debb5
--- /dev/null
+++ b/papers/items/2026-2604-27699-bridging-values-and-behavior-a-hierarchical-framework-for-proactive-embodied-age.md
@@ -0,0 +1,64 @@
+# Paper: Bridging Values and Behavior: A Hierarchical Framework for Proactive Embodied Agents
+
+---
+type: paper
+title: "Bridging Values and Behavior: A Hierarchical Framework for Proactive Embodied Agents"
+authors: Chunhui Zhang, Yuxuan Wang, Aoyang Qin, Yi-Long Lu, Kunlun Wu, Yizhou Wang, Wei Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.27699
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-30
+updated_at: 2026-04-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - embodied-agent
+  - memory
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, agent-safety, embodied-agent, memory, planning, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.27699
diff --git a/papers/items/2026-2604-27859-rethinking-agentic-reinforcement-learning-in-large-language-models.md b/papers/items/2026-2604-27859-rethinking-agentic-reinforcement-learning-in-large-language-models.md
new file mode 100644
index 0000000..49f09b6
--- /dev/null
+++ b/papers/items/2026-2604-27859-rethinking-agentic-reinforcement-learning-in-large-language-models.md
@@ -0,0 +1,62 @@
+# Paper: Rethinking Agentic Reinforcement Learning In Large Language Models
+
+---
+type: paper
+title: Rethinking Agentic Reinforcement Learning In Large Language Models
+authors: Fangming Cui, Ruixiao Zhu, Cheng Fang, Sunan Li, Jiahong Li
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.27859
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-30
+updated_at: 2026-05-15
+status: queued
+relevance: high
+topics:
+  - memory
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.ET
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: memory, planning, reasoning, tool-use
+- arXiv categories: cs.AI, cs.ET
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.27859
diff --git a/papers/items/2026-2604-28157-flashrt-towards-computationally-and-memory-efficient-red-teaming-for-prompt-inje.md b/papers/items/2026-2604-28157-flashrt-towards-computationally-and-memory-efficient-red-teaming-for-prompt-inje.md
new file mode 100644
index 0000000..bb39447
--- /dev/null
+++ b/papers/items/2026-2604-28157-flashrt-towards-computationally-and-memory-efficient-red-teaming-for-prompt-inje.md
@@ -0,0 +1,62 @@
+# Paper: FlashRT: Towards Computationally and Memory Efficient Red-Teaming for Prompt Injection and Knowledge Corruption
+
+---
+type: paper
+title: "FlashRT: Towards Computationally and Memory Efficient Red-Teaming for Prompt Injection and Knowledge Corruption"
+authors: Yanting Wang, Chenlong Yin, Ying Chen, Jinyuan Jia
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2604.28157
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-30
+updated_at: 2026-04-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, agent-safety, memory, rag, tool-use
+- arXiv categories: cs.CR
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2604.28157
diff --git a/papers/items/2026-2605-00081-alignment-contracts-for-agentic-security-systems.md b/papers/items/2026-2605-00081-alignment-contracts-for-agentic-security-systems.md
new file mode 100644
index 0000000..5c5269f
--- /dev/null
+++ b/papers/items/2026-2605-00081-alignment-contracts-for-agentic-security-systems.md
@@ -0,0 +1,63 @@
+# Paper: Alignment Contracts for Agentic Security Systems
+
+---
+type: paper
+title: Alignment Contracts for Agentic Security Systems
+authors: Isaac David, Marco Guarnieri, Arthur Gervais
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.00081
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-30
+updated_at: 2026-04-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.LO
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, planning, tool-use, workflow-agent
+- arXiv categories: cs.CR, cs.LO
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.00081
diff --git a/papers/items/2026-2605-00741-self-adaptive-multi-agent-llm-based-security-pattern-selection-for-iot-systems.md b/papers/items/2026-2605-00741-self-adaptive-multi-agent-llm-based-security-pattern-selection-for-iot-systems.md
new file mode 100644
index 0000000..bc500cc
--- /dev/null
+++ b/papers/items/2026-2605-00741-self-adaptive-multi-agent-llm-based-security-pattern-selection-for-iot-systems.md
@@ -0,0 +1,62 @@
+# Paper: Self-Adaptive Multi-Agent LLM-Based Security Pattern Selection for IoT Systems
+
+---
+type: paper
+title: Self-Adaptive Multi-Agent LLM-Based Security Pattern Selection for IoT Systems
+authors: Saeid Jamshidi, Foutse Khomh, Carol Fung, Kawser Wazed Nafi
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.00741
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-01
+updated_at: 2026-05-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, multi-agent, reasoning, tool-use
+- arXiv categories: cs.CR
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.00741
diff --git a/papers/items/2026-2605-00845-graph-query-generation-with-constraint-guided-large-language-agents.md b/papers/items/2026-2605-00845-graph-query-generation-with-constraint-guided-large-language-agents.md
new file mode 100644
index 0000000..a5e1758
--- /dev/null
+++ b/papers/items/2026-2605-00845-graph-query-generation-with-constraint-guided-large-language-agents.md
@@ -0,0 +1,63 @@
+# Paper: Graph Query Generation with Constraint-guided Large Language Agents
+
+---
+type: paper
+title: Graph Query Generation with Constraint-guided Large Language Agents
+authors: Mengying Wang, Nicolaas Jedema, Rahul Pandey, RaviKiran Krishnan, Jens Lehmann, Yinghui Wu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.00845
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-09
+updated_at: 2026-04-09
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - reasoning
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.DB
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, computer-use, reasoning, workflow-agent
+- arXiv categories: cs.DB, cs.AI, cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.00845
diff --git a/papers/items/2026-2605-01101-virtual-speech-therapist-a-clinician-in-the-loop-ai-speech-therapy-agent-for-per.md b/papers/items/2026-2605-01101-virtual-speech-therapist-a-clinician-in-the-loop-ai-speech-therapy-agent-for-per.md
new file mode 100644
index 0000000..78129f3
--- /dev/null
+++ b/papers/items/2026-2605-01101-virtual-speech-therapist-a-clinician-in-the-loop-ai-speech-therapy-agent-for-per.md
@@ -0,0 +1,68 @@
+# Paper: Virtual Speech Therapist: A Clinician-in-the-Loop AI Speech Therapy Agent for Personalized and Supervised Therapy
+
+---
+type: paper
+title: "Virtual Speech Therapist: A Clinician-in-the-Loop AI Speech Therapy Agent for Personalized and Supervised Therapy"
+authors: Shakeel Sheikh, Patrick Marmaroli, MD Sahidullah, Slim Ouni, Fabrice Hirsch, Goncalo Leal, Bjorn W Schuller
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.01101
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-01
+updated_at: 2026-06-15
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - multi-agent
+  - planning
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+  - cs.SD
+  - eess.AS
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, agent-safety, computer-use, multi-agent, planning, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI, cs.CL, cs.SD, eess.AS
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.01101
diff --git a/papers/items/2026-2605-01644-toward-a-principled-framework-for-agent-safety-measurement.md b/papers/items/2026-2605-01644-toward-a-principled-framework-for-agent-safety-measurement.md
new file mode 100644
index 0000000..abddee1
--- /dev/null
+++ b/papers/items/2026-2605-01644-toward-a-principled-framework-for-agent-safety-measurement.md
@@ -0,0 +1,61 @@
+# Paper: Toward a Principled Framework for Agent Safety Measurement
+
+---
+type: paper
+title: Toward a Principled Framework for Agent Safety Measurement
+authors: Shuyi Lin, Anshuman Suri, Alina Oprea, Cheng Tan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.01644
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-02
+updated_at: 2026-05-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, coding-agent, tool-use
+- arXiv categories: cs.CR
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.01644
diff --git a/papers/items/2026-2605-02240-physicianbench-evaluating-llm-agents-in-real-world-ehr-environments.md b/papers/items/2026-2605-02240-physicianbench-evaluating-llm-agents-in-real-world-ehr-environments.md
new file mode 100644
index 0000000..8c3d085
--- /dev/null
+++ b/papers/items/2026-2605-02240-physicianbench-evaluating-llm-agents-in-real-world-ehr-environments.md
@@ -0,0 +1,63 @@
+# Paper: PhysicianBench: Evaluating LLM Agents in Real-World EHR Environments
+
+---
+type: paper
+title: "PhysicianBench: Evaluating LLM Agents in Real-World EHR Environments"
+authors: Ruoqi Liu, Imran Q. Mohiuddin, Austin J. Schoeffler, Kavita Renduchintala, Ashwin Nayak, Prasantha L. Vemu, Shivam C. Vedak, Kameron C. Black, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.02240
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-04
+updated_at: 2026-05-04
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, planning, rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.02240
diff --git a/papers/items/2026-2605-03242-enhancing-agent-safety-judgment-controlled-benchmark-rewriting-and-analogical-re.md b/papers/items/2026-2605-03242-enhancing-agent-safety-judgment-controlled-benchmark-rewriting-and-analogical-re.md
new file mode 100644
index 0000000..036c251
--- /dev/null
+++ b/papers/items/2026-2605-03242-enhancing-agent-safety-judgment-controlled-benchmark-rewriting-and-analogical-re.md
@@ -0,0 +1,64 @@
+# Paper: Enhancing Agent Safety Judgment: Controlled Benchmark Rewriting and Analogical Reasoning for Deceptive Out-of-Distribution Scenarios
+
+---
+type: paper
+title: "Enhancing Agent Safety Judgment: Controlled Benchmark Rewriting and Analogical Reasoning for Deceptive Out-of-Distribution Scenarios"
+authors: Zuoyu Zhang, Yancheng Zhu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.03242
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-05
+updated_at: 2026-05-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - multi-agent
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 22
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, computer-use, multi-agent, rag, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 22
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.03242
diff --git a/papers/items/2026-2605-03312-memflow-intent-driven-memory-orchestration-for-small-language-model-agents.md b/papers/items/2026-2605-03312-memflow-intent-driven-memory-orchestration-for-small-language-model-agents.md
new file mode 100644
index 0000000..98f0689
--- /dev/null
+++ b/papers/items/2026-2605-03312-memflow-intent-driven-memory-orchestration-for-small-language-model-agents.md
@@ -0,0 +1,63 @@
+# Paper: MemFlow: Intent-Driven Memory Orchestration for Small Language Model Agents
+
+---
+type: paper
+title: "MemFlow: Intent-Driven Memory Orchestration for Small Language Model Agents"
+authors: Jiayi Chen, Yingcong Li, Guiling Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.03312
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-05
+updated_at: 2026-05-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, memory, planning, rag, reasoning, tool-use
+- arXiv categories: cs.MA
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.03312
diff --git a/papers/items/2026-2605-03328-llm-adam-a-generalizable-llm-agent-framework-for-pre-print-anomaly-detection-in-.md b/papers/items/2026-2605-03328-llm-adam-a-generalizable-llm-agent-framework-for-pre-print-anomaly-detection-in-.md
new file mode 100644
index 0000000..a6ba0ff
--- /dev/null
+++ b/papers/items/2026-2605-03328-llm-adam-a-generalizable-llm-agent-framework-for-pre-print-anomaly-detection-in-.md
@@ -0,0 +1,61 @@
+# Paper: LLM-ADAM: A Generalizable LLM Agent Framework for Pre-Print Anomaly Detection in Additive Manufacturing
+
+---
+type: paper
+title: "LLM-ADAM: A Generalizable LLM Agent Framework for Pre-Print Anomaly Detection in Additive Manufacturing"
+authors: Ahmadreza Eslaminia, Chuhan Cai, Cameron Smith, Ruo-Syuan Mei, Shichen Li, Rajiv Malhotra, Klara Nahrstedt, Chenhui Shao
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.03328
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-05
+updated_at: 2026-05-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, computer-use, planning
+- arXiv categories: cs.LG, cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.03328
diff --git a/papers/items/2026-2605-03505-lats-rca-language-agent-tree-search-for-root-cause-analysis-in-microservices.md b/papers/items/2026-2605-03505-lats-rca-language-agent-tree-search-for-root-cause-analysis-in-microservices.md
new file mode 100644
index 0000000..d9f2bb4
--- /dev/null
+++ b/papers/items/2026-2605-03505-lats-rca-language-agent-tree-search-for-root-cause-analysis-in-microservices.md
@@ -0,0 +1,61 @@
+# Paper: LATS-RCA: Language Agent Tree Search for Root Cause Analysis in Microservices
+
+---
+type: paper
+title: "LATS-RCA: Language Agent Tree Search for Root Cause Analysis in Microservices"
+authors: Alexander Naakka, Yuqing Wang, Mika V Mäntylä
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.03505
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-05
+updated_at: 2026-06-12
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, computer-use, rag, reasoning
+- arXiv categories: cs.SE
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.03505
diff --git a/papers/items/2026-2605-04107-tscg-deterministic-tool-schema-compilation-for-agentic-llm-deployments.md b/papers/items/2026-2605-04107-tscg-deterministic-tool-schema-compilation-for-agentic-llm-deployments.md
new file mode 100644
index 0000000..875e38b
--- /dev/null
+++ b/papers/items/2026-2605-04107-tscg-deterministic-tool-schema-compilation-for-agentic-llm-deployments.md
@@ -0,0 +1,62 @@
+# Paper: TSCG: Deterministic Tool-Schema Compilation for Agentic LLM Deployments
+
+---
+type: paper
+title: "TSCG: Deterministic Tool-Schema Compilation for Agentic LLM Deployments"
+authors: Furkan Sakizli
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.04107
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-04
+updated_at: 2026-05-04
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, computer-use, tool-use
+- arXiv categories: cs.SE, cs.AI, cs.CL
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.04107
diff --git a/papers/items/2026-2605-04808-decodingtrust-agent-platform-dtap-a-controllable-and-interactive-red-teaming-pla.md b/papers/items/2026-2605-04808-decodingtrust-agent-platform-dtap-a-controllable-and-interactive-red-teaming-pla.md
new file mode 100644
index 0000000..a2f1d7b
--- /dev/null
+++ b/papers/items/2026-2605-04808-decodingtrust-agent-platform-dtap-a-controllable-and-interactive-red-teaming-pla.md
@@ -0,0 +1,64 @@
+# Paper: DecodingTrust-Agent Platform (DTap): A Controllable and Interactive Red-Teaming Platform for AI Agents
+
+---
+type: paper
+title: "DecodingTrust-Agent Platform (DTap): A Controllable and Interactive Red-Teaming Platform for AI Agents"
+authors: Zhaorun Chen, Xun Liu, Haibo Tong, Chengquan Guo, Yuzhou Nie, Jiawei Zhang, Mintong Kang, Chejian Xu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.04808
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-06
+updated_at: 2026-05-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - planning
+  - tool-use
+  - workflow-agent
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, coding-agent, planning, tool-use, workflow-agent, world-model
+- arXiv categories: cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.04808
diff --git a/papers/items/2026-2605-05242-beyond-semantic-similarity-rethinking-retrieval-for-agentic-search-via-direct-co.md b/papers/items/2026-2605-05242-beyond-semantic-similarity-rethinking-retrieval-for-agentic-search-via-direct-co.md
new file mode 100644
index 0000000..74b26d7
--- /dev/null
+++ b/papers/items/2026-2605-05242-beyond-semantic-similarity-rethinking-retrieval-for-agentic-search-via-direct-co.md
@@ -0,0 +1,63 @@
+# Paper: Beyond Semantic Similarity: Rethinking Retrieval for Agentic Search via Direct Corpus Interaction
+
+---
+type: paper
+title: "Beyond Semantic Similarity: Rethinking Retrieval for Agentic Search via Direct Corpus Interaction"
+authors: Zhuofeng Li, Haoxiang Zhang, Cong Wei, Pan Lu, Ping Nie, Yi Lu, Yuyang Bai, Shangbin Feng, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.05242
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-03
+updated_at: 2026-05-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.IR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, planning, rag, reasoning, tool-use
+- arXiv categories: cs.IR, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.05242
diff --git a/papers/items/2026-2605-05704-safeharbor-hierarchical-memory-augmented-guardrail-for-llm-agent-safety.md b/papers/items/2026-2605-05704-safeharbor-hierarchical-memory-augmented-guardrail-for-llm-agent-safety.md
new file mode 100644
index 0000000..80a955b
--- /dev/null
+++ b/papers/items/2026-2605-05704-safeharbor-hierarchical-memory-augmented-guardrail-for-llm-agent-safety.md
@@ -0,0 +1,63 @@
+# Paper: SafeHarbor: Hierarchical Memory-Augmented Guardrail for LLM Agent Safety
+
+---
+type: paper
+title: "SafeHarbor: Hierarchical Memory-Augmented Guardrail for LLM Agent Safety"
+authors: Zhe Liu, Zonghao Ying, Wenxin Zhang, Quanchen Zou, Deyue Zhang, Dongdong Yang, Xiangzheng Zhang, Hao Peng
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.05704
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-07
+updated_at: 2026-05-22
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - computer-use
+  - memory
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 22
+collection_queries: agent-safety, autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety, autonomous-agent-llm
+- inferred topics: agent-safety, computer-use, memory, reasoning, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 22
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.05704
diff --git a/papers/items/2026-2605-05716-more-is-not-always-better-cross-component-interference-in-llm-agent-scaffolding.md b/papers/items/2026-2605-05716-more-is-not-always-better-cross-component-interference-in-llm-agent-scaffolding.md
new file mode 100644
index 0000000..a7127f9
--- /dev/null
+++ b/papers/items/2026-2605-05716-more-is-not-always-better-cross-component-interference-in-llm-agent-scaffolding.md
@@ -0,0 +1,64 @@
+# Paper: More Is Not Always Better: Cross-Component Interference in LLM Agent Scaffolding
+
+---
+type: paper
+title: "More Is Not Always Better: Cross-Component Interference in LLM Agent Scaffolding"
+authors: Ming Liu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.05716
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-07
+updated_at: 2026-05-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, memory, planning, rag, reasoning, tool-use
+- arXiv categories: cs.AI, cs.CL
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.05716
diff --git a/papers/items/2026-2605-06078-milestone-guided-policy-learning-for-long-horizon-language-agents.md b/papers/items/2026-2605-06078-milestone-guided-policy-learning-for-long-horizon-language-agents.md
new file mode 100644
index 0000000..69577bb
--- /dev/null
+++ b/papers/items/2026-2605-06078-milestone-guided-policy-learning-for-long-horizon-language-agents.md
@@ -0,0 +1,63 @@
+# Paper: Milestone-Guided Policy Learning for Long-Horizon Language Agents
+
+---
+type: paper
+title: Milestone-Guided Policy Learning for Long-Horizon Language Agents
+authors: Zixuan Wang, Yuchen Yan, Hongxing Li, Teng Pan, Dingming Li, Ruiqing Zhang, Weiming Lu, Jun Xiao, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.06078
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-07
+updated_at: 2026-05-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, computer-use, planning, rag, tool-use
+- arXiv categories: cs.CL, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.06078
diff --git a/papers/items/2026-2605-06713-agentic-ai-and-the-industrialization-of-cyber-offense-forecast-consequences-and-.md b/papers/items/2026-2605-06713-agentic-ai-and-the-industrialization-of-cyber-offense-forecast-consequences-and-.md
new file mode 100644
index 0000000..474c789
--- /dev/null
+++ b/papers/items/2026-2605-06713-agentic-ai-and-the-industrialization-of-cyber-offense-forecast-consequences-and-.md
@@ -0,0 +1,64 @@
+# Paper: Agentic AI and the Industrialization of Cyber Offense: Forecast, Consequences, and Defensive Priorities for Enterprises and the Mittelstand
+
+---
+type: paper
+title: "Agentic AI and the Industrialization of Cyber Offense: Forecast, Consequences, and Defensive Priorities for Enterprises and the Mittelstand"
+authors: Christopher Koch
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.06713
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-06
+updated_at: 2026-05-06
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - computer-use
+  - planning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+  - cs.HC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-safety, planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety, planning-agent
+- inferred topics: agent-safety, computer-use, planning, tool-use, workflow-agent
+- arXiv categories: cs.CR, cs.AI, cs.HC
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.06713
diff --git a/papers/items/2026-2605-06716-from-storage-to-experience-a-survey-on-the-evolution-of-llm-agent-memory-mechani.md b/papers/items/2026-2605-06716-from-storage-to-experience-a-survey-on-the-evolution-of-llm-agent-memory-mechani.md
new file mode 100644
index 0000000..11700a1
--- /dev/null
+++ b/papers/items/2026-2605-06716-from-storage-to-experience-a-survey-on-the-evolution-of-llm-agent-memory-mechani.md
@@ -0,0 +1,63 @@
+# Paper: From Storage to Experience: A Survey on the Evolution of LLM Agent Memory Mechanisms
+
+---
+type: paper
+title: "From Storage to Experience: A Survey on the Evolution of LLM Agent Memory Mechanisms"
+authors: Jinghao Luo, Yuchen Tian, Chuxue Cao, Ziyang Luo, Hongzhan Lin, Kaixin Li, Chuyi Kong, Ruichao Yang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.06716
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-07
+updated_at: 2026-05-07
+status: queued
+relevance: high
+topics:
+  - memory
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: memory, planning, rag, reasoning, tool-use
+- arXiv categories: cs.AI, cs.CL
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.06716
diff --git a/papers/items/2026-2605-06737-a-self-healing-framework-for-reliable-llm-based-autonomous-agents.md b/papers/items/2026-2605-06737-a-self-healing-framework-for-reliable-llm-based-autonomous-agents.md
new file mode 100644
index 0000000..dd167a7
--- /dev/null
+++ b/papers/items/2026-2605-06737-a-self-healing-framework-for-reliable-llm-based-autonomous-agents.md
@@ -0,0 +1,64 @@
+# Paper: A Self-Healing Framework for Reliable LLM-Based Autonomous Agents
+
+---
+type: paper
+title: A Self-Healing Framework for Reliable LLM-Based Autonomous Agents
+authors: Cheonsu Jeong, Younggun Shin
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.06737
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-07
+updated_at: 2026-05-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - multi-agent
+  - planning
+  - reasoning
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, computer-use, multi-agent, planning, reasoning, workflow-agent
+- arXiv categories: cs.SE, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.06737
diff --git a/papers/items/2026-2605-06812-towards-security-auditable-llm-agents-a-unified-graph-representation.md b/papers/items/2026-2605-06812-towards-security-auditable-llm-agents-a-unified-graph-representation.md
new file mode 100644
index 0000000..ee08d14
--- /dev/null
+++ b/papers/items/2026-2605-06812-towards-security-auditable-llm-agents-a-unified-graph-representation.md
@@ -0,0 +1,64 @@
+# Paper: Towards Security-Auditable LLM Agents: A Unified Graph Representation
+
+---
+type: paper
+title: "Towards Security-Auditable LLM Agents: A Unified Graph Representation"
+authors: Chaofan Li, Lyuye Zhang, Jintao Zhai, Siyue Feng, Xichun Yang, Huahao Wang, Shihan Dou, Yu Ji, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.06812
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-07
+updated_at: 2026-05-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - multi-agent
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 21
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, memory, multi-agent, rag, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 21
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.06812
diff --git a/papers/items/2026-2605-06869-agentick-a-unified-benchmark-for-general-sequential-decision-making-agents.md b/papers/items/2026-2605-06869-agentick-a-unified-benchmark-for-general-sequential-decision-making-agents.md
new file mode 100644
index 0000000..ae45339
--- /dev/null
+++ b/papers/items/2026-2605-06869-agentick-a-unified-benchmark-for-general-sequential-decision-making-agents.md
@@ -0,0 +1,64 @@
+# Paper: Agentick: A Unified Benchmark for General Sequential Decision-Making Agents
+
+---
+type: paper
+title: "Agentick: A Unified Benchmark for General Sequential Decision-Making Agents"
+authors: Roger Creus Castanyer, Pablo Samuel Castro, Glen Berseth
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.06869
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-07
+updated_at: 2026-05-12
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - multi-agent
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 23
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, coding-agent, multi-agent, planning, rag, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 23
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.06869
diff --git a/papers/items/2026-2605-06890-beyond-the-black-box-interpretability-of-agentic-ai-tool-use.md b/papers/items/2026-2605-06890-beyond-the-black-box-interpretability-of-agentic-ai-tool-use.md
new file mode 100644
index 0000000..12685a9
--- /dev/null
+++ b/papers/items/2026-2605-06890-beyond-the-black-box-interpretability-of-agentic-ai-tool-use.md
@@ -0,0 +1,63 @@
+# Paper: Beyond the Black Box: Interpretability of Agentic AI Tool Use
+
+---
+type: paper
+title: "Beyond the Black Box: Interpretability of Agentic AI Tool Use"
+authors: Hariom Tatsat, Ariye Shater
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.06890
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-07
+updated_at: 2026-07-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, agent-safety, planning, tool-use, workflow-agent
+- arXiv categories: cs.AI, cs.MA
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.06890
diff --git a/papers/items/2026-2605-06957-learning-and-reusing-policy-decompositions-for-hierarchical-generalized-planning.md b/papers/items/2026-2605-06957-learning-and-reusing-policy-decompositions-for-hierarchical-generalized-planning.md
new file mode 100644
index 0000000..582cde8
--- /dev/null
+++ b/papers/items/2026-2605-06957-learning-and-reusing-policy-decompositions-for-hierarchical-generalized-planning.md
@@ -0,0 +1,60 @@
+# Paper: Learning and Reusing Policy Decompositions for Hierarchical Generalized Planning with LLM Agents
+
+---
+type: paper
+title: Learning and Reusing Policy Decompositions for Hierarchical Generalized Planning with LLM Agents
+authors: Shirin Sohrabi, Haritha Ananthakrishnan, Harsha Kokel, Kavitha Srinivas, Michael Katz
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.06957
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-07
+updated_at: 2026-05-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, planning, rag
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.06957
diff --git a/papers/items/2026-2605-06992-why-does-agentic-safety-fail-to-generalize-across-tasks.md b/papers/items/2026-2605-06992-why-does-agentic-safety-fail-to-generalize-across-tasks.md
new file mode 100644
index 0000000..e2612a5
--- /dev/null
+++ b/papers/items/2026-2605-06992-why-does-agentic-safety-fail-to-generalize-across-tasks.md
@@ -0,0 +1,60 @@
+# Paper: Why Does Agentic Safety Fail to Generalize Across Tasks?
+
+---
+type: paper
+title: Why Does Agentic Safety Fail to Generalize Across Tasks?
+authors: Yonatan Slutzky, Yotam Alexander, Tomer Slor, Yoav Nagel, Nadav Cohen
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.06992
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-07
+updated_at: 2026-05-07
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - embodied-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - stat.ML
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-safety, embodied-agent
+- arXiv categories: cs.LG, stat.ML
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.06992
diff --git a/papers/items/2026-2605-07112-switchcraft-ai-model-router-for-agentic-tool-calling.md b/papers/items/2026-2605-07112-switchcraft-ai-model-router-for-agentic-tool-calling.md
new file mode 100644
index 0000000..982ba2e
--- /dev/null
+++ b/papers/items/2026-2605-07112-switchcraft-ai-model-router-for-agentic-tool-calling.md
@@ -0,0 +1,61 @@
+# Paper: Switchcraft: AI Model Router for Agentic Tool Calling
+
+---
+type: paper
+title: "Switchcraft: AI Model Router for Agentic Tool Calling"
+authors: Sharad Agarwal, Pooria Namyar, Alec Wolman, Rahul Ambavat, Ankur Gupta, Qizheng Zhang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.07112
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-08
+updated_at: 2026-05-08
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, reasoning, tool-use
+- arXiv categories: cs.AI, cs.MA
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.07112
diff --git a/papers/items/2026-2605-07251-can-agents-price-a-reaction-evaluating-llms-on-chemical-cost-reasoning.md b/papers/items/2026-2605-07251-can-agents-price-a-reaction-evaluating-llms-on-chemical-cost-reasoning.md
new file mode 100644
index 0000000..f406c1a
--- /dev/null
+++ b/papers/items/2026-2605-07251-can-agents-price-a-reaction-evaluating-llms-on-chemical-cost-reasoning.md
@@ -0,0 +1,62 @@
+# Paper: Can Agents Price a Reaction? Evaluating LLMs on Chemical Cost Reasoning
+
+---
+type: paper
+title: Can Agents Price a Reaction? Evaluating LLMs on Chemical Cost Reasoning
+authors: Yuyang Wu, Yue Huang, Shuaike Shen, Xujian Wang, Shuhao Zhang, Qiyao Xue, Weichen Liu, Runtian Gao, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.07251
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-08
+updated_at: 2026-05-08
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, planning, rag, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.07251
diff --git a/papers/items/2026-2605-07830-cybiasbench-benchmarking-bias-in-llm-agents-for-cyber-attack-scenarios.md b/papers/items/2026-2605-07830-cybiasbench-benchmarking-bias-in-llm-agents-for-cyber-attack-scenarios.md
new file mode 100644
index 0000000..1e66faa
--- /dev/null
+++ b/papers/items/2026-2605-07830-cybiasbench-benchmarking-bias-in-llm-agents-for-cyber-attack-scenarios.md
@@ -0,0 +1,60 @@
+# Paper: CyBiasBench: Benchmarking Bias in LLM Agents for Cyber-Attack Scenarios
+
+---
+type: paper
+title: "CyBiasBench: Benchmarking Bias in LLM Agents for Cyber-Attack Scenarios"
+authors: Taein Lim, Seongyong Ju, Munhyeok Kim, Hyunjun Kim, Hoki Kim
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.07830
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-08
+updated_at: 2026-05-08
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, agent-safety
+- arXiv categories: cs.CR, cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.07830
diff --git a/papers/items/2026-2605-08374-memq-integrating-q-learning-into-self-evolving-memory-agents-over-provenance-dag.md b/papers/items/2026-2605-08374-memq-integrating-q-learning-into-self-evolving-memory-agents-over-provenance-dag.md
new file mode 100644
index 0000000..44a2c77
--- /dev/null
+++ b/papers/items/2026-2605-08374-memq-integrating-q-learning-into-self-evolving-memory-agents-over-provenance-dag.md
@@ -0,0 +1,64 @@
+# Paper: MemQ: Integrating Q-Learning into Self-Evolving Memory Agents over Provenance DAGs
+
+---
+type: paper
+title: "MemQ: Integrating Q-Learning into Self-Evolving Memory Agents over Provenance DAGs"
+authors: Junwei Liao, Haoting Shi, Ruiwen Zhou, Jiaqian Wang, Shengtao Zhang, Wei Zhang, Ying Wen, Zhiyu Li, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.08374
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-08
+updated_at: 2026-05-14
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - embodied-agent
+  - memory
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, computer-use, embodied-agent, memory, rag, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.08374
diff --git a/papers/items/2026-2605-08442-defense-effectiveness-across-architectural-layers-a-mechanistic-evaluation-of-pe.md b/papers/items/2026-2605-08442-defense-effectiveness-across-architectural-layers-a-mechanistic-evaluation-of-pe.md
new file mode 100644
index 0000000..ac8ead1
--- /dev/null
+++ b/papers/items/2026-2605-08442-defense-effectiveness-across-architectural-layers-a-mechanistic-evaluation-of-pe.md
@@ -0,0 +1,66 @@
+# Paper: Defense effectiveness across architectural layers: a mechanistic evaluation of persistent memory attacks on stateful LLM agents
+
+---
+type: paper
+title: "Defense effectiveness across architectural layers: a mechanistic evaluation of persistent memory attacks on stateful LLM agents"
+authors: Jun Wen Leong
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.08442
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-08
+updated_at: 2026-07-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - memory
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 23
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, computer-use, memory, rag, reasoning, tool-use
+- arXiv categories: cs.CR, cs.AI, cs.LG
+- collection score: 23
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.08442
diff --git a/papers/items/2026-2605-08763-when-llms-team-up-a-coordinated-attack-framework-for-automated-cyber-intrusions.md b/papers/items/2026-2605-08763-when-llms-team-up-a-coordinated-attack-framework-for-automated-cyber-intrusions.md
new file mode 100644
index 0000000..bd2ce35
--- /dev/null
+++ b/papers/items/2026-2605-08763-when-llms-team-up-a-coordinated-attack-framework-for-automated-cyber-intrusions.md
@@ -0,0 +1,64 @@
+# Paper: When LLMs Team Up: A Coordinated Attack Framework for Automated Cyber Intrusions
+
+---
+type: paper
+title: "When LLMs Team Up: A Coordinated Attack Framework for Automated Cyber Intrusions"
+authors: Minfeng Qi, Tianqing Zhu, Zijie Xu, Congcong Zhu, Qin Wang, Wanlei Zhou
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.08763
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-09
+updated_at: 2026-05-09
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - planning
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, agent-safety, multi-agent, planning, rag, tool-use, workflow-agent
+- arXiv categories: cs.CR
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.08763
diff --git a/papers/items/2026-2605-08876-otora-a-unified-red-teaming-framework-for-reasoning-level-denial-of-service-in-l.md b/papers/items/2026-2605-08876-otora-a-unified-red-teaming-framework-for-reasoning-level-denial-of-service-in-l.md
new file mode 100644
index 0000000..78edad4
--- /dev/null
+++ b/papers/items/2026-2605-08876-otora-a-unified-red-teaming-framework-for-reasoning-level-denial-of-service-in-l.md
@@ -0,0 +1,60 @@
+# Paper: OTora: A Unified Red Teaming Framework for Reasoning-Level Denial-of-Service in LLM Agents
+
+---
+type: paper
+title: "OTora: A Unified Red Teaming Framework for Reasoning-Level Denial-of-Service in LLM Agents"
+authors: Xinyu Li, Ronghui Mu, Lin Li, Tianjin Huang, Gaojie Jin
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.08876
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-09
+updated_at: 2026-06-07
+status: queued
+relevance: high
+topics:
+  - computer-use
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: computer-use, reasoning, tool-use
+- arXiv categories: cs.LG
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.08876
diff --git a/papers/items/2026-2605-08964-trustworthy-ai-ensuring-reliability-and-accountability-from-models-to-agents.md b/papers/items/2026-2605-08964-trustworthy-ai-ensuring-reliability-and-accountability-from-models-to-agents.md
new file mode 100644
index 0000000..652faf1
--- /dev/null
+++ b/papers/items/2026-2605-08964-trustworthy-ai-ensuring-reliability-and-accountability-from-models-to-agents.md
@@ -0,0 +1,63 @@
+# Paper: Trustworthy AI: Ensuring Reliability and Accountability from Models to Agents
+
+---
+type: paper
+title: "Trustworthy AI: Ensuring Reliability and Accountability from Models to Agents"
+authors: Carol Xuan Long
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.08964
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-09
+updated_at: 2026-05-09
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - multi-agent
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, agent-safety, coding-agent, multi-agent, rag, tool-use
+- arXiv categories: cs.LG
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.08964
diff --git a/papers/items/2026-2605-09168-civex-causal-intervention-verification-for-language-agents.md b/papers/items/2026-2605-09168-civex-causal-intervention-verification-for-language-agents.md
new file mode 100644
index 0000000..6102580
--- /dev/null
+++ b/papers/items/2026-2605-09168-civex-causal-intervention-verification-for-language-agents.md
@@ -0,0 +1,62 @@
+# Paper: CIVeX: Causal Intervention Verification for Language Agents
+
+---
+type: paper
+title: "CIVeX: Causal Intervention Verification for Language Agents"
+authors: Fabio Rovai
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.09168
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-09
+updated_at: 2026-05-09
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-safety, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI, cs.LG
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.09168
diff --git a/papers/items/2026-2605-09692-causal-state-binding-predicts-action-control-in-language-agents.md b/papers/items/2026-2605-09692-causal-state-binding-predicts-action-control-in-language-agents.md
new file mode 100644
index 0000000..b033172
--- /dev/null
+++ b/papers/items/2026-2605-09692-causal-state-binding-predicts-action-control-in-language-agents.md
@@ -0,0 +1,62 @@
+# Paper: Causal state binding predicts action control in language agents
+
+---
+type: paper
+title: Causal state binding predicts action control in language agents
+authors: Xiao Jia
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.09692
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-10
+updated_at: 2026-06-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - memory
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, coding-agent, memory, planning, tool-use
+- arXiv categories: cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.09692
diff --git a/papers/items/2026-2605-10365-agent-valuebench-a-comprehensive-benchmark-for-evaluating-agent-values.md b/papers/items/2026-2605-10365-agent-valuebench-a-comprehensive-benchmark-for-evaluating-agent-values.md
new file mode 100644
index 0000000..bf5c525
--- /dev/null
+++ b/papers/items/2026-2605-10365-agent-valuebench-a-comprehensive-benchmark-for-evaluating-agent-values.md
@@ -0,0 +1,60 @@
+# Paper: Agent-ValueBench: A Comprehensive Benchmark for Evaluating Agent Values
+
+---
+type: paper
+title: "Agent-ValueBench: A Comprehensive Benchmark for Evaluating Agent Values"
+authors: Haonan Dong, Qiguan Feng, Kehan Jiang, Haoran Ye, Xin Zhang, Guojie Song
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.10365
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-11
+updated_at: 2026-05-11
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, agent-safety, tool-use
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.10365
diff --git a/papers/items/2026-2605-10763-matra-modeling-the-attack-surface-of-agentic-ai-systems-openclaw-case-study.md b/papers/items/2026-2605-10763-matra-modeling-the-attack-surface-of-agentic-ai-systems-openclaw-case-study.md
new file mode 100644
index 0000000..c2076c4
--- /dev/null
+++ b/papers/items/2026-2605-10763-matra-modeling-the-attack-surface-of-agentic-ai-systems-openclaw-case-study.md
@@ -0,0 +1,62 @@
+# Paper: MATRA: Modeling the Attack Surface of Agentic AI Systems -- OpenClaw Case Study
+
+---
+type: paper
+title: "MATRA: Modeling the Attack Surface of Agentic AI Systems -- OpenClaw Case Study"
+authors: Tim Van hamme, Thomas Vissers, Javier Carnerero-Cano, Mario Fritz, Emil C. Lupu, Lieven Desmet, Dinil Mon Divakaran
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.10763
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-11
+updated_at: 2026-05-11
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, agent-safety, rag, tool-use
+- arXiv categories: cs.AI, cs.CR
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.10763
diff --git a/papers/items/2026-2605-10779-litmus-benchmarking-behavioral-jailbreaks-of-llm-agents-in-real-os-environments.md b/papers/items/2026-2605-10779-litmus-benchmarking-behavioral-jailbreaks-of-llm-agents-in-real-os-environments.md
new file mode 100644
index 0000000..88e23c6
--- /dev/null
+++ b/papers/items/2026-2605-10779-litmus-benchmarking-behavioral-jailbreaks-of-llm-agents-in-real-os-environments.md
@@ -0,0 +1,62 @@
+# Paper: LITMUS: Benchmarking Behavioral Jailbreaks of LLM Agents in Real OS Environments
+
+---
+type: paper
+title: "LITMUS: Benchmarking Behavioral Jailbreaks of LLM Agents in Real OS Environments"
+authors: Chiyu Zhang, Huiqin Yang, Bendong Jiang, Xiaolei Zhang, Yiran Zhao, Ruyi Chen, Lu Zhou, Xiaogang Xu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.10779
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-11
+updated_at: 2026-05-11
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, agent-safety, multi-agent, tool-use
+- arXiv categories: cs.CR, cs.CL
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.10779
diff --git a/papers/items/2026-2605-10870-remember-the-decision-not-the-description-a-rate-distortion-framework-for-agent-.md b/papers/items/2026-2605-10870-remember-the-decision-not-the-description-a-rate-distortion-framework-for-agent-.md
new file mode 100644
index 0000000..e1dce16
--- /dev/null
+++ b/papers/items/2026-2605-10870-remember-the-decision-not-the-description-a-rate-distortion-framework-for-agent-.md
@@ -0,0 +1,60 @@
+# Paper: Remember the Decision, Not the Description: A Rate-Distortion Framework for Agent Memory
+
+---
+type: paper
+title: "Remember the Decision, Not the Description: A Rate-Distortion Framework for Agent Memory"
+authors: Mingxi Zou, Zhihan Guo, Langzhang Liang, Zhuo Wang, Qifan Wang, Qingsong Wen, Irwin King, Lizhen Qu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.10870
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-11
+updated_at: 2026-05-11
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, memory, planning
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.10870
diff --git a/papers/items/2026-2605-11039-the-granularity-mismatch-in-agent-security-argument-level-provenance-solves-enfo.md b/papers/items/2026-2605-11039-the-granularity-mismatch-in-agent-security-argument-level-provenance-solves-enfo.md
new file mode 100644
index 0000000..7133dbc
--- /dev/null
+++ b/papers/items/2026-2605-11039-the-granularity-mismatch-in-agent-security-argument-level-provenance-solves-enfo.md
@@ -0,0 +1,65 @@
+# Paper: The Granularity Mismatch in Agent Security: Argument-Level Provenance Solves Enforcement and Isolates the LLM Reasoning Bottleneck
+
+---
+type: paper
+title: "The Granularity Mismatch in Agent Security: Argument-Level Provenance Solves Enforcement and Isolates the LLM Reasoning Bottleneck"
+authors: Linfeng Fan, Ziwei Li, Yuan Tian, Yichen Wang, Rongsheng Li, Xiong Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.11039
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-11
+updated_at: 2026-05-11
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, planning, rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.CR, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.11039
diff --git a/papers/items/2026-2605-11225-pivot-bridging-planning-and-execution-in-llm-agents-via-trajectory-refinement.md b/papers/items/2026-2605-11225-pivot-bridging-planning-and-execution-in-llm-agents-via-trajectory-refinement.md
new file mode 100644
index 0000000..d2a2745
--- /dev/null
+++ b/papers/items/2026-2605-11225-pivot-bridging-planning-and-execution-in-llm-agents-via-trajectory-refinement.md
@@ -0,0 +1,64 @@
+# Paper: PIVOT: Bridging Planning and Execution in LLM Agents via Trajectory Refinement
+
+---
+type: paper
+title: "PIVOT: Bridging Planning and Execution in LLM Agents via Trajectory Refinement"
+authors: Tuo Zhang, Alin-Ionut Popa, Yan Xu, Rui Song, Dimitrios Dimitriadis
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.11225
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-11
+updated_at: 2026-05-11
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.LG
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: autonomous-agent-llm, planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm, planning-agent
+- inferred topics: agent-evaluation, agent-safety, coding-agent, planning, tool-use
+- arXiv categories: cs.AI, cs.LG, cs.MA
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.11225
diff --git a/papers/items/2026-2605-11388-deep-reasoning-in-general-purpose-agents-via-structured-meta-cognition.md b/papers/items/2026-2605-11388-deep-reasoning-in-general-purpose-agents-via-structured-meta-cognition.md
new file mode 100644
index 0000000..04a3ad9
--- /dev/null
+++ b/papers/items/2026-2605-11388-deep-reasoning-in-general-purpose-agents-via-structured-meta-cognition.md
@@ -0,0 +1,63 @@
+# Paper: Deep Reasoning in General Purpose Agents via Structured Meta-Cognition
+
+---
+type: paper
+title: Deep Reasoning in General Purpose Agents via Structured Meta-Cognition
+authors: Dean Light, Michael Theologitis, Kshitish Ghate, Shuyue Stella Li, Benjamin Newman, Chirag Shah, Aylin Caliskan, Pang Wei Koh, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.11388
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-12
+updated_at: 2026-05-12
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, computer-use, planning, rag, reasoning
+- arXiv categories: cs.CL, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.11388
diff --git a/papers/items/2026-2605-11534-prism-planning-and-reasoning-with-intent-in-simulated-embodied-environments.md b/papers/items/2026-2605-11534-prism-planning-and-reasoning-with-intent-in-simulated-embodied-environments.md
new file mode 100644
index 0000000..ec11f56
--- /dev/null
+++ b/papers/items/2026-2605-11534-prism-planning-and-reasoning-with-intent-in-simulated-embodied-environments.md
@@ -0,0 +1,64 @@
+# Paper: PRISM: : Planning and Reasoning with Intent in Simulated Embodied Environments
+
+---
+type: paper
+title: "PRISM: : Planning and Reasoning with Intent in Simulated Embodied Environments"
+authors: Yunn Kang Lim, Pengzhan Sun, Ziyi Bai, Xun Xu, Angela Yao, Xulei Yang, Shijie Li
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.11534
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-12
+updated_at: 2026-05-12
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - embodied-agent
+  - memory
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.RO
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, embodied-agent, memory, planning, rag, reasoning, tool-use
+- arXiv categories: cs.RO
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.11534
diff --git a/papers/items/2026-2605-11633-can-llm-agents-respond-to-disasters-benchmarking-heterogeneous-geospatial-reason.md b/papers/items/2026-2605-11633-can-llm-agents-respond-to-disasters-benchmarking-heterogeneous-geospatial-reason.md
new file mode 100644
index 0000000..736ff21
--- /dev/null
+++ b/papers/items/2026-2605-11633-can-llm-agents-respond-to-disasters-benchmarking-heterogeneous-geospatial-reason.md
@@ -0,0 +1,63 @@
+# Paper: Can LLM Agents Respond to Disasters? Benchmarking Heterogeneous Geospatial Reasoning in Emergency Operations
+
+---
+type: paper
+title: Can LLM Agents Respond to Disasters? Benchmarking Heterogeneous Geospatial Reasoning in Emergency Operations
+authors: Junjue Wang, Weihao Xuan, Heli Qi, Pengyu Dai, Kunyi Liu, Hongruixuan Chen, Zhuo Zheng, Junshi Xia, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.11633
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-12
+updated_at: 2026-05-12
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 21
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, planning, rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 21
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.11633
diff --git a/papers/items/2026-2605-11882-on-policy-self-evolution-via-failure-trajectories-for-agentic-safety-alignment.md b/papers/items/2026-2605-11882-on-policy-self-evolution-via-failure-trajectories-for-agentic-safety-alignment.md
new file mode 100644
index 0000000..b5fc85b
--- /dev/null
+++ b/papers/items/2026-2605-11882-on-policy-self-evolution-via-failure-trajectories-for-agentic-safety-alignment.md
@@ -0,0 +1,60 @@
+# Paper: On-Policy Self-Evolution via Failure Trajectories for Agentic Safety Alignment
+
+---
+type: paper
+title: On-Policy Self-Evolution via Failure Trajectories for Agentic Safety Alignment
+authors: Bo Yin, Qi Li, Xinchao Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.11882
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-12
+updated_at: 2026-05-12
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-safety, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.11882
diff --git a/papers/items/2026-2605-11928-when-simulation-lies-a-sim-to-real-benchmark-and-domain-randomized-rl-recipe-for.md b/papers/items/2026-2605-11928-when-simulation-lies-a-sim-to-real-benchmark-and-domain-randomized-rl-recipe-for.md
new file mode 100644
index 0000000..44945d3
--- /dev/null
+++ b/papers/items/2026-2605-11928-when-simulation-lies-a-sim-to-real-benchmark-and-domain-randomized-rl-recipe-for.md
@@ -0,0 +1,60 @@
+# Paper: When Simulation Lies: A Sim-to-Real Benchmark and Domain-Randomized RL Recipe for Tool-Use Agents
+
+---
+type: paper
+title: "When Simulation Lies: A Sim-to-Real Benchmark and Domain-Randomized RL Recipe for Tool-Use Agents"
+authors: Xiaolin Zhou, Aojie Yuan, Zheng Luo, Zipeng Ling, Xixiao Pan, Yicheng Gao, Haiyue Zhang, Jiate Li, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.11928
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-12
+updated_at: 2026-05-12
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: function-calling, language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling, language-agent
+- inferred topics: agent-evaluation, tool-use, world-model
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.11928
diff --git a/papers/items/2026-2605-11946-counterfactual-trace-auditing-of-llm-agent-skills.md b/papers/items/2026-2605-11946-counterfactual-trace-auditing-of-llm-agent-skills.md
new file mode 100644
index 0000000..b48f819
--- /dev/null
+++ b/papers/items/2026-2605-11946-counterfactual-trace-auditing-of-llm-agent-skills.md
@@ -0,0 +1,61 @@
+# Paper: Counterfactual Trace Auditing of LLM Agent Skills
+
+---
+type: paper
+title: Counterfactual Trace Auditing of LLM Agent Skills
+authors: Xiaolin Zhou, Jinbo Liu, Li Li, Ryan A. Rossi, Xiyang Hu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.11946
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-12
+updated_at: 2026-05-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - planning
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, coding-agent, planning, rag
+- arXiv categories: cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.11946
diff --git a/papers/items/2026-2605-12015-skillsafetybench-evaluating-agent-safety-under-skill-facing-attack-surfaces.md b/papers/items/2026-2605-12015-skillsafetybench-evaluating-agent-safety-under-skill-facing-attack-surfaces.md
new file mode 100644
index 0000000..bf9c70d
--- /dev/null
+++ b/papers/items/2026-2605-12015-skillsafetybench-evaluating-agent-safety-under-skill-facing-attack-surfaces.md
@@ -0,0 +1,68 @@
+# Paper: SkillSafetyBench: Evaluating Agent Safety under Skill-Facing Attack Surfaces
+
+---
+type: paper
+title: "SkillSafetyBench: Evaluating Agent Safety under Skill-Facing Attack Surfaces"
+authors: Chang Jin, An Wang, Zeming Wei, Kai Wang, Biaojie Zeng, Qiaosheng Zhang, Chao Yang, Jingjing Qu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.12015
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-12
+updated_at: 2026-05-27
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - memory
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+  - cs.CL
+  - cs.LG
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, computer-use, memory, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.CR, cs.AI, cs.CL, cs.LG, cs.MA
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.12015
diff --git a/papers/items/2026-2605-12061-sage-a-self-evolving-agentic-graph-memory-engine-for-structure-aware-associative.md b/papers/items/2026-2605-12061-sage-a-self-evolving-agentic-graph-memory-engine-for-structure-aware-associative.md
new file mode 100644
index 0000000..4f8ed94
--- /dev/null
+++ b/papers/items/2026-2605-12061-sage-a-self-evolving-agentic-graph-memory-engine-for-structure-aware-associative.md
@@ -0,0 +1,62 @@
+# Paper: SAGE: A Self-Evolving Agentic Graph-Memory Engine for Structure-Aware Associative Memory
+
+---
+type: paper
+title: "SAGE: A Self-Evolving Agentic Graph-Memory Engine for Structure-Aware Associative Memory"
+authors: Juntong Wang, Haoyue Zhao, guanghui Pan, Xiyuan Wang, Yanbo Wang, Qiyan Deng, Muhan Zhang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.12061
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-12
+updated_at: 2026-05-12
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, memory, planning, rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.12061
diff --git a/papers/items/2026-2605-12260-prism-pareto-efficient-retrieval-over-intent-aware-structured-memory-for-long-ho.md b/papers/items/2026-2605-12260-prism-pareto-efficient-retrieval-over-intent-aware-structured-memory-for-long-ho.md
new file mode 100644
index 0000000..05944ee
--- /dev/null
+++ b/papers/items/2026-2605-12260-prism-pareto-efficient-retrieval-over-intent-aware-structured-memory-for-long-ho.md
@@ -0,0 +1,62 @@
+# Paper: PRISM: Pareto-Efficient Retrieval over Intent-Aware Structured Memory for Long-Horizon Agents
+
+---
+type: paper
+title: "PRISM: Pareto-Efficient Retrieval over Intent-Aware Structured Memory for Long-Horizon Agents"
+authors: Jingyi Peng, Zhongwei Wan, Weiting Liu, Qiuzhuang Sun
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.12260
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-12
+updated_at: 2026-05-22
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, memory, planning, rag, tool-use
+- arXiv categories: cs.CL
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.12260
diff --git a/papers/items/2026-2605-13481-personalai-2-0-enhancing-knowledge-graph-traversal-retrieval-with-planning-mecha.md b/papers/items/2026-2605-13481-personalai-2-0-enhancing-knowledge-graph-traversal-retrieval-with-planning-mecha.md
new file mode 100644
index 0000000..2023c24
--- /dev/null
+++ b/papers/items/2026-2605-13481-personalai-2-0-enhancing-knowledge-graph-traversal-retrieval-with-planning-mecha.md
@@ -0,0 +1,62 @@
+# Paper: PersonalAI 2.0: Enhancing knowledge graph traversal/retrieval with planning mechanism for Personalized LLM Agents
+
+---
+type: paper
+title: "PersonalAI 2.0: Enhancing knowledge graph traversal/retrieval with planning mechanism for Personalized LLM Agents"
+authors: Mikhail Menschikov, Matvey Iskornev, Alexander Kharitonov, Alina Bogdanova, Mikhail Belkin, Ekaterina Lisitsyna, Artyom Sosedka, Victoria Dochkina, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.13481
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-13
+updated_at: 2026-05-13
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, computer-use, planning, rag, reasoning
+- arXiv categories: cs.CL
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.13481
diff --git a/papers/items/2026-2605-13542-realicu-do-llm-agents-understand-long-context-icu-data-a-benchmark-beyond-behavi.md b/papers/items/2026-2605-13542-realicu-do-llm-agents-understand-long-context-icu-data-a-benchmark-beyond-behavi.md
new file mode 100644
index 0000000..2614017
--- /dev/null
+++ b/papers/items/2026-2605-13542-realicu-do-llm-agents-understand-long-context-icu-data-a-benchmark-beyond-behavi.md
@@ -0,0 +1,66 @@
+# Paper: RealICU: Do LLM Agents Understand Long-Context ICU Data? A Benchmark Beyond Behavior Imitation
+
+---
+type: paper
+title: "RealICU: Do LLM Agents Understand Long-Context ICU Data? A Benchmark Beyond Behavior Imitation"
+authors: Chengzhi Shen, Weixiang Shen, Tobias Susetzky, Chen, Chen, Jun Li, Yuyuan Liu, Xuepeng Zhang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.13542
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-13
+updated_at: 2026-05-13
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+  - cs.LG
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, agent-safety, memory, planning, reasoning, tool-use
+- arXiv categories: cs.AI, cs.CL, cs.LG, cs.MA
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.13542
diff --git a/papers/items/2026-2605-13618-openaaas-an-open-agent-as-a-service-framework-for-distributed-materials-informat.md b/papers/items/2026-2605-13618-openaaas-an-open-agent-as-a-service-framework-for-distributed-materials-informat.md
new file mode 100644
index 0000000..2c29cd0
--- /dev/null
+++ b/papers/items/2026-2605-13618-openaaas-an-open-agent-as-a-service-framework-for-distributed-materials-informat.md
@@ -0,0 +1,64 @@
+# Paper: OpenAaaS: An Open Agent-as-a-Service Framework for Distributed Materials-Informatics Research
+
+---
+type: paper
+title: "OpenAaaS: An Open Agent-as-a-Service Framework for Distributed Materials-Informatics Research"
+authors: Peng Kang, Bixuan Li, Xiaoya Huang, Shuo Shi, Weiqiao Zhou, Zhen Li, Yu Liu, Lei Zheng
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.13618
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-13
+updated_at: 2026-05-13
+status: queued
+relevance: high
+topics:
+  - memory
+  - multi-agent
+  - planning
+  - rag
+  - reasoning
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cond-mat.mtrl-sci
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: memory, multi-agent, planning, rag, reasoning, workflow-agent
+- arXiv categories: cond-mat.mtrl-sci, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.13618
diff --git a/papers/items/2026-2605-13716-skillops-managing-llm-agent-skill-libraries-as-self-maintaining-software-ecosyst.md b/papers/items/2026-2605-13716-skillops-managing-llm-agent-skill-libraries-as-self-maintaining-software-ecosyst.md
new file mode 100644
index 0000000..c86bf2f
--- /dev/null
+++ b/papers/items/2026-2605-13716-skillops-managing-llm-agent-skill-libraries-as-self-maintaining-software-ecosyst.md
@@ -0,0 +1,62 @@
+# Paper: SkillOps: Managing LLM Agent Skill Libraries as Self-Maintaining Software Ecosystems
+
+---
+type: paper
+title: "SkillOps: Managing LLM Agent Skill Libraries as Self-Maintaining Software Ecosystems"
+authors: Hongji Pu, Xinyuan Song, Liang Zhao
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.13716
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-13
+updated_at: 2026-05-13
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, agent-safety, planning, rag
+- arXiv categories: cs.SE, cs.MA
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.13716
diff --git a/papers/items/2026-2605-14126-reinforcement-learning-for-tool-calling-agents-in-fast-healthcare-interoperabili.md b/papers/items/2026-2605-14126-reinforcement-learning-for-tool-calling-agents-in-fast-healthcare-interoperabili.md
new file mode 100644
index 0000000..2a2d025
--- /dev/null
+++ b/papers/items/2026-2605-14126-reinforcement-learning-for-tool-calling-agents-in-fast-healthcare-interoperabili.md
@@ -0,0 +1,63 @@
+# Paper: Reinforcement Learning for Tool-Calling Agents in Fast Healthcare Interoperability Resources (FHIR)
+
+---
+type: paper
+title: Reinforcement Learning for Tool-Calling Agents in Fast Healthcare Interoperability Resources (FHIR)
+authors: Marius S. Knorr, Robert Müller, Jan P. Bremer, Nils Schweingruber
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.14126
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-13
+updated_at: 2026-05-13
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, planning, rag, reasoning, tool-use
+- arXiv categories: cs.LG, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.14126
diff --git a/papers/items/2026-2605-14290-web-agents-should-adopt-the-plan-then-execute-paradigm.md b/papers/items/2026-2605-14290-web-agents-should-adopt-the-plan-then-execute-paradigm.md
new file mode 100644
index 0000000..71b4521
--- /dev/null
+++ b/papers/items/2026-2605-14290-web-agents-should-adopt-the-plan-then-execute-paradigm.md
@@ -0,0 +1,65 @@
+# Paper: Web Agents Should Adopt the Plan-Then-Execute Paradigm
+
+---
+type: paper
+title: Web Agents Should Adopt the Plan-Then-Execute Paradigm
+authors: Julien Piet, Annabella Chow, Yiwei Hou, Muxi Lyu, Sylvie Venuto, Jinhao Zhu, Raluca Ada Popa, David Wagner
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.14290
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-14
+updated_at: 2026-05-14
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+  - cs.CL
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, agent-safety, computer-use, planning, tool-use
+- arXiv categories: cs.CR, cs.AI, cs.CL, cs.SE
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.14290
diff --git a/papers/items/2026-2605-14322-are-agents-ready-to-teach-a-multi-stage-benchmark-for-real-world-teaching-workfl.md b/papers/items/2026-2605-14322-are-agents-ready-to-teach-a-multi-stage-benchmark-for-real-world-teaching-workfl.md
new file mode 100644
index 0000000..bccc2fc
--- /dev/null
+++ b/papers/items/2026-2605-14322-are-agents-ready-to-teach-a-multi-stage-benchmark-for-real-world-teaching-workfl.md
@@ -0,0 +1,60 @@
+# Paper: Are Agents Ready to Teach? A Multi-Stage Benchmark for Real-World Teaching Workflows
+
+---
+type: paper
+title: Are Agents Ready to Teach? A Multi-Stage Benchmark for Real-World Teaching Workflows
+authors: Zixin Chen, Peng Liu, Rui Sheng, Haobo Li, Jianhong Tu, Xiaodong Deng, Kashun Shum, Dayiheng Liu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.14322
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-14
+updated_at: 2026-05-20
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, tool-use, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.14322
diff --git a/papers/items/2026-2605-14421-memlineage-lineage-guided-enforcement-for-llm-agent-memory.md b/papers/items/2026-2605-14421-memlineage-lineage-guided-enforcement-for-llm-agent-memory.md
new file mode 100644
index 0000000..618a86b
--- /dev/null
+++ b/papers/items/2026-2605-14421-memlineage-lineage-guided-enforcement-for-llm-agent-memory.md
@@ -0,0 +1,62 @@
+# Paper: MemLineage: Lineage-Guided Enforcement for LLM Agent Memory
+
+---
+type: paper
+title: "MemLineage: Lineage-Guided Enforcement for LLM Agent Memory"
+authors: Ciyan Ouyang, Rui Hou
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.14421
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-14
+updated_at: 2026-05-14
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, computer-use, memory, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.14421
diff --git a/papers/items/2026-2605-14460-exploiting-llm-agent-supply-chains-via-payload-less-skills.md b/papers/items/2026-2605-14460-exploiting-llm-agent-supply-chains-via-payload-less-skills.md
new file mode 100644
index 0000000..f5a4d2f
--- /dev/null
+++ b/papers/items/2026-2605-14460-exploiting-llm-agent-supply-chains-via-payload-less-skills.md
@@ -0,0 +1,62 @@
+# Paper: Exploiting LLM Agent Supply Chains via Payload-less Skills
+
+---
+type: paper
+title: Exploiting LLM Agent Supply Chains via Payload-less Skills
+authors: Xinyu Liu, Yukai Zhao, Xing Hu, Xin Xia
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.14460
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-14
+updated_at: 2026-05-14
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, agent-safety, coding-agent, tool-use
+- arXiv categories: cs.CR, cs.SE
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.14460
diff --git a/papers/items/2026-2605-14498-groupmembench-benchmarking-llm-agent-memory-in-multi-party-conversations.md b/papers/items/2026-2605-14498-groupmembench-benchmarking-llm-agent-memory-in-multi-party-conversations.md
new file mode 100644
index 0000000..b3283f4
--- /dev/null
+++ b/papers/items/2026-2605-14498-groupmembench-benchmarking-llm-agent-memory-in-multi-party-conversations.md
@@ -0,0 +1,62 @@
+# Paper: GroupMemBench: Benchmarking LLM Agent Memory in Multi-Party Conversations
+
+---
+type: paper
+title: "GroupMemBench: Benchmarking LLM Agent Memory in Multi-Party Conversations"
+authors: Jingbo Yang, Kwei-Herng Lai, Xiaowen Wang, Shiyu Chang, Yaar Harari, Evgeniy Gabrilovich
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.14498
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-14
+updated_at: 2026-05-16
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 21
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, computer-use, memory, rag, reasoning
+- arXiv categories: cs.CL
+- collection score: 21
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.14498
diff --git a/papers/items/2026-2605-14527-lang2mlip-end-to-end-language-to-machine-learning-interatomic-potential-developm.md b/papers/items/2026-2605-14527-lang2mlip-end-to-end-language-to-machine-learning-interatomic-potential-developm.md
new file mode 100644
index 0000000..da57fe0
--- /dev/null
+++ b/papers/items/2026-2605-14527-lang2mlip-end-to-end-language-to-machine-learning-interatomic-potential-developm.md
@@ -0,0 +1,64 @@
+# Paper: Lang2MLIP: End-to-End Language-to-Machine Learning Interatomic Potential Development with Autonomous Agentic Workflows
+
+---
+type: paper
+title: "Lang2MLIP: End-to-End Language-to-Machine Learning Interatomic Potential Development with Autonomous Agentic Workflows"
+authors: Wenwen Li, Yuki Orimo, Nontawat Charoenphakdee
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.14527
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-14
+updated_at: 2026-05-14
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - tool-use
+  - workflow-agent
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cond-mat.mtrl-sci
+  - physics.comp-ph
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, multi-agent, tool-use, workflow-agent, world-model
+- arXiv categories: cs.LG, cond-mat.mtrl-sci, physics.comp-ph
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.14527
diff --git a/papers/items/2026-2605-14892-beyond-individual-intelligence-surveying-collaboration-failure-attribution-and-s.md b/papers/items/2026-2605-14892-beyond-individual-intelligence-surveying-collaboration-failure-attribution-and-s.md
new file mode 100644
index 0000000..ad1ee88
--- /dev/null
+++ b/papers/items/2026-2605-14892-beyond-individual-intelligence-surveying-collaboration-failure-attribution-and-s.md
@@ -0,0 +1,63 @@
+# Paper: Beyond Individual Intelligence: Surveying Collaboration, Failure Attribution, and Self-Evolution in LLM-based Multi-Agent Systems
+
+---
+type: paper
+title: "Beyond Individual Intelligence: Surveying Collaboration, Failure Attribution, and Self-Evolution in LLM-based Multi-Agent Systems"
+authors: Shihao Qi, Jie Ma, Rui Xing, Wei Guo, Xiao Huang, Zhitao Gao, Jianhao Deng, Jun Liu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.14892
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-14
+updated_at: 2026-05-15
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - multi-agent
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-safety, multi-agent, planning, rag, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.14892
diff --git a/papers/items/2026-2605-14906-memlens-benchmarking-multimodal-long-term-memory-in-large-vision-language-models.md b/papers/items/2026-2605-14906-memlens-benchmarking-multimodal-long-term-memory-in-large-vision-language-models.md
new file mode 100644
index 0000000..e00145d
--- /dev/null
+++ b/papers/items/2026-2605-14906-memlens-benchmarking-multimodal-long-term-memory-in-large-vision-language-models.md
@@ -0,0 +1,62 @@
+# Paper: MemLens: Benchmarking Multimodal Long-Term Memory in Large Vision-Language Models
+
+---
+type: paper
+title: "MemLens: Benchmarking Multimodal Long-Term Memory in Large Vision-Language Models"
+authors: Xiyu Ren, Zhaowei Wang, Yiming Du, Zhongwei Xie, Chi Liu, Xinlin Yang, Haoyue Feng, Wenjun Pan, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.14906
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-14
+updated_at: 2026-05-14
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, rag, reasoning, tool-use
+- arXiv categories: cs.CV
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.14906
diff --git a/papers/items/2026-2605-14932-toward-securing-ai-agents-like-operating-systems.md b/papers/items/2026-2605-14932-toward-securing-ai-agents-like-operating-systems.md
new file mode 100644
index 0000000..ddae9c5
--- /dev/null
+++ b/papers/items/2026-2605-14932-toward-securing-ai-agents-like-operating-systems.md
@@ -0,0 +1,61 @@
+# Paper: Toward Securing AI Agents Like Operating Systems
+
+---
+type: paper
+title: Toward Securing AI Agents Like Operating Systems
+authors: Lukas Pirch, Micha Horlboge, Patrick Großmann, Syeda Mahnur Asif, Klim Kireev, Thorsten Holz, Konrad Rieck
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.14932
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-14
+updated_at: 2026-05-14
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, agent-safety, computer-use, tool-use
+- arXiv categories: cs.CR
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.14932
diff --git a/papers/items/2026-2605-15040-orchard-an-open-source-agentic-modeling-framework.md b/papers/items/2026-2605-15040-orchard-an-open-source-agentic-modeling-framework.md
new file mode 100644
index 0000000..1f1c5d0
--- /dev/null
+++ b/papers/items/2026-2605-15040-orchard-an-open-source-agentic-modeling-framework.md
@@ -0,0 +1,64 @@
+# Paper: Orchard: An Open-Source Agentic Modeling Framework
+
+---
+type: paper
+title: "Orchard: An Open-Source Agentic Modeling Framework"
+authors: Baolin Peng, Wenlin Yao, Qianhui Wu, Hao Cheng, Xiao Yu, Rui Yang, Tao Ge, Alessandro Sordoni, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.15040
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-14
+updated_at: 2026-05-21
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - computer-use
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, coding-agent, computer-use, planning, reasoning, tool-use
+- arXiv categories: cs.AI, cs.CL
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.15040
diff --git a/papers/items/2026-2605-15128-memeye-a-visual-centric-evaluation-framework-for-multimodal-agent-memory.md b/papers/items/2026-2605-15128-memeye-a-visual-centric-evaluation-framework-for-multimodal-agent-memory.md
new file mode 100644
index 0000000..4106b53
--- /dev/null
+++ b/papers/items/2026-2605-15128-memeye-a-visual-centric-evaluation-framework-for-multimodal-agent-memory.md
@@ -0,0 +1,63 @@
+# Paper: MemEye: A Visual-Centric Evaluation Framework for Multimodal Agent Memory
+
+---
+type: paper
+title: "MemEye: A Visual-Centric Evaluation Framework for Multimodal Agent Memory"
+authors: Minghao Guo, Qingyue Jiao, Zeru Shi, Yihao Quan, Boxuan Zhang, Danrui Li, Liwei Che, Wujiang Xu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.15128
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-14
+updated_at: 2026-05-14
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+  - cs.CL
+  - cs.IR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, reasoning, tool-use
+- arXiv categories: cs.CV, cs.CL, cs.IR
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.15128
diff --git a/papers/items/2026-2605-15206-agentstop-terminating-local-ai-agents-early-to-save-energy-in-consumer-devices.md b/papers/items/2026-2605-15206-agentstop-terminating-local-ai-agents-early-to-save-energy-in-consumer-devices.md
new file mode 100644
index 0000000..3cc7f8c
--- /dev/null
+++ b/papers/items/2026-2605-15206-agentstop-terminating-local-ai-agents-early-to-save-energy-in-consumer-devices.md
@@ -0,0 +1,65 @@
+# Paper: AgentStop: Terminating Local AI Agents Early to Save Energy in Consumer Devices
+
+---
+type: paper
+title: "AgentStop: Terminating Local AI Agents Early to Save Energy in Consumer Devices"
+authors: Dzung Pham, Kleomenis Katevas, Ali Shahin Shamsabadi, Hamed Haddadi
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.15206
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-01
+updated_at: 2026-05-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+  - cs.DC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, coding-agent, rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.LG, cs.AI, cs.DC
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.15206
diff --git a/papers/items/2026-2605-15625-colpackagent-agent-skill-guided-hard-particle-monte-carlo-workflows-for-colloida.md b/papers/items/2026-2605-15625-colpackagent-agent-skill-guided-hard-particle-monte-carlo-workflows-for-colloida.md
new file mode 100644
index 0000000..82b9bee
--- /dev/null
+++ b/papers/items/2026-2605-15625-colpackagent-agent-skill-guided-hard-particle-monte-carlo-workflows-for-colloida.md
@@ -0,0 +1,64 @@
+# Paper: ColPackAgent: Agent-Skill-Guided Hard-Particle Monte Carlo Workflows for Colloidal Packing
+
+---
+type: paper
+title: "ColPackAgent: Agent-Skill-Guided Hard-Particle Monte Carlo Workflows for Colloidal Packing"
+authors: Lijie Ding, Changwoo Do
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.15625
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-15
+updated_at: 2026-05-15
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - tool-use
+  - workflow-agent
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cond-mat.soft
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, computer-use, planning, tool-use, workflow-agent, world-model
+- arXiv categories: cs.AI, cond-mat.soft
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.15625
diff --git a/papers/items/2026-2605-15701-h-mem-a-novel-memory-mechanism-for-evolving-and-retrieving-agent-memory-via-a-hy.md b/papers/items/2026-2605-15701-h-mem-a-novel-memory-mechanism-for-evolving-and-retrieving-agent-memory-via-a-hy.md
new file mode 100644
index 0000000..9c96b08
--- /dev/null
+++ b/papers/items/2026-2605-15701-h-mem-a-novel-memory-mechanism-for-evolving-and-retrieving-agent-memory-via-a-hy.md
@@ -0,0 +1,61 @@
+# Paper: H-Mem: A Novel Memory Mechanism for Evolving and Retrieving Agent Memory via a Hybrid Structure
+
+---
+type: paper
+title: "H-Mem: A Novel Memory Mechanism for Evolving and Retrieving Agent Memory via a Hybrid Structure"
+authors: Jiawei Yu, Yixiang Fang, Xilin Liu, Yuchi Ma
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.15701
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-15
+updated_at: 2026-05-15
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, rag
+- arXiv categories: cs.CL, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.15701
diff --git a/papers/items/2026-2605-15710-smmbench-a-benchmark-for-source-distributed-multimodal-agent-memory.md b/papers/items/2026-2605-15710-smmbench-a-benchmark-for-source-distributed-multimodal-agent-memory.md
new file mode 100644
index 0000000..dcc3809
--- /dev/null
+++ b/papers/items/2026-2605-15710-smmbench-a-benchmark-for-source-distributed-multimodal-agent-memory.md
@@ -0,0 +1,62 @@
+# Paper: SMMBench: A Benchmark for Source-Distributed Multimodal Agent Memory
+
+---
+type: paper
+title: "SMMBench: A Benchmark for Source-Distributed Multimodal Agent Memory"
+authors: Huacan Chai, Yukai Wang, Yingxuan Yang, Dan Peng, Yuanyi Song, Zhihui Fu, Weiwen Liu, Jianghao Lin, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.15710
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-15
+updated_at: 2026-05-15
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, rag, reasoning, tool-use
+- arXiv categories: cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.15710
diff --git a/papers/items/2026-2605-15759-dimmem-dimensional-structuring-for-efficient-long-term-agent-memory.md b/papers/items/2026-2605-15759-dimmem-dimensional-structuring-for-efficient-long-term-agent-memory.md
new file mode 100644
index 0000000..ee3ee42
--- /dev/null
+++ b/papers/items/2026-2605-15759-dimmem-dimensional-structuring-for-efficient-long-term-agent-memory.md
@@ -0,0 +1,61 @@
+# Paper: DimMem: Dimensional Structuring for Efficient Long-Term Agent Memory
+
+---
+type: paper
+title: "DimMem: Dimensional Structuring for Efficient Long-Term Agent Memory"
+authors: Wentao Qiu, Haotian Hu, Fanyi Wang, Jinwei Kong, Yu Zhang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.15759
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-15
+updated_at: 2026-05-24
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, rag, tool-use
+- arXiv categories: cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.15759
diff --git a/papers/items/2026-2605-16233-forge-self-evolving-agent-memory-with-no-weight-updates-via-population-broadcast.md b/papers/items/2026-2605-16233-forge-self-evolving-agent-memory-with-no-weight-updates-via-population-broadcast.md
new file mode 100644
index 0000000..58deff7
--- /dev/null
+++ b/papers/items/2026-2605-16233-forge-self-evolving-agent-memory-with-no-weight-updates-via-population-broadcast.md
@@ -0,0 +1,65 @@
+# Paper: FORGE: Self-Evolving Agent Memory With No Weight Updates via Population Broadcast
+
+---
+type: paper
+title: "FORGE: Self-Evolving Agent Memory With No Weight Updates via Population Broadcast"
+authors: Igor Bogdanov, Chung-Horng Lung, Thomas Kunz, Jie Gao, Adrian Taylor, Marzia Zaman
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.16233
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-15
+updated_at: 2026-05-15
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+  - cs.LG
+  - cs.MA
+  - eess.SY
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, rag, reasoning
+- arXiv categories: cs.AI, cs.CL, cs.LG, cs.MA, eess.SY
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.16233
diff --git a/papers/items/2026-2605-16481-visual-agentic-memory-enabling-online-long-video-understanding-via-online-indexi.md b/papers/items/2026-2605-16481-visual-agentic-memory-enabling-online-long-video-understanding-via-online-indexi.md
new file mode 100644
index 0000000..09beba6
--- /dev/null
+++ b/papers/items/2026-2605-16481-visual-agentic-memory-enabling-online-long-video-understanding-via-online-indexi.md
@@ -0,0 +1,63 @@
+# Paper: Visual Agentic Memory: Enabling Online Long Video Understanding via Online Indexing, Hierarchical Memory, and Agentic Retrieval
+
+---
+type: paper
+title: "Visual Agentic Memory: Enabling Online Long Video Understanding via Online Indexing, Hierarchical Memory, and Agentic Retrieval"
+authors: Aiden Yiliu Li, Nels Numan, Anthony Steed
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.16481
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-15
+updated_at: 2026-05-15
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, planning, rag, reasoning
+- arXiv categories: cs.CV, cs.AI
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.16481
diff --git a/papers/items/2026-2605-16821-multi-paradigm-agent-interaction-in-practice-a-systematic-analysis-of-generator-.md b/papers/items/2026-2605-16821-multi-paradigm-agent-interaction-in-practice-a-systematic-analysis-of-generator-.md
new file mode 100644
index 0000000..024ca54
--- /dev/null
+++ b/papers/items/2026-2605-16821-multi-paradigm-agent-interaction-in-practice-a-systematic-analysis-of-generator-.md
@@ -0,0 +1,63 @@
+# Paper: Multi-Paradigm Agent Interaction in Practice:A Systematic Analysis of Generator-Evaluator, ReAct Loop,and Adversarial Evaluation in the buddyMe Framework
+
+---
+type: paper
+title: "Multi-Paradigm Agent Interaction in Practice:A Systematic Analysis of Generator-Evaluator, ReAct Loop,and Adversarial Evaluation in the buddyMe Framework"
+authors: Xiaohua Wang, Chao Han, Kai Yu, XiaoLiang Xu, Liang Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.16821
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-16
+updated_at: 2026-05-16
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - multi-agent
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, computer-use, memory, multi-agent, planning, tool-use
+- arXiv categories: cs.AI
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.16821
diff --git a/papers/items/2026-2605-17075-a-red-teaming-framework-for-evaluating-robustness-of-ai-enabled-security-orchest.md b/papers/items/2026-2605-17075-a-red-teaming-framework-for-evaluating-robustness-of-ai-enabled-security-orchest.md
new file mode 100644
index 0000000..e463db9
--- /dev/null
+++ b/papers/items/2026-2605-17075-a-red-teaming-framework-for-evaluating-robustness-of-ai-enabled-security-orchest.md
@@ -0,0 +1,63 @@
+# Paper: A Red Teaming Framework for Evaluating Robustness of AI-enabled Security Orchestration, Automation, and Response Systems
+
+---
+type: paper
+title: A Red Teaming Framework for Evaluating Robustness of AI-enabled Security Orchestration, Automation, and Response Systems
+authors: Ayan Javeed Shaikh, Nathaniel D. Bastian, Ankit Shah
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.17075
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-16
+updated_at: 2026-05-16
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - tool-use
+  - workflow-agent
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, agent-safety, planning, tool-use, workflow-agent, world-model
+- arXiv categories: cs.CR
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.17075
diff --git a/papers/items/2026-2605-17348-taming-zombie-agents-a-markov-state-aware-framework-for-resilient-multi-agent-ev.md b/papers/items/2026-2605-17348-taming-zombie-agents-a-markov-state-aware-framework-for-resilient-multi-agent-ev.md
new file mode 100644
index 0000000..5505ded
--- /dev/null
+++ b/papers/items/2026-2605-17348-taming-zombie-agents-a-markov-state-aware-framework-for-resilient-multi-agent-ev.md
@@ -0,0 +1,61 @@
+# Paper: Taming "Zombie'' Agents: A Markov State-Aware Framework for Resilient Multi-Agent Evolution
+
+---
+type: paper
+title: "Taming \"Zombie'' Agents: A Markov State-Aware Framework for Resilient Multi-Agent Evolution"
+authors: Taolin Zhang, Pukun Zhao, Qizhou Chen, Jiuheng Wan, Chen Chen, Xiaofeng He, Chengyu Wang, Richang Hong
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.17348
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-17
+updated_at: 2026-05-17
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - memory
+  - multi-agent
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-safety, memory, multi-agent, reasoning
+- arXiv categories: cs.CL
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.17348
diff --git a/papers/items/2026-2605-17453-trust-no-tool-evaluating-and-defending-llm-agents-under-untrusted-tool-feedback.md b/papers/items/2026-2605-17453-trust-no-tool-evaluating-and-defending-llm-agents-under-untrusted-tool-feedback.md
new file mode 100644
index 0000000..57dc52d
--- /dev/null
+++ b/papers/items/2026-2605-17453-trust-no-tool-evaluating-and-defending-llm-agents-under-untrusted-tool-feedback.md
@@ -0,0 +1,61 @@
+# Paper: Trust No Tool: Evaluating and Defending LLM Agents under Untrusted Tool Feedback
+
+---
+type: paper
+title: "Trust No Tool: Evaluating and Defending LLM Agents under Untrusted Tool Feedback"
+authors: Lecheng Yan, Ruizhe Li, Xicheng Han, Wenxi Li, Binwu Wang, Longyue Wang, Chenyang Lyu, Guanhua Chen
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.17453
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-17
+updated_at: 2026-05-17
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, tool-use
+- arXiv categories: cs.CR, cs.CL
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.17453
diff --git a/papers/items/2026-2605-17625-episodic-semantic-memory-architecture-for-long-horizon-scientific-agents.md b/papers/items/2026-2605-17625-episodic-semantic-memory-architecture-for-long-horizon-scientific-agents.md
new file mode 100644
index 0000000..4789f2f
--- /dev/null
+++ b/papers/items/2026-2605-17625-episodic-semantic-memory-architecture-for-long-horizon-scientific-agents.md
@@ -0,0 +1,64 @@
+# Paper: Episodic-Semantic Memory Architecture for Long-Horizon Scientific Agents
+
+---
+type: paper
+title: Episodic-Semantic Memory Architecture for Long-Horizon Scientific Agents
+authors: Nikola Milosevic
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.17625
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-17
+updated_at: 2026-05-17
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, planning, rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.17625
diff --git a/papers/items/2026-2605-18284-commitdistill-a-lightweight-knowledge-centric-memory-layer-for-software-reposito.md b/papers/items/2026-2605-18284-commitdistill-a-lightweight-knowledge-centric-memory-layer-for-software-reposito.md
new file mode 100644
index 0000000..da3a429
--- /dev/null
+++ b/papers/items/2026-2605-18284-commitdistill-a-lightweight-knowledge-centric-memory-layer-for-software-reposito.md
@@ -0,0 +1,64 @@
+# Paper: CommitDistill: A Lightweight Knowledge-Centric Memory Layer for Software Repositories
+
+---
+type: paper
+title: "CommitDistill: A Lightweight Knowledge-Centric Memory Layer for Software Repositories"
+authors: Divya Chukkapalli, Thejesh Avula, Aditya Aggarwal, Harsimran Singh, Amith Tallanki
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.18284
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-18
+updated_at: 2026-05-18
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - computer-use
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, coding-agent, computer-use, memory, rag, tool-use
+- arXiv categories: cs.SE, cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.18284
diff --git a/papers/items/2026-2605-18502-the-distance-based-formation-controller-design-for-multi-agent-systems-in-port-h.md b/papers/items/2026-2605-18502-the-distance-based-formation-controller-design-for-multi-agent-systems-in-port-h.md
new file mode 100644
index 0000000..90d9c7b
--- /dev/null
+++ b/papers/items/2026-2605-18502-the-distance-based-formation-controller-design-for-multi-agent-systems-in-port-h.md
@@ -0,0 +1,62 @@
+# Paper: The distance-based formation controller design for multi-agent systems in port-Hamiltonian form
+
+---
+type: paper
+title: The distance-based formation controller design for multi-agent systems in port-Hamiltonian form
+authors: Jingyi Zhao, Yongxin Wu, Héctor García de Marina, Yuhu Wu, Yann Le Gorrec
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.18502
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-18
+updated_at: 2026-05-18
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - math.OC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, multi-agent, tool-use, world-model
+- arXiv categories: math.OC
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.18502
diff --git a/papers/items/2026-2605-18652-mementogui-learning-agentic-multimodal-memory-control-for-long-horizon-gui-agent.md b/papers/items/2026-2605-18652-mementogui-learning-agentic-multimodal-memory-control-for-long-horizon-gui-agent.md
new file mode 100644
index 0000000..2e342f0
--- /dev/null
+++ b/papers/items/2026-2605-18652-mementogui-learning-agentic-multimodal-memory-control-for-long-horizon-gui-agent.md
@@ -0,0 +1,63 @@
+# Paper: MementoGUI: Learning Agentic Multimodal Memory Control for Long-Horizon GUI Agents
+
+---
+type: paper
+title: "MementoGUI: Learning Agentic Multimodal Memory Control for Long-Horizon GUI Agents"
+authors: Ziyun Zeng, Hang Hua, Bocheng Zou, Mu Cai, Rogerio Feris, Jiebo Luo
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.18652
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-18
+updated_at: 2026-05-18
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 22
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, computer-use, memory, planning, rag, tool-use
+- arXiv categories: cs.CV
+- collection score: 22
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.18652
diff --git a/papers/items/2026-2605-18672-position-a-three-layer-probabilistic-assume-guarantee-architecture-is-structural.md b/papers/items/2026-2605-18672-position-a-three-layer-probabilistic-assume-guarantee-architecture-is-structural.md
new file mode 100644
index 0000000..35e89a7
--- /dev/null
+++ b/papers/items/2026-2605-18672-position-a-three-layer-probabilistic-assume-guarantee-architecture-is-structural.md
@@ -0,0 +1,60 @@
+# Paper: Position: A Three-Layer Probabilistic Assume-Guarantee Architecture Is Structurally Required for Safe LLM Agent Deployment
+
+---
+type: paper
+title: "Position: A Three-Layer Probabilistic Assume-Guarantee Architecture Is Structurally Required for Safe LLM Agent Deployment"
+authors: S. Bensalem, Y. Dong, M. Franzle, X. Huang, J. Kroger, D. Nickovic, A. Nouri, R. Roy, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.18672
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-18
+updated_at: 2026-05-18
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - multi-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-safety, multi-agent, tool-use
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.18672
diff --git a/papers/items/2026-2605-18930-oep-poisoning-self-evolving-llm-agents-via-locally-correct-but-non-transferable-.md b/papers/items/2026-2605-18930-oep-poisoning-self-evolving-llm-agents-via-locally-correct-but-non-transferable-.md
new file mode 100644
index 0000000..be25b2a
--- /dev/null
+++ b/papers/items/2026-2605-18930-oep-poisoning-self-evolving-llm-agents-via-locally-correct-but-non-transferable-.md
@@ -0,0 +1,63 @@
+# Paper: OEP: Poisoning Self-Evolving LLM Agents via Locally Correct but Non-Transferable Experiences
+
+---
+type: paper
+title: "OEP: Poisoning Self-Evolving LLM Agents via Locally Correct but Non-Transferable Experiences"
+authors: Kaixiang Wang, Jiong Lou, Zhaojiacheng Zhou, Jie Li
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.18930
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-18
+updated_at: 2026-05-18
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, agent-safety, memory, reasoning
+- arXiv categories: cs.CR, cs.AI, cs.LG
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.18930
diff --git a/papers/items/2026-2605-19604-formal-skill-programmable-runtime-skills-for-efficient-and-accurate-llm-agents.md b/papers/items/2026-2605-19604-formal-skill-programmable-runtime-skills-for-efficient-and-accurate-llm-agents.md
new file mode 100644
index 0000000..05c2627
--- /dev/null
+++ b/papers/items/2026-2605-19604-formal-skill-programmable-runtime-skills-for-efficient-and-accurate-llm-agents.md
@@ -0,0 +1,61 @@
+# Paper: Formal Skill: Programmable Runtime Skills for Efficient and Accurate LLM Agents
+
+---
+type: paper
+title: "Formal Skill: Programmable Runtime Skills for Efficient and Accurate LLM Agents"
+authors: Xi Zhang, Meijun Gao, Yuntian Zhao, Xinyu Tan, Yilun Yao, Feiyu Wang, Yanshu Wang, Dingsiyi, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.19604
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-19
+updated_at: 2026-05-19
+status: queued
+relevance: high
+topics:
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.19604
diff --git a/papers/items/2026-2605-19952-rethinking-how-to-remember-beyond-atomic-facts-in-lifelong-llm-agent-memory.md b/papers/items/2026-2605-19952-rethinking-how-to-remember-beyond-atomic-facts-in-lifelong-llm-agent-memory.md
new file mode 100644
index 0000000..426bb31
--- /dev/null
+++ b/papers/items/2026-2605-19952-rethinking-how-to-remember-beyond-atomic-facts-in-lifelong-llm-agent-memory.md
@@ -0,0 +1,62 @@
+# Paper: Rethinking How to Remember: Beyond Atomic Facts in Lifelong LLM Agent Memory
+
+---
+type: paper
+title: "Rethinking How to Remember: Beyond Atomic Facts in Lifelong LLM Agent Memory"
+authors: Jingwei Sun, Jianing Zhu, Jiangchao Yao, Tongliang Liu, Bo Han
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.19952
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-19
+updated_at: 2026-05-19
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, rag, reasoning, tool-use
+- arXiv categories: cs.CL
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.19952
diff --git a/papers/items/2026-2605-20306-wildroadbench-a-wild-aerial-road-damage-grounding-benchmark-for-vision-language-.md b/papers/items/2026-2605-20306-wildroadbench-a-wild-aerial-road-damage-grounding-benchmark-for-vision-language-.md
new file mode 100644
index 0000000..71e1345
--- /dev/null
+++ b/papers/items/2026-2605-20306-wildroadbench-a-wild-aerial-road-damage-grounding-benchmark-for-vision-language-.md
@@ -0,0 +1,63 @@
+# Paper: WildRoadBench: A Wild Aerial Road-Damage Grounding Benchmark for Vision-Language Models and Autonomous Agents
+
+---
+type: paper
+title: "WildRoadBench: A Wild Aerial Road-Damage Grounding Benchmark for Vision-Language Models and Autonomous Agents"
+authors: Bingnan Liu, Chenhang Cui, Rui Huang, Jiani Luo, Zhirong Shen, Tinghao Wang, Xiande Huang, Lingbei Meng, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.20306
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-19
+updated_at: 2026-06-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, coding-agent, rag, reasoning, tool-use
+- arXiv categories: cs.CV, cs.LG
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.20306
diff --git a/papers/items/2026-2605-20315-mix-quant-quantized-prefilling-precise-decoding-for-agentic-llms.md b/papers/items/2026-2605-20315-mix-quant-quantized-prefilling-precise-decoding-for-agentic-llms.md
new file mode 100644
index 0000000..378455e
--- /dev/null
+++ b/papers/items/2026-2605-20315-mix-quant-quantized-prefilling-precise-decoding-for-agentic-llms.md
@@ -0,0 +1,64 @@
+# Paper: Mix-Quant: Quantized Prefilling, Precise Decoding for Agentic LLMs
+
+---
+type: paper
+title: "Mix-Quant: Quantized Prefilling, Precise Decoding for Agentic LLMs"
+authors: Haiquan Lu, Zigeng Chen, Gongfan Fang, Xinyin Ma, Xinchao Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.20315
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-19
+updated_at: 2026-05-19
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - memory
+  - planning
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 21
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, coding-agent, memory, planning, rag, tool-use, workflow-agent
+- arXiv categories: cs.CL
+- collection score: 21
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.20315
diff --git a/papers/items/2026-2605-20616-auto-dreamer-learning-offline-memory-consolidation-for-language-agents.md b/papers/items/2026-2605-20616-auto-dreamer-learning-offline-memory-consolidation-for-language-agents.md
new file mode 100644
index 0000000..cf3e852
--- /dev/null
+++ b/papers/items/2026-2605-20616-auto-dreamer-learning-offline-memory-consolidation-for-language-agents.md
@@ -0,0 +1,61 @@
+# Paper: Auto-Dreamer: Learning Offline Memory Consolidation for Language Agents
+
+---
+type: paper
+title: "Auto-Dreamer: Learning Offline Memory Consolidation for Language Agents"
+authors: Chongrui Ye, Yuxiang Liu, Yu Wang, Haofei Yu, Yining Zhao, Ge Liu, Julian McAuley, Jiaxuan You
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.20616
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-20
+updated_at: 2026-05-20
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-memory, language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory, language-agent
+- inferred topics: agent-evaluation, memory, rag, tool-use
+- arXiv categories: cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.20616
diff --git a/papers/items/2026-2605-20833-memgym-a-long-horizon-memory-environment-for-llm-agents.md b/papers/items/2026-2605-20833-memgym-a-long-horizon-memory-environment-for-llm-agents.md
new file mode 100644
index 0000000..6bbe775
--- /dev/null
+++ b/papers/items/2026-2605-20833-memgym-a-long-horizon-memory-environment-for-llm-agents.md
@@ -0,0 +1,66 @@
+# Paper: MemGym: a Long-Horizon Memory Environment for LLM Agents
+
+---
+type: paper
+title: "MemGym: a Long-Horizon Memory Environment for LLM Agents"
+authors: Wujiang Xu, Yu Wang, Kai Mei, Kaiqu Liang, Zhenting Wang, Mingyu Jin, Han Zhang, Shi-Xiong Zhang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.20833
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-20
+updated_at: 2026-05-20
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - computer-use
+  - embodied-agent
+  - memory
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 26
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, coding-agent, computer-use, embodied-agent, memory, planning, rag, reasoning, tool-use
+- arXiv categories: cs.CL
+- collection score: 26
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.20833
diff --git a/papers/items/2026-2605-20874-governance-by-construction-for-generalist-agents.md b/papers/items/2026-2605-20874-governance-by-construction-for-generalist-agents.md
new file mode 100644
index 0000000..81431bb
--- /dev/null
+++ b/papers/items/2026-2605-20874-governance-by-construction-for-generalist-agents.md
@@ -0,0 +1,64 @@
+# Paper: Governance by Construction for Generalist Agents
+
+---
+type: paper
+title: Governance by Construction for Generalist Agents
+authors: Segev Shlomov, Iftach Shoham, Alon Oved, Ido Levy, Sami Marreed, Harold Ship, Offer Akrabi, Sergey Zeltyn, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.20874
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-20
+updated_at: 2026-05-20
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - computer-use
+  - planning
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-safety, computer-use, planning, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI, cs.SE
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.20874
diff --git a/papers/items/2026-2605-21240-apex-autonomous-policy-exploration-for-self-evolving-llm-agents.md b/papers/items/2026-2605-21240-apex-autonomous-policy-exploration-for-self-evolving-llm-agents.md
new file mode 100644
index 0000000..0cf88b5
--- /dev/null
+++ b/papers/items/2026-2605-21240-apex-autonomous-policy-exploration-for-self-evolving-llm-agents.md
@@ -0,0 +1,63 @@
+# Paper: APEX: Autonomous Policy Exploration for Self-Evolving LLM Agents
+
+---
+type: paper
+title: "APEX: Autonomous Policy Exploration for Self-Evolving LLM Agents"
+authors: Yibo Li, Jiashuo Yang, Zhi Zheng, Zhiyuan Hu, Yuan Sui, Shizun Wang, Yufei He, Bryan Hooi
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.21240
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-20
+updated_at: 2026-05-20
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, memory, planning, reasoning, tool-use
+- arXiv categories: cs.LG, cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.21240
diff --git a/papers/items/2026-2605-21740-smdd-bench-can-llms-solve-real-world-small-molecule-drug-design-tasks.md b/papers/items/2026-2605-21740-smdd-bench-can-llms-solve-real-world-small-molecule-drug-design-tasks.md
new file mode 100644
index 0000000..6c686fa
--- /dev/null
+++ b/papers/items/2026-2605-21740-smdd-bench-can-llms-solve-real-world-small-molecule-drug-design-tasks.md
@@ -0,0 +1,62 @@
+# Paper: SMDD-Bench: Can LLMs Solve Real-World Small Molecule Drug Design Tasks?
+
+---
+type: paper
+title: "SMDD-Bench: Can LLMs Solve Real-World Small Molecule Drug Design Tasks?"
+authors: Kevin Han, Renfei Zhang, Kathy Wei, Hamed Mahdavi, Niloofar Mireshghallah, Amir Barati Farimani
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.21740
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-20
+updated_at: 2026-05-24
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, planning, rag, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.21740
diff --git a/papers/items/2026-2605-22154-idlespec-exploiting-idle-time-via-speculative-planning-for-llm-agents.md b/papers/items/2026-2605-22154-idlespec-exploiting-idle-time-via-speculative-planning-for-llm-agents.md
new file mode 100644
index 0000000..c880a67
--- /dev/null
+++ b/papers/items/2026-2605-22154-idlespec-exploiting-idle-time-via-speculative-planning-for-llm-agents.md
@@ -0,0 +1,63 @@
+# Paper: IdleSpec: Exploiting Idle Time via Speculative Planning for LLM Agents
+
+---
+type: paper
+title: "IdleSpec: Exploiting Idle Time via Speculative Planning for LLM Agents"
+authors: Daewon Choi, Kyunghyun Park, Woomin Song, Saket Dingliwal, Sai Muralidhar Jayanthi, Jinwoo Shin, Aram Galstyan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.22154
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-21
+updated_at: 2026-05-21
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, computer-use, planning, rag, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.22154
diff --git a/papers/items/2026-2605-22321-benchmarking-autonomous-agents-against-temporal-spatial-and-semantic-evasions.md b/papers/items/2026-2605-22321-benchmarking-autonomous-agents-against-temporal-spatial-and-semantic-evasions.md
new file mode 100644
index 0000000..7093c62
--- /dev/null
+++ b/papers/items/2026-2605-22321-benchmarking-autonomous-agents-against-temporal-spatial-and-semantic-evasions.md
@@ -0,0 +1,64 @@
+# Paper: Benchmarking Autonomous Agents against Temporal, Spatial, and Semantic Evasions
+
+---
+type: paper
+title: Benchmarking Autonomous Agents against Temporal, Spatial, and Semantic Evasions
+authors: Jianan Ma, Xiaohu Du, Ruixiao Lin, Yaoxiang Bian, Jialuo Chen, Jingyi Wang, Xiaofang Yang, Shiwen Cui, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.22321
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-21
+updated_at: 2026-05-21
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, agent-safety, memory, rag, tool-use
+- arXiv categories: cs.CR, cs.AI, cs.SE
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.22321
diff --git a/papers/items/2026-2605-22643-boiling-the-frog-a-multi-turn-benchmark-for-agentic-safety.md b/papers/items/2026-2605-22643-boiling-the-frog-a-multi-turn-benchmark-for-agentic-safety.md
new file mode 100644
index 0000000..6536f70
--- /dev/null
+++ b/papers/items/2026-2605-22643-boiling-the-frog-a-multi-turn-benchmark-for-agentic-safety.md
@@ -0,0 +1,63 @@
+# Paper: Boiling the Frog: A Multi-Turn Benchmark for Agentic Safety
+
+---
+type: paper
+title: "Boiling the Frog: A Multi-Turn Benchmark for Agentic Safety"
+authors: Piercosma Bisconti, Matteo Prandi, Federico Pierucci, Federico Sartore, Enrico Panai, Laura Caroli, Yue Zhu, Adam Leon Smith, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.22643
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-21
+updated_at: 2026-05-22
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, memory, rag, tool-use, workflow-agent
+- arXiv categories: cs.CL
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.22643
diff --git a/papers/items/2026-2605-23067-what-training-data-teaches-rl-memory-agents-an-empirical-study-of-curriculum-eff.md b/papers/items/2026-2605-23067-what-training-data-teaches-rl-memory-agents-an-empirical-study-of-curriculum-eff.md
new file mode 100644
index 0000000..f0abc66
--- /dev/null
+++ b/papers/items/2026-2605-23067-what-training-data-teaches-rl-memory-agents-an-empirical-study-of-curriculum-eff.md
@@ -0,0 +1,60 @@
+# Paper: What Training Data Teaches RL Memory Agents: An Empirical Study of Curriculum Effects in Memory-Augmented QA
+
+---
+type: paper
+title: "What Training Data Teaches RL Memory Agents: An Empirical Study of Curriculum Effects in Memory-Augmented QA"
+authors: Xinjie He, Zhiyuan Lin, Su Liu, Jialun Wu, Qiyang Xie, Weikai Zhou, Shuai Xiao
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.23067
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-21
+updated_at: 2026-05-21
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, reasoning
+- arXiv categories: cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.23067
diff --git a/papers/items/2026-2605-23574-push-your-agent-measuring-and-enforcing-quantitative-goal-persistence-in-long-ho.md b/papers/items/2026-2605-23574-push-your-agent-measuring-and-enforcing-quantitative-goal-persistence-in-long-ho.md
new file mode 100644
index 0000000..6e5b25d
--- /dev/null
+++ b/papers/items/2026-2605-23574-push-your-agent-measuring-and-enforcing-quantitative-goal-persistence-in-long-ho.md
@@ -0,0 +1,64 @@
+# Paper: Push Your Agent: Measuring and Enforcing Quantitative Goal Persistence in Long-Horizon LLM Agents
+
+---
+type: paper
+title: "Push Your Agent: Measuring and Enforcing Quantitative Goal Persistence in Long-Horizon LLM Agents"
+authors: Yuandao Cai, Yuzhang Zhu, Liyou Gao, Wensheng Tang, Shengchao Qin
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.23574
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-22
+updated_at: 2026-05-22
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, coding-agent, planning, rag, reasoning, tool-use
+- arXiv categories: cs.LG, cs.SE
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.23574
diff --git a/papers/items/2026-2605-23636-rf-instrument-agent-rfia-empowering-rf-instruments-with-natural-language-underst.md b/papers/items/2026-2605-23636-rf-instrument-agent-rfia-empowering-rf-instruments-with-natural-language-underst.md
new file mode 100644
index 0000000..64277a6
--- /dev/null
+++ b/papers/items/2026-2605-23636-rf-instrument-agent-rfia-empowering-rf-instruments-with-natural-language-underst.md
@@ -0,0 +1,63 @@
+# Paper: RF Instrument Agent (RFIA): Empowering RF Instruments with Natural Language Understanding, Scheduling and Execution of Complex Tasks
+
+---
+type: paper
+title: "RF Instrument Agent (RFIA): Empowering RF Instruments with Natural Language Understanding, Scheduling and Execution of Complex Tasks"
+authors: Chunhui Li, Wei Fan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.23636
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-22
+updated_at: 2026-05-22
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - eess.SY
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, agent-safety, planning, rag, tool-use, workflow-agent
+- arXiv categories: eess.SY
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.23636
diff --git a/papers/items/2026-2605-23723-memaudit-post-hoc-auditing-of-poisoned-agent-memory-via-causal-attribution-and-s.md b/papers/items/2026-2605-23723-memaudit-post-hoc-auditing-of-poisoned-agent-memory-via-causal-attribution-and-s.md
new file mode 100644
index 0000000..163fcb0
--- /dev/null
+++ b/papers/items/2026-2605-23723-memaudit-post-hoc-auditing-of-poisoned-agent-memory-via-causal-attribution-and-s.md
@@ -0,0 +1,63 @@
+# Paper: MemAudit: Post-hoc Auditing of Poisoned Agent Memory via Causal Attribution and Structural Anomaly Detection
+
+---
+type: paper
+title: "MemAudit: Post-hoc Auditing of Poisoned Agent Memory via Causal Attribution and Structural Anomaly Detection"
+authors: Zhewen Tan, Yilun Yao, Huiyan Jin, Wenhan Yu, Guoan Wang, Mengyuan Fan, liang lu, Feng Liu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.23723
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-22
+updated_at: 2026-05-22
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, agent-safety, memory, planning, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.23723
diff --git a/papers/items/2026-2605-23899-from-raw-experience-to-skill-consumption-a-systematic-study-of-model-generated-a.md b/papers/items/2026-2605-23899-from-raw-experience-to-skill-consumption-a-systematic-study-of-model-generated-a.md
new file mode 100644
index 0000000..a33eed6
--- /dev/null
+++ b/papers/items/2026-2605-23899-from-raw-experience-to-skill-consumption-a-systematic-study-of-model-generated-a.md
@@ -0,0 +1,62 @@
+# Paper: From Raw Experience to Skill Consumption: A Systematic Study of Model-Generated Agent Skills
+
+---
+type: paper
+title: "From Raw Experience to Skill Consumption: A Systematic Study of Model-Generated Agent Skills"
+authors: Zisu Huang, Jingwen Xu, Yifan Yang, Ziyang Gong, Qihao Yang, Muzhao Tian, Xiaohua Wang, Changze Lv, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.23899
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-22
+updated_at: 2026-05-22
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - computer-use
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, coding-agent, computer-use, rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.23899
diff --git a/papers/items/2026-2605-23986-memforest-an-efficient-agent-memory-system-with-hierarchical-temporal-indexing.md b/papers/items/2026-2605-23986-memforest-an-efficient-agent-memory-system-with-hierarchical-temporal-indexing.md
new file mode 100644
index 0000000..63c3914
--- /dev/null
+++ b/papers/items/2026-2605-23986-memforest-an-efficient-agent-memory-system-with-hierarchical-temporal-indexing.md
@@ -0,0 +1,62 @@
+# Paper: MemForest: An Efficient Agent Memory System with Hierarchical Temporal Indexing
+
+---
+type: paper
+title: "MemForest: An Efficient Agent Memory System with Hierarchical Temporal Indexing"
+authors: Han Chen, Zining Zhang, Wenqi Pei, Bingsheng He, Ming Wu, Jason Zeng, Michael Heinrich, Wei Wu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.23986
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-16
+updated_at: 2026-05-16
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.DB
+  - cs.AI
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, tool-use
+- arXiv categories: cs.DB, cs.AI, cs.MA
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.23986
diff --git a/papers/items/2026-2605-24069-when-the-manual-lies-a-realistic-benchmark-to-evaluate-mcp-poisoning-attacks-for.md b/papers/items/2026-2605-24069-when-the-manual-lies-a-realistic-benchmark-to-evaluate-mcp-poisoning-attacks-for.md
new file mode 100644
index 0000000..42ba980
--- /dev/null
+++ b/papers/items/2026-2605-24069-when-the-manual-lies-a-realistic-benchmark-to-evaluate-mcp-poisoning-attacks-for.md
@@ -0,0 +1,62 @@
+# Paper: When the Manual Lies: A Realistic Benchmark to Evaluate MCP Poisoning Attacks for LLM Agents
+
+---
+type: paper
+title: "When the Manual Lies: A Realistic Benchmark to Evaluate MCP Poisoning Attacks for LLM Agents"
+authors: Shi Liu, Xuehai Tang, Xikang Yang, Liang Lin, Biyu Zhou, Wenjie Xiao, Wantao Liu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.24069
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-22
+updated_at: 2026-05-22
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, agent-safety, planning, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.24069
diff --git a/papers/items/2026-2605-24216-agent-tom-learning-to-monitor-autonomous-llm-agents-via-theory-of-mind-reasoning.md b/papers/items/2026-2605-24216-agent-tom-learning-to-monitor-autonomous-llm-agents-via-theory-of-mind-reasoning.md
new file mode 100644
index 0000000..e51e28a
--- /dev/null
+++ b/papers/items/2026-2605-24216-agent-tom-learning-to-monitor-autonomous-llm-agents-via-theory-of-mind-reasoning.md
@@ -0,0 +1,67 @@
+# Paper: Agent-ToM: Learning to Monitor Autonomous LLM Agents via Theory-of-Mind Reasoning
+
+---
+type: paper
+title: "Agent-ToM: Learning to Monitor Autonomous LLM Agents via Theory-of-Mind Reasoning"
+authors: Nesreen K. Ahmed, Nima Nafisi
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.24216
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-22
+updated_at: 2026-05-22
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - memory
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+  - cs.CL
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, agent-safety, computer-use, memory, planning, reasoning, tool-use
+- arXiv categories: cs.LG, cs.AI, cs.CL, cs.CR
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.24216
diff --git a/papers/items/2026-2605-24219-beyond-final-answers-auditing-trajectory-level-hallucinations-in-multi-agent-ind.md b/papers/items/2026-2605-24219-beyond-final-answers-auditing-trajectory-level-hallucinations-in-multi-agent-ind.md
new file mode 100644
index 0000000..a349412
--- /dev/null
+++ b/papers/items/2026-2605-24219-beyond-final-answers-auditing-trajectory-level-hallucinations-in-multi-agent-ind.md
@@ -0,0 +1,61 @@
+# Paper: Beyond Final Answers: Auditing Trajectory-Level Hallucinations in Multi-Agent Industrial Workflows
+
+---
+type: paper
+title: "Beyond Final Answers: Auditing Trajectory-Level Hallucinations in Multi-Agent Industrial Workflows"
+authors: Harshada Badave, Santosh Borse, Andrea Gomez, Harshitha Narahari, Sara Carter, Vishwa Bhatt, Aishani Rachakonda, Shuxin Lin, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.24219
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-22
+updated_at: 2026-05-26
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, multi-agent, tool-use, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.24219
diff --git a/papers/items/2026-2605-24220-polar-agentic-rl-on-any-harness-at-scale.md b/papers/items/2026-2605-24220-polar-agentic-rl-on-any-harness-at-scale.md
new file mode 100644
index 0000000..4b29ebc
--- /dev/null
+++ b/papers/items/2026-2605-24220-polar-agentic-rl-on-any-harness-at-scale.md
@@ -0,0 +1,61 @@
+# Paper: Polar: Agentic RL on Any Harness at Scale
+
+---
+type: paper
+title: "Polar: Agentic RL on Any Harness at Scale"
+authors: Binfeng Xu, Hao Zhang, Shaokun Zhang, Songyang Han, Mingjie Liu, Jian Hu, Shizhe Diao, Zhenghui Jin, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.24220
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-22
+updated_at: 2026-05-22
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - multi-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.DC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, coding-agent, multi-agent, tool-use
+- arXiv categories: cs.DC
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.24220
diff --git a/papers/items/2026-2605-24309-reframing-llm-agent-security-as-an-agent-human-interaction-problem.md b/papers/items/2026-2605-24309-reframing-llm-agent-security-as-an-agent-human-interaction-problem.md
new file mode 100644
index 0000000..f4873f7
--- /dev/null
+++ b/papers/items/2026-2605-24309-reframing-llm-agent-security-as-an-agent-human-interaction-problem.md
@@ -0,0 +1,60 @@
+# Paper: Reframing LLM Agent Security as an Agent-Human Interaction Problem
+
+---
+type: paper
+title: Reframing LLM Agent Security as an Agent-Human Interaction Problem
+authors: Peiran Wang, Ying Li, Yuan Tian
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.24309
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-23
+updated_at: 2026-05-23
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, tool-use
+- arXiv categories: cs.CR
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.24309
diff --git a/papers/items/2026-2605-24659-iterinject-indirect-prompt-injection-against-llm-agents-via-feedback-guided-iter.md b/papers/items/2026-2605-24659-iterinject-indirect-prompt-injection-against-llm-agents-via-feedback-guided-iter.md
new file mode 100644
index 0000000..1e7b845
--- /dev/null
+++ b/papers/items/2026-2605-24659-iterinject-indirect-prompt-injection-against-llm-agents-via-feedback-guided-iter.md
@@ -0,0 +1,62 @@
+# Paper: IterInject: Indirect Prompt Injection Against LLM Agents via Feedback-Guided Iterative Optimization
+
+---
+type: paper
+title: "IterInject: Indirect Prompt Injection Against LLM Agents via Feedback-Guided Iterative Optimization"
+authors: Zixuan Chen, Jiaxiang Chen, Li Luo, Ke Xu, Xiaoxiang Huang, Tanfeng Sun, Xinghao Jiang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.24659
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-23
+updated_at: 2026-05-23
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - coding-agent
+  - computer-use
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-safety, coding-agent, computer-use, planning, tool-use
+- arXiv categories: cs.LG
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.24659
diff --git a/papers/items/2026-2605-24812-core-code-collaborative-reinforcement-learning-for-code-generation.md b/papers/items/2026-2605-24812-core-code-collaborative-reinforcement-learning-for-code-generation.md
new file mode 100644
index 0000000..5466229
--- /dev/null
+++ b/papers/items/2026-2605-24812-core-code-collaborative-reinforcement-learning-for-code-generation.md
@@ -0,0 +1,64 @@
+# Paper: CoRe-Code: Collaborative Reinforcement Learning for Code Generation
+
+---
+type: paper
+title: "CoRe-Code: Collaborative Reinforcement Learning for Code Generation"
+authors: Zhihao Dou, Qinjian Zhao, Zhongwei Wan, Xiaoyu Xia, Sumon Biswas
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.24812
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-24
+updated_at: 2026-05-24
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - memory
+  - multi-agent
+  - planning
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, agent-safety, coding-agent, memory, multi-agent, planning, rag
+- arXiv categories: cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.24812
diff --git a/papers/items/2026-2605-25141-llm-agent-based-renewable-energy-forecasting-using-edge-and-iot-data-a-review-of.md b/papers/items/2026-2605-25141-llm-agent-based-renewable-energy-forecasting-using-edge-and-iot-data-a-review-of.md
new file mode 100644
index 0000000..c3a61e3
--- /dev/null
+++ b/papers/items/2026-2605-25141-llm-agent-based-renewable-energy-forecasting-using-edge-and-iot-data-a-review-of.md
@@ -0,0 +1,65 @@
+# Paper: LLM Agent Based Renewable Energy Forecasting Using Edge and IoT Data A Review of Solar Wind Weather and Grid Aware Decision Support
+
+---
+type: paper
+title: LLM Agent Based Renewable Energy Forecasting Using Edge and IoT Data A Review of Solar Wind Weather and Grid Aware Decision Support
+authors: Pavan Manjunath, Thomas Pruefer
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.25141
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-24
+updated_at: 2026-05-24
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, computer-use, planning, rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.CL, cs.AI
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.25141
diff --git a/papers/items/2026-2605-25200-grouptravelbench-benchmarking-llm-agents-on-multi-person-travel-planning.md b/papers/items/2026-2605-25200-grouptravelbench-benchmarking-llm-agents-on-multi-person-travel-planning.md
new file mode 100644
index 0000000..89221a6
--- /dev/null
+++ b/papers/items/2026-2605-25200-grouptravelbench-benchmarking-llm-agents-on-multi-person-travel-planning.md
@@ -0,0 +1,61 @@
+# Paper: GroupTravelBench: Benchmarking LLM Agents on Multi-Person Travel Planning
+
+---
+type: paper
+title: "GroupTravelBench: Benchmarking LLM Agents on Multi-Person Travel Planning"
+authors: Xiang Cheng, Yulan Hu, Lulu Zheng, Zheng Pan, Xin Li, Yong Liu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.25200
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-24
+updated_at: 2026-06-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, planning, reasoning, tool-use
+- arXiv categories: cs.CL
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.25200
diff --git a/papers/items/2026-2605-25310-tool-call-dependency-structure-is-linearly-decodable-in-llm-agent-residual-strea.md b/papers/items/2026-2605-25310-tool-call-dependency-structure-is-linearly-decodable-in-llm-agent-residual-strea.md
new file mode 100644
index 0000000..5fde601
--- /dev/null
+++ b/papers/items/2026-2605-25310-tool-call-dependency-structure-is-linearly-decodable-in-llm-agent-residual-strea.md
@@ -0,0 +1,60 @@
+# Paper: Tool-Call Dependency Structure is Linearly Decodable in LLM Agent Residual Streams
+
+---
+type: paper
+title: Tool-Call Dependency Structure is Linearly Decodable in LLM Agent Residual Streams
+authors: Tianda Sun, Dimitar Kazakov
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.25310
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-25
+updated_at: 2026-05-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, planning, tool-use
+- arXiv categories: cs.CL
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.25310
diff --git a/papers/items/2026-2605-25393-decision-making-with-lightweight-confidence-aware-language-model-for-autonomous-.md b/papers/items/2026-2605-25393-decision-making-with-lightweight-confidence-aware-language-model-for-autonomous-.md
new file mode 100644
index 0000000..2614908
--- /dev/null
+++ b/papers/items/2026-2605-25393-decision-making-with-lightweight-confidence-aware-language-model-for-autonomous-.md
@@ -0,0 +1,64 @@
+# Paper: Decision-Making with Lightweight Confidence-Aware Language Model for Autonomous Driving
+
+---
+type: paper
+title: Decision-Making with Lightweight Confidence-Aware Language Model for Autonomous Driving
+authors: Ruoyu Yao, Ruiguo Zhong, Pei Liu, Mingxing Peng, Rui Yang, Jun Ma
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.25393
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-25
+updated_at: 2026-05-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.RO
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, multi-agent, planning, rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.RO
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.25393
diff --git a/papers/items/2026-2605-25435-security-of-openclaw-agents-fundamentals-attacks-and-countermeasures.md b/papers/items/2026-2605-25435-security-of-openclaw-agents-fundamentals-attacks-and-countermeasures.md
new file mode 100644
index 0000000..02198f6
--- /dev/null
+++ b/papers/items/2026-2605-25435-security-of-openclaw-agents-fundamentals-attacks-and-countermeasures.md
@@ -0,0 +1,63 @@
+# Paper: Security of OpenClaw Agents: Fundamentals, Attacks, and Countermeasures
+
+---
+type: paper
+title: "Security of OpenClaw Agents: Fundamentals, Attacks, and Countermeasures"
+authors: Yuntao Wang, Jianle Ba, Han Liu, Yanghe Pan, Jintao Wei, Zhou Su, Tom H. Luan, Linkang Du
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.25435
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-25
+updated_at: 2026-05-25
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - computer-use
+  - memory
+  - multi-agent
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-safety, computer-use, memory, multi-agent, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.25435
diff --git a/papers/items/2026-2605-25920-can-llms-time-travel-enhancing-temporal-consistency-in-legal-agentic-search-thro.md b/papers/items/2026-2605-25920-can-llms-time-travel-enhancing-temporal-consistency-in-legal-agentic-search-thro.md
new file mode 100644
index 0000000..205fac8
--- /dev/null
+++ b/papers/items/2026-2605-25920-can-llms-time-travel-enhancing-temporal-consistency-in-legal-agentic-search-thro.md
@@ -0,0 +1,61 @@
+# Paper: Can LLMs Time Travel? Enhancing Temporal Consistency in Legal Agentic Search through Reinforcement Learning
+
+---
+type: paper
+title: Can LLMs Time Travel? Enhancing Temporal Consistency in Legal Agentic Search through Reinforcement Learning
+authors: Wei Fan, Yining Zhou, Mufan Zhang, Yanbing Weng, Yiran HU, Tianshi Zheng, Baixuan Xu, Chunyang Li, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.25920
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-25
+updated_at: 2026-05-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, rag, reasoning
+- arXiv categories: cs.CL, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.25920
diff --git a/papers/items/2026-2605-26165-tool-schema-compression-enables-agentic-rag-under-constrained-context-budgets.md b/papers/items/2026-2605-26165-tool-schema-compression-enables-agentic-rag-under-constrained-context-budgets.md
new file mode 100644
index 0000000..834c68d
--- /dev/null
+++ b/papers/items/2026-2605-26165-tool-schema-compression-enables-agentic-rag-under-constrained-context-budgets.md
@@ -0,0 +1,62 @@
+# Paper: Tool-Schema Compression Enables Agentic RAG Under Constrained Context Budgets
+
+---
+type: paper
+title: Tool-Schema Compression Enables Agentic RAG Under Constrained Context Budgets
+authors: Furkan Sakizli
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.26165
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-24
+updated_at: 2026-05-24
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, rag, tool-use
+- arXiv categories: cs.SE, cs.AI, cs.CL
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.26165
diff --git a/papers/items/2026-2605-26252-is-agent-memory-a-database-rethinking-data-foundations-for-long-term-ai-agent-me.md b/papers/items/2026-2605-26252-is-agent-memory-a-database-rethinking-data-foundations-for-long-term-ai-agent-me.md
new file mode 100644
index 0000000..c642aa2
--- /dev/null
+++ b/papers/items/2026-2605-26252-is-agent-memory-a-database-rethinking-data-foundations-for-long-term-ai-agent-me.md
@@ -0,0 +1,62 @@
+# Paper: Is Agent Memory a Database? Rethinking Data Foundations for Long-Term AI Agent Memory
+
+---
+type: paper
+title: Is Agent Memory a Database? Rethinking Data Foundations for Long-Term AI Agent Memory
+authors: Abdelghny Orogat, Essam Mansour
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.26252
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-25
+updated_at: 2026-05-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.DB
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, rag, tool-use
+- arXiv categories: cs.AI, cs.DB
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.26252
diff --git a/papers/items/2026-2605-26305-experiments-in-agentic-ai-for-science.md b/papers/items/2026-2605-26305-experiments-in-agentic-ai-for-science.md
new file mode 100644
index 0000000..54dece0
--- /dev/null
+++ b/papers/items/2026-2605-26305-experiments-in-agentic-ai-for-science.md
@@ -0,0 +1,63 @@
+# Paper: Experiments in Agentic AI for Science
+
+---
+type: paper
+title: Experiments in Agentic AI for Science
+authors: Judy Fox, Geoffrey Fox
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.26305
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-25
+updated_at: 2026-05-29
+status: queued
+relevance: high
+topics:
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - eess.SY
+  - hep-ph
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: autonomous-agent-llm, rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm, rag-agent
+- inferred topics: rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI, eess.SY, hep-ph
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.26305
diff --git a/papers/items/2026-2605-26497-aligning-provenance-with-authorization-a-dual-graph-defense-for-llm-agents.md b/papers/items/2026-2605-26497-aligning-provenance-with-authorization-a-dual-graph-defense-for-llm-agents.md
new file mode 100644
index 0000000..39c99bb
--- /dev/null
+++ b/papers/items/2026-2605-26497-aligning-provenance-with-authorization-a-dual-graph-defense-for-llm-agents.md
@@ -0,0 +1,60 @@
+# Paper: Aligning Provenance with Authorization: A Dual-Graph Defense for LLM Agents
+
+---
+type: paper
+title: "Aligning Provenance with Authorization: A Dual-Graph Defense for LLM Agents"
+authors: Peiran Wang, Ying Li, Yuan Tian
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.26497
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-26
+updated_at: 2026-05-26
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-safety, reasoning, tool-use
+- arXiv categories: cs.CR
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.26497
diff --git a/papers/items/2026-2605-26720-towards-feedback-to-plan-decisions-for-self-evolving-llm-agents-in-cuda-kernel-g.md b/papers/items/2026-2605-26720-towards-feedback-to-plan-decisions-for-self-evolving-llm-agents-in-cuda-kernel-g.md
new file mode 100644
index 0000000..60b1ddd
--- /dev/null
+++ b/papers/items/2026-2605-26720-towards-feedback-to-plan-decisions-for-self-evolving-llm-agents-in-cuda-kernel-g.md
@@ -0,0 +1,61 @@
+# Paper: Towards Feedback-to-Plan Decisions for Self-Evolving LLM Agents in CUDA Kernel Generation
+
+---
+type: paper
+title: Towards Feedback-to-Plan Decisions for Self-Evolving LLM Agents in CUDA Kernel Generation
+authors: Yee Hin Chong, Jiaming Wu, Youhui Zhang, Peng Qu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.26720
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-26
+updated_at: 2026-05-26
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, planning, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.26720
diff --git a/papers/items/2026-2605-26926-from-norms-to-indicators-n2i-rag-an-agentic-retrieval-augmented-generation-frame.md b/papers/items/2026-2605-26926-from-norms-to-indicators-n2i-rag-an-agentic-retrieval-augmented-generation-frame.md
new file mode 100644
index 0000000..416c24a
--- /dev/null
+++ b/papers/items/2026-2605-26926-from-norms-to-indicators-n2i-rag-an-agentic-retrieval-augmented-generation-frame.md
@@ -0,0 +1,61 @@
+# Paper: From Norms to Indicators (N2I-RAG): An Agentic Retrieval-Augmented Generation Framework for Legal Indicator Computation
+
+---
+type: paper
+title: "From Norms to Indicators (N2I-RAG): An Agentic Retrieval-Augmented Generation Framework for Legal Indicator Computation"
+authors: Youssef Al Mouatamid, Marie Bonnin, Jihad Zahir
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.26926
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-26
+updated_at: 2026-05-26
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, agent-safety, planning, rag
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.26926
diff --git a/papers/items/2026-2605-27123-rethinking-agentic-rag-toward-llm-driven-logical-retrieval-beyond-embeddings.md b/papers/items/2026-2605-27123-rethinking-agentic-rag-toward-llm-driven-logical-retrieval-beyond-embeddings.md
new file mode 100644
index 0000000..c9b69cd
--- /dev/null
+++ b/papers/items/2026-2605-27123-rethinking-agentic-rag-toward-llm-driven-logical-retrieval-beyond-embeddings.md
@@ -0,0 +1,60 @@
+# Paper: Rethinking Agentic RAG: Toward LLM-Driven Logical Retrieval Beyond Embeddings
+
+---
+type: paper
+title: "Rethinking Agentic RAG: Toward LLM-Driven Logical Retrieval Beyond Embeddings"
+authors: Yuqi Zeng, Qixiang Deng, Yulei Wan, Ruiquan Jiang, Xiaoqing Zheng, Xuanjing Huang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.27123
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-26
+updated_at: 2026-05-26
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.IR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, computer-use, rag
+- arXiv categories: cs.IR
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.27123
diff --git a/papers/items/2026-2605-27134-scaling-benchmarking-and-reasoning-of-vision-language-agents-for-mobile-gui-navi.md b/papers/items/2026-2605-27134-scaling-benchmarking-and-reasoning-of-vision-language-agents-for-mobile-gui-navi.md
new file mode 100644
index 0000000..1674411
--- /dev/null
+++ b/papers/items/2026-2605-27134-scaling-benchmarking-and-reasoning-of-vision-language-agents-for-mobile-gui-navi.md
@@ -0,0 +1,63 @@
+# Paper: Scaling, Benchmarking, and Reasoning of Vision-Language Agents for Mobile GUI Navigation
+
+---
+type: paper
+title: Scaling, Benchmarking, and Reasoning of Vision-Language Agents for Mobile GUI Navigation
+authors: Heng Qu, Yike Liu, Renren Jin, Wenzong Zhang, Pengzhi Gao, Wei Liu, Jian Luan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.27134
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-26
+updated_at: 2026-05-26
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - embodied-agent
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, computer-use, embodied-agent, rag, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.27134
diff --git a/papers/items/2026-2605-27240-enpmr-bench-benchmarking-proactive-memory-retrieval-for-emotional-support-agents.md b/papers/items/2026-2605-27240-enpmr-bench-benchmarking-proactive-memory-retrieval-for-emotional-support-agents.md
new file mode 100644
index 0000000..2a02d49
--- /dev/null
+++ b/papers/items/2026-2605-27240-enpmr-bench-benchmarking-proactive-memory-retrieval-for-emotional-support-agents.md
@@ -0,0 +1,62 @@
+# Paper: ENPMR-Bench: Benchmarking Proactive Memory Retrieval for Emotional Support Agents
+
+---
+type: paper
+title: "ENPMR-Bench: Benchmarking Proactive Memory Retrieval for Emotional Support Agents"
+authors: Xing Fu, Yulin Hu, Mengtong Ji, Haozhen Li, Yixin Sun, Weixiang Zhao, Yanyan Zhao, Bing Qin
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.27240
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-26
+updated_at: 2026-05-26
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, agent-safety, memory, rag, tool-use
+- arXiv categories: cs.CL
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.27240
diff --git a/papers/items/2026-2605-27333-finharness-an-inline-lifecycle-safety-harness-for-finance-llm-agents.md b/papers/items/2026-2605-27333-finharness-an-inline-lifecycle-safety-harness-for-finance-llm-agents.md
new file mode 100644
index 0000000..30f9464
--- /dev/null
+++ b/papers/items/2026-2605-27333-finharness-an-inline-lifecycle-safety-harness-for-finance-llm-agents.md
@@ -0,0 +1,62 @@
+# Paper: FinHarness: An Inline Lifecycle Safety Harness for Finance LLM Agents
+
+---
+type: paper
+title: "FinHarness: An Inline Lifecycle Safety Harness for Finance LLM Agents"
+authors: Haoxuan Jia, Yang Liu, Bin Chong, Yingguang Yang, Yancheng Chen, Jiayu Liang, Qian Li, Hanning Lu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.27333
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-26
+updated_at: 2026-05-26
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, agent-safety, planning, tool-use, workflow-agent
+- arXiv categories: cs.CL
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.27333
diff --git a/papers/items/2026-2605-27366-muse-autoskill-self-evolving-agents-via-skill-creation-memory-management-and-eva.md b/papers/items/2026-2605-27366-muse-autoskill-self-evolving-agents-via-skill-creation-memory-management-and-eva.md
new file mode 100644
index 0000000..624d49e
--- /dev/null
+++ b/papers/items/2026-2605-27366-muse-autoskill-self-evolving-agents-via-skill-creation-memory-management-and-eva.md
@@ -0,0 +1,62 @@
+# Paper: MUSE-Autoskill: Self-Evolving Agents via Skill Creation, Memory, Management, and Evaluation
+
+---
+type: paper
+title: "MUSE-Autoskill: Self-Evolving Agents via Skill Creation, Memory, Management, and Evaluation"
+authors: Huawei Lin, Peng Li, Jie Song, Fuxin Jiang, Tieying Zhang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.27366
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-26
+updated_at: 2026-07-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+  - cs.LG
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory
+- arXiv categories: cs.AI, cs.CL, cs.LG, cs.MA
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.27366
diff --git a/papers/items/2026-2605-27690-traces-proactive-safety-auditing-for-multi-turn-llm-agents-via-trajectory-state-.md b/papers/items/2026-2605-27690-traces-proactive-safety-auditing-for-multi-turn-llm-agents-via-trajectory-state-.md
new file mode 100644
index 0000000..cf14f11
--- /dev/null
+++ b/papers/items/2026-2605-27690-traces-proactive-safety-auditing-for-multi-turn-llm-agents-via-trajectory-state-.md
@@ -0,0 +1,63 @@
+# Paper: TRACES: Proactive Safety Auditing for Multi-Turn LLM Agents via Trajectory-State Modeling
+
+---
+type: paper
+title: "TRACES: Proactive Safety Auditing for Multi-Turn LLM Agents via Trajectory-State Modeling"
+authors: Jiaqian Li, Yanshu Li, Boxuan Zhang, Ruixiang Tang, Kuan-Hao Huang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.27690
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-26
+updated_at: 2026-05-26
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, computer-use, planning, tool-use
+- arXiv categories: cs.CL, cs.LG
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.27690
diff --git a/papers/items/2026-2605-27762-peam-parametric-embodied-agent-memory-through-contrastive-internalization-of-exp.md b/papers/items/2026-2605-27762-peam-parametric-embodied-agent-memory-through-contrastive-internalization-of-exp.md
new file mode 100644
index 0000000..d556259
--- /dev/null
+++ b/papers/items/2026-2605-27762-peam-parametric-embodied-agent-memory-through-contrastive-internalization-of-exp.md
@@ -0,0 +1,64 @@
+# Paper: PEAM: Parametric Embodied Agent Memory through Contrastive Internalization of Experience in Minecraft
+
+---
+type: paper
+title: "PEAM: Parametric Embodied Agent Memory through Contrastive Internalization of Experience in Minecraft"
+authors: Yuchen Guo, Junli Gong, Weicheng Wang, Hongmin Cai, Yiu-ming Cheung, Weifeng Su
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.27762
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-26
+updated_at: 2026-06-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - embodied-agent
+  - memory
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, embodied-agent, memory, planning, rag, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.27762
diff --git a/papers/items/2026-2605-27825-mrmmia-membership-inference-attacks-on-memory-in-chat-agents.md b/papers/items/2026-2605-27825-mrmmia-membership-inference-attacks-on-memory-in-chat-agents.md
new file mode 100644
index 0000000..8e274a3
--- /dev/null
+++ b/papers/items/2026-2605-27825-mrmmia-membership-inference-attacks-on-memory-in-chat-agents.md
@@ -0,0 +1,63 @@
+# Paper: MRMMIA: Membership Inference Attacks on Memory in Chat Agents
+
+---
+type: paper
+title: "MRMMIA: Membership Inference Attacks on Memory in Chat Agents"
+authors: Kai Chen, Yan Pang, Tianhao Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.27825
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-27
+updated_at: 2026-05-27
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, agent-safety, memory, rag, tool-use
+- arXiv categories: cs.CR, cs.LG
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.27825
diff --git a/papers/items/2026-2605-27935-do-agents-think-deeper-a-mechanistic-investigation-of-layer-wise-dynamics-in-seq.md b/papers/items/2026-2605-27935-do-agents-think-deeper-a-mechanistic-investigation-of-layer-wise-dynamics-in-seq.md
new file mode 100644
index 0000000..2912a9c
--- /dev/null
+++ b/papers/items/2026-2605-27935-do-agents-think-deeper-a-mechanistic-investigation-of-layer-wise-dynamics-in-seq.md
@@ -0,0 +1,60 @@
+# Paper: Do Agents Think Deeper? A Mechanistic Investigation of Layer-Wise Dynamics in Sequential Planning
+
+---
+type: paper
+title: Do Agents Think Deeper? A Mechanistic Investigation of Layer-Wise Dynamics in Sequential Planning
+authors: Zhenyu Cui, Xiangzhong Luo
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.27935
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-27
+updated_at: 2026-05-27
+status: queued
+relevance: high
+topics:
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: autonomous-agent-llm, planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm, planning-agent
+- inferred topics: planning, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.27935
diff --git a/papers/items/2026-2605-28046-memcog-from-memory-as-tool-to-memory-as-cognition-in-conversational-agents.md b/papers/items/2026-2605-28046-memcog-from-memory-as-tool-to-memory-as-cognition-in-conversational-agents.md
new file mode 100644
index 0000000..06275ea
--- /dev/null
+++ b/papers/items/2026-2605-28046-memcog-from-memory-as-tool-to-memory-as-cognition-in-conversational-agents.md
@@ -0,0 +1,64 @@
+# Paper: MemCog: From Memory-as-Tool to Memory-as-Cognition in Conversational Agents
+
+---
+type: paper
+title: "MemCog: From Memory-as-Tool to Memory-as-Cognition in Conversational Agents"
+authors: Zihan Li, Xingyu Fan, Feifei Li, Wenhui Que
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.28046
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-27
+updated_at: 2026-05-27
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - embodied-agent
+  - memory
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, embodied-agent, memory, rag, reasoning, tool-use
+- arXiv categories: cs.AI, cs.CL
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.28046
diff --git a/papers/items/2026-2605-28120-legalgraphrag-multi-agent-graph-retrieval-augmented-generation-for-reliable-lega.md b/papers/items/2026-2605-28120-legalgraphrag-multi-agent-graph-retrieval-augmented-generation-for-reliable-lega.md
new file mode 100644
index 0000000..cacea7a
--- /dev/null
+++ b/papers/items/2026-2605-28120-legalgraphrag-multi-agent-graph-retrieval-augmented-generation-for-reliable-lega.md
@@ -0,0 +1,64 @@
+# Paper: LegalGraphRAG: Multi-Agent Graph Retrieval-Augmented Generation for Reliable Legal Reasoning
+
+---
+type: paper
+title: "LegalGraphRAG: Multi-Agent Graph Retrieval-Augmented Generation for Reliable Legal Reasoning"
+authors: Zerui Chen, Qinggang Zhang, Zhishang Xiang, Zhimin Wei, Linfeng Gao, Xiao Huang, Zhihong Zhang, Jinsong Su
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.28120
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-27
+updated_at: 2026-05-27
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, multi-agent, rag, reasoning, tool-use
+- arXiv categories: cs.CL, cs.AI, cs.MA
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.28120
diff --git a/papers/items/2026-2605-28175-mixture-of-experts-knowledge-graph-retrieval-augmented-generation-for-multi-agen.md b/papers/items/2026-2605-28175-mixture-of-experts-knowledge-graph-retrieval-augmented-generation-for-multi-agen.md
new file mode 100644
index 0000000..d6b6f58
--- /dev/null
+++ b/papers/items/2026-2605-28175-mixture-of-experts-knowledge-graph-retrieval-augmented-generation-for-multi-agen.md
@@ -0,0 +1,61 @@
+# Paper: Mixture-of-Experts Knowledge Graph Retrieval-Augmented Generation for Multi-Agent LLM-based Recommendation
+
+---
+type: paper
+title: Mixture-of-Experts Knowledge Graph Retrieval-Augmented Generation for Multi-Agent LLM-based Recommendation
+authors: Shijie Wang, Chengyi Liu, Yujuan Ding, Shanru Lin, See-Kiong Ng, Xu Xin, Wenqi Fan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.28175
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-27
+updated_at: 2026-05-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.IR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, agent-safety, multi-agent, rag
+- arXiv categories: cs.IR
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.28175
diff --git a/papers/items/2026-2605-28424-skill0-5-joint-skill-internalization-and-utilization-for-out-of-distribution-gen.md b/papers/items/2026-2605-28424-skill0-5-joint-skill-internalization-and-utilization-for-out-of-distribution-gen.md
new file mode 100644
index 0000000..88ee003
--- /dev/null
+++ b/papers/items/2026-2605-28424-skill0-5-joint-skill-internalization-and-utilization-for-out-of-distribution-gen.md
@@ -0,0 +1,59 @@
+# Paper: Skill0.5: Joint Skill Internalization and Utilization for Out-of-Distribution Generalization in Agentic Reinforcement Learning
+
+---
+type: paper
+title: "Skill0.5: Joint Skill Internalization and Utilization for Out-of-Distribution Generalization in Agentic Reinforcement Learning"
+authors: Jiapeng Zhu, Jianxiang Yu, Yibo Zhao, Chengcheng Han, Qi Gu, Xunliang Cai, Xiang Li, Weining Qian
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.28424
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-27
+updated_at: 2026-05-27
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - memory
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-safety, memory
+- arXiv categories: cs.CL
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.28424
diff --git a/papers/items/2026-2605-28607-adaptive-multimodal-agents-based-framework-for-automatic-workflow-execution.md b/papers/items/2026-2605-28607-adaptive-multimodal-agents-based-framework-for-automatic-workflow-execution.md
new file mode 100644
index 0000000..b5f859f
--- /dev/null
+++ b/papers/items/2026-2605-28607-adaptive-multimodal-agents-based-framework-for-automatic-workflow-execution.md
@@ -0,0 +1,65 @@
+# Paper: Adaptive Multimodal Agents-Based Framework for Automatic Workflow Execution
+
+---
+type: paper
+title: Adaptive Multimodal Agents-Based Framework for Automatic Workflow Execution
+authors: Susanna Cifani, Mario Luca Bernardi, Marta Cimitile
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.28607
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-27
+updated_at: 2026-05-27
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - embodied-agent
+  - multi-agent
+  - planning
+  - rag
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, computer-use, embodied-agent, multi-agent, planning, rag, workflow-agent
+- arXiv categories: cs.AI, cs.CL
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.28607
diff --git a/papers/items/2026-2605-28617-lacuna-safe-agents-as-recursive-program-holes.md b/papers/items/2026-2605-28617-lacuna-safe-agents-as-recursive-program-holes.md
new file mode 100644
index 0000000..3a95878
--- /dev/null
+++ b/papers/items/2026-2605-28617-lacuna-safe-agents-as-recursive-program-holes.md
@@ -0,0 +1,63 @@
+# Paper: LACUNA: Safe Agents as Recursive Program Holes
+
+---
+type: paper
+title: "LACUNA: Safe Agents as Recursive Program Holes"
+authors: Yaoyu Zhao, Yichen Xu, Oliver Bračevac, Cao Nguyen Pham, Frank Zhengqing Wu, Martin Odersky
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.28617
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-27
+updated_at: 2026-05-27
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.PL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, agent-safety, planning, rag, tool-use
+- arXiv categories: cs.AI, cs.PL
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.28617
diff --git a/papers/items/2026-2605-28787-do-agents-need-semantic-metadata-a-comparative-study-in-agentic-data-retrieval.md b/papers/items/2026-2605-28787-do-agents-need-semantic-metadata-a-comparative-study-in-agentic-data-retrieval.md
new file mode 100644
index 0000000..ef06505
--- /dev/null
+++ b/papers/items/2026-2605-28787-do-agents-need-semantic-metadata-a-comparative-study-in-agentic-data-retrieval.md
@@ -0,0 +1,62 @@
+# Paper: Do Agents Need Semantic Metadata? A Comparative Study in Agentic Data Retrieval
+
+---
+type: paper
+title: Do Agents Need Semantic Metadata? A Comparative Study in Agentic Data Retrieval
+authors: Shiyu Chen, Tarfah Alrashed, Alon Halevy, Natasha Noy
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.28787
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-27
+updated_at: 2026-05-27
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.IR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, rag, tool-use, workflow-agent
+- arXiv categories: cs.IR, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.28787
diff --git a/papers/items/2026-2605-28835-genesisfunc-multi-agent-data-generation-for-accurate-and-generalizable-function-.md b/papers/items/2026-2605-28835-genesisfunc-multi-agent-data-generation-for-accurate-and-generalizable-function-.md
new file mode 100644
index 0000000..bc60b7a
--- /dev/null
+++ b/papers/items/2026-2605-28835-genesisfunc-multi-agent-data-generation-for-accurate-and-generalizable-function-.md
@@ -0,0 +1,62 @@
+# Paper: GenesisFunc: Multi-Agent Data Generation for Accurate and Generalizable Function-Calling
+
+---
+type: paper
+title: "GenesisFunc: Multi-Agent Data Generation for Accurate and Generalizable Function-Calling"
+authors: Hao-Xiang Xu, Chong Deng, Jiaqing Liu, Wen Wang, Qian Chen, Lujia Bao, Xiangang Li, Zhen-Hua Ling
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.28835
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-10
+updated_at: 2026-04-10
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, multi-agent, rag, tool-use
+- arXiv categories: cs.CL, cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.28835
diff --git a/papers/items/2026-2605-28850-representation-signatures-and-risk-feedback-alignment-in-llm-trading-agents.md b/papers/items/2026-2605-28850-representation-signatures-and-risk-feedback-alignment-in-llm-trading-agents.md
new file mode 100644
index 0000000..c6e1cd4
--- /dev/null
+++ b/papers/items/2026-2605-28850-representation-signatures-and-risk-feedback-alignment-in-llm-trading-agents.md
@@ -0,0 +1,65 @@
+# Paper: Representation Signatures and Risk-Feedback Alignment in LLM Trading Agents
+
+---
+type: paper
+title: Representation Signatures and Risk-Feedback Alignment in LLM Trading Agents
+authors: Weicheng Xue
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.28850
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-16
+updated_at: 2026-05-30
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - coding-agent
+  - memory
+  - planning
+  - reasoning
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - q-fin.CP
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-safety, coding-agent, memory, planning, reasoning, tool-use, world-model
+- arXiv categories: cs.LG, q-fin.CP
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.28850
diff --git a/papers/items/2026-2605-29341-worldmemarena-evaluating-multimodal-agent-memory-through-action-world-interactio.md b/papers/items/2026-2605-29341-worldmemarena-evaluating-multimodal-agent-memory-through-action-world-interactio.md
new file mode 100644
index 0000000..1db343e
--- /dev/null
+++ b/papers/items/2026-2605-29341-worldmemarena-evaluating-multimodal-agent-memory-through-action-world-interactio.md
@@ -0,0 +1,63 @@
+# Paper: WorldMemArena: Evaluating Multimodal Agent Memory Through Action-World Interaction
+
+---
+type: paper
+title: "WorldMemArena: Evaluating Multimodal Agent Memory Through Action-World Interaction"
+authors: Chengzhi Liu, Yuzhe Yang, Sophia Xiao Pu, Yepeng Liu, Lin Long, Yichen Guo, Nuo Chen, Zhaotian Weng, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.29341
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-28
+updated_at: 2026-06-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: agent-memory, rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory, rag-agent
+- inferred topics: agent-evaluation, memory, planning, rag, tool-use
+- arXiv categories: cs.CV, cs.CL
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.29341
diff --git a/papers/items/2026-2605-29630-entity-collision-a-stratified-protocol-for-attributing-retrieval-lift-in-agent-m.md b/papers/items/2026-2605-29630-entity-collision-a-stratified-protocol-for-attributing-retrieval-lift-in-agent-m.md
new file mode 100644
index 0000000..1f8cd06
--- /dev/null
+++ b/papers/items/2026-2605-29630-entity-collision-a-stratified-protocol-for-attributing-retrieval-lift-in-agent-m.md
@@ -0,0 +1,63 @@
+# Paper: Entity-Collision: A Stratified Protocol for Attributing Retrieval Lift in Agent Memory
+
+---
+type: paper
+title: "Entity-Collision: A Stratified Protocol for Attributing Retrieval Lift in Agent Memory"
+authors: Youwang Deng
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.29630
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-28
+updated_at: 2026-05-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+  - cs.IR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, rag, tool-use
+- arXiv categories: cs.CL, cs.AI, cs.IR
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.29630
diff --git a/papers/items/2026-2605-29640-vikingmem-a-memory-base-management-system-for-stateful-llm-based-applications.md b/papers/items/2026-2605-29640-vikingmem-a-memory-base-management-system-for-stateful-llm-based-applications.md
new file mode 100644
index 0000000..5dbe583
--- /dev/null
+++ b/papers/items/2026-2605-29640-vikingmem-a-memory-base-management-system-for-stateful-llm-based-applications.md
@@ -0,0 +1,61 @@
+# Paper: VikingMem: A Memory Base Management System for Stateful LLM-based Applications
+
+---
+type: paper
+title: "VikingMem: A Memory Base Management System for Stateful LLM-based Applications"
+authors: Jiajie Fu, Junwen Chen, Mengzhao Wang, Aoxiang He, Maojia Sheng, Xiangyu Ke, Yifan Zhu, Yunjun Gao
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.29640
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-28
+updated_at: 2026-06-12
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.29640
diff --git a/papers/items/2026-2605-29653-ptcg-bench-can-llm-agents-master-pok-mon-trading-card-game.md b/papers/items/2026-2605-29653-ptcg-bench-can-llm-agents-master-pok-mon-trading-card-game.md
new file mode 100644
index 0000000..7c2aa3f
--- /dev/null
+++ b/papers/items/2026-2605-29653-ptcg-bench-can-llm-agents-master-pok-mon-trading-card-game.md
@@ -0,0 +1,58 @@
+# Paper: PTCG-Bench: Can LLM Agents Master Pokémon Trading Card Game?
+
+---
+type: paper
+title: "PTCG-Bench: Can LLM Agents Master Pokémon Trading Card Game?"
+authors: Dongdong Hua, Yifei Sun, Renhong Huang, Feng Gao, Chunping Wang, Yang Yang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.29653
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-28
+updated_at: 2026-05-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-evaluation, autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, autonomous-agent-llm
+- inferred topics: agent-evaluation
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.29653
diff --git a/papers/items/2026-2605-29676-notation-matters-a-benchmark-study-of-token-optimized-formats-in-agentic-ai-syst.md b/papers/items/2026-2605-29676-notation-matters-a-benchmark-study-of-token-optimized-formats-in-agentic-ai-syst.md
new file mode 100644
index 0000000..b022476
--- /dev/null
+++ b/papers/items/2026-2605-29676-notation-matters-a-benchmark-study-of-token-optimized-formats-in-agentic-ai-syst.md
@@ -0,0 +1,60 @@
+# Paper: Notation Matters: A Benchmark Study of Token-Optimized Formats in Agentic AI Systems
+
+---
+type: paper
+title: "Notation Matters: A Benchmark Study of Token-Optimized Formats in Agentic AI Systems"
+authors: Lorenz Kutschka, Bernhard Geiger
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.29676
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-28
+updated_at: 2026-06-17
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, tool-use
+- arXiv categories: cs.AI, cs.CL
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.29676
diff --git a/papers/items/2026-2605-29790-evolve-as-a-team-collaborative-self-evolution-for-llm-based-multi-agent-systems.md b/papers/items/2026-2605-29790-evolve-as-a-team-collaborative-self-evolution-for-llm-based-multi-agent-systems.md
new file mode 100644
index 0000000..cf4dac7
--- /dev/null
+++ b/papers/items/2026-2605-29790-evolve-as-a-team-collaborative-self-evolution-for-llm-based-multi-agent-systems.md
@@ -0,0 +1,61 @@
+# Paper: Evolve as a Team: Collaborative Self-Evolution for LLM-based Multi-Agent Systems
+
+---
+type: paper
+title: "Evolve as a Team: Collaborative Self-Evolution for LLM-based Multi-Agent Systems"
+authors: Zhezheng Hao, Tianfu Wang, Huanshuo Dong, Ziyan Liu, Hong Wang, Xiankun Lin, Qiang Lin, Can Wang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.29790
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-28
+updated_at: 2026-05-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - planning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, multi-agent, planning
+- arXiv categories: cs.MA, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.29790
diff --git a/papers/items/2026-2605-29801-agentdog-1-5-a-lightweight-and-scalable-alignment-framework-for-ai-agent-safety-.md b/papers/items/2026-2605-29801-agentdog-1-5-a-lightweight-and-scalable-alignment-framework-for-ai-agent-safety-.md
new file mode 100644
index 0000000..b9a63a5
--- /dev/null
+++ b/papers/items/2026-2605-29801-agentdog-1-5-a-lightweight-and-scalable-alignment-framework-for-ai-agent-safety-.md
@@ -0,0 +1,63 @@
+# Paper: AgentDoG 1.5: A Lightweight and Scalable Alignment Framework for AI Agent Safety and Security
+
+---
+type: paper
+title: "AgentDoG 1.5: A Lightweight and Scalable Alignment Framework for AI Agent Safety and Security"
+authors: Dongrui Liu, Yu Li, Zhonghao Yang, Peng Wang, Guanxu Chen, Yuejin Xie, Qinghua Mao, Wanying Qu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.29801
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-28
+updated_at: 2026-05-28
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - computer-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+  - cs.CR
+  - cs.CV
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-safety, computer-use
+- arXiv categories: cs.AI, cs.CL, cs.CR, cs.CV, cs.LG
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.29801
diff --git a/papers/items/2026-2605-29861-towards-verifiable-multimodal-deep-research-a-multi-agent-harness-for-interleave.md b/papers/items/2026-2605-29861-towards-verifiable-multimodal-deep-research-a-multi-agent-harness-for-interleave.md
new file mode 100644
index 0000000..cc7b111
--- /dev/null
+++ b/papers/items/2026-2605-29861-towards-verifiable-multimodal-deep-research-a-multi-agent-harness-for-interleave.md
@@ -0,0 +1,66 @@
+# Paper: Towards Verifiable Multimodal Deep Research: A Multi-Agent Harness for Interleaved Report Generation
+
+---
+type: paper
+title: "Towards Verifiable Multimodal Deep Research: A Multi-Agent Harness for Interleaved Report Generation"
+authors: Chenghao Zhang, Guanting Dong, Yufan Liu, Tong Zhao, Xiaoxi Li, Zhicheng Dou
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.29861
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-28
+updated_at: 2026-06-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - multi-agent
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 22
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, memory, multi-agent, planning, rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.CL, cs.AI
+- collection score: 22
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.29861
diff --git a/papers/items/2026-2605-29960-hijacking-agent-memory-stealthy-trojan-attacks-through-conversational-interactio.md b/papers/items/2026-2605-29960-hijacking-agent-memory-stealthy-trojan-attacks-through-conversational-interactio.md
new file mode 100644
index 0000000..2619939
--- /dev/null
+++ b/papers/items/2026-2605-29960-hijacking-agent-memory-stealthy-trojan-attacks-through-conversational-interactio.md
@@ -0,0 +1,62 @@
+# Paper: Hijacking Agent Memory: Stealthy Trojan Attacks Through Conversational Interaction
+
+---
+type: paper
+title: "Hijacking Agent Memory: Stealthy Trojan Attacks Through Conversational Interaction"
+authors: Hongtao Wang, Se Yang, Yu Chen, Puzhuo Liu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.29960
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-28
+updated_at: 2026-05-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, rag, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.29960
diff --git a/papers/items/2026-2605-30058-heart-bench-do-llm-agents-exhibit-human-like-psychology.md b/papers/items/2026-2605-30058-heart-bench-do-llm-agents-exhibit-human-like-psychology.md
new file mode 100644
index 0000000..b313460
--- /dev/null
+++ b/papers/items/2026-2605-30058-heart-bench-do-llm-agents-exhibit-human-like-psychology.md
@@ -0,0 +1,63 @@
+# Paper: HEART-Bench: Do LLM Agents Exhibit Human-like Psychology?
+
+---
+type: paper
+title: "HEART-Bench: Do LLM Agents Exhibit Human-like Psychology?"
+authors: Weihan Peng, Chenxu Zhang, Qianao Wang, Yuling Shi, Heng Lian, Qihong Mao, Jiahao Pang, Chunliang Feng, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.30058
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-28
+updated_at: 2026-05-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, computer-use, memory, planning, reasoning, tool-use
+- arXiv categories: cs.CL
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.30058
diff --git a/papers/items/2026-2605-30090-directorbench-diagnosing-long-form-video-generation-with-personalized-multi-agen.md b/papers/items/2026-2605-30090-directorbench-diagnosing-long-form-video-generation-with-personalized-multi-agen.md
new file mode 100644
index 0000000..b74e36b
--- /dev/null
+++ b/papers/items/2026-2605-30090-directorbench-diagnosing-long-form-video-generation-with-personalized-multi-agen.md
@@ -0,0 +1,64 @@
+# Paper: DirectorBench: Diagnosing Long-Form Video Generation with Personalized Multi-Agent Evaluation
+
+---
+type: paper
+title: "DirectorBench: Diagnosing Long-Form Video Generation with Personalized Multi-Agent Evaluation"
+authors: Jiamin Chen, Qianben Chen, Jiawen Zhang, Yidi Wu, Yuchen Li, Xiaokun Zhang, Wangchunshu Zhou, Chen Ma
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.30090
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-28
+updated_at: 2026-05-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.CV
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, agent-safety, multi-agent, rag, tool-use, workflow-agent
+- arXiv categories: cs.CL, cs.CV
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.30090
diff --git a/papers/items/2026-2605-30407-exploring-autonomous-agentic-data-engineering-for-model-specialization.md b/papers/items/2026-2605-30407-exploring-autonomous-agentic-data-engineering-for-model-specialization.md
new file mode 100644
index 0000000..398fdf6
--- /dev/null
+++ b/papers/items/2026-2605-30407-exploring-autonomous-agentic-data-engineering-for-model-specialization.md
@@ -0,0 +1,64 @@
+# Paper: Exploring Autonomous Agentic Data Engineering for Model Specialization
+
+---
+type: paper
+title: Exploring Autonomous Agentic Data Engineering for Model Specialization
+authors: Yujie Luo, Xiangyuan Ru, Jingsheng Zheng, Jingjing Wang, Yuqi Zhu, Jintian Zhang, Runnan Fang, Kewei Xu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.30407
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-28
+updated_at: 2026-06-08
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+  - cs.IR
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, computer-use, planning, workflow-agent
+- arXiv categories: cs.CL, cs.AI, cs.IR, cs.LG
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.30407
diff --git a/papers/items/2026-2605-30604-an-organization-scoped-llm-agent-runtime-architecture-for-regulated-cybersecurit.md b/papers/items/2026-2605-30604-an-organization-scoped-llm-agent-runtime-architecture-for-regulated-cybersecurit.md
new file mode 100644
index 0000000..b52a5c3
--- /dev/null
+++ b/papers/items/2026-2605-30604-an-organization-scoped-llm-agent-runtime-architecture-for-regulated-cybersecurit.md
@@ -0,0 +1,67 @@
+# Paper: An Organization-Scoped LLM Agent Runtime Architecture for Regulated Cybersecurity Operations
+
+---
+type: paper
+title: An Organization-Scoped LLM Agent Runtime Architecture for Regulated Cybersecurity Operations
+authors: George Fatouros, Georgios Makridis, George Kousiouris, John Soldatos, Dimosthenis Kyriazis
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.30604
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-28
+updated_at: 2026-05-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - planning
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+  - cs.CL
+  - cs.IR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, agent-safety, memory, planning, rag, tool-use, workflow-agent
+- arXiv categories: cs.CR, cs.AI, cs.CL, cs.IR
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.30604
diff --git a/papers/items/2026-2605-30690-elasticmem-latent-memory-as-a-learnable-resource-for-llm-agents.md b/papers/items/2026-2605-30690-elasticmem-latent-memory-as-a-learnable-resource-for-llm-agents.md
new file mode 100644
index 0000000..f3c3923
--- /dev/null
+++ b/papers/items/2026-2605-30690-elasticmem-latent-memory-as-a-learnable-resource-for-llm-agents.md
@@ -0,0 +1,63 @@
+# Paper: ElasticMem: Latent Memory as a Learnable Resource for LLM Agents
+
+---
+type: paper
+title: "ElasticMem: Latent Memory as a Learnable Resource for LLM Agents"
+authors: Tao Feng, Chongrui Ye, Tianyang Luo, Jingjun Xu, Xueqiang Xu, Haozhen Zhang, Ge Liu, Jiaxuan You
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.30690
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-29
+updated_at: 2026-05-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - embodied-agent
+  - memory
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, embodied-agent, memory, planning, rag, tool-use
+- arXiv categories: cs.CL
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.30690
diff --git a/papers/items/2026-2605-30711-sage-a-novelty-gate-for-efficient-memory-evolution-in-agentic-llms.md b/papers/items/2026-2605-30711-sage-a-novelty-gate-for-efficient-memory-evolution-in-agentic-llms.md
new file mode 100644
index 0000000..41e7977
--- /dev/null
+++ b/papers/items/2026-2605-30711-sage-a-novelty-gate-for-efficient-memory-evolution-in-agentic-llms.md
@@ -0,0 +1,65 @@
+# Paper: SAGE: A Novelty Gate for Efficient Memory Evolution in Agentic LLMs
+
+---
+type: paper
+title: "SAGE: A Novelty Gate for Efficient Memory Evolution in Agentic LLMs"
+authors: Sijia Wang, Dhanajit Brahma, Ricardo Henao
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.30711
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-29
+updated_at: 2026-06-18
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+  - cs.LG
+  - stat.ML
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, rag, reasoning, tool-use
+- arXiv categories: cs.CL, cs.AI, cs.LG, stat.ML
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.30711
diff --git a/papers/items/2026-2605-30858-forecastcompass-guiding-agentic-forecasting-with-adaptive-factor-memory.md b/papers/items/2026-2605-30858-forecastcompass-guiding-agentic-forecasting-with-adaptive-factor-memory.md
new file mode 100644
index 0000000..6f16300
--- /dev/null
+++ b/papers/items/2026-2605-30858-forecastcompass-guiding-agentic-forecasting-with-adaptive-factor-memory.md
@@ -0,0 +1,63 @@
+# Paper: ForecastCompass: Guiding Agentic Forecasting with Adaptive Factor Memory
+
+---
+type: paper
+title: "ForecastCompass: Guiding Agentic Forecasting with Adaptive Factor Memory"
+authors: Yurui Chang, Yongkang Du, Yuanpu Cao, Jinghui Chen, Lu Lin
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.30858
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-29
+updated_at: 2026-05-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, computer-use, memory, rag, reasoning, tool-use
+- arXiv categories: cs.LG
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.30858
diff --git a/papers/items/2026-2605-30883-trace-task-aware-adaptive-self-evolving-agentic-jailbreaking.md b/papers/items/2026-2605-30883-trace-task-aware-adaptive-self-evolving-agentic-jailbreaking.md
new file mode 100644
index 0000000..278e265
--- /dev/null
+++ b/papers/items/2026-2605-30883-trace-task-aware-adaptive-self-evolving-agentic-jailbreaking.md
@@ -0,0 +1,65 @@
+# Paper: TRACE: Task-Aware Adaptive Self-Evolving Agentic Jailbreaking
+
+---
+type: paper
+title: "TRACE: Task-Aware Adaptive Self-Evolving Agentic Jailbreaking"
+authors: Churui Zeng, Weiwei Qi, Kedong Xiu, Tianhang Zheng, Chaochao Lu, Liang He, Zhan Qin, Kui Ren
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.30883
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-29
+updated_at: 2026-05-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - computer-use
+  - planning
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, agent-safety, coding-agent, computer-use, planning, rag, tool-use, workflow-agent
+- arXiv categories: cs.CR
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.30883
diff --git a/papers/items/2026-2605-30907-bluefin-benchmarking-llm-agents-on-financial-spreadsheets.md b/papers/items/2026-2605-30907-bluefin-benchmarking-llm-agents-on-financial-spreadsheets.md
new file mode 100644
index 0000000..08dc35f
--- /dev/null
+++ b/papers/items/2026-2605-30907-bluefin-benchmarking-llm-agents-on-financial-spreadsheets.md
@@ -0,0 +1,62 @@
+# Paper: BlueFin: Benchmarking LLM Agents on Financial Spreadsheets
+
+---
+type: paper
+title: "BlueFin: Benchmarking LLM Agents on Financial Spreadsheets"
+authors: Srivatsa Kundurthy, Clara Na, Colton Moraine, Anoushka Mohta, Case Winter, George Fang, John Ling, Emma Strubell, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.30907
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-29
+updated_at: 2026-05-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+  - cs.CL
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, rag
+- arXiv categories: cs.SE, cs.AI, cs.CL, cs.LG
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.30907
diff --git a/papers/items/2026-2605-30947-extending-ai-for-research-to-the-humanities-a-multi-agent-framework-for-evidence.md b/papers/items/2026-2605-30947-extending-ai-for-research-to-the-humanities-a-multi-agent-framework-for-evidence.md
new file mode 100644
index 0000000..1290886
--- /dev/null
+++ b/papers/items/2026-2605-30947-extending-ai-for-research-to-the-humanities-a-multi-agent-framework-for-evidence.md
@@ -0,0 +1,62 @@
+# Paper: Extending AI for Research to the Humanities: A Multi-Agent Framework for Evidence-Grounded Scholarship
+
+---
+type: paper
+title: "Extending AI for Research to the Humanities: A Multi-Agent Framework for Evidence-Grounded Scholarship"
+authors: Yating Pan, Jiajun Zhang, Jun Wang, Qi Su
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.30947
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-29
+updated_at: 2026-06-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, multi-agent, rag, reasoning, tool-use
+- arXiv categories: cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.30947
diff --git a/papers/items/2026-2605-31075-task-focused-memorization-for-multimodal-agents.md b/papers/items/2026-2605-31075-task-focused-memorization-for-multimodal-agents.md
new file mode 100644
index 0000000..05d8647
--- /dev/null
+++ b/papers/items/2026-2605-31075-task-focused-memorization-for-multimodal-agents.md
@@ -0,0 +1,61 @@
+# Paper: Task-Focused Memorization for Multimodal Agents
+
+---
+type: paper
+title: Task-Focused Memorization for Multimodal Agents
+authors: Tao Zou, Yichen He, Tian Qiu, Yuan Lin, Hang Li
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.31075
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-29
+updated_at: 2026-05-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - embodied-agent
+  - memory
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, computer-use, embodied-agent, memory
+- arXiv categories: cs.CV
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.31075
diff --git a/papers/items/2026-2605-31268-mellum2-technical-report.md b/papers/items/2026-2605-31268-mellum2-technical-report.md
new file mode 100644
index 0000000..b344be8
--- /dev/null
+++ b/papers/items/2026-2605-31268-mellum2-technical-report.md
@@ -0,0 +1,62 @@
+# Paper: Mellum2 Technical Report
+
+---
+type: paper
+title: Mellum2 Technical Report
+authors: Marko Kojic, Ivan Bondyrev, Aral de Moor, Joseph Shtok, Petr Borovlev, Kseniia Lysaniuk, Madeeswaran Kannan, Ivan Dolgov, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.31268
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-29
+updated_at: 2026-05-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, agent-safety, coding-agent, reasoning, tool-use
+- arXiv categories: cs.CL
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.31268
diff --git a/papers/items/2026-2605-31278-industrializing-prediction-powered-inference-the-glide-library-for-reliable-gena.md b/papers/items/2026-2605-31278-industrializing-prediction-powered-inference-the-glide-library-for-reliable-gena.md
new file mode 100644
index 0000000..044722f
--- /dev/null
+++ b/papers/items/2026-2605-31278-industrializing-prediction-powered-inference-the-glide-library-for-reliable-gena.md
@@ -0,0 +1,61 @@
+# Paper: Industrializing Prediction-Powered Inference: The GLIDE Library for Reliable GenAI and Agentic Systems Evaluation
+
+---
+type: paper
+title: "Industrializing Prediction-Powered Inference: The GLIDE Library for Reliable GenAI and Agentic Systems Evaluation"
+authors: Grégoire Martinon, Ibrahim Merad, Mohammed Raki
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.31278
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-29
+updated_at: 2026-06-04
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.LG
+  - stat.ME
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, tool-use
+- arXiv categories: cs.AI, cs.LG, stat.ME
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.31278
diff --git a/papers/items/2026-2605-31308-tracegraph-shared-decision-landscapes-for-diagnosing-and-improving-agent-traject.md b/papers/items/2026-2605-31308-tracegraph-shared-decision-landscapes-for-diagnosing-and-improving-agent-traject.md
new file mode 100644
index 0000000..d6049e1
--- /dev/null
+++ b/papers/items/2026-2605-31308-tracegraph-shared-decision-landscapes-for-diagnosing-and-improving-agent-traject.md
@@ -0,0 +1,62 @@
+# Paper: TraceGraph: Shared Decision Landscapes for Diagnosing and Improving Agent Trajectories
+
+---
+type: paper
+title: "TraceGraph: Shared Decision Landscapes for Diagnosing and Improving Agent Trajectories"
+authors: Junjie Nian, Kang Chen, Ge Zhang, Yixin Cao, Yugang Jiang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.31308
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-29
+updated_at: 2026-05-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - computer-use
+  - embodied-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, coding-agent, computer-use, embodied-agent, tool-use
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.31308
diff --git a/papers/items/2026-2605-31377-dynatree-dynamic-agentic-retrieval-tree-for-time-sensitive-news-retrieval.md b/papers/items/2026-2605-31377-dynatree-dynamic-agentic-retrieval-tree-for-time-sensitive-news-retrieval.md
new file mode 100644
index 0000000..d8ba424
--- /dev/null
+++ b/papers/items/2026-2605-31377-dynatree-dynamic-agentic-retrieval-tree-for-time-sensitive-news-retrieval.md
@@ -0,0 +1,63 @@
+# Paper: DynaTree: Dynamic Agentic Retrieval Tree for Time-Sensitive News Retrieval
+
+---
+type: paper
+title: "DynaTree: Dynamic Agentic Retrieval Tree for Time-Sensitive News Retrieval"
+authors: Siyuan Qi, Xinyuan Wang, Yingxuan Yang, Haochuan Guo, Jianghao Lin, Weiwen Liu, Yong Yu, Weinan Zhang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2605.31377
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-29
+updated_at: 2026-05-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.IR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, planning, rag, reasoning, tool-use
+- arXiv categories: cs.IR, cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2605.31377
diff --git a/papers/items/2026-2606-00198-bagen-are-llm-agents-budget-aware.md b/papers/items/2026-2606-00198-bagen-are-llm-agents-budget-aware.md
new file mode 100644
index 0000000..db31aa8
--- /dev/null
+++ b/papers/items/2026-2606-00198-bagen-are-llm-agents-budget-aware.md
@@ -0,0 +1,63 @@
+# Paper: BAGEN: Are LLM Agents Budget-Aware?
+
+---
+type: paper
+title: "BAGEN: Are LLM Agents Budget-Aware?"
+authors: Yuxiang Lin, Zihan Wang, Mengyang Liu, Yuxuan Shan, Longju Bai, Junyao Zhang, Xing Jin, Boshan Chen, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.00198
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-29
+updated_at: 2026-05-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, planning, rag, tool-use
+- arXiv categories: cs.LG, cs.AI, cs.CL
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.00198
diff --git a/papers/items/2026-2606-00341-rogue-misaligned-agent-behavior-arising-from-ordinary-computer-use.md b/papers/items/2026-2606-00341-rogue-misaligned-agent-behavior-arising-from-ordinary-computer-use.md
new file mode 100644
index 0000000..ce9d712
--- /dev/null
+++ b/papers/items/2026-2606-00341-rogue-misaligned-agent-behavior-arising-from-ordinary-computer-use.md
@@ -0,0 +1,63 @@
+# Paper: ROGUE: Misaligned Agent Behavior Arising from Ordinary Computer Use
+
+---
+type: paper
+title: "ROGUE: Misaligned Agent Behavior Arising from Ordinary Computer Use"
+authors: Jeremy Tien, Abishek Anand, Yu-Rou Tuan, Yuchen Shen, J. Zico Kolter, Aran Nayebi
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.00341
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-29
+updated_at: 2026-05-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, computer-use, tool-use, workflow-agent
+- arXiv categories: cs.LG, cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.00341
diff --git a/papers/items/2026-2606-00610-memgraphrag-memory-based-multi-agent-system-for-graph-retrieval-augmented-genera.md b/papers/items/2026-2606-00610-memgraphrag-memory-based-multi-agent-system-for-graph-retrieval-augmented-genera.md
new file mode 100644
index 0000000..4d3552d
--- /dev/null
+++ b/papers/items/2026-2606-00610-memgraphrag-memory-based-multi-agent-system-for-graph-retrieval-augmented-genera.md
@@ -0,0 +1,65 @@
+# Paper: MemGraphRAG: Memory-based Multi-Agent System for Graph Retrieval-Augmented Generation
+
+---
+type: paper
+title: "MemGraphRAG: Memory-based Multi-Agent System for Graph Retrieval-Augmented Generation"
+authors: Chuanjie Wu, Zhishang Xiang, Yunbo Tang, Zerui Chen, Qinggang Zhang, Jinsong Su
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.00610
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-30
+updated_at: 2026-05-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - multi-agent
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.IR
+  - cs.AI
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, memory, multi-agent, rag, reasoning, tool-use
+- arXiv categories: cs.IR, cs.AI, cs.MA
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.00610
diff --git a/papers/items/2026-2606-00611-trace-trajectory-risk-aware-compression-for-long-horizon-agent-safety.md b/papers/items/2026-2606-00611-trace-trajectory-risk-aware-compression-for-long-horizon-agent-safety.md
new file mode 100644
index 0000000..980fd6a
--- /dev/null
+++ b/papers/items/2026-2606-00611-trace-trajectory-risk-aware-compression-for-long-horizon-agent-safety.md
@@ -0,0 +1,60 @@
+# Paper: TRACE: Trajectory Risk-Aware Compression for Long-Horizon Agent Safety
+
+---
+type: paper
+title: "TRACE: Trajectory Risk-Aware Compression for Long-Horizon Agent Safety"
+authors: Zhepei Hong, Lin Wang, Liting Li, Haokai Ma, Junfeng Fang, Fei Shen, Dan Zhang, Xiang Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.00611
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-30
+updated_at: 2026-05-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, planning
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.00611
diff --git a/papers/items/2026-2606-00619-mempro-agentic-memory-systems-as-evolvable-programs.md b/papers/items/2026-2606-00619-mempro-agentic-memory-systems-as-evolvable-programs.md
new file mode 100644
index 0000000..da627d2
--- /dev/null
+++ b/papers/items/2026-2606-00619-mempro-agentic-memory-systems-as-evolvable-programs.md
@@ -0,0 +1,63 @@
+# Paper: MemPro: Agentic Memory Systems as Evolvable Programs
+
+---
+type: paper
+title: "MemPro: Agentic Memory Systems as Evolvable Programs"
+authors: Qingshan Liu, Guoqing Wang, Wen Wu, Jingqi Huang, Xinqi Tao, Dejia Song, Jie Zhou, Liang He
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.00619
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-30
+updated_at: 2026-05-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - planning
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, computer-use, memory, planning, rag
+- arXiv categories: cs.CL, cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.00619
diff --git a/papers/items/2026-2606-00644-foresci-evaluating-llm-agents-for-forward-looking-ai-research-judgment.md b/papers/items/2026-2606-00644-foresci-evaluating-llm-agents-for-forward-looking-ai-research-judgment.md
new file mode 100644
index 0000000..e9338b6
--- /dev/null
+++ b/papers/items/2026-2606-00644-foresci-evaluating-llm-agents-for-forward-looking-ai-research-judgment.md
@@ -0,0 +1,59 @@
+# Paper: ForeSci: Evaluating LLM Agents for Forward-Looking AI Research Judgment
+
+---
+type: paper
+title: "ForeSci: Evaluating LLM Agents for Forward-Looking AI Research Judgment"
+authors: Qiuyu Tian, Haojie Yin, Yingce Xia, Youyong Kong, Zequn Liu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.00644
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-30
+updated_at: 2026-06-04
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, rag
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.00644
diff --git a/papers/items/2026-2606-00756-comic-collaborative-memory-and-insights-circulation-for-long-horizon-llm-agents-.md b/papers/items/2026-2606-00756-comic-collaborative-memory-and-insights-circulation-for-long-horizon-llm-agents-.md
new file mode 100644
index 0000000..3fb5028
--- /dev/null
+++ b/papers/items/2026-2606-00756-comic-collaborative-memory-and-insights-circulation-for-long-horizon-llm-agents-.md
@@ -0,0 +1,64 @@
+# Paper: CoMIC: Collaborative Memory and Insights Circulation for Long-Horizon LLM Agents in Cloud-Edge Systems
+
+---
+type: paper
+title: "CoMIC: Collaborative Memory and Insights Circulation for Long-Horizon LLM Agents in Cloud-Edge Systems"
+authors: Yannan Wang, Longli Yang, Zhen Liu, Abhishek Kumar, Carsten Maple
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.00756
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-30
+updated_at: 2026-05-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 21
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, computer-use, memory, planning, rag, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 21
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.00756
diff --git a/papers/items/2026-2606-00914-adversarial-feeds-steer-llm-agent-decisions-against-their-defaults.md b/papers/items/2026-2606-00914-adversarial-feeds-steer-llm-agent-decisions-against-their-defaults.md
new file mode 100644
index 0000000..594dd89
--- /dev/null
+++ b/papers/items/2026-2606-00914-adversarial-feeds-steer-llm-agent-decisions-against-their-defaults.md
@@ -0,0 +1,63 @@
+# Paper: Adversarial Feeds Steer LLM Agent Decisions Against Their Defaults
+
+---
+type: paper
+title: Adversarial Feeds Steer LLM Agent Decisions Against Their Defaults
+authors: Rana Muhammad Usman
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.00914
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-30
+updated_at: 2026-05-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, agent-safety, rag, tool-use
+- arXiv categories: cs.AI, cs.CL, cs.CR
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.00914
diff --git a/papers/items/2026-2606-00915-autonomous-agentic-design-for-photonics.md b/papers/items/2026-2606-00915-autonomous-agentic-design-for-photonics.md
new file mode 100644
index 0000000..c18c6f1
--- /dev/null
+++ b/papers/items/2026-2606-00915-autonomous-agentic-design-for-photonics.md
@@ -0,0 +1,61 @@
+# Paper: Autonomous agentic design for photonics
+
+---
+type: paper
+title: Autonomous agentic design for photonics
+authors: Prashanta Kharel, Amin Khavasi, Xinzhong Chen, Tyler W. Hughes
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.00915
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-30
+updated_at: 2026-05-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - physics.optics
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, computer-use, tool-use, world-model
+- arXiv categories: physics.optics
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.00915
diff --git a/papers/items/2026-2606-00922-a-machine-to-machine-knowledge-guided-llm-agent-for-generalizable-radiotherapy-t.md b/papers/items/2026-2606-00922-a-machine-to-machine-knowledge-guided-llm-agent-for-generalizable-radiotherapy-t.md
new file mode 100644
index 0000000..229ec03
--- /dev/null
+++ b/papers/items/2026-2606-00922-a-machine-to-machine-knowledge-guided-llm-agent-for-generalizable-radiotherapy-t.md
@@ -0,0 +1,63 @@
+# Paper: A Machine-to-Machine Knowledge-Guided LLM Agent for Generalizable Radiotherapy Treatment Planning
+
+---
+type: paper
+title: A Machine-to-Machine Knowledge-Guided LLM Agent for Generalizable Radiotherapy Treatment Planning
+authors: Md Mainul Abrar, Xun Jia, Yujie Chi
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.00922
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-30
+updated_at: 2026-05-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - planning
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - physics.med-ph
+  - cs.RO
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, agent-safety, computer-use, planning, reasoning
+- arXiv categories: physics.med-ph, cs.RO
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.00922
diff --git a/papers/items/2026-2606-00939-fincom-a-financial-multi-agent-demo-with-disagree-or-commit-deliberation.md b/papers/items/2026-2606-00939-fincom-a-financial-multi-agent-demo-with-disagree-or-commit-deliberation.md
new file mode 100644
index 0000000..c73cd08
--- /dev/null
+++ b/papers/items/2026-2606-00939-fincom-a-financial-multi-agent-demo-with-disagree-or-commit-deliberation.md
@@ -0,0 +1,63 @@
+# Paper: FinCom: A Financial Multi-Agent Demo with Disagree-or-Commit Deliberation
+
+---
+type: paper
+title: "FinCom: A Financial Multi-Agent Demo with Disagree-or-Commit Deliberation"
+authors: Chao Peter Yang, Zixiao Tan, Kaisen Yao, Ziyu Zhou, Eleanor Jiang, Michael Wu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.00939
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-31
+updated_at: 2026-05-31
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, agent-safety, multi-agent, rag, reasoning, tool-use
+- arXiv categories: cs.MA
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.00939
diff --git a/papers/items/2026-2606-01041-expweaver-llm-agents-learn-from-experience-via-latent-rag.md b/papers/items/2026-2606-01041-expweaver-llm-agents-learn-from-experience-via-latent-rag.md
new file mode 100644
index 0000000..fa572dc
--- /dev/null
+++ b/papers/items/2026-2606-01041-expweaver-llm-agents-learn-from-experience-via-latent-rag.md
@@ -0,0 +1,63 @@
+# Paper: ExpWeaver: LLM Agents Learn from Experience via Latent RAG
+
+---
+type: paper
+title: "ExpWeaver: LLM Agents Learn from Experience via Latent RAG"
+authors: Tao Feng, Tianyang Luo, Jingjun Xu, Zhigang Hua, Yan Xie, Shuang Yang, Ge Liu, Jiaxuan You
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.01041
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-31
+updated_at: 2026-05-31
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: planning-agent, rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent, rag-agent
+- inferred topics: agent-evaluation, coding-agent, planning, rag, reasoning, tool-use
+- arXiv categories: cs.CL
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.01041
diff --git a/papers/items/2026-2606-01138-memorywire-a-vendor-neutral-wire-format-for-agent-memory-operations.md b/papers/items/2026-2606-01138-memorywire-a-vendor-neutral-wire-format-for-agent-memory-operations.md
new file mode 100644
index 0000000..5f5343a
--- /dev/null
+++ b/papers/items/2026-2606-01138-memorywire-a-vendor-neutral-wire-format-for-agent-memory-operations.md
@@ -0,0 +1,63 @@
+# Paper: memorywire: A Vendor-Neutral Wire Format for Agent Memory Operations
+
+---
+type: paper
+title: "memorywire: A Vendor-Neutral Wire Format for Agent Memory Operations"
+authors: Thamilvendhan Munirathinam
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.01138
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-31
+updated_at: 2026-06-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+  - cs.DC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, rag, workflow-agent
+- arXiv categories: cs.CR, cs.AI, cs.DC
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.01138
diff --git a/papers/items/2026-2606-01166-braveguard-from-open-world-threats-to-safer-computer-use-agents.md b/papers/items/2026-2606-01166-braveguard-from-open-world-threats-to-safer-computer-use-agents.md
new file mode 100644
index 0000000..0dc6d21
--- /dev/null
+++ b/papers/items/2026-2606-01166-braveguard-from-open-world-threats-to-safer-computer-use-agents.md
@@ -0,0 +1,63 @@
+# Paper: BraveGuard: From Open-World Threats to Safer Computer-Use Agents
+
+---
+type: paper
+title: "BraveGuard: From Open-World Threats to Safer Computer-Use Agents"
+authors: Yunhao Feng, Xiaohu Du, Xinhao Deng, Yifan Ding, Ming Wen, Yixu Wang, Yuxiang Xie, Baihui Zheng, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.01166
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-31
+updated_at: 2026-06-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, computer-use, rag, tool-use
+- arXiv categories: cs.CR, cs.CL
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.01166
diff --git a/papers/items/2026-2606-01185-skill-issues-data-centric-optimization-of-lakehouse-agents.md b/papers/items/2026-2606-01185-skill-issues-data-centric-optimization-of-lakehouse-agents.md
new file mode 100644
index 0000000..615662b
--- /dev/null
+++ b/papers/items/2026-2606-01185-skill-issues-data-centric-optimization-of-lakehouse-agents.md
@@ -0,0 +1,63 @@
+# Paper: "Skill issues'': data-centric optimization of lakehouse agents
+
+---
+type: paper
+title: "\"Skill issues'': data-centric optimization of lakehouse agents"
+authors: Nicole Rose Schneider, Davide Ghilardi, Giacomo Piccinini, Jacopo Tagliabue
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.01185
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-31
+updated_at: 2026-05-31
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - planning
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, coding-agent, planning, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.01185
diff --git a/papers/items/2026-2606-01199-can-llm-agents-sustain-long-horizon-organizational-dynamics.md b/papers/items/2026-2606-01199-can-llm-agents-sustain-long-horizon-organizational-dynamics.md
new file mode 100644
index 0000000..2bec0ce
--- /dev/null
+++ b/papers/items/2026-2606-01199-can-llm-agents-sustain-long-horizon-organizational-dynamics.md
@@ -0,0 +1,64 @@
+# Paper: Can LLM Agents Sustain Long-Horizon Organizational Dynamics?
+
+---
+type: paper
+title: Can LLM Agents Sustain Long-Horizon Organizational Dynamics?
+authors: Xuancheng Zhu, Yang Yue, Shuaibing Wan, Zihan Dou, Xiaohan Zhang, Yongrui Liu, Guoshun Nan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.01199
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-31
+updated_at: 2026-05-31
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - multi-agent
+  - planning
+  - rag
+  - workflow-agent
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 22
+collection_queries: language-agent, planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent, planning-agent
+- inferred topics: agent-evaluation, memory, multi-agent, planning, rag, workflow-agent, world-model
+- arXiv categories: cs.AI
+- collection score: 22
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.01199
diff --git a/papers/items/2026-2606-01222-rag-driven-multi-agent-llm-framework-with-task-decomposition-for-beyond-5g-auto-.md b/papers/items/2026-2606-01222-rag-driven-multi-agent-llm-framework-with-task-decomposition-for-beyond-5g-auto-.md
new file mode 100644
index 0000000..b3d39e3
--- /dev/null
+++ b/papers/items/2026-2606-01222-rag-driven-multi-agent-llm-framework-with-task-decomposition-for-beyond-5g-auto-.md
@@ -0,0 +1,62 @@
+# Paper: RAG-driven Multi-Agent LLM Framework with Task Decomposition for Beyond 5G Auto-Configuration
+
+---
+type: paper
+title: RAG-driven Multi-Agent LLM Framework with Task Decomposition for Beyond 5G Auto-Configuration
+authors: İrşat Emin Sarıdaş, Onur Salan, Ali Görçin, Ibrahim Hokelek, Hakan Ali Çırpan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.01222
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-31
+updated_at: 2026-05-31
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - planning
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - eess.SP
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, multi-agent, planning, rag, reasoning
+- arXiv categories: eess.SP
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.01222
diff --git a/papers/items/2026-2606-01385-bridging-requirements-and-architecture-multi-agent-orchestration-with-external-k.md b/papers/items/2026-2606-01385-bridging-requirements-and-architecture-multi-agent-orchestration-with-external-k.md
new file mode 100644
index 0000000..6864f29
--- /dev/null
+++ b/papers/items/2026-2606-01385-bridging-requirements-and-architecture-multi-agent-orchestration-with-external-k.md
@@ -0,0 +1,65 @@
+# Paper: Bridging Requirements and Architecture: Multi-Agent Orchestration with External Knowledge and Hierarchical Memory
+
+---
+type: paper
+title: "Bridging Requirements and Architecture: Multi-Agent Orchestration with External Knowledge and Hierarchical Memory"
+authors: Ruiyin Li, Yiran Zhang, Xiyu Zhou, Yangxiao Cai, Peng Liang, Weisong Sun, Jifeng Xuan, Zhi Jin, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.01385
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-31
+updated_at: 2026-05-31
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - memory
+  - multi-agent
+  - rag
+  - reasoning
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 21
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, coding-agent, memory, multi-agent, rag, reasoning, workflow-agent
+- arXiv categories: cs.SE, cs.AI
+- collection score: 21
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.01385
diff --git a/papers/items/2026-2606-01416-self-healing-agentic-orchestrators-for-reliable-tool-augmented-large-language-mo.md b/papers/items/2026-2606-01416-self-healing-agentic-orchestrators-for-reliable-tool-augmented-large-language-mo.md
new file mode 100644
index 0000000..0898ee6
--- /dev/null
+++ b/papers/items/2026-2606-01416-self-healing-agentic-orchestrators-for-reliable-tool-augmented-large-language-mo.md
@@ -0,0 +1,65 @@
+# Paper: Self-Healing Agentic Orchestrators for Reliable Tool-Augmented Large Language Model Systems
+
+---
+type: paper
+title: Self-Healing Agentic Orchestrators for Reliable Tool-Augmented Large Language Model Systems
+authors: Rahul Suresh Babu, Adarsh Agrawal
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.01416
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-31
+updated_at: 2026-05-31
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, computer-use, memory, planning, rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.01416
diff --git a/papers/items/2026-2606-01528-joint-agent-memory-and-exploration-learning-via-novelty-signals.md b/papers/items/2026-2606-01528-joint-agent-memory-and-exploration-learning-via-novelty-signals.md
new file mode 100644
index 0000000..17d40f9
--- /dev/null
+++ b/papers/items/2026-2606-01528-joint-agent-memory-and-exploration-learning-via-novelty-signals.md
@@ -0,0 +1,62 @@
+# Paper: Joint Agent Memory and Exploration Learning via Novelty Signals
+
+---
+type: paper
+title: Joint Agent Memory and Exploration Learning via Novelty Signals
+authors: Shizuo Tian, Xiaohong Weng, Rui Kong, Yuxuan Chen, Guohong Liu, Yuebing Song, Jiacheng Liu, Yuchen Li, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.01528
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-01
+updated_at: 2026-06-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, computer-use, memory, rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.01528
diff --git a/papers/items/2026-2606-01613-techrag-evidence-gated-multimodal-agentic-rag-for-technical-literature-reasoning.md b/papers/items/2026-2606-01613-techrag-evidence-gated-multimodal-agentic-rag-for-technical-literature-reasoning.md
new file mode 100644
index 0000000..ee26931
--- /dev/null
+++ b/papers/items/2026-2606-01613-techrag-evidence-gated-multimodal-agentic-rag-for-technical-literature-reasoning.md
@@ -0,0 +1,67 @@
+# Paper: TechRAG: Evidence-Gated Multimodal Agentic RAG for Technical Literature Reasoning
+
+---
+type: paper
+title: "TechRAG: Evidence-Gated Multimodal Agentic RAG for Technical Literature Reasoning"
+authors: Kanwar Bharat Singh
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.01613
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-01
+updated_at: 2026-06-13
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - computer-use
+  - multi-agent
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.IR
+  - cs.AI
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, coding-agent, computer-use, multi-agent, planning, rag, reasoning, tool-use
+- arXiv categories: cs.IR, cs.AI, cs.MA
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.01613
diff --git a/papers/items/2026-2606-01815-crab-bench-evaluating-llm-agents-under-complex-task-dependencies-and-human-align.md b/papers/items/2026-2606-01815-crab-bench-evaluating-llm-agents-under-complex-task-dependencies-and-human-align.md
new file mode 100644
index 0000000..77a63b5
--- /dev/null
+++ b/papers/items/2026-2606-01815-crab-bench-evaluating-llm-agents-under-complex-task-dependencies-and-human-align.md
@@ -0,0 +1,60 @@
+# Paper: CRAB-Bench: Evaluating LLM Agents under Complex Task Dependencies and Human-aligned User Simulation
+
+---
+type: paper
+title: "CRAB-Bench: Evaluating LLM Agents under Complex Task Dependencies and Human-aligned User Simulation"
+authors: Danqing Wang, Akshay Sivaraman, Lei Li
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.01815
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-01
+updated_at: 2026-06-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, tool-use, world-model
+- arXiv categories: cs.CL
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.01815
diff --git a/papers/items/2026-2606-01961-automedbench-towards-medical-autoresearch-with-agentic-ai-models.md b/papers/items/2026-2606-01961-automedbench-towards-medical-autoresearch-with-agentic-ai-models.md
new file mode 100644
index 0000000..42b53c3
--- /dev/null
+++ b/papers/items/2026-2606-01961-automedbench-towards-medical-autoresearch-with-agentic-ai-models.md
@@ -0,0 +1,61 @@
+# Paper: AutoMedBench: Towards Medical AutoResearch with Agentic AI Models
+
+---
+type: paper
+title: "AutoMedBench: Towards Medical AutoResearch with Agentic AI Models"
+authors: Junqi Liu, Selena Song, Yuhan Wang, Jiawei Mao, Hardy Chen, Xiaoke Huang, Tianhao Qi, Pengfei Guo, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.01961
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-01
+updated_at: 2026-06-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, planning, rag, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.01961
diff --git a/papers/items/2026-2606-02109-badger-bridging-agentic-and-deterministic-evaluation-for-generative-enterprise-r.md b/papers/items/2026-2606-02109-badger-bridging-agentic-and-deterministic-evaluation-for-generative-enterprise-r.md
new file mode 100644
index 0000000..888206d
--- /dev/null
+++ b/papers/items/2026-2606-02109-badger-bridging-agentic-and-deterministic-evaluation-for-generative-enterprise-r.md
@@ -0,0 +1,63 @@
+# Paper: BADGER: Bridging Agentic and Deterministic Evaluation for Generative Enterprise Reasoning
+
+---
+type: paper
+title: "BADGER: Bridging Agentic and Deterministic Evaluation for Generative Enterprise Reasoning"
+authors: Shannon Serrao, Soumitra Chatterjee, Dorina Strori, Abhishek Sharma, Nathan Miller
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.02109
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-01
+updated_at: 2026-06-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, agent-safety, rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.02109
diff --git a/papers/items/2026-2606-02302-seclaw-spec-driven-security-task-synthesis-for-evaluating-autonomous-agents.md b/papers/items/2026-2606-02302-seclaw-spec-driven-security-task-synthesis-for-evaluating-autonomous-agents.md
new file mode 100644
index 0000000..0a50009
--- /dev/null
+++ b/papers/items/2026-2606-02302-seclaw-spec-driven-security-task-synthesis-for-evaluating-autonomous-agents.md
@@ -0,0 +1,64 @@
+# Paper: SeClaw: Spec-Driven Security Task Synthesis for Evaluating Autonomous Agents
+
+---
+type: paper
+title: "SeClaw: Spec-Driven Security Task Synthesis for Evaluating Autonomous Agents"
+authors: Hao Cheng, Changtao Miao, Tianle Song, Yin Wu, He Liu, Erjia Xiao, Junchi Chen, Xiaoyu Shi, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.02302
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-01
+updated_at: 2026-06-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: agent-safety, autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety, autonomous-agent-llm
+- inferred topics: agent-evaluation, agent-safety, memory, rag, tool-use, workflow-agent
+- arXiv categories: cs.CR, cs.AI
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.02302
diff --git a/papers/items/2026-2606-02372-comap-co-evolving-world-models-and-agent-policies-for-llm-agents.md b/papers/items/2026-2606-02372-comap-co-evolving-world-models-and-agent-policies-for-llm-agents.md
new file mode 100644
index 0000000..b4414f5
--- /dev/null
+++ b/papers/items/2026-2606-02372-comap-co-evolving-world-models-and-agent-policies-for-llm-agents.md
@@ -0,0 +1,64 @@
+# Paper: COMAP: Co-Evolving World Models and Agent Policies for LLM Agents
+
+---
+type: paper
+title: "COMAP: Co-Evolving World Models and Agent Policies for LLM Agents"
+authors: Youwei Liu, Jian Wang, Hanlin Wang, Wenjie Li
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.02372
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-01
+updated_at: 2026-06-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - embodied-agent
+  - planning
+  - reasoning
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: language-agent, planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent, planning-agent
+- inferred topics: agent-evaluation, embodied-agent, planning, reasoning, tool-use, world-model
+- arXiv categories: cs.AI, cs.CL
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.02372
diff --git a/papers/items/2026-2606-02380-spade-bench-evaluating-spontaneous-strategic-deception-in-agents-via-plan-action.md b/papers/items/2026-2606-02380-spade-bench-evaluating-spontaneous-strategic-deception-in-agents-via-plan-action.md
new file mode 100644
index 0000000..f19eafe
--- /dev/null
+++ b/papers/items/2026-2606-02380-spade-bench-evaluating-spontaneous-strategic-deception-in-agents-via-plan-action.md
@@ -0,0 +1,63 @@
+# Paper: SPADE-Bench: Evaluating Spontaneous Strategic Deception in Agents via Plan-Action Divergence
+
+---
+type: paper
+title: "SPADE-Bench: Evaluating Spontaneous Strategic Deception in Agents via Plan-Action Divergence"
+authors: Yuyan Bu, Haowei Li, Qirui Zheng, Bowen Dong, Kaiyue Yang, Jiaming Ji, Yingshui Tan, Wenxin Li, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.02380
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-01
+updated_at: 2026-06-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, computer-use, planning, tool-use
+- arXiv categories: cs.CL, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.02380
diff --git a/papers/items/2026-2606-02388-policy-and-world-modeling-co-training-for-language-agents.md b/papers/items/2026-2606-02388-policy-and-world-modeling-co-training-for-language-agents.md
new file mode 100644
index 0000000..26e8e90
--- /dev/null
+++ b/papers/items/2026-2606-02388-policy-and-world-modeling-co-training-for-language-agents.md
@@ -0,0 +1,61 @@
+# Paper: Policy and World Modeling Co-Training for Language Agents
+
+---
+type: paper
+title: Policy and World Modeling Co-Training for Language Agents
+authors: Ning Lu, Baijiong Lin, Shengcai Liu, Jiahao Wu, Haoze Lv, Yanbin Wei, Lingting Zhu, Shengju Qian, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.02388
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-01
+updated_at: 2026-06-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, tool-use, world-model
+- arXiv categories: cs.LG, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.02388
diff --git a/papers/items/2026-2606-02404-k-browsecomp-a-web-browsing-agent-benchmark-grounded-in-korean-contexts.md b/papers/items/2026-2606-02404-k-browsecomp-a-web-browsing-agent-benchmark-grounded-in-korean-contexts.md
new file mode 100644
index 0000000..3d4e911
--- /dev/null
+++ b/papers/items/2026-2606-02404-k-browsecomp-a-web-browsing-agent-benchmark-grounded-in-korean-contexts.md
@@ -0,0 +1,59 @@
+# Paper: K-BrowseComp: A Web Browsing Agent Benchmark Grounded in Korean Contexts
+
+---
+type: paper
+title: "K-BrowseComp: A Web Browsing Agent Benchmark Grounded in Korean Contexts"
+authors: Nahyun Lee, Dongkeun Yoon, Guijin Son, Geewook Kim, Dayoon Ko, Jeonghun Park, Haneul Yoo, Jaewon Cho, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.02404
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-01
+updated_at: 2026-06-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, reasoning
+- arXiv categories: cs.CL
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.02404
diff --git a/papers/items/2026-2606-02461-agentcl-toward-rigorous-evaluation-of-continual-learning-in-language-agents.md b/papers/items/2026-2606-02461-agentcl-toward-rigorous-evaluation-of-continual-learning-in-language-agents.md
new file mode 100644
index 0000000..df889da
--- /dev/null
+++ b/papers/items/2026-2606-02461-agentcl-toward-rigorous-evaluation-of-continual-learning-in-language-agents.md
@@ -0,0 +1,66 @@
+# Paper: AgentCL: Toward Rigorous Evaluation of Continual Learning in Language Agents
+
+---
+type: paper
+title: "AgentCL: Toward Rigorous Evaluation of Continual Learning in Language Agents"
+authors: Yiheng Shu, Bernal Jiménez Gutiérrez, Saisri Padmaja Jonnalagedda, Yuguang Yao, Huan Sun, Yu Su
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.02461
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-01
+updated_at: 2026-06-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - computer-use
+  - memory
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 21
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, coding-agent, computer-use, memory, rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI, cs.CL
+- collection score: 21
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.02461
diff --git a/papers/items/2026-2606-02497-bridging-the-last-mile-of-time-series-forecasting-with-llm-agents.md b/papers/items/2026-2606-02497-bridging-the-last-mile-of-time-series-forecasting-with-llm-agents.md
new file mode 100644
index 0000000..58fe290
--- /dev/null
+++ b/papers/items/2026-2606-02497-bridging-the-last-mile-of-time-series-forecasting-with-llm-agents.md
@@ -0,0 +1,62 @@
+# Paper: Bridging the Last Mile of Time Series Forecasting with LLM Agents
+
+---
+type: paper
+title: Bridging the Last Mile of Time Series Forecasting with LLM Agents
+authors: Yuhua Liao, Zetian Wang, Qiangqiang Nie, Zhenhua Zhang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.02497
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-01
+updated_at: 2026-06-01
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - memory
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-safety, memory, planning, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.02497
diff --git a/papers/items/2026-2606-02812-traj-evolve-a-self-evolving-multi-agent-system-for-patient-trajectory-modeling-i.md b/papers/items/2026-2606-02812-traj-evolve-a-self-evolving-multi-agent-system-for-patient-trajectory-modeling-i.md
new file mode 100644
index 0000000..7bca052
--- /dev/null
+++ b/papers/items/2026-2606-02812-traj-evolve-a-self-evolving-multi-agent-system-for-patient-trajectory-modeling-i.md
@@ -0,0 +1,64 @@
+# Paper: Traj-Evolve: A Self-Evolving Multi-Agent System for Patient Trajectory Modeling in Lung Cancer Early Detection
+
+---
+type: paper
+title: "Traj-Evolve: A Self-Evolving Multi-Agent System for Patient Trajectory Modeling in Lung Cancer Early Detection"
+authors: Sihang Zeng, Matthew Thompson, Ruth Etzioni, Meliha Yetisgen
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.02812
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-01
+updated_at: 2026-06-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - multi-agent
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, agent-safety, memory, multi-agent, rag, reasoning
+- arXiv categories: cs.AI, cs.CL
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.02812
diff --git a/papers/items/2026-2606-02965-what-benchmarks-don-t-measure-the-case-for-evaluating-abstention-competence-in-a.md b/papers/items/2026-2606-02965-what-benchmarks-don-t-measure-the-case-for-evaluating-abstention-competence-in-a.md
new file mode 100644
index 0000000..83be11b
--- /dev/null
+++ b/papers/items/2026-2606-02965-what-benchmarks-don-t-measure-the-case-for-evaluating-abstention-competence-in-a.md
@@ -0,0 +1,62 @@
+# Paper: What Benchmarks Don't Measure: The Case for Evaluating Abstention Competence in Autonomous Agents
+
+---
+type: paper
+title: "What Benchmarks Don't Measure: The Case for Evaluating Abstention Competence in Autonomous Agents"
+authors: Victor Ojewale, Suresh Venkatasubramanian
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.02965
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-01
+updated_at: 2026-06-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, agent-safety, computer-use, tool-use, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.02965
diff --git a/papers/items/2026-2606-03108-evotrainer-co-evolving-llm-policies-and-training-harnesses-for-autonomous-agenti.md b/papers/items/2026-2606-03108-evotrainer-co-evolving-llm-policies-and-training-harnesses-for-autonomous-agenti.md
new file mode 100644
index 0000000..b7fccf1
--- /dev/null
+++ b/papers/items/2026-2606-03108-evotrainer-co-evolving-llm-policies-and-training-harnesses-for-autonomous-agenti.md
@@ -0,0 +1,61 @@
+# Paper: EvoTrainer: Co-Evolving LLM Policies and Training Harnesses for Autonomous Agentic Reinforcement Learning
+
+---
+type: paper
+title: "EvoTrainer: Co-Evolving LLM Policies and Training Harnesses for Autonomous Agentic Reinforcement Learning"
+authors: Guhong Chen, Yingcheng Shi, Yongbin Li, Binhua Li, Xander Xu, Hu Wei, Shiwen Ni, Min Yang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.03108
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-02
+updated_at: 2026-06-12
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - planning
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, coding-agent, planning, reasoning
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.03108
diff --git a/papers/items/2026-2606-03135-uncertainty-aware-clarification-in-llm-agents-with-information-gain.md b/papers/items/2026-2606-03135-uncertainty-aware-clarification-in-llm-agents-with-information-gain.md
new file mode 100644
index 0000000..30e1f7d
--- /dev/null
+++ b/papers/items/2026-2606-03135-uncertainty-aware-clarification-in-llm-agents-with-information-gain.md
@@ -0,0 +1,61 @@
+# Paper: Uncertainty-Aware Clarification in LLM Agents with Information Gain
+
+---
+type: paper
+title: Uncertainty-Aware Clarification in LLM Agents with Information Gain
+authors: Mengyi Deng, Zhiwei Li, Xin Li, Tingyu Zhu, Ying Zhao, Zhijiang Guo, Wei Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.03135
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-02
+updated_at: 2026-06-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, computer-use, rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.03135
diff --git a/papers/items/2026-2606-03157-clinicalmc-a-benchmark-for-multi-course-clinical-decision-making-with-large-lang.md b/papers/items/2026-2606-03157-clinicalmc-a-benchmark-for-multi-course-clinical-decision-making-with-large-lang.md
new file mode 100644
index 0000000..d0025b2
--- /dev/null
+++ b/papers/items/2026-2606-03157-clinicalmc-a-benchmark-for-multi-course-clinical-decision-making-with-large-lang.md
@@ -0,0 +1,60 @@
+# Paper: ClinicalMC: A Benchmark for Multi-Course Clinical Decision-Making with Large Language Models
+
+---
+type: paper
+title: "ClinicalMC: A Benchmark for Multi-Course Clinical Decision-Making with Large Language Models"
+authors: Ruihui Hou, Siyi Zhu, Ziyue Huai, Guangya Yu, Yongqi Fan, Chunming Wang, Tong Ruan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.03157
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-02
+updated_at: 2026-06-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, multi-agent, rag
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.03157
diff --git a/papers/items/2026-2606-03197-memtrain-self-supervised-context-memory-training.md b/papers/items/2026-2606-03197-memtrain-self-supervised-context-memory-training.md
new file mode 100644
index 0000000..6184812
--- /dev/null
+++ b/papers/items/2026-2606-03197-memtrain-self-supervised-context-memory-training.md
@@ -0,0 +1,63 @@
+# Paper: MemTrain: Self-Supervised Context Memory Training
+
+---
+type: paper
+title: "MemTrain: Self-Supervised Context Memory Training"
+authors: Ziheng Li, Xingrun Xing, Haoqing Wang, Zhi-Hong Deng, Yehui Tang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.03197
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-02
+updated_at: 2026-06-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, planning, rag, reasoning, tool-use
+- arXiv categories: cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.03197
diff --git a/papers/items/2026-2606-03329-infomem-training-long-context-memory-agents-with-answer-conditioned-information-.md b/papers/items/2026-2606-03329-infomem-training-long-context-memory-agents-with-answer-conditioned-information-.md
new file mode 100644
index 0000000..1be7e73
--- /dev/null
+++ b/papers/items/2026-2606-03329-infomem-training-long-context-memory-agents-with-answer-conditioned-information-.md
@@ -0,0 +1,61 @@
+# Paper: InfoMem: Training Long-Context Memory Agents with Answer-Conditioned Information Gain
+
+---
+type: paper
+title: "InfoMem: Training Long-Context Memory Agents with Answer-Conditioned Information Gain"
+authors: Tiancheng Han, Yong Li, Wuzhou Yu, Qiaosheng Zhang, Wenqi Shao
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.03329
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-02
+updated_at: 2026-06-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.03329
diff --git a/papers/items/2026-2606-03374-emem-a-hybrid-spatio-temporal-memory-system-for-embodied-agents.md b/papers/items/2026-2606-03374-emem-a-hybrid-spatio-temporal-memory-system-for-embodied-agents.md
new file mode 100644
index 0000000..0b422d9
--- /dev/null
+++ b/papers/items/2026-2606-03374-emem-a-hybrid-spatio-temporal-memory-system-for-embodied-agents.md
@@ -0,0 +1,63 @@
+# Paper: eMEM: A Hybrid Spatio-Temporal Memory System For Embodied Agents
+
+---
+type: paper
+title: "eMEM: A Hybrid Spatio-Temporal Memory System For Embodied Agents"
+authors: A. Haroon Rasheed, Maria Kabtoul
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.03374
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-02
+updated_at: 2026-06-22
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - embodied-agent
+  - memory
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.RO
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agent-memory, rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory, rag-agent
+- inferred topics: agent-evaluation, embodied-agent, memory, planning, rag, tool-use
+- arXiv categories: cs.RO
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.03374
diff --git a/papers/items/2026-2606-03544-sage-a-quantitative-evaluation-of-socialized-evolution-in-agent-ecosystems.md b/papers/items/2026-2606-03544-sage-a-quantitative-evaluation-of-socialized-evolution-in-agent-ecosystems.md
new file mode 100644
index 0000000..9e8f29c
--- /dev/null
+++ b/papers/items/2026-2606-03544-sage-a-quantitative-evaluation-of-socialized-evolution-in-agent-ecosystems.md
@@ -0,0 +1,62 @@
+# Paper: SAGE: A Quantitative Evaluation of Socialized Evolution in Agent Ecosystems
+
+---
+type: paper
+title: "SAGE: A Quantitative Evaluation of Socialized Evolution in Agent Ecosystems"
+authors: Linyue Pan, Yaoming Zhu, Lin Qiu, Xuezhi Cao, Xunliang Cai
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.03544
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-02
+updated_at: 2026-06-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, planning, reasoning, tool-use
+- arXiv categories: cs.AI, cs.CL
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.03544
diff --git a/papers/items/2026-2606-03657-diagnosing-knowledge-gaps-in-llm-tool-use-an-agentic-benchmark-for-novel-api-acq.md b/papers/items/2026-2606-03657-diagnosing-knowledge-gaps-in-llm-tool-use-an-agentic-benchmark-for-novel-api-acq.md
new file mode 100644
index 0000000..327b146
--- /dev/null
+++ b/papers/items/2026-2606-03657-diagnosing-knowledge-gaps-in-llm-tool-use-an-agentic-benchmark-for-novel-api-acq.md
@@ -0,0 +1,61 @@
+# Paper: Diagnosing Knowledge Gaps in LLM Tool Use: An Agentic Benchmark for Novel API Acquisition
+
+---
+type: paper
+title: "Diagnosing Knowledge Gaps in LLM Tool Use: An Agentic Benchmark for Novel API Acquisition"
+authors: Jinnuo Liu, Yue Peng, Jinhan Niu, Hongyi Wen
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.03657
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-02
+updated_at: 2026-06-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, coding-agent, rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.03657
diff --git a/papers/items/2026-2606-03895-agent-libos-a-runtime-substrate-for-capability-controlled-self-evolving-llm-agen.md b/papers/items/2026-2606-03895-agent-libos-a-runtime-substrate-for-capability-controlled-self-evolving-llm-agen.md
new file mode 100644
index 0000000..1edc35d
--- /dev/null
+++ b/papers/items/2026-2606-03895-agent-libos-a-runtime-substrate-for-capability-controlled-self-evolving-llm-agen.md
@@ -0,0 +1,64 @@
+# Paper: Agent libOS: A Runtime Substrate for Capability-Controlled Self-Evolving LLM Agents
+
+---
+type: paper
+title: "Agent libOS: A Runtime Substrate for Capability-Controlled Self-Evolving LLM Agents"
+authors: Yingqi Zhang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.03895
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-02
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.OS
+  - cs.AI
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, agent-safety, memory, planning, tool-use
+- arXiv categories: cs.OS, cs.AI, cs.CR
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.03895
diff --git a/papers/items/2026-2606-04051-rubas-rubric-based-reinforcement-learning-for-agent-safety.md b/papers/items/2026-2606-04051-rubas-rubric-based-reinforcement-learning-for-agent-safety.md
new file mode 100644
index 0000000..88bb4d8
--- /dev/null
+++ b/papers/items/2026-2606-04051-rubas-rubric-based-reinforcement-learning-for-agent-safety.md
@@ -0,0 +1,62 @@
+# Paper: RUBAS: Rubric-Based Reinforcement Learning for Agent Safety
+
+---
+type: paper
+title: "RUBAS: Rubric-Based Reinforcement Learning for Agent Safety"
+authors: Xian Qi Loye, Qinglin Su, Zhexin Zhang, Shiyao Cui, Qi Zhu, Fei Mi, Hongning Wang, Minlie Huang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.04051
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-02
+updated_at: 2026-06-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, tool-use
+- arXiv categories: cs.LG, cs.AI, cs.CR
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.04051
diff --git a/papers/items/2026-2606-04120-salimory-orchestrating-cognitive-memory-for-conversational-agents.md b/papers/items/2026-2606-04120-salimory-orchestrating-cognitive-memory-for-conversational-agents.md
new file mode 100644
index 0000000..6a3947d
--- /dev/null
+++ b/papers/items/2026-2606-04120-salimory-orchestrating-cognitive-memory-for-conversational-agents.md
@@ -0,0 +1,63 @@
+# Paper: SaliMory: Orchestrating Cognitive Memory for Conversational Agents
+
+---
+type: paper
+title: "SaliMory: Orchestrating Cognitive Memory for Conversational Agents"
+authors: Kai Zhang, Xinyuan Zhang, Hongda Jiang, Shiun-Zu Kuo, Hyokun Yun, Ejaz Ahmed, Shereen Oraby, Ziyun Li, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.04120
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-02
+updated_at: 2026-06-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, rag, reasoning, tool-use
+- arXiv categories: cs.CL, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.04120
diff --git a/papers/items/2026-2606-04296-the-saturation-trap-and-the-subjectivity-of-intervention-timing-why-affect-based.md b/papers/items/2026-2606-04296-the-saturation-trap-and-the-subjectivity-of-intervention-timing-why-affect-based.md
new file mode 100644
index 0000000..b93c416
--- /dev/null
+++ b/papers/items/2026-2606-04296-the-saturation-trap-and-the-subjectivity-of-intervention-timing-why-affect-based.md
@@ -0,0 +1,63 @@
+# Paper: The Saturation Trap and the Subjectivity of Intervention Timing: Why Affect-Based Triggers and LLM Judges Fail to Time Interventions on Autonomous Agents
+
+---
+type: paper
+title: "The Saturation Trap and the Subjectivity of Intervention Timing: Why Affect-Based Triggers and LLM Judges Fail to Time Interventions on Autonomous Agents"
+authors: Manvendra Modgil
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.04296
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-02
+updated_at: 2026-06-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, agent-safety, coding-agent, planning, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.04296
diff --git a/papers/items/2026-2606-04315-exploring-cross-scenario-generality-of-agentic-memory-systems-diagnostics-and-a-.md b/papers/items/2026-2606-04315-exploring-cross-scenario-generality-of-agentic-memory-systems-diagnostics-and-a-.md
new file mode 100644
index 0000000..d4355c8
--- /dev/null
+++ b/papers/items/2026-2606-04315-exploring-cross-scenario-generality-of-agentic-memory-systems-diagnostics-and-a-.md
@@ -0,0 +1,62 @@
+# Paper: Exploring Cross-Scenario Generality of Agentic Memory Systems: Diagnostics and a Strong Baseline
+
+---
+type: paper
+title: "Exploring Cross-Scenario Generality of Agentic Memory Systems: Diagnostics and a Strong Baseline"
+authors: Zhikai Chen, Jialiang Gu, Junyu Yin, Xianxuan Long, Shenglai Zeng, Xiaoze Liu, Kai Guo, Keren Zhou, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.04315
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-03
+updated_at: 2026-06-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, planning, rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.04315
diff --git a/papers/items/2026-2606-04555-temporal-order-matters-for-agentic-memory-segment-trees-for-long-horizon-agents.md b/papers/items/2026-2606-04555-temporal-order-matters-for-agentic-memory-segment-trees-for-long-horizon-agents.md
new file mode 100644
index 0000000..2ea33af
--- /dev/null
+++ b/papers/items/2026-2606-04555-temporal-order-matters-for-agentic-memory-segment-trees-for-long-horizon-agents.md
@@ -0,0 +1,62 @@
+# Paper: Temporal Order Matters for Agentic Memory: Segment Trees for Long-Horizon Agents
+
+---
+type: paper
+title: "Temporal Order Matters for Agentic Memory: Segment Trees for Long-Horizon Agents"
+authors: Yifan Simon Liu, Liam Gallagher, Faeze Moradi Kalarde, Jiazhou Liang, Armin Toroghi, Scott Sanner
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.04555
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-03
+updated_at: 2026-06-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, planning, rag
+- arXiv categories: cs.CL, cs.AI
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.04555
diff --git a/papers/items/2026-2606-04599-plan-first-judge-later-run-better-a-dmaic-inspired-agentic-system-for-industrial.md b/papers/items/2026-2606-04599-plan-first-judge-later-run-better-a-dmaic-inspired-agentic-system-for-industrial.md
new file mode 100644
index 0000000..31d6b70
--- /dev/null
+++ b/papers/items/2026-2606-04599-plan-first-judge-later-run-better-a-dmaic-inspired-agentic-system-for-industrial.md
@@ -0,0 +1,63 @@
+# Paper: Plan First, Judge Later, Run Better: A DMAIC-Inspired Agentic System for Industrial Anomaly Detection
+
+---
+type: paper
+title: "Plan First, Judge Later, Run Better: A DMAIC-Inspired Agentic System for Industrial Anomaly Detection"
+authors: Yongzi Yu, Ao Li, Le Wang, Ziyue Li, Fugee Tsung, Yuxuan Liang, Man Li
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.04599
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-03
+updated_at: 2026-06-03
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - multi-agent
+  - planning
+  - rag
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-safety, multi-agent, planning, rag, workflow-agent
+- arXiv categories: cs.AI, cs.CE
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.04599
diff --git a/papers/items/2026-2606-04628-rampart-registry-based-agentic-memory-with-priority-aware-runtime-transformation.md b/papers/items/2026-2606-04628-rampart-registry-based-agentic-memory-with-priority-aware-runtime-transformation.md
new file mode 100644
index 0000000..b6ba548
--- /dev/null
+++ b/papers/items/2026-2606-04628-rampart-registry-based-agentic-memory-with-priority-aware-runtime-transformation.md
@@ -0,0 +1,59 @@
+# Paper: RAMPART: Registry-based Agentic Memory with Priority-Aware Runtime Transformation
+
+---
+type: paper
+title: "RAMPART: Registry-based Agentic Memory with Priority-Aware Runtime Transformation"
+authors: Nikodem Tomczak
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.04628
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-03
+updated_at: 2026-06-03
+status: queued
+relevance: high
+topics:
+  - memory
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: memory
+- arXiv categories: cs.CL, cs.MA
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.04628
diff --git a/papers/items/2026-2606-04780-personatree-structured-lifecycle-memory-for-person-understanding-in-llm-agents.md b/papers/items/2026-2606-04780-personatree-structured-lifecycle-memory-for-person-understanding-in-llm-agents.md
new file mode 100644
index 0000000..4350e46
--- /dev/null
+++ b/papers/items/2026-2606-04780-personatree-structured-lifecycle-memory-for-person-understanding-in-llm-agents.md
@@ -0,0 +1,63 @@
+# Paper: PersonaTree: Structured Lifecycle Memory for Person Understanding in LLM Agents
+
+---
+type: paper
+title: "PersonaTree: Structured Lifecycle Memory for Person Understanding in LLM Agents"
+authors: Yubo Hou, Jingwei Song, Hongbo Zhang, Zhisheng Chen, Bang Xiao, Tao Wan, Zengchang Qin
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.04780
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-03
+updated_at: 2026-06-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, agent-safety, computer-use, memory, rag, tool-use
+- arXiv categories: cs.CL
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.04780
diff --git a/papers/items/2026-2606-04874-agent-planning-benchmark-a-diagnostic-framework-for-planning-capabilities-in-llm.md b/papers/items/2026-2606-04874-agent-planning-benchmark-a-diagnostic-framework-for-planning-capabilities-in-llm.md
new file mode 100644
index 0000000..34ef3f4
--- /dev/null
+++ b/papers/items/2026-2606-04874-agent-planning-benchmark-a-diagnostic-framework-for-planning-capabilities-in-llm.md
@@ -0,0 +1,61 @@
+# Paper: Agent Planning Benchmark: A Diagnostic Framework for Planning Capabilities in LLM Agents
+
+---
+type: paper
+title: "Agent Planning Benchmark: A Diagnostic Framework for Planning Capabilities in LLM Agents"
+authors: Haoyu Sun, Wenxuan Wang, Mingyang Song, Jujie He, Weinan Zhang, Yang Liu, Yang Yang, Yu Cheng
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.04874
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-03
+updated_at: 2026-06-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agent-evaluation, planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, planning-agent
+- inferred topics: agent-evaluation, computer-use, planning, tool-use
+- arXiv categories: cs.CL
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.04874
diff --git a/papers/items/2026-2606-04990-from-agent-traces-to-trust-a-survey-of-evidence-tracing-and-execution-provenance.md b/papers/items/2026-2606-04990-from-agent-traces-to-trust-a-survey-of-evidence-tracing-and-execution-provenance.md
new file mode 100644
index 0000000..650e480
--- /dev/null
+++ b/papers/items/2026-2606-04990-from-agent-traces-to-trust-a-survey-of-evidence-tracing-and-execution-provenance.md
@@ -0,0 +1,65 @@
+# Paper: From Agent Traces to Trust: A Survey of Evidence Tracing and Execution Provenance in LLM Agents
+
+---
+type: paper
+title: "From Agent Traces to Trust: A Survey of Evidence Tracing and Execution Provenance in LLM Agents"
+authors: Yiqi Wang, Jiaqi Zhang, Taotao Cai, Zirui Liu, Qingqiang Sun, Zequn Sun, Zhangkai Wu, Manqing Dong, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.04990
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-03
+updated_at: 2026-06-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - multi-agent
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 21
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, agent-safety, memory, multi-agent, planning, rag, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 21
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.04990
diff --git a/papers/items/2026-2606-05241-search-time-contamination-in-deep-research-agents-measuring-performance-inflatio.md b/papers/items/2026-2606-05241-search-time-contamination-in-deep-research-agents-measuring-performance-inflatio.md
new file mode 100644
index 0000000..8222b99
--- /dev/null
+++ b/papers/items/2026-2606-05241-search-time-contamination-in-deep-research-agents-measuring-performance-inflatio.md
@@ -0,0 +1,61 @@
+# Paper: Search-Time Contamination in Deep Research Agents: Measuring Performance Inflation in Public Benchmark Evaluation
+
+---
+type: paper
+title: "Search-Time Contamination in Deep Research Agents: Measuring Performance Inflation in Public Benchmark Evaluation"
+authors: Yongjie Wang, Xinyue Zhang, Kunhong Yao, Zhiwei Zeng, Kaisong Song, Jun Lin, Zhiqi Shen
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.05241
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-03
+updated_at: 2026-06-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, rag, reasoning
+- arXiv categories: cs.CR, cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.05241
diff --git a/papers/items/2026-2606-05263-policy-conditioned-counterfactual-credit-for-verifiable-reinforcement-learning-o.md b/papers/items/2026-2606-05263-policy-conditioned-counterfactual-credit-for-verifiable-reinforcement-learning-o.md
new file mode 100644
index 0000000..4a4b99e
--- /dev/null
+++ b/papers/items/2026-2606-05263-policy-conditioned-counterfactual-credit-for-verifiable-reinforcement-learning-o.md
@@ -0,0 +1,63 @@
+# Paper: Policy-Conditioned Counterfactual Credit for Verifiable Reinforcement Learning of Long-Horizon Language Agents
+
+---
+type: paper
+title: Policy-Conditioned Counterfactual Credit for Verifiable Reinforcement Learning of Long-Horizon Language Agents
+authors: Renwei Meng
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.05263
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-03
+updated_at: 2026-06-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, planning, rag, reasoning, tool-use
+- arXiv categories: cs.LG, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.05263
diff --git a/papers/items/2026-2606-05414-when-evidence-is-sparse-weakly-supervised-early-failure-alerting-in-dialogs-and-.md b/papers/items/2026-2606-05414-when-evidence-is-sparse-weakly-supervised-early-failure-alerting-in-dialogs-and-.md
new file mode 100644
index 0000000..fe1795f
--- /dev/null
+++ b/papers/items/2026-2606-05414-when-evidence-is-sparse-weakly-supervised-early-failure-alerting-in-dialogs-and-.md
@@ -0,0 +1,65 @@
+# Paper: When Evidence is Sparse: Weakly Supervised Early Failure Alerting in Dialogs and LLM-Agent Trajectories
+
+---
+type: paper
+title: "When Evidence is Sparse: Weakly Supervised Early Failure Alerting in Dialogs and LLM-Agent Trajectories"
+authors: Avinash Baidya, Xinran Liang, Ruocheng Guo, Xiang Gao, Kamalika Das
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.05414
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-03
+updated_at: 2026-06-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+  - cs.HC
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, agent-safety, planning, rag, tool-use
+- arXiv categories: cs.CL, cs.AI, cs.HC, cs.LG
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.05414
diff --git a/papers/items/2026-2606-05436-ten-headache-specialists-versus-artificial-intelligence-for-clinical-literature-.md b/papers/items/2026-2606-05436-ten-headache-specialists-versus-artificial-intelligence-for-clinical-literature-.md
new file mode 100644
index 0000000..4fa4b5f
--- /dev/null
+++ b/papers/items/2026-2606-05436-ten-headache-specialists-versus-artificial-intelligence-for-clinical-literature-.md
@@ -0,0 +1,63 @@
+# Paper: Ten Headache Specialists versus Artificial Intelligence for Clinical Literature Summarization: A Critical Evaluation and Comparison
+
+---
+type: paper
+title: "Ten Headache Specialists versus Artificial Intelligence for Clinical Literature Summarization: A Critical Evaluation and Comparison"
+authors: Alejandro Lozano, Keiko Ihara, Ping-Hao Yang, Carrie E. Robertson, Jennifer Stern, Allan Purdy, Hsiangkuo Yuan, Pengfei Zhang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.05436
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-03
+updated_at: 2026-06-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+  - cs.IR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, computer-use, rag, tool-use
+- arXiv categories: cs.AI, cs.CL, cs.IR
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.05436
diff --git a/papers/items/2026-2606-05463-psebench-a-controllable-and-verifiable-benchmark-for-evaluating-llms-in-patient-.md b/papers/items/2026-2606-05463-psebench-a-controllable-and-verifiable-benchmark-for-evaluating-llms-in-patient-.md
new file mode 100644
index 0000000..f9667f9
--- /dev/null
+++ b/papers/items/2026-2606-05463-psebench-a-controllable-and-verifiable-benchmark-for-evaluating-llms-in-patient-.md
@@ -0,0 +1,62 @@
+# Paper: PSEBench: A Controllable and Verifiable Benchmark for Evaluating LLMs in Patient Safety Event Triage
+
+---
+type: paper
+title: "PSEBench: A Controllable and Verifiable Benchmark for Evaluating LLMs in Patient Safety Event Triage"
+authors: Keqi Han, Ryan Young, Annabel Strauss, Lindsey Hughes, Katharine M. Nesbitt, Nicole Schueler, Che Ngufor, Carl Yang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.05463
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-03
+updated_at: 2026-06-09
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, agent-safety, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.05463
diff --git a/papers/items/2026-2606-05548-adk-arena-evaluating-agent-development-kits-via-llm-as-a-developer.md b/papers/items/2026-2606-05548-adk-arena-evaluating-agent-development-kits-via-llm-as-a-developer.md
new file mode 100644
index 0000000..4235ebb
--- /dev/null
+++ b/papers/items/2026-2606-05548-adk-arena-evaluating-agent-development-kits-via-llm-as-a-developer.md
@@ -0,0 +1,61 @@
+# Paper: ADK Arena: Evaluating Agent Development Kits via LLM-as-a-Developer
+
+---
+type: paper
+title: "ADK Arena: Evaluating Agent Development Kits via LLM-as-a-Developer"
+authors: Jintao Huang, Xiaomin Li, Gaurav Mittal, Yu Hu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.05548
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-04
+updated_at: 2026-06-04
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-evaluation, autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, autonomous-agent-llm
+- inferred topics: agent-evaluation, coding-agent, tool-use
+- arXiv categories: cs.SE, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.05548
diff --git a/papers/items/2026-2606-05558-autoregressive-diffusion-world-models-for-off-policy-evaluation-of-llm-agents.md b/papers/items/2026-2606-05558-autoregressive-diffusion-world-models-for-off-policy-evaluation-of-llm-agents.md
new file mode 100644
index 0000000..e524eff
--- /dev/null
+++ b/papers/items/2026-2606-05558-autoregressive-diffusion-world-models-for-off-policy-evaluation-of-llm-agents.md
@@ -0,0 +1,62 @@
+# Paper: Autoregressive Diffusion World Models for Off-Policy Evaluation of LLM Agents
+
+---
+type: paper
+title: Autoregressive Diffusion World Models for Off-Policy Evaluation of LLM Agents
+authors: Kaixuan Liu, Guojun Xiong, Weinan Zhang, Shengpu Tang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.05558
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-04
+updated_at: 2026-06-04
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, agent-safety, computer-use, tool-use, world-model
+- arXiv categories: cs.LG
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.05558
diff --git a/papers/items/2026-2606-05622-adaplanbench-evaluating-adaptive-planning-in-large-language-model-agents-under-w.md b/papers/items/2026-2606-05622-adaplanbench-evaluating-adaptive-planning-in-large-language-model-agents-under-w.md
new file mode 100644
index 0000000..be1e410
--- /dev/null
+++ b/papers/items/2026-2606-05622-adaplanbench-evaluating-adaptive-planning-in-large-language-model-agents-under-w.md
@@ -0,0 +1,61 @@
+# Paper: AdaPlanBench: Evaluating Adaptive Planning in Large Language Model Agents under World and User Constraints
+
+---
+type: paper
+title: "AdaPlanBench: Evaluating Adaptive Planning in Large Language Model Agents under World and User Constraints"
+authors: Jiayu Liu, Cheng Qian, Zhenhailong Wang, Bingxuan Li, Jiateng Liu, Heng Wang, Jeonghwan Kim, Yumeng Wang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.05622
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-04
+updated_at: 2026-06-04
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, planning, rag, tool-use
+- arXiv categories: cs.CL
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.05622
diff --git a/papers/items/2026-2606-05658-agent-orchestrated-adaptive-rag-a-comparative-study-on-structured-and-multi-hop-.md b/papers/items/2026-2606-05658-agent-orchestrated-adaptive-rag-a-comparative-study-on-structured-and-multi-hop-.md
new file mode 100644
index 0000000..bee8723
--- /dev/null
+++ b/papers/items/2026-2606-05658-agent-orchestrated-adaptive-rag-a-comparative-study-on-structured-and-multi-hop-.md
@@ -0,0 +1,61 @@
+# Paper: Agent-Orchestrated Adaptive RAG: A Comparative Study on Structured and Multi-Hop Retrieval
+
+---
+type: paper
+title: "Agent-Orchestrated Adaptive RAG: A Comparative Study on Structured and Multi-Hop Retrieval"
+authors: Anuj Maharjan, Devinder Kaur, Richard Molyet
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.05658
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-04
+updated_at: 2026-06-04
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.IR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, rag, reasoning
+- arXiv categories: cs.IR, cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.05658
diff --git a/papers/items/2026-2606-05684-adamem-test-time-adaptive-memory-for-language-agents.md b/papers/items/2026-2606-05684-adamem-test-time-adaptive-memory-for-language-agents.md
new file mode 100644
index 0000000..99642ad
--- /dev/null
+++ b/papers/items/2026-2606-05684-adamem-test-time-adaptive-memory-for-language-agents.md
@@ -0,0 +1,63 @@
+# Paper: AdaMEM: Test-Time Adaptive Memory for Language Agents
+
+---
+type: paper
+title: "AdaMEM: Test-Time Adaptive Memory for Language Agents"
+authors: Yunxiang Zhang, Yiheng Li, Ali Payani, Lu Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.05684
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-04
+updated_at: 2026-06-04
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - planning
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agent-memory, language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory, language-agent
+- inferred topics: agent-evaluation, computer-use, memory, planning, rag, reasoning
+- arXiv categories: cs.AI
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.05684
diff --git a/papers/items/2026-2606-05711-beyond-tokens-a-unified-framework-for-latent-communication-in-llm-based-multi-ag.md b/papers/items/2026-2606-05711-beyond-tokens-a-unified-framework-for-latent-communication-in-llm-based-multi-ag.md
new file mode 100644
index 0000000..ccf7307
--- /dev/null
+++ b/papers/items/2026-2606-05711-beyond-tokens-a-unified-framework-for-latent-communication-in-llm-based-multi-ag.md
@@ -0,0 +1,63 @@
+# Paper: Beyond tokens: a unified framework for latent communication in LLM-based multi-agent systems
+
+---
+type: paper
+title: "Beyond tokens: a unified framework for latent communication in LLM-based multi-agent systems"
+authors: Yingzhuo Liu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.05711
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-04
+updated_at: 2026-06-05
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - computer-use
+  - multi-agent
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-safety, computer-use, multi-agent, planning, reasoning, tool-use
+- arXiv categories: cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.05711
diff --git a/papers/items/2026-2606-05805-from-risk-classification-to-action-plan-remediation-a-guardrail-feedback-driven-.md b/papers/items/2026-2606-05805-from-risk-classification-to-action-plan-remediation-a-guardrail-feedback-driven-.md
new file mode 100644
index 0000000..d0299a4
--- /dev/null
+++ b/papers/items/2026-2606-05805-from-risk-classification-to-action-plan-remediation-a-guardrail-feedback-driven-.md
@@ -0,0 +1,63 @@
+# Paper: From Risk Classification to Action Plan Remediation: A Guardrail Feedback Driven Framework for LLM Agents
+
+---
+type: paper
+title: "From Risk Classification to Action Plan Remediation: A Guardrail Feedback Driven Framework for LLM Agents"
+authors: Yuhao Sun, Jiacheng Zhang, Shaanan Cohney, Zhexin Zhang, Feng Liu, Xingliang Yuan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.05805
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-04
+updated_at: 2026-06-04
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, agent-safety, computer-use, planning, rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.05805
diff --git a/papers/items/2026-2606-06054-beyond-similarity-trustworthy-memory-search-for-personal-ai-agents.md b/papers/items/2026-2606-06054-beyond-similarity-trustworthy-memory-search-for-personal-ai-agents.md
new file mode 100644
index 0000000..2f5a271
--- /dev/null
+++ b/papers/items/2026-2606-06054-beyond-similarity-trustworthy-memory-search-for-personal-ai-agents.md
@@ -0,0 +1,60 @@
+# Paper: Beyond Similarity: Trustworthy Memory Search for Personal AI Agents
+
+---
+type: paper
+title: "Beyond Similarity: Trustworthy Memory Search for Personal AI Agents"
+authors: Jiawen Zhang, Kejia Chen, Jiachen Ma, Yangfan Hu, Lipeng He, Yechao Zhang, Jian Liu, Xiaohu Yang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.06054
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-04
+updated_at: 2026-06-04
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, tool-use
+- arXiv categories: cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.06054
diff --git a/papers/items/2026-2606-06090-beyond-semantic-organization-memory-as-execution-state-management-for-long-horiz.md b/papers/items/2026-2606-06090-beyond-semantic-organization-memory-as-execution-state-management-for-long-horiz.md
new file mode 100644
index 0000000..386cc93
--- /dev/null
+++ b/papers/items/2026-2606-06090-beyond-semantic-organization-memory-as-execution-state-management-for-long-horiz.md
@@ -0,0 +1,62 @@
+# Paper: Beyond Semantic Organization: Memory as Execution State Management for Long-Horizon Agents
+
+---
+type: paper
+title: "Beyond Semantic Organization: Memory as Execution State Management for Long-Horizon Agents"
+authors: Yaoqi Chen, Haibin Lai, Yuru Feng, Chuyu Han, Qianxi Zhang, Baotong Lu, Menghao Li, Xinjiang Wang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.06090
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-04
+updated_at: 2026-06-04
+status: queued
+relevance: high
+topics:
+  - computer-use
+  - memory
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-memory, rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory, rag-agent
+- inferred topics: computer-use, memory, planning, rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.06090
diff --git a/papers/items/2026-2606-06388-humans-almanac-a-human-collaboration-dataset-of-action-level-mental-model-annota.md b/papers/items/2026-2606-06388-humans-almanac-a-human-collaboration-dataset-of-action-level-mental-model-annota.md
new file mode 100644
index 0000000..a2852f7
--- /dev/null
+++ b/papers/items/2026-2606-06388-humans-almanac-a-human-collaboration-dataset-of-action-level-mental-model-annota.md
@@ -0,0 +1,64 @@
+# Paper: Humans' ALMANAC: A Human Collaboration Dataset of Action-Level Mental Model Annotations for Agent Collaboration
+
+---
+type: paper
+title: "Humans' ALMANAC: A Human Collaboration Dataset of Action-Level Mental Model Annotations for Agent Collaboration"
+authors: Jiaju Chen, Yuxuan Lu, Jiayi Su, Chaoran Chen, Songlin Xiao, Zheng Zhang, Yun Wang, Yunyao Li, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.06388
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-04
+updated_at: 2026-06-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - multi-agent
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, computer-use, multi-agent, planning, reasoning, tool-use
+- arXiv categories: cs.AI, cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.06388
diff --git a/papers/items/2026-2606-06399-collabsim-a-cscw-grounded-methodology-for-investigating-collaborative-competence.md b/papers/items/2026-2606-06399-collabsim-a-cscw-grounded-methodology-for-investigating-collaborative-competence.md
new file mode 100644
index 0000000..1865988
--- /dev/null
+++ b/papers/items/2026-2606-06399-collabsim-a-cscw-grounded-methodology-for-investigating-collaborative-competence.md
@@ -0,0 +1,64 @@
+# Paper: CollabSim: A CSCW-Grounded Methodology for Investigating Collaborative Competence of LLM Agents through Controlled Multi-Agent Experiments
+
+---
+type: paper
+title: "CollabSim: A CSCW-Grounded Methodology for Investigating Collaborative Competence of LLM Agents through Controlled Multi-Agent Experiments"
+authors: Jiaju Chen, Bo Sun, Yuxuan Lu, Yun Wang, Dakuo Wang, Bingsheng Yao
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.06399
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-04
+updated_at: 2026-06-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - planning
+  - reasoning
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 22
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, agent-safety, multi-agent, planning, reasoning, tool-use, world-model
+- arXiv categories: cs.CL
+- collection score: 22
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.06399
diff --git a/papers/items/2026-2606-06448-agent-memory-characterization-and-system-implications-of-stateful-long-horizon-w.md b/papers/items/2026-2606-06448-agent-memory-characterization-and-system-implications-of-stateful-long-horizon-w.md
new file mode 100644
index 0000000..cb0ac4c
--- /dev/null
+++ b/papers/items/2026-2606-06448-agent-memory-characterization-and-system-implications-of-stateful-long-horizon-w.md
@@ -0,0 +1,63 @@
+# Paper: Agent Memory: Characterization and System Implications of Stateful Long-Horizon Workloads
+
+---
+type: paper
+title: "Agent Memory: Characterization and System Implications of Stateful Long-Horizon Workloads"
+authors: Yasmine Omri, Ziyu Gan, Zachary Broveak, Robin Geens, Zexue He, Alex Pentland, Marian Verhelst, Tsachy Weissman, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.06448
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-04
+updated_at: 2026-06-04
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, planning, rag, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.06448
diff --git a/papers/items/2026-2606-06462-benchmark-everything-everywhere-all-at-once.md b/papers/items/2026-2606-06462-benchmark-everything-everywhere-all-at-once.md
new file mode 100644
index 0000000..e9f45c2
--- /dev/null
+++ b/papers/items/2026-2606-06462-benchmark-everything-everywhere-all-at-once.md
@@ -0,0 +1,61 @@
+# Paper: Benchmark Everything Everywhere All at Once
+
+---
+type: paper
+title: Benchmark Everything Everywhere All at Once
+authors: Shiyun Xiong, Dongming Wu, Peiwen Sun, Yuang Ai, Bokang Yang, Wencheng Han, Xiao-Hui Li, Xiangyu Yue
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.06462
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-04
+updated_at: 2026-06-04
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, coding-agent, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.06462
diff --git a/papers/items/2026-2606-06473-mlevolve-a-self-evolving-framework-for-automated-machine-learning-algorithm-disc.md b/papers/items/2026-2606-06473-mlevolve-a-self-evolving-framework-for-automated-machine-learning-algorithm-disc.md
new file mode 100644
index 0000000..4ce7160
--- /dev/null
+++ b/papers/items/2026-2606-06473-mlevolve-a-self-evolving-framework-for-automated-machine-learning-algorithm-disc.md
@@ -0,0 +1,64 @@
+# Paper: MLEvolve: A Self-Evolving Framework for Automated Machine Learning Algorithm Discovery
+
+---
+type: paper
+title: "MLEvolve: A Self-Evolving Framework for Automated Machine Learning Algorithm Discovery"
+authors: Shangheng Du, Xiangchao Yan, Jinxin Shi, Zongsheng Cao, Shiyang Feng, Zichen Liang, Boyuan Sun, Tianshuo Peng, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.06473
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-04
+updated_at: 2026-06-04
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - memory
+  - multi-agent
+  - planning
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, coding-agent, memory, multi-agent, planning, rag
+- arXiv categories: cs.AI, cs.CL
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.06473
diff --git a/papers/items/2026-2606-07314-qbuglm-an-agentic-benchmarking-framework-for-llm-based-quantum-software-debuggin.md b/papers/items/2026-2606-07314-qbuglm-an-agentic-benchmarking-framework-for-llm-based-quantum-software-debuggin.md
new file mode 100644
index 0000000..defc7de
--- /dev/null
+++ b/papers/items/2026-2606-07314-qbuglm-an-agentic-benchmarking-framework-for-llm-based-quantum-software-debuggin.md
@@ -0,0 +1,64 @@
+# Paper: QBugLM: An Agentic Benchmarking Framework for LLM-based Quantum Software Debugging
+
+---
+type: paper
+title: "QBugLM: An Agentic Benchmarking Framework for LLM-based Quantum Software Debugging"
+authors: An B. B. Pham, Hoa T. Nguyen, Muhammad Usman
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.07314
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-05
+updated_at: 2026-06-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - multi-agent
+  - reasoning
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.ET
+  - quant-ph
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, coding-agent, multi-agent, reasoning, world-model
+- arXiv categories: cs.SE, cs.ET, quant-ph
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.07314
diff --git a/papers/items/2026-2606-07379-do-coding-agents-deceive-us-detecting-and-preventing-cheating-via-capped-evaluat.md b/papers/items/2026-2606-07379-do-coding-agents-deceive-us-detecting-and-preventing-cheating-via-capped-evaluat.md
new file mode 100644
index 0000000..eef31c5
--- /dev/null
+++ b/papers/items/2026-2606-07379-do-coding-agents-deceive-us-detecting-and-preventing-cheating-via-capped-evaluat.md
@@ -0,0 +1,63 @@
+# Paper: Do Coding Agents Deceive Us? Detecting and Preventing Cheating via Capped Evaluation with Randomized Tests
+
+---
+type: paper
+title: Do Coding Agents Deceive Us? Detecting and Preventing Cheating via Capped Evaluation with Randomized Tests
+authors: Thanawat Lodkaew, Johannes Ackermann, Soichiro Nishimori, Nontawat Charoenphakdee, Masashi Sugiyama, Takashi Ishida
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.07379
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-05
+updated_at: 2026-06-08
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+  - cs.CL
+  - stat.ME
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, coding-agent, rag
+- arXiv categories: cs.LG, cs.AI, cs.CL, stat.ME
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.07379
diff --git a/papers/items/2026-2606-07402-m-3-exam-benchmarking-multimodal-memory-for-realistic-user-agent-interactions.md b/papers/items/2026-2606-07402-m-3-exam-benchmarking-multimodal-memory-for-realistic-user-agent-interactions.md
new file mode 100644
index 0000000..cb63ea4
--- /dev/null
+++ b/papers/items/2026-2606-07402-m-3-exam-benchmarking-multimodal-memory-for-realistic-user-agent-interactions.md
@@ -0,0 +1,62 @@
+# Paper: M$^3$Exam: Benchmarking Multimodal Memory for Realistic User-Agent Interactions
+
+---
+type: paper
+title: "M$^3$Exam: Benchmarking Multimodal Memory for Realistic User-Agent Interactions"
+authors: Zhengjun Huang, Wenxuan Liu, Zhoujin Tian, Wei Chen, Junle Chen, Yuqian Wu, Fangyuan Zhang, Qintian Guo, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.07402
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-05
+updated_at: 2026-06-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, memory, rag, reasoning, tool-use
+- arXiv categories: cs.CL
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.07402
diff --git a/papers/items/2026-2606-07591-researchclawbench-a-benchmark-for-end-to-end-autonomous-scientific-research.md b/papers/items/2026-2606-07591-researchclawbench-a-benchmark-for-end-to-end-autonomous-scientific-research.md
new file mode 100644
index 0000000..8d44ed4
--- /dev/null
+++ b/papers/items/2026-2606-07591-researchclawbench-a-benchmark-for-end-to-end-autonomous-scientific-research.md
@@ -0,0 +1,62 @@
+# Paper: ResearchClawBench: A Benchmark for End-to-End Autonomous Scientific Research
+
+---
+type: paper
+title: "ResearchClawBench: A Benchmark for End-to-End Autonomous Scientific Research"
+authors: Wanghan Xu, Shuo Li, Tianlin Ye, Qinglong Cao, Yixin Chen, Hengjian Gao, Yiheng Wang, Qi Li, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.07591
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-28
+updated_at: 2026-07-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, coding-agent, rag
+- arXiv categories: cs.LG, cs.AI, cs.CL
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.07591
diff --git a/papers/items/2026-2606-07595-visualleakbench-reproducible-action-boundary-propagation-failures-in-vision-lang.md b/papers/items/2026-2606-07595-visualleakbench-reproducible-action-boundary-propagation-failures-in-vision-lang.md
new file mode 100644
index 0000000..e34b6e6
--- /dev/null
+++ b/papers/items/2026-2606-07595-visualleakbench-reproducible-action-boundary-propagation-failures-in-vision-lang.md
@@ -0,0 +1,64 @@
+# Paper: VisualLeakBench: Reproducible Action-Boundary Propagation Failures in Vision-Language Agents
+
+---
+type: paper
+title: "VisualLeakBench: Reproducible Action-Boundary Propagation Failures in Vision-Language Agents"
+authors: Youting Wang, Yuan Tang, Yitian Qian, Chen Zhao
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.07595
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-29
+updated_at: 2026-05-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+  - cs.AI
+  - cs.IR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, agent-safety, memory, tool-use, workflow-agent
+- arXiv categories: cs.CV, cs.AI, cs.IR
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.07595
diff --git a/papers/items/2026-2606-07682-swe-marathon-can-agents-autonomously-complete-ultra-long-horizon-software-work.md b/papers/items/2026-2606-07682-swe-marathon-can-agents-autonomously-complete-ultra-long-horizon-software-work.md
new file mode 100644
index 0000000..703996f
--- /dev/null
+++ b/papers/items/2026-2606-07682-swe-marathon-can-agents-autonomously-complete-ultra-long-horizon-software-work.md
@@ -0,0 +1,65 @@
+# Paper: SWE-Marathon: Can Agents Autonomously Complete Ultra-Long-Horizon Software Work?
+
+---
+type: paper
+title: "SWE-Marathon: Can Agents Autonomously Complete Ultra-Long-Horizon Software Work?"
+authors: Rishi Desai, Jesse Hu, Joan Cabezas, Neel Harsola, Pratyush Shukla, Roey Ben Chaim, Adnan El Assadi, Omkaar Mukund Kamath, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.07682
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-05
+updated_at: 2026-06-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - memory
+  - planning
+  - rag
+  - reasoning
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, coding-agent, memory, planning, rag, reasoning, workflow-agent
+- arXiv categories: cs.SE, cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.07682
diff --git a/papers/items/2026-2606-07711-rosetta-memory-adaptive-memory-for-cross-llm-agents.md b/papers/items/2026-2606-07711-rosetta-memory-adaptive-memory-for-cross-llm-agents.md
new file mode 100644
index 0000000..d2c86e8
--- /dev/null
+++ b/papers/items/2026-2606-07711-rosetta-memory-adaptive-memory-for-cross-llm-agents.md
@@ -0,0 +1,62 @@
+# Paper: Rosetta Memory: Adaptive Memory for Cross-LLM Agents
+
+---
+type: paper
+title: "Rosetta Memory: Adaptive Memory for Cross-LLM Agents"
+authors: Hao Yang, Shiqi Shen, Haoxuan Li, Zhipeng Wang, Zhi Gong, Xu Chen
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.07711
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-05
+updated_at: 2026-06-05
+status: queued
+relevance: high
+topics:
+  - coding-agent
+  - memory
+  - planning
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: coding-agent, memory, planning, reasoning
+- arXiv categories: cs.LG, cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.07711
diff --git a/papers/items/2026-2606-07836-agentic-multi-fidelity-learning-of-quasiparticle-and-excitonic-properties.md b/papers/items/2026-2606-07836-agentic-multi-fidelity-learning-of-quasiparticle-and-excitonic-properties.md
new file mode 100644
index 0000000..e2af279
--- /dev/null
+++ b/papers/items/2026-2606-07836-agentic-multi-fidelity-learning-of-quasiparticle-and-excitonic-properties.md
@@ -0,0 +1,66 @@
+# Paper: Agentic multi-fidelity learning of quasiparticle and excitonic properties
+
+---
+type: paper
+title: Agentic multi-fidelity learning of quasiparticle and excitonic properties
+authors: Arnab Neogi, Aaron Forde, Christopher A. Lane, Sergei Tretiak, Jian-Xin Zhu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.07836
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-05
+updated_at: 2026-06-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - rag
+  - workflow-agent
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cond-mat.mtrl-sci
+  - cond-mat.stat-mech
+  - cs.AI
+  - physics.comp-ph
+  - quant-ph
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, computer-use, rag, workflow-agent, world-model
+- arXiv categories: cond-mat.mtrl-sci, cond-mat.stat-mech, cs.AI, physics.comp-ph, quant-ph
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.07836
diff --git a/papers/items/2026-2606-07867-the-cold-start-safety-gap-in-llm-agents.md b/papers/items/2026-2606-07867-the-cold-start-safety-gap-in-llm-agents.md
new file mode 100644
index 0000000..8d65402
--- /dev/null
+++ b/papers/items/2026-2606-07867-the-cold-start-safety-gap-in-llm-agents.md
@@ -0,0 +1,60 @@
+# Paper: The Cold-Start Safety Gap in LLM Agents
+
+---
+type: paper
+title: The Cold-Start Safety Gap in LLM Agents
+authors: Chung-En Sun, Linbo Liu, Tsui-Wei Weng
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.07867
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-05
+updated_at: 2026-06-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, tool-use
+- arXiv categories: cs.CL
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.07867
diff --git a/papers/items/2026-2606-08162-silent-failure-in-llm-agent-systems-the-entropy-principle-and-the-inevitable-dis.md b/papers/items/2026-2606-08162-silent-failure-in-llm-agent-systems-the-entropy-principle-and-the-inevitable-dis.md
new file mode 100644
index 0000000..ff77c21
--- /dev/null
+++ b/papers/items/2026-2606-08162-silent-failure-in-llm-agent-systems-the-entropy-principle-and-the-inevitable-dis.md
@@ -0,0 +1,59 @@
+# Paper: Silent Failure in LLM Agent Systems: The Entropy Principle and the Inevitable Disorder of Autonomous Agents
+
+---
+type: paper
+title: "Silent Failure in LLM Agent Systems: The Entropy Principle and the Inevitable Disorder of Autonomous Agents"
+authors: Dexing Liu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.08162
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-06
+updated_at: 2026-06-06
+status: queued
+relevance: high
+topics:
+  - memory
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: memory, tool-use
+- arXiv categories: cs.MA
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.08162
diff --git a/papers/items/2026-2606-08172-the-governance-of-human-llm-interaction-safety-gating-civility-steering-and-affe.md b/papers/items/2026-2606-08172-the-governance-of-human-llm-interaction-safety-gating-civility-steering-and-affe.md
new file mode 100644
index 0000000..4745151
--- /dev/null
+++ b/papers/items/2026-2606-08172-the-governance-of-human-llm-interaction-safety-gating-civility-steering-and-affe.md
@@ -0,0 +1,65 @@
+# Paper: The Governance of Human-LLM Interaction: Safety Gating, Civility Steering, and Affective Default Lock-In
+
+---
+type: paper
+title: "The Governance of Human-LLM Interaction: Safety Gating, Civility Steering, and Affective Default Lock-In"
+authors: Manuele Reani, Hongjian Zhang, Hongyu Tian
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.08172
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-06
+updated_at: 2026-06-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - multi-agent
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.HC
+  - cs.AI
+  - cs.CY
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, agent-safety, computer-use, multi-agent, planning, tool-use
+- arXiv categories: cs.HC, cs.AI, cs.CY
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.08172
diff --git a/papers/items/2026-2606-08274-toward-human-centered-multi-agent-systems-integrating-cognition-culture-values-a.md b/papers/items/2026-2606-08274-toward-human-centered-multi-agent-systems-integrating-cognition-culture-values-a.md
new file mode 100644
index 0000000..9e31382
--- /dev/null
+++ b/papers/items/2026-2606-08274-toward-human-centered-multi-agent-systems-integrating-cognition-culture-values-a.md
@@ -0,0 +1,64 @@
+# Paper: Toward Human-Centered Multi-Agent Systems: Integrating Cognition, Culture, Values, and Cooperation in AI Agents
+
+---
+type: paper
+title: "Toward Human-Centered Multi-Agent Systems: Integrating Cognition, Culture, Values, and Cooperation in AI Agents"
+authors: Safia Baloch, Rahemeen Khan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.08274
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-06
+updated_at: 2026-06-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - multi-agent
+  - planning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 24
+collection_queries: autonomous-agent-llm, planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm, planning-agent
+- inferred topics: agent-evaluation, agent-safety, computer-use, multi-agent, planning, tool-use, workflow-agent
+- arXiv categories: cs.MA
+- collection score: 24
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.08274
diff --git a/papers/items/2026-2606-08340-benchmarking-open-ended-multi-agent-coordination-in-language-agents.md b/papers/items/2026-2606-08340-benchmarking-open-ended-multi-agent-coordination-in-language-agents.md
new file mode 100644
index 0000000..6301bc4
--- /dev/null
+++ b/papers/items/2026-2606-08340-benchmarking-open-ended-multi-agent-coordination-in-language-agents.md
@@ -0,0 +1,66 @@
+# Paper: Benchmarking Open-Ended Multi-Agent Coordination in Language Agents
+
+---
+type: paper
+title: Benchmarking Open-Ended Multi-Agent Coordination in Language Agents
+authors: Kale-ab Abebe Tessera, Andras Szecsenyi, Cameron Barker, Alexander Rutherford, Davide Paglieri, Aidan Scannell, Henry Gouk, Elliot J. Crowley, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.08340
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-06
+updated_at: 2026-06-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - multi-agent
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.LG
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 28
+collection_queries: autonomous-agent-llm, language-agent, planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm, language-agent, planning-agent
+- inferred topics: agent-evaluation, memory, multi-agent, planning, rag, reasoning, tool-use
+- arXiv categories: cs.AI, cs.LG, cs.MA
+- collection score: 28
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.08340
diff --git a/papers/items/2026-2606-08531-vesta-a-fully-automated-scenario-generation-and-safety-evaluation-framework-for-.md b/papers/items/2026-2606-08531-vesta-a-fully-automated-scenario-generation-and-safety-evaluation-framework-for-.md
new file mode 100644
index 0000000..0dcd981
--- /dev/null
+++ b/papers/items/2026-2606-08531-vesta-a-fully-automated-scenario-generation-and-safety-evaluation-framework-for-.md
@@ -0,0 +1,62 @@
+# Paper: VESTA: A Fully Automated Scenario Generation and Safety Evaluation Framework for LLM Agents
+
+---
+type: paper
+title: "VESTA: A Fully Automated Scenario Generation and Safety Evaluation Framework for LLM Agents"
+authors: Lu Jia, Haibo Tong, Feifei Zhao, Jindong Li, Dongqi Liang, Ping Wu, Qian Zhang, Yi Zeng
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.08531
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-07
+updated_at: 2026-06-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, memory, rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.08531
diff --git a/papers/items/2026-2606-08625-from-holistic-evaluation-to-structured-criteria-rubrics-across-the-evolving-llm-.md b/papers/items/2026-2606-08625-from-holistic-evaluation-to-structured-criteria-rubrics-across-the-evolving-llm-.md
new file mode 100644
index 0000000..685bb68
--- /dev/null
+++ b/papers/items/2026-2606-08625-from-holistic-evaluation-to-structured-criteria-rubrics-across-the-evolving-llm-.md
@@ -0,0 +1,62 @@
+# Paper: From Holistic Evaluation to Structured Criteria: Rubrics Across the Evolving LLM Landscape
+
+---
+type: paper
+title: "From Holistic Evaluation to Structured Criteria: Rubrics Across the Evolving LLM Landscape"
+authors: Hao Chen, Ziyu Han, Yukun Yan, Qingfu Zhu, Maosong Sun, Wanxiang Che
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.08625
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-07
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, agent-safety, computer-use, reasoning, tool-use
+- arXiv categories: cs.CL
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.08625
diff --git a/papers/items/2026-2606-08790-rails-verification-native-clearing-for-agentic-commerce.md b/papers/items/2026-2606-08790-rails-verification-native-clearing-for-agentic-commerce.md
new file mode 100644
index 0000000..f854cd8
--- /dev/null
+++ b/papers/items/2026-2606-08790-rails-verification-native-clearing-for-agentic-commerce.md
@@ -0,0 +1,62 @@
+# Paper: RAILS: Verification-Native Clearing For Agentic Commerce
+
+---
+type: paper
+title: "RAILS: Verification-Native Clearing For Agentic Commerce"
+authors: Adrian de Valois-Franklin, Alex Bogdan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.08790
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-07
+updated_at: 2026-06-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CR
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, agent-safety, tool-use
+- arXiv categories: cs.AI, cs.CR, cs.MA
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.08790
diff --git a/papers/items/2026-2606-08960-hardening-agent-benchmarks-with-adversarial-hacker-fixer-loops.md b/papers/items/2026-2606-08960-hardening-agent-benchmarks-with-adversarial-hacker-fixer-loops.md
new file mode 100644
index 0000000..115a9da
--- /dev/null
+++ b/papers/items/2026-2606-08960-hardening-agent-benchmarks-with-adversarial-hacker-fixer-loops.md
@@ -0,0 +1,62 @@
+# Paper: Hardening Agent Benchmarks with Adversarial Hacker-Fixer Loops
+
+---
+type: paper
+title: Hardening Agent Benchmarks with Adversarial Hacker-Fixer Loops
+authors: Ziqian Zhong, Ivgeni Segal, Ivan Bercovich, Shashwat Saxena, Kexun Zhang, Aditi Raghunathan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.08960
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-08
+updated_at: 2026-06-08
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+  - cs.LG
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, reasoning
+- arXiv categories: cs.CR, cs.AI, cs.LG, cs.MA
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.08960
diff --git a/papers/items/2026-2606-09037-a-multi-agent-system-for-ipmsm-design-optimization-via-an-fea-ai-hybrid-approach.md b/papers/items/2026-2606-09037-a-multi-agent-system-for-ipmsm-design-optimization-via-an-fea-ai-hybrid-approach.md
new file mode 100644
index 0000000..bfb53bd
--- /dev/null
+++ b/papers/items/2026-2606-09037-a-multi-agent-system-for-ipmsm-design-optimization-via-an-fea-ai-hybrid-approach.md
@@ -0,0 +1,64 @@
+# Paper: A Multi-Agent System for IPMSM Design Optimization via an FEA-AI Hybrid Approach
+
+---
+type: paper
+title: A Multi-Agent System for IPMSM Design Optimization via an FEA-AI Hybrid Approach
+authors: Jinseong Han, Sunwoong Yang, Namwoo Kang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.09037
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-08
+updated_at: 2026-06-08
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - planning
+  - rag
+  - reasoning
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, multi-agent, planning, rag, reasoning, workflow-agent
+- arXiv categories: cs.AI, cs.MA
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.09037
diff --git a/papers/items/2026-2606-09071-reflect-intervention-supported-error-attribution-for-silent-failures-in-llm-agen.md b/papers/items/2026-2606-09071-reflect-intervention-supported-error-attribution-for-silent-failures-in-llm-agen.md
new file mode 100644
index 0000000..bc0f4d1
--- /dev/null
+++ b/papers/items/2026-2606-09071-reflect-intervention-supported-error-attribution-for-silent-failures-in-llm-agen.md
@@ -0,0 +1,61 @@
+# Paper: REFLECT: Intervention-Supported Error Attribution for Silent Failures in LLM Agent Traces
+
+---
+type: paper
+title: "REFLECT: Intervention-Supported Error Attribution for Silent Failures in LLM Agent Traces"
+authors: Xiaofeng Lin, Yingxu Wang, Tung Sum Thomas Kwok, Daniel Guo, Sahil Arun Nale, Charles Fleming, Guang Cheng
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.09071
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-08
+updated_at: 2026-06-08
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, planning, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.09071
diff --git a/papers/items/2026-2606-09198-mass-deep-research-for-social-sciences-with-memory-augmented-social-simulation.md b/papers/items/2026-2606-09198-mass-deep-research-for-social-sciences-with-memory-augmented-social-simulation.md
new file mode 100644
index 0000000..16997d7
--- /dev/null
+++ b/papers/items/2026-2606-09198-mass-deep-research-for-social-sciences-with-memory-augmented-social-simulation.md
@@ -0,0 +1,63 @@
+# Paper: MASS: Deep Research for Social Sciences with Memory-Augmented Social Simulation
+
+---
+type: paper
+title: "MASS: Deep Research for Social Sciences with Memory-Augmented Social Simulation"
+authors: Yongrui Liu, Deyi Xiong
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.09198
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-08
+updated_at: 2026-06-08
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - planning
+  - rag
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, computer-use, memory, planning, rag, world-model
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.09198
diff --git a/papers/items/2026-2606-09316-anything2skill-compiling-external-knowledge-into-reusable-skills-for-agents.md b/papers/items/2026-2606-09316-anything2skill-compiling-external-knowledge-into-reusable-skills-for-agents.md
new file mode 100644
index 0000000..d4f7a73
--- /dev/null
+++ b/papers/items/2026-2606-09316-anything2skill-compiling-external-knowledge-into-reusable-skills-for-agents.md
@@ -0,0 +1,64 @@
+# Paper: Anything2Skill: Compiling External Knowledge into Reusable Skills for Agents
+
+---
+type: paper
+title: "Anything2Skill: Compiling External Knowledge into Reusable Skills for Agents"
+authors: Qianjun Pan, Yutao Yang, Junsong Li, Jie Zhou, Kai Chen, Xin Li, Qin Chen, Liang He
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.09316
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-08
+updated_at: 2026-06-19
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - planning
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, computer-use, memory, planning, rag, tool-use, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.09316
diff --git a/papers/items/2026-2606-09399-runagent-superbrowser-a-theory-of-autonomous-web-navigation-grounded-in-human-br.md b/papers/items/2026-2606-09399-runagent-superbrowser-a-theory-of-autonomous-web-navigation-grounded-in-human-br.md
new file mode 100644
index 0000000..917d32a
--- /dev/null
+++ b/papers/items/2026-2606-09399-runagent-superbrowser-a-theory-of-autonomous-web-navigation-grounded-in-human-br.md
@@ -0,0 +1,64 @@
+# Paper: RunAgent SuperBrowser: A Theory of Autonomous Web Navigation Grounded in Human Browsing Behaviour
+
+---
+type: paper
+title: "RunAgent SuperBrowser: A Theory of Autonomous Web Navigation Grounded in Human Browsing Behaviour"
+authors: Radeen Mostafa, Sawradip Saha
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.09399
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-08
+updated_at: 2026-06-08
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - embodied-agent
+  - memory
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, computer-use, embodied-agent, memory, planning, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.09399
diff --git a/papers/items/2026-2606-09426-weavebench-a-long-horizon-real-world-benchmark-for-computer-use-agents-with-hybr.md b/papers/items/2026-2606-09426-weavebench-a-long-horizon-real-world-benchmark-for-computer-use-agents-with-hybr.md
new file mode 100644
index 0000000..fcc8566
--- /dev/null
+++ b/papers/items/2026-2606-09426-weavebench-a-long-horizon-real-world-benchmark-for-computer-use-agents-with-hybr.md
@@ -0,0 +1,61 @@
+# Paper: WeaveBench: A Long-Horizon, Real-World Benchmark for Computer-Use Agents with Hybrid Interfaces
+
+---
+type: paper
+title: "WeaveBench: A Long-Horizon, Real-World Benchmark for Computer-Use Agents with Hybrid Interfaces"
+authors: Wanli Li, Bowen Zhou, Yunyao Yu, Zhou Xu, Yifan Yang, Dongsheng Li, Caihua Shan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.09426
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-08
+updated_at: 2026-07-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, computer-use, planning, tool-use
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.09426
diff --git a/papers/items/2026-2606-09447-aliyunconsoleagent-training-web-agents-in-real-world-cloud-environments-via-dist.md b/papers/items/2026-2606-09447-aliyunconsoleagent-training-web-agents-in-real-world-cloud-environments-via-dist.md
new file mode 100644
index 0000000..d497311
--- /dev/null
+++ b/papers/items/2026-2606-09447-aliyunconsoleagent-training-web-agents-in-real-world-cloud-environments-via-dist.md
@@ -0,0 +1,61 @@
+# Paper: AliyunConsoleAgent: Training Web Agents in Real-World Cloud Environments via Distillation and Reinforcement Learning
+
+---
+type: paper
+title: "AliyunConsoleAgent: Training Web Agents in Real-World Cloud Environments via Distillation and Reinforcement Learning"
+authors: Bojie Rong, Zheyu Shen, Qiaoping Wang, Pengfei Kang, Yang Xu, Yawen Wei, Hanyu Wu, Zhi Zhao, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.09447
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-08
+updated_at: 2026-06-08
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, computer-use, rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.09447
diff --git a/papers/items/2026-2606-09483-memory-beyond-recall-a-dual-process-cognitive-memory-system-for-self-evolving-ll.md b/papers/items/2026-2606-09483-memory-beyond-recall-a-dual-process-cognitive-memory-system-for-self-evolving-ll.md
new file mode 100644
index 0000000..8867652
--- /dev/null
+++ b/papers/items/2026-2606-09483-memory-beyond-recall-a-dual-process-cognitive-memory-system-for-self-evolving-ll.md
@@ -0,0 +1,63 @@
+# Paper: Memory Beyond Recall: A Dual-Process Cognitive Memory System for Self-Evolving LLM Agents
+
+---
+type: paper
+title: "Memory Beyond Recall: A Dual-Process Cognitive Memory System for Self-Evolving LLM Agents"
+authors: Tianxiang Fei, Mingyang Song, Mao Zheng, Xiang Yu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.09483
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-08
+updated_at: 2026-06-08
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, rag, reasoning, tool-use
+- arXiv categories: cs.CL, cs.AI
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.09483
diff --git a/papers/items/2026-2606-09549-secureclaw-clawing-back-control-of-llm-agents.md b/papers/items/2026-2606-09549-secureclaw-clawing-back-control-of-llm-agents.md
new file mode 100644
index 0000000..3a555ba
--- /dev/null
+++ b/papers/items/2026-2606-09549-secureclaw-clawing-back-control-of-llm-agents.md
@@ -0,0 +1,62 @@
+# Paper: SecureClaw: Clawing Back Control of LLM Agents
+
+---
+type: paper
+title: "SecureClaw: Clawing Back Control of LLM Agents"
+authors: Yuhan Ma, Stefan Schmid
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.09549
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-08
+updated_at: 2026-06-08
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-safety, planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety, planning-agent
+- inferred topics: agent-evaluation, agent-safety, planning, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.09549
diff --git a/papers/items/2026-2606-09738-hdsl-a-hierarchical-domain-specific-language-for-structured-3d-indoor-scene-gene.md b/papers/items/2026-2606-09738-hdsl-a-hierarchical-domain-specific-language-for-structured-3d-indoor-scene-gene.md
new file mode 100644
index 0000000..50b40cf
--- /dev/null
+++ b/papers/items/2026-2606-09738-hdsl-a-hierarchical-domain-specific-language-for-structured-3d-indoor-scene-gene.md
@@ -0,0 +1,62 @@
+# Paper: HDSL: A Hierarchical Domain-Specific Language for Structured 3D Indoor Scene Generation and Localized Editing with LLM Agents
+
+---
+type: paper
+title: "HDSL: A Hierarchical Domain-Specific Language for Structured 3D Indoor Scene Generation and Localized Editing with LLM Agents"
+authors: Letian Li, Chao Shen, Shuzhao Xie, Chenghao Gu, ZhengXiao He, Yu Meng, Xin Yang, Wenyuan Jiang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.09738
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-08
+updated_at: 2026-06-08
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - planning
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, agent-safety, coding-agent, planning, rag
+- arXiv categories: cs.CV
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.09738
diff --git a/papers/items/2026-2606-09764-iosworld-a-benchmark-for-personally-intelligent-phone-agents.md b/papers/items/2026-2606-09764-iosworld-a-benchmark-for-personally-intelligent-phone-agents.md
new file mode 100644
index 0000000..c6ccf19
--- /dev/null
+++ b/papers/items/2026-2606-09764-iosworld-a-benchmark-for-personally-intelligent-phone-agents.md
@@ -0,0 +1,62 @@
+# Paper: iOSWorld: A Benchmark for Personally Intelligent Phone Agents
+
+---
+type: paper
+title: "iOSWorld: A Benchmark for Personally Intelligent Phone Agents"
+authors: Lawrence Keunho Jang, Mareks Woodside, Geronimo Carom, Andrew Keunwoo Jang, Jing Yu Koh, Ruslan Salakhutdinov
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.09764
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-08
+updated_at: 2026-06-08
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-evaluation, web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, web-gui-agent
+- inferred topics: agent-evaluation, computer-use, memory, tool-use
+- arXiv categories: cs.LG, cs.CL
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.09764
diff --git a/papers/items/2026-2606-09774-auto-configuring-scientific-simulators-with-lightweight-coding-agent-adapters.md b/papers/items/2026-2606-09774-auto-configuring-scientific-simulators-with-lightweight-coding-agent-adapters.md
new file mode 100644
index 0000000..0e24092
--- /dev/null
+++ b/papers/items/2026-2606-09774-auto-configuring-scientific-simulators-with-lightweight-coding-agent-adapters.md
@@ -0,0 +1,65 @@
+# Paper: Auto-Configuring Scientific Simulators with Lightweight Coding-Agent Adapters
+
+---
+type: paper
+title: Auto-Configuring Scientific Simulators with Lightweight Coding-Agent Adapters
+authors: Matthew Ho, Brian Liu, Jixuan Chen, Audrey Wang, Lianhui Qin
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.09774
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-08
+updated_at: 2026-06-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - memory
+  - rag
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, agent-safety, coding-agent, memory, rag, tool-use, world-model
+- arXiv categories: cs.AI, cs.CL
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.09774
diff --git a/papers/items/2026-2606-09863-from-confident-closing-to-silent-failure-characterizing-false-success-in-llm-age.md b/papers/items/2026-2606-09863-from-confident-closing-to-silent-failure-characterizing-false-success-in-llm-age.md
new file mode 100644
index 0000000..eb906ba
--- /dev/null
+++ b/papers/items/2026-2606-09863-from-confident-closing-to-silent-failure-characterizing-false-success-in-llm-age.md
@@ -0,0 +1,60 @@
+# Paper: From Confident Closing to Silent Failure: Characterizing False Success in LLM Agents
+
+---
+type: paper
+title: "From Confident Closing to Silent Failure: Characterizing False Success in LLM Agents"
+authors: Laksh Advani
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.09863
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-01
+updated_at: 2026-06-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, coding-agent, tool-use
+- arXiv categories: cs.LG
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.09863
diff --git a/papers/items/2026-2606-09961-3spo-state-score-supervised-policy-optimization-for-llm-agents.md b/papers/items/2026-2606-09961-3spo-state-score-supervised-policy-optimization-for-llm-agents.md
new file mode 100644
index 0000000..b6cb798
--- /dev/null
+++ b/papers/items/2026-2606-09961-3spo-state-score-supervised-policy-optimization-for-llm-agents.md
@@ -0,0 +1,61 @@
+# Paper: 3SPO: State-Score-Supervised Policy Optimization for LLM Agents
+
+---
+type: paper
+title: "3SPO: State-Score-Supervised Policy Optimization for LLM Agents"
+authors: Yu Han, Kailing Li, Yang Jiao, Yulin Dai, Yuqian Fu, Linhai Zhuo, Tianwen Qian
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.09961
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-08
+updated_at: 2026-06-08
+status: queued
+relevance: high
+topics:
+  - computer-use
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: computer-use, planning, tool-use
+- arXiv categories: cs.LG, cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.09961
diff --git a/papers/items/2026-2606-10209-less-context-better-agents-efficient-context-engineering-for-long-horizon-tool-u.md b/papers/items/2026-2606-10209-less-context-better-agents-efficient-context-engineering-for-long-horizon-tool-u.md
new file mode 100644
index 0000000..4a12c8b
--- /dev/null
+++ b/papers/items/2026-2606-10209-less-context-better-agents-efficient-context-engineering-for-long-horizon-tool-u.md
@@ -0,0 +1,64 @@
+# Paper: Less Context, Better Agents: Efficient Context Engineering for Long-Horizon Tool-Using LLM Agents
+
+---
+type: paper
+title: "Less Context, Better Agents: Efficient Context Engineering for Long-Horizon Tool-Using LLM Agents"
+authors: Abhilasha Lodha, Mahsa Pahlavikhah Varnosfaderani, Abir Chakraborty, Abhinav Mithal
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.10209
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-08
+updated_at: 2026-06-08
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.LG
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, planning, rag, tool-use, workflow-agent
+- arXiv categories: cs.AI, cs.LG, cs.SE
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.10209
diff --git a/papers/items/2026-2606-10304-mirage-a-polarity-flipping-encoding-subspace-in-llm-agents.md b/papers/items/2026-2606-10304-mirage-a-polarity-flipping-encoding-subspace-in-llm-agents.md
new file mode 100644
index 0000000..34c9782
--- /dev/null
+++ b/papers/items/2026-2606-10304-mirage-a-polarity-flipping-encoding-subspace-in-llm-agents.md
@@ -0,0 +1,63 @@
+# Paper: MIRAGE: A Polarity-Flipping Encoding Subspace in LLM Agents
+
+---
+type: paper
+title: "MIRAGE: A Polarity-Flipping Encoding Subspace in LLM Agents"
+authors: Pratibha Revankar, Kargi Chauhan, Jihye Kim, Sadiba Nusrat Nur, Vincent Siu, Chenguang Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.10304
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-09
+updated_at: 2026-06-09
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - computer-use
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, coding-agent, computer-use, planning, rag, tool-use
+- arXiv categories: cs.CL
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.10304
diff --git a/papers/items/2026-2606-10316-tabclaw-an-interactive-and-self-evolving-agent-for-spreadsheet-manipulation-and-.md b/papers/items/2026-2606-10316-tabclaw-an-interactive-and-self-evolving-agent-for-spreadsheet-manipulation-and-.md
new file mode 100644
index 0000000..eb5744f
--- /dev/null
+++ b/papers/items/2026-2606-10316-tabclaw-an-interactive-and-self-evolving-agent-for-spreadsheet-manipulation-and-.md
@@ -0,0 +1,63 @@
+# Paper: TabClaw: An Interactive and Self-Evolving Agent for Spreadsheet Manipulation and Table Reasoning
+
+---
+type: paper
+title: "TabClaw: An Interactive and Self-Evolving Agent for Spreadsheet Manipulation and Table Reasoning"
+authors: Mingyue Cheng, Shuo Yu, Daoyu Wang, Qingchuan Li, Xiaoyu Tao, Qingyang Mao, Yitong Zhou, Qi Liu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.10316
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-09
+updated_at: 2026-06-09
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, memory, planning, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.10316
diff --git a/papers/items/2026-2606-10381-agentic-hybrid-rag-for-evidence-grounded-muon-collider-analysis.md b/papers/items/2026-2606-10381-agentic-hybrid-rag-for-evidence-grounded-muon-collider-analysis.md
new file mode 100644
index 0000000..9d70272
--- /dev/null
+++ b/papers/items/2026-2606-10381-agentic-hybrid-rag-for-evidence-grounded-muon-collider-analysis.md
@@ -0,0 +1,66 @@
+# Paper: Agentic Hybrid RAG for Evidence-Grounded Muon Collider Analysis
+
+---
+type: paper
+title: Agentic Hybrid RAG for Evidence-Grounded Muon Collider Analysis
+authors: Ruobing Jiang, Dawei Fu, Cheng Jiang, Tianyi Yang, Zijian Wang, Youpeng Wu, Yong Ban, Yajun Mao, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.10381
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-09
+updated_at: 2026-06-09
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - hep-ex
+  - cs.AI
+  - cs.CL
+  - cs.IR
+  - physics.ins-det
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, rag, reasoning, tool-use, workflow-agent
+- arXiv categories: hep-ex, cs.AI, cs.CL, cs.IR, physics.ins-det
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.10381
diff --git a/papers/items/2026-2606-10394-stage-claw-automated-state-based-agent-benchmarking-for-realistic-scenarios.md b/papers/items/2026-2606-10394-stage-claw-automated-state-based-agent-benchmarking-for-realistic-scenarios.md
new file mode 100644
index 0000000..d1fd181
--- /dev/null
+++ b/papers/items/2026-2606-10394-stage-claw-automated-state-based-agent-benchmarking-for-realistic-scenarios.md
@@ -0,0 +1,59 @@
+# Paper: STAGE-Claw: Automated State-based Agent Benchmarking for Realistic Scenarios
+
+---
+type: paper
+title: "STAGE-Claw: Automated State-based Agent Benchmarking for Realistic Scenarios"
+authors: Sirui Liang, Bohan Yu, Peiyu Wang, Shiguang Guo, Wenxing Hu, Pengfei Cao, Jian Zhao, Cao Liu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.10394
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-09
+updated_at: 2026-06-09
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, tool-use
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.10394
diff --git a/papers/items/2026-2606-10423-webchallenger-a-reliable-and-efficient-generalist-web-agent.md b/papers/items/2026-2606-10423-webchallenger-a-reliable-and-efficient-generalist-web-agent.md
new file mode 100644
index 0000000..f6512fb
--- /dev/null
+++ b/papers/items/2026-2606-10423-webchallenger-a-reliable-and-efficient-generalist-web-agent.md
@@ -0,0 +1,63 @@
+# Paper: WebChallenger: A Reliable and Efficient Generalist Web Agent
+
+---
+type: paper
+title: "WebChallenger: A Reliable and Efficient Generalist Web Agent"
+authors: Jayoo Hwang, Xiaowen Zhang, Vedant Padwal
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.10423
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-09
+updated_at: 2026-06-09
+status: queued
+relevance: high
+topics:
+  - computer-use
+  - embodied-agent
+  - memory
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: computer-use, embodied-agent, memory, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.CL
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.10423
diff --git a/papers/items/2026-2606-10507-hipif-hierarchical-planning-and-information-folding-for-long-horizon-llm-agent-l.md b/papers/items/2026-2606-10507-hipif-hierarchical-planning-and-information-folding-for-long-horizon-llm-agent-l.md
new file mode 100644
index 0000000..ed3f590
--- /dev/null
+++ b/papers/items/2026-2606-10507-hipif-hierarchical-planning-and-information-folding-for-long-horizon-llm-agent-l.md
@@ -0,0 +1,62 @@
+# Paper: HIPIF: Hierarchical Planning and Information Folding for Long-Horizon LLM Agent Learning
+
+---
+type: paper
+title: "HIPIF: Hierarchical Planning and Information Folding for Long-Horizon LLM Agent Learning"
+authors: Juncheng Diao, Zhicong Lu, Peiguang Li, Yongwei Zhou, Changyuan Tian, Qingbin Li, Rongxiang Weng, Jingang Wang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.10507
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-09
+updated_at: 2026-06-09
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - planning
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agent-evaluation, autonomous-agent-llm, planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, autonomous-agent-llm, planning-agent
+- inferred topics: agent-evaluation, computer-use, memory, planning, reasoning
+- arXiv categories: cs.AI
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.10507
diff --git a/papers/items/2026-2606-10532-activemem-distributed-active-memory-for-long-horizon-llm-reasoning.md b/papers/items/2026-2606-10532-activemem-distributed-active-memory-for-long-horizon-llm-reasoning.md
new file mode 100644
index 0000000..31803ee
--- /dev/null
+++ b/papers/items/2026-2606-10532-activemem-distributed-active-memory-for-long-horizon-llm-reasoning.md
@@ -0,0 +1,62 @@
+# Paper: ActiveMem: Distributed Active Memory for Long-Horizon LLM Reasoning
+
+---
+type: paper
+title: "ActiveMem: Distributed Active Memory for Long-Horizon LLM Reasoning"
+authors: Yunhan Jiang, Wenbin Duan, Shasha Guo, Liang Pang, Xiaoqian Sun, Huawei Shen
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.10532
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-09
+updated_at: 2026-06-09
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - memory
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-safety, memory, planning, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.10532
diff --git a/papers/items/2026-2606-10577-agenticnav-zero-shot-vision-and-language-navigation-as-a-tool-calling-harness.md b/papers/items/2026-2606-10577-agenticnav-zero-shot-vision-and-language-navigation-as-a-tool-calling-harness.md
new file mode 100644
index 0000000..4b4bf1a
--- /dev/null
+++ b/papers/items/2026-2606-10577-agenticnav-zero-shot-vision-and-language-navigation-as-a-tool-calling-harness.md
@@ -0,0 +1,62 @@
+# Paper: AgenticNav: Zero-Shot Vision-and-Language Navigation as a Tool-Calling Harness
+
+---
+type: paper
+title: "AgenticNav: Zero-Shot Vision-and-Language Navigation as a Tool-Calling Harness"
+authors: Yijian Li, Changze Li, Hantian Shi, Jiaying Luo, Jiyuan Cai, Ming Yang, Tong Qin
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.10577
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-09
+updated_at: 2026-06-09
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - embodied-agent
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.RO
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, embodied-agent, memory, rag, tool-use
+- arXiv categories: cs.RO
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.10577
diff --git a/papers/items/2026-2606-10616-learning-what-to-remember-observability-safe-memory-retention-via-constrained-op.md b/papers/items/2026-2606-10616-learning-what-to-remember-observability-safe-memory-retention-via-constrained-op.md
new file mode 100644
index 0000000..af8d5a7
--- /dev/null
+++ b/papers/items/2026-2606-10616-learning-what-to-remember-observability-safe-memory-retention-via-constrained-op.md
@@ -0,0 +1,63 @@
+# Paper: Learning What to Remember: Observability-Safe Memory Retention via Constrained Optimization for Long-Horizon Language Agents
+
+---
+type: paper
+title: "Learning What to Remember: Observability-Safe Memory Retention via Constrained Optimization for Long-Horizon Language Agents"
+authors: Qingcan Kang, Liu Mingyang, Shixiong Kai, Kaichao Liang, Tao Zhong, Mingxuan Yuan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.10616
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-09
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, computer-use, memory, planning, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.10616
diff --git a/papers/items/2026-2606-10677-infini-memory-maintainable-topic-documents-for-long-term-llm-agent-memory.md b/papers/items/2026-2606-10677-infini-memory-maintainable-topic-documents-for-long-term-llm-agent-memory.md
new file mode 100644
index 0000000..062e995
--- /dev/null
+++ b/papers/items/2026-2606-10677-infini-memory-maintainable-topic-documents-for-long-term-llm-agent-memory.md
@@ -0,0 +1,62 @@
+# Paper: Infini Memory: Maintainable Topic Documents for Long-Term LLM Agent Memory
+
+---
+type: paper
+title: "Infini Memory: Maintainable Topic Documents for Long-Term LLM Agent Memory"
+authors: Suozhao Ji, Baodong Wu, Zehao Wang, Lei Xia, Qingping Li, Ruisong Wang, Wenbo Ding, Zhenhua Zhu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.10677
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-09
+updated_at: 2026-06-09
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, rag, tool-use
+- arXiv categories: cs.AI, cs.CL
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.10677
diff --git a/papers/items/2026-2606-10684-divide-and-cooperate-role-decomposed-multi-agent-llm-training-with-cross-agent-l.md b/papers/items/2026-2606-10684-divide-and-cooperate-role-decomposed-multi-agent-llm-training-with-cross-agent-l.md
new file mode 100644
index 0000000..f14f7b8
--- /dev/null
+++ b/papers/items/2026-2606-10684-divide-and-cooperate-role-decomposed-multi-agent-llm-training-with-cross-agent-l.md
@@ -0,0 +1,62 @@
+# Paper: Divide and Cooperate: Role-Decomposed Multi-Agent LLM Training with Cross-Agent Learning Signals
+
+---
+type: paper
+title: "Divide and Cooperate: Role-Decomposed Multi-Agent LLM Training with Cross-Agent Learning Signals"
+authors: Jaewan Park, Solbee Cho, Jay-Yoon Lee
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.10684
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-09
+updated_at: 2026-06-09
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, multi-agent, reasoning, tool-use
+- arXiv categories: cs.LG, cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.10684
diff --git a/papers/items/2026-2606-10742-memvenom-triggered-poisoning-of-multimodal-memories-in-web-agents.md b/papers/items/2026-2606-10742-memvenom-triggered-poisoning-of-multimodal-memories-in-web-agents.md
new file mode 100644
index 0000000..3bf1c73
--- /dev/null
+++ b/papers/items/2026-2606-10742-memvenom-triggered-poisoning-of-multimodal-memories-in-web-agents.md
@@ -0,0 +1,64 @@
+# Paper: MemVenom: Triggered Poisoning of Multimodal Memories in Web Agents
+
+---
+type: paper
+title: "MemVenom: Triggered Poisoning of Multimodal Memories in Web Agents"
+authors: Yv Zhang, Hao Sun, Hao Fang, Kuofeng Gao, Fan Mo, Bin Chen, Shu-Tao Xia, Yaowei Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.10742
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-09
+updated_at: 2026-06-09
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - planning
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, computer-use, memory, planning, rag, reasoning
+- arXiv categories: cs.CR, cs.LG
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.10742
diff --git a/papers/items/2026-2606-10749-toward-secure-llm-agents-threat-surfaces-attacks-defenses-and-evaluation.md b/papers/items/2026-2606-10749-toward-secure-llm-agents-threat-surfaces-attacks-defenses-and-evaluation.md
new file mode 100644
index 0000000..ac102e1
--- /dev/null
+++ b/papers/items/2026-2606-10749-toward-secure-llm-agents-threat-surfaces-attacks-defenses-and-evaluation.md
@@ -0,0 +1,65 @@
+# Paper: Toward Secure LLM Agents: Threat Surfaces, Attacks, Defenses, and Evaluation
+
+---
+type: paper
+title: "Toward Secure LLM Agents: Threat Surfaces, Attacks, Defenses, and Evaluation"
+authors: Yuchen Ling, Shengcheng Yu, Zhenyu Chen, Chunrong Fang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.10749
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-09
+updated_at: 2026-06-09
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - multi-agent
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 25
+collection_queries: agent-safety, planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety, planning-agent
+- inferred topics: agent-evaluation, agent-safety, memory, multi-agent, planning, rag, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 25
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.10749
diff --git a/papers/items/2026-2606-10921-trace-only-what-you-need-structure-aware-on-demand-hypergraph-memory-for-long-do.md b/papers/items/2026-2606-10921-trace-only-what-you-need-structure-aware-on-demand-hypergraph-memory-for-long-do.md
new file mode 100644
index 0000000..c3f196a
--- /dev/null
+++ b/papers/items/2026-2606-10921-trace-only-what-you-need-structure-aware-on-demand-hypergraph-memory-for-long-do.md
@@ -0,0 +1,64 @@
+# Paper: Trace Only What You Need: Structure-Aware On-Demand Hypergraph Memory for Long-Document Question Answering
+
+---
+type: paper
+title: "Trace Only What You Need: Structure-Aware On-Demand Hypergraph Memory for Long-Document Question Answering"
+authors: Xiangjun Zai, Xingyu Tan, Chen Chen, Xiaoyang Wang, Wenjie Zhang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.10921
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-09
+updated_at: 2026-06-09
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - multi-agent
+  - planning
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, computer-use, memory, multi-agent, planning, rag, reasoning
+- arXiv categories: cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.10921
diff --git a/papers/items/2026-2606-10933-frontier-coding-agents-use-metaprogramming-to-adapt-to-unfamiliar-programming-la.md b/papers/items/2026-2606-10933-frontier-coding-agents-use-metaprogramming-to-adapt-to-unfamiliar-programming-la.md
new file mode 100644
index 0000000..5e596b6
--- /dev/null
+++ b/papers/items/2026-2606-10933-frontier-coding-agents-use-metaprogramming-to-adapt-to-unfamiliar-programming-la.md
@@ -0,0 +1,61 @@
+# Paper: Frontier Coding Agents Use Metaprogramming to Adapt to Unfamiliar Programming Languages
+
+---
+type: paper
+title: Frontier Coding Agents Use Metaprogramming to Adapt to Unfamiliar Programming Languages
+authors: Aman Sharma, Sushrut Thorat, Paras Chopra
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.10933
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-09
+updated_at: 2026-06-09
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - computer-use
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, coding-agent, computer-use, tool-use
+- arXiv categories: cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.10933
diff --git a/papers/items/2026-2606-11042-workflow-gym-towards-long-horizon-evaluation-of-computer-use-agentic-tasks-in-re.md b/papers/items/2026-2606-11042-workflow-gym-towards-long-horizon-evaluation-of-computer-use-agentic-tasks-in-re.md
new file mode 100644
index 0000000..bedaa64
--- /dev/null
+++ b/papers/items/2026-2606-11042-workflow-gym-towards-long-horizon-evaluation-of-computer-use-agentic-tasks-in-re.md
@@ -0,0 +1,62 @@
+# Paper: Workflow-GYM: Towards Long-Horizon Evaluation of Computer-use Agentic tasks in Real-World Professional Fields
+
+---
+type: paper
+title: "Workflow-GYM: Towards Long-Horizon Evaluation of Computer-use Agentic tasks in Real-World Professional Fields"
+authors: Liya Zhu, Jingzhe Ding, Jian Zhang, Jianbo Xue, Shihao Liang, Ge Zhang, Yi Zhu, Duju Zeng, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.11042
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-09
+updated_at: 2026-06-11
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, computer-use, planning, tool-use, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.11042
diff --git a/papers/items/2026-2606-11078-a-history-aware-visually-grounded-critic-for-computer-use-agents.md b/papers/items/2026-2606-11078-a-history-aware-visually-grounded-critic-for-computer-use-agents.md
new file mode 100644
index 0000000..04af856
--- /dev/null
+++ b/papers/items/2026-2606-11078-a-history-aware-visually-grounded-critic-for-computer-use-agents.md
@@ -0,0 +1,64 @@
+# Paper: A History-Aware Visually Grounded Critic for Computer Use Agents
+
+---
+type: paper
+title: A History-Aware Visually Grounded Critic for Computer Use Agents
+authors: Jaewoo Lee, Zaid Khan, Archiki Prasad, Justin Chih-Yao Chen, Supriyo Chakraborty, Kartik Balasubramaniam, Sambit Sahu, Elias Stengel-Eskin, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.11078
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-09
+updated_at: 2026-06-09
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+  - cs.CV
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, computer-use, planning, rag, tool-use
+- arXiv categories: cs.AI, cs.CL, cs.CV
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.11078
diff --git a/papers/items/2026-2606-11079-vista-a-versatile-interactive-user-simulation-toolkit-for-agent-evaluation.md b/papers/items/2026-2606-11079-vista-a-versatile-interactive-user-simulation-toolkit-for-agent-evaluation.md
new file mode 100644
index 0000000..b1a3c83
--- /dev/null
+++ b/papers/items/2026-2606-11079-vista-a-versatile-interactive-user-simulation-toolkit-for-agent-evaluation.md
@@ -0,0 +1,61 @@
+# Paper: VISTA: A Versatile Interactive User Simulation Toolkit for Agent Evaluation
+
+---
+type: paper
+title: "VISTA: A Versatile Interactive User Simulation Toolkit for Agent Evaluation"
+authors: Yunan Lu, Ryan Shea, Yusen Zhang, Zhou Yu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.11079
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-09
+updated_at: 2026-06-09
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, rag, tool-use, world-model
+- arXiv categories: cs.CL
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.11079
diff --git a/papers/items/2026-2606-11119-trace-a-unified-rollout-budget-allocation-framework-for-efficient-agentic-reinfo.md b/papers/items/2026-2606-11119-trace-a-unified-rollout-budget-allocation-framework-for-efficient-agentic-reinfo.md
new file mode 100644
index 0000000..dc1fde5
--- /dev/null
+++ b/papers/items/2026-2606-11119-trace-a-unified-rollout-budget-allocation-framework-for-efficient-agentic-reinfo.md
@@ -0,0 +1,64 @@
+# Paper: TRACE: A Unified Rollout Budget Allocation Framework for Efficient Agentic Reinforcement Learning
+
+---
+type: paper
+title: "TRACE: A Unified Rollout Budget Allocation Framework for Efficient Agentic Reinforcement Learning"
+authors: Heming Zou, Qi Wang, Yun Qu, Yuhang Jiang, Lizhou Cai, Yixiu Mao, Ru Peng, Xin Xu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.11119
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-09
+updated_at: 2026-06-09
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, computer-use, rag, reasoning, tool-use
+- arXiv categories: cs.LG, cs.AI, cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.11119
diff --git a/papers/items/2026-2606-11176-data-journalist-agent-transforming-data-into-verifiable-multimodal-stories.md b/papers/items/2026-2606-11176-data-journalist-agent-transforming-data-into-verifiable-multimodal-stories.md
new file mode 100644
index 0000000..5df92da
--- /dev/null
+++ b/papers/items/2026-2606-11176-data-journalist-agent-transforming-data-into-verifiable-multimodal-stories.md
@@ -0,0 +1,66 @@
+# Paper: Data Journalist Agent: Transforming Data into Verifiable Multimodal Stories
+
+---
+type: paper
+title: "Data Journalist Agent: Transforming Data into Verifiable Multimodal Stories"
+authors: Kevin Qinghong Lin, Batu EI, Yuhong Shi, Pan Lu, Philip Torr, James Zou
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.11176
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-09
+updated_at: 2026-06-09
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - multi-agent
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+  - cs.CL
+  - cs.CY
+  - cs.HC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, coding-agent, multi-agent, rag, reasoning, tool-use
+- arXiv categories: cs.CV, cs.CL, cs.CY, cs.HC
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.11176
diff --git a/papers/items/2026-2606-11349-knowing-when-to-ask-self-gated-clarification-for-hierarchical-language-agents.md b/papers/items/2026-2606-11349-knowing-when-to-ask-self-gated-clarification-for-hierarchical-language-agents.md
new file mode 100644
index 0000000..a57230f
--- /dev/null
+++ b/papers/items/2026-2606-11349-knowing-when-to-ask-self-gated-clarification-for-hierarchical-language-agents.md
@@ -0,0 +1,62 @@
+# Paper: Knowing When to Ask: Self-Gated Clarification for Hierarchical Language Agents
+
+---
+type: paper
+title: "Knowing When to Ask: Self-Gated Clarification for Hierarchical Language Agents"
+authors: Aijing Gao, Yiming Kang, Mengdie Flora Wang, Jae Oh Woo
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.11349
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-09
+updated_at: 2026-06-12
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - embodied-agent
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.HC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, embodied-agent, reasoning, tool-use
+- arXiv categories: cs.AI, cs.HC
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.11349
diff --git a/papers/items/2026-2606-11354-a-zero-shot-multi-agent-framework-for-human-building-interaction-via-programmati.md b/papers/items/2026-2606-11354-a-zero-shot-multi-agent-framework-for-human-building-interaction-via-programmati.md
new file mode 100644
index 0000000..b68a7a9
--- /dev/null
+++ b/papers/items/2026-2606-11354-a-zero-shot-multi-agent-framework-for-human-building-interaction-via-programmati.md
@@ -0,0 +1,64 @@
+# Paper: A Zero-Shot Multi-Agent Framework for Human-Building Interaction via Programmatic Reasoning
+
+---
+type: paper
+title: A Zero-Shot Multi-Agent Framework for Human-Building Interaction via Programmatic Reasoning
+authors: Yuqi Wang, Gulai Shen, Ali Mehmani
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.11354
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-09
+updated_at: 2026-06-09
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - coding-agent
+  - multi-agent
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.ET
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-safety, coding-agent, multi-agent, planning, rag, reasoning, tool-use
+- arXiv categories: cs.ET
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.11354
diff --git a/papers/items/2026-2606-11680-organize-then-retrieve-hierarchical-memory-navigation-for-efficient-agents.md b/papers/items/2026-2606-11680-organize-then-retrieve-hierarchical-memory-navigation-for-efficient-agents.md
new file mode 100644
index 0000000..bd295c1
--- /dev/null
+++ b/papers/items/2026-2606-11680-organize-then-retrieve-hierarchical-memory-navigation-for-efficient-agents.md
@@ -0,0 +1,66 @@
+# Paper: Organize then Retrieve: Hierarchical Memory Navigation for Efficient Agents
+
+---
+type: paper
+title: "Organize then Retrieve: Hierarchical Memory Navigation for Efficient Agents"
+authors: Hao-Lun Hsu, Nikki Lijing Kuang, Boyi Liu, Zhewei Yao, Yuxiong He
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.11680
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-10
+updated_at: 2026-06-10
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - embodied-agent
+  - memory
+  - planning
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, computer-use, embodied-agent, memory, planning, rag, reasoning
+- arXiv categories: cs.AI, cs.CL, cs.LG
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.11680
diff --git a/papers/items/2026-2606-11688-goal-autopilot-a-verifiable-anti-fabrication-firewall-for-unattended-long-horizo.md b/papers/items/2026-2606-11688-goal-autopilot-a-verifiable-anti-fabrication-firewall-for-unattended-long-horizo.md
new file mode 100644
index 0000000..5f55e22
--- /dev/null
+++ b/papers/items/2026-2606-11688-goal-autopilot-a-verifiable-anti-fabrication-firewall-for-unattended-long-horizo.md
@@ -0,0 +1,62 @@
+# Paper: Goal-Autopilot: A Verifiable Anti-Fabrication Firewall for Unattended Long-Horizon Agents
+
+---
+type: paper
+title: "Goal-Autopilot: A Verifiable Anti-Fabrication Firewall for Unattended Long-Horizon Agents"
+authors: Youwang Deng
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.11688
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-10
+updated_at: 2026-06-10
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - planning
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, coding-agent, planning, rag
+- arXiv categories: cs.CL, cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.11688
diff --git a/papers/items/2026-2606-11702-medcta-a-benchmark-for-clinical-tool-agents.md b/papers/items/2026-2606-11702-medcta-a-benchmark-for-clinical-tool-agents.md
new file mode 100644
index 0000000..e3972c5
--- /dev/null
+++ b/papers/items/2026-2606-11702-medcta-a-benchmark-for-clinical-tool-agents.md
@@ -0,0 +1,63 @@
+# Paper: MedCTA: A Benchmark for Clinical Tool Agents
+
+---
+type: paper
+title: "MedCTA: A Benchmark for Clinical Tool Agents"
+authors: Tajamul Ashraf, Hyewon Jeong, Fida Mohammad Thoker, Bernard Ghanem
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.11702
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-10
+updated_at: 2026-06-10
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, planning, rag, tool-use
+- arXiv categories: cs.CV, cs.AI, cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.11702
diff --git a/papers/items/2026-2606-11869-agents-all-the-way-down-a-methodology-for-building-custom-ai-agents-from-substra.md b/papers/items/2026-2606-11869-agents-all-the-way-down-a-methodology-for-building-custom-ai-agents-from-substra.md
new file mode 100644
index 0000000..672f172
--- /dev/null
+++ b/papers/items/2026-2606-11869-agents-all-the-way-down-a-methodology-for-building-custom-ai-agents-from-substra.md
@@ -0,0 +1,62 @@
+# Paper: Agents All the Way Down; A Methodology for Building Custom AI Agents from Substrate to Production
+
+---
+type: paper
+title: Agents All the Way Down; A Methodology for Building Custom AI Agents from Substrate to Production
+authors: Marc Alier Forment, Juanan Pereira, Francisco José García-Peñalvo, María José Casañ Guerrero
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.11869
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-10
+updated_at: 2026-06-10
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - coding-agent
+  - multi-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-safety, coding-agent, multi-agent, tool-use
+- arXiv categories: cs.SE, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.11869
diff --git a/papers/items/2026-2606-12195-internvideo3-agentify-foundation-models-with-multimodal-contextual-reasoning.md b/papers/items/2026-2606-12195-internvideo3-agentify-foundation-models-with-multimodal-contextual-reasoning.md
new file mode 100644
index 0000000..96f8adc
--- /dev/null
+++ b/papers/items/2026-2606-12195-internvideo3-agentify-foundation-models-with-multimodal-contextual-reasoning.md
@@ -0,0 +1,63 @@
+# Paper: InternVideo3: Agentify Foundation Models with Multimodal Contextual Reasoning
+
+---
+type: paper
+title: "InternVideo3: Agentify Foundation Models with Multimodal Contextual Reasoning"
+authors: Ziang Yan, Sheng Xia, Jiashuo Yu, Yue Wu, Tianxiang Jiang, Songze Li, Kanghui Tian, Yicheng Xu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.12195
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-10
+updated_at: 2026-06-10
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, memory, planning, rag, reasoning, tool-use
+- arXiv categories: cs.CV
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.12195
diff --git a/papers/items/2026-2606-12320-a-five-plane-reference-architecture-for-runtime-governance-of-production-ai-agen.md b/papers/items/2026-2606-12320-a-five-plane-reference-architecture-for-runtime-governance-of-production-ai-agen.md
new file mode 100644
index 0000000..24519e3
--- /dev/null
+++ b/papers/items/2026-2606-12320-a-five-plane-reference-architecture-for-runtime-governance-of-production-ai-agen.md
@@ -0,0 +1,67 @@
+# Paper: A Five-Plane Reference Architecture for Runtime Governance of Production AI Agents
+
+---
+type: paper
+title: A Five-Plane Reference Architecture for Runtime Governance of Production AI Agents
+authors: Krti Tallam
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.12320
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-10
+updated_at: 2026-06-10
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - planning
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CC
+  - cs.CR
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, agent-safety, memory, planning, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI, cs.CC, cs.CR, cs.SE
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.12320
diff --git a/papers/items/2026-2606-12341-ocelot-inference-leakage-budgets-for-privacy-preserving-llm-agents.md b/papers/items/2026-2606-12341-ocelot-inference-leakage-budgets-for-privacy-preserving-llm-agents.md
new file mode 100644
index 0000000..57d3ea8
--- /dev/null
+++ b/papers/items/2026-2606-12341-ocelot-inference-leakage-budgets-for-privacy-preserving-llm-agents.md
@@ -0,0 +1,61 @@
+# Paper: OCELOT: Inference-Leakage Budgets for Privacy-Preserving LLM Agents
+
+---
+type: paper
+title: "OCELOT: Inference-Leakage Budgets for Privacy-Preserving LLM Agents"
+authors: Jin Xie, Songze Li
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.12341
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-10
+updated_at: 2026-06-10
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, agent-safety, reasoning, tool-use
+- arXiv categories: cs.CR
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.12341
diff --git a/papers/items/2026-2606-12344-claw-swe-bench-a-benchmark-for-evaluating-openclaw-style-agent-harnesses-on-codi.md b/papers/items/2026-2606-12344-claw-swe-bench-a-benchmark-for-evaluating-openclaw-style-agent-harnesses-on-codi.md
new file mode 100644
index 0000000..5d10d73
--- /dev/null
+++ b/papers/items/2026-2606-12344-claw-swe-bench-a-benchmark-for-evaluating-openclaw-style-agent-harnesses-on-codi.md
@@ -0,0 +1,61 @@
+# Paper: Claw-SWE-Bench: A Benchmark for Evaluating OpenClaw-style Agent Harnesses on Coding Tasks
+
+---
+type: paper
+title: "Claw-SWE-Bench: A Benchmark for Evaluating OpenClaw-style Agent Harnesses on Coding Tasks"
+authors: Mengyu Zheng, Kai Han, Boxun Li, Haiyang Xu, Yuchuan Tian, Wei He, Hang Zhou, Jianyuan Guo, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.12344
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-10
+updated_at: 2026-06-10
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, coding-agent, tool-use
+- arXiv categories: cs.LG, cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.12344
diff --git a/papers/items/2026-2606-12384-appo-agentic-procedural-policy-optimization.md b/papers/items/2026-2606-12384-appo-agentic-procedural-policy-optimization.md
new file mode 100644
index 0000000..d0227ed
--- /dev/null
+++ b/papers/items/2026-2606-12384-appo-agentic-procedural-policy-optimization.md
@@ -0,0 +1,61 @@
+# Paper: APPO: Agentic Procedural Policy Optimization
+
+---
+type: paper
+title: "APPO: Agentic Procedural Policy Optimization"
+authors: Xucong Wang, Ziyu Ma, Yong Wang, Yuxiang Ji, Shidong Yang, Guanhua Chen, Pengkun Wang, Xiangxiang Chu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.12384
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-10
+updated_at: 2026-06-10
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, tool-use, workflow-agent
+- arXiv categories: cs.LG, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.12384
diff --git a/papers/items/2026-2606-12563-arbor-tree-search-as-a-cognition-layer-for-autonomous-agents.md b/papers/items/2026-2606-12563-arbor-tree-search-as-a-cognition-layer-for-autonomous-agents.md
new file mode 100644
index 0000000..ef8d0e9
--- /dev/null
+++ b/papers/items/2026-2606-12563-arbor-tree-search-as-a-cognition-layer-for-autonomous-agents.md
@@ -0,0 +1,61 @@
+# Paper: Arbor: Tree Search as a Cognition Layer for Autonomous Agents
+
+---
+type: paper
+title: "Arbor: Tree Search as a Cognition Layer for Autonomous Agents"
+authors: Neha Prakriya, Chaojun Hou, Zheng Gong, Huasha Zhao, Xi Zhao, Mou Li, Zhenyu Gu, Emad Barsoum
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.12563
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-10
+updated_at: 2026-06-10
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - multi-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, memory, multi-agent, tool-use
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.12563
diff --git a/papers/items/2026-2606-12586-beyond-attack-success-rate-examining-trigger-leakage-in-vision-language-agentic-.md b/papers/items/2026-2606-12586-beyond-attack-success-rate-examining-trigger-leakage-in-vision-language-agentic-.md
new file mode 100644
index 0000000..96b1d51
--- /dev/null
+++ b/papers/items/2026-2606-12586-beyond-attack-success-rate-examining-trigger-leakage-in-vision-language-agentic-.md
@@ -0,0 +1,62 @@
+# Paper: Beyond Attack Success Rate: Examining Trigger Leakage in Vision-Language Agentic Systems
+
+---
+type: paper
+title: "Beyond Attack Success Rate: Examining Trigger Leakage in Vision-Language Agentic Systems"
+authors: Jiamin Chang, Salil Kanhere, Piotr Koniusz, Jason, Xue, Hammond Pearce
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.12586
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-10
+updated_at: 2026-06-10
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - embodied-agent
+  - planning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: language-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent, tool-use
+- inferred topics: agent-evaluation, embodied-agent, planning, tool-use, workflow-agent
+- arXiv categories: cs.CR
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.12586
diff --git a/papers/items/2026-2606-12634-keep-policy-gradient-in-charge-sibling-guided-credit-distillation-for-long-horiz.md b/papers/items/2026-2606-12634-keep-policy-gradient-in-charge-sibling-guided-credit-distillation-for-long-horiz.md
new file mode 100644
index 0000000..842fd4a
--- /dev/null
+++ b/papers/items/2026-2606-12634-keep-policy-gradient-in-charge-sibling-guided-credit-distillation-for-long-horiz.md
@@ -0,0 +1,64 @@
+# Paper: Keep Policy Gradient in Charge: Sibling-Guided Credit Distillation for Long-Horizon Tool-Use Agents
+
+---
+type: paper
+title: "Keep Policy Gradient in Charge: Sibling-Guided Credit Distillation for Long-Horizon Tool-Use Agents"
+authors: Tianyu Ding, Jianhong Xin, Juan Pablo De la Cruz Weinstein
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.12634
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-10
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, computer-use, planning, reasoning, tool-use
+- arXiv categories: cs.LG, cs.AI, cs.CL
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.12634
diff --git a/papers/items/2026-2606-12657-trajgenagent-a-hierarchical-llm-agent-for-human-mobility-trajectory-generation.md b/papers/items/2026-2606-12657-trajgenagent-a-hierarchical-llm-agent-for-human-mobility-trajectory-generation.md
new file mode 100644
index 0000000..a7fc4fa
--- /dev/null
+++ b/papers/items/2026-2606-12657-trajgenagent-a-hierarchical-llm-agent-for-human-mobility-trajectory-generation.md
@@ -0,0 +1,65 @@
+# Paper: TrajGenAgent: A Hierarchical LLM Agent for Human Mobility Trajectory Generation
+
+---
+type: paper
+title: "TrajGenAgent: A Hierarchical LLM Agent for Human Mobility Trajectory Generation"
+authors: Siyu Li, Toan Tran, Lingyi Zhao, Khurram Shafique, Li Xiong
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.12657
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-10
+updated_at: 2026-06-10
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - reasoning
+  - workflow-agent
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.DB
+  - cs.RO
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, planning, rag, reasoning, workflow-agent, world-model
+- arXiv categories: cs.AI, cs.DB, cs.RO
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.12657
diff --git a/papers/items/2026-2606-12674-evoflux-inference-time-evolution-of-executable-tool-workflows-for-compact-agents.md b/papers/items/2026-2606-12674-evoflux-inference-time-evolution-of-executable-tool-workflows-for-compact-agents.md
new file mode 100644
index 0000000..6cf5ffa
--- /dev/null
+++ b/papers/items/2026-2606-12674-evoflux-inference-time-evolution-of-executable-tool-workflows-for-compact-agents.md
@@ -0,0 +1,62 @@
+# Paper: Evoflux: Inference-Time Evolution of Executable Tool Workflows for Compact Agents
+
+---
+type: paper
+title: "Evoflux: Inference-Time Evolution of Executable Tool Workflows for Compact Agents"
+authors: Kushal Raj Bhandari, Ling Yue, Ching-Yun Ko, Dhaval Patel, Shaowu Pan, Pin-Yu Chen, Jianxi Gao
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.12674
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-10
+updated_at: 2026-06-10
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - computer-use
+  - planning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: function-calling, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling, tool-use
+- inferred topics: agent-safety, computer-use, planning, tool-use, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.12674
diff --git a/papers/items/2026-2606-12703-smsr-certified-defence-against-runtime-memory-poisoning-in-persistent-llm-agent-.md b/papers/items/2026-2606-12703-smsr-certified-defence-against-runtime-memory-poisoning-in-persistent-llm-agent-.md
new file mode 100644
index 0000000..407fe1a
--- /dev/null
+++ b/papers/items/2026-2606-12703-smsr-certified-defence-against-runtime-memory-poisoning-in-persistent-llm-agent-.md
@@ -0,0 +1,63 @@
+# Paper: SMSR: Certified Defence Against Runtime Memory Poisoning in Persistent LLM Agent Systems
+
+---
+type: paper
+title: "SMSR: Certified Defence Against Runtime Memory Poisoning in Persistent LLM Agent Systems"
+authors: Tarun Sharma
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.12703
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-10
+updated_at: 2026-06-10
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, memory, rag, workflow-agent
+- arXiv categories: cs.CR, cs.AI, cs.LG
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.12703
diff --git a/papers/items/2026-2606-12780-proplay-procedural-world-models-for-self-evolving-llm-agents.md b/papers/items/2026-2606-12780-proplay-procedural-world-models-for-self-evolving-llm-agents.md
new file mode 100644
index 0000000..0c2c539
--- /dev/null
+++ b/papers/items/2026-2606-12780-proplay-procedural-world-models-for-self-evolving-llm-agents.md
@@ -0,0 +1,64 @@
+# Paper: ProPlay: Procedural World Models for Self-Evolving LLM Agents
+
+---
+type: paper
+title: "ProPlay: Procedural World Models for Self-Evolving LLM Agents"
+authors: Yijun Ma, Zehong Wang, Yiyang Li, Ziming Li, Xiaoguang Guo, Weixiang Sun, Chuxu Zhang, Yanfang Ye
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.12780
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-11
+updated_at: 2026-06-11
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - planning
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, computer-use, memory, planning, tool-use, world-model
+- arXiv categories: cs.LG, cs.CL
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.12780
diff --git a/papers/items/2026-2606-12837-lohosearch-benchmarking-long-horizon-search-agents-beyond-the-human-difficulty-c.md b/papers/items/2026-2606-12837-lohosearch-benchmarking-long-horizon-search-agents-beyond-the-human-difficulty-c.md
new file mode 100644
index 0000000..0b82351
--- /dev/null
+++ b/papers/items/2026-2606-12837-lohosearch-benchmarking-long-horizon-search-agents-beyond-the-human-difficulty-c.md
@@ -0,0 +1,61 @@
+# Paper: LoHoSearch: Benchmarking Long-Horizon Search Agents Beyond the Human Difficulty Ceiling
+
+---
+type: paper
+title: "LoHoSearch: Benchmarking Long-Horizon Search Agents Beyond the Human Difficulty Ceiling"
+authors: Jiarui Zhao, Rongzhi Zhang, Lingchuan Liu, Hao Yang, Xunliang Cai, Xi Su
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.12837
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-11
+updated_at: 2026-06-17
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, planning, reasoning, tool-use
+- arXiv categories: cs.CL
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.12837
diff --git a/papers/items/2026-2606-12945-learning-what-to-remember-a-cognitively-grounded-multi-factor-value-model-for-ag.md b/papers/items/2026-2606-12945-learning-what-to-remember-a-cognitively-grounded-multi-factor-value-model-for-ag.md
new file mode 100644
index 0000000..523d13d
--- /dev/null
+++ b/papers/items/2026-2606-12945-learning-what-to-remember-a-cognitively-grounded-multi-factor-value-model-for-ag.md
@@ -0,0 +1,64 @@
+# Paper: Learning What to Remember: A Cognitively Grounded Multi-Factor Value Model for Agentic Memory
+
+---
+type: paper
+title: "Learning What to Remember: A Cognitively Grounded Multi-Factor Value Model for Agentic Memory"
+authors: Zhibao Chen, Qian Cheng
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.12945
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-11
+updated_at: 2026-06-20
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - memory
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 22
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, agent-safety, coding-agent, memory, planning, rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 22
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.12945
diff --git a/papers/items/2026-2606-13148-terrabench-can-agents-reason-over-heterogeneous-earth-system-data.md b/papers/items/2026-2606-13148-terrabench-can-agents-reason-over-heterogeneous-earth-system-data.md
new file mode 100644
index 0000000..9195c80
--- /dev/null
+++ b/papers/items/2026-2606-13148-terrabench-can-agents-reason-over-heterogeneous-earth-system-data.md
@@ -0,0 +1,64 @@
+# Paper: TerraBench: Can Agents Reason Over Heterogeneous Earth-System Data?
+
+---
+type: paper
+title: "TerraBench: Can Agents Reason Over Heterogeneous Earth-System Data?"
+authors: Dat Tien Nguyen, Thao Nguyen, Fadillah Adamsyah Maani, Huy M. Le, Muhammad Umer Sheikh, Numan Saeed, Muhammad Haris Khan, Salman Khan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.13148
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-11
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, planning, rag, reasoning, tool-use, workflow-agent, world-model
+- arXiv categories: cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.13148
diff --git a/papers/items/2026-2606-13177-memrefine-llm-guided-compression-for-long-term-agent-memory.md b/papers/items/2026-2606-13177-memrefine-llm-guided-compression-for-long-term-agent-memory.md
new file mode 100644
index 0000000..63ffb88
--- /dev/null
+++ b/papers/items/2026-2606-13177-memrefine-llm-guided-compression-for-long-term-agent-memory.md
@@ -0,0 +1,64 @@
+# Paper: MemRefine: LLM-Guided Compression for Long-Term Agent Memory
+
+---
+type: paper
+title: "MemRefine: LLM-Guided Compression for Long-Term Agent Memory"
+authors: Minjae Kim, Jinheon Baek, Soyeong Jeong, Sung Ju Hwang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.13177
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-11
+updated_at: 2026-06-11
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, computer-use, memory, rag, tool-use
+- arXiv categories: cs.CL, cs.AI, cs.LG
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.13177
diff --git a/papers/items/2026-2606-13192-reasoning-for-mobile-user-experience-with-multimodal-llms-task-benchmark-and-app.md b/papers/items/2026-2606-13192-reasoning-for-mobile-user-experience-with-multimodal-llms-task-benchmark-and-app.md
new file mode 100644
index 0000000..a422e35
--- /dev/null
+++ b/papers/items/2026-2606-13192-reasoning-for-mobile-user-experience-with-multimodal-llms-task-benchmark-and-app.md
@@ -0,0 +1,62 @@
+# Paper: Reasoning for Mobile User Experience with Multimodal LLMs: Task, Benchmark, and Approach
+
+---
+type: paper
+title: "Reasoning for Mobile User Experience with Multimodal LLMs: Task, Benchmark, and Approach"
+authors: Ruichao Mao, Zhou Fang, Teng Guo, Hao Yang, Yaping Li, Shaohua Peng, Maji Huang, Xiaoyu Lin, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.13192
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-11
+updated_at: 2026-06-11
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, computer-use, rag, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.13192
diff --git a/papers/items/2026-2606-13317-skillcat-contrastive-assessment-and-topology-aware-skill-self-evolution-for-llm-.md b/papers/items/2026-2606-13317-skillcat-contrastive-assessment-and-topology-aware-skill-self-evolution-for-llm-.md
new file mode 100644
index 0000000..7176efb
--- /dev/null
+++ b/papers/items/2026-2606-13317-skillcat-contrastive-assessment-and-topology-aware-skill-self-evolution-for-llm-.md
@@ -0,0 +1,60 @@
+# Paper: SkillCAT: Contrastive Assessment and Topology-Aware Skill Self-Evolution for LLM Agents
+
+---
+type: paper
+title: "SkillCAT: Contrastive Assessment and Topology-Aware Skill Self-Evolution for LLM Agents"
+authors: Kunfeng Chen, Qihuang Zhong, Juhua Liu, Bo Du
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.13317
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-11
+updated_at: 2026-06-11
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, rag, tool-use
+- arXiv categories: cs.CL
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.13317
diff --git a/papers/items/2026-2606-13385-who-pays-the-price-stakeholder-centric-prompt-injection-benchmarking-for-real-wo.md b/papers/items/2026-2606-13385-who-pays-the-price-stakeholder-centric-prompt-injection-benchmarking-for-real-wo.md
new file mode 100644
index 0000000..5568ab3
--- /dev/null
+++ b/papers/items/2026-2606-13385-who-pays-the-price-stakeholder-centric-prompt-injection-benchmarking-for-real-wo.md
@@ -0,0 +1,65 @@
+# Paper: Who Pays the Price? Stakeholder-Centric Prompt Injection Benchmarking for Real-world Web Agents
+
+---
+type: paper
+title: Who Pays the Price? Stakeholder-Centric Prompt Injection Benchmarking for Real-world Web Agents
+authors: Zihao Wang, Yiming Li, Yutong Wu, Zheyu Liu, Kangjie Chen, Fok Kar Wai, Pin-Yu Chen, Vrizlynn L. L. Thing, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.13385
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-11
+updated_at: 2026-06-11
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+  - cs.CY
+  - cs.HC
+  - cs.MM
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, agent-safety, computer-use, tool-use
+- arXiv categories: cs.CR, cs.AI, cs.CY, cs.HC, cs.MM
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.13385
diff --git a/papers/items/2026-2606-13602-epibench-verifiable-evaluation-of-ai-agents-on-epigenomics-analysis.md b/papers/items/2026-2606-13602-epibench-verifiable-evaluation-of-ai-agents-on-epigenomics-analysis.md
new file mode 100644
index 0000000..dcfc0f2
--- /dev/null
+++ b/papers/items/2026-2606-13602-epibench-verifiable-evaluation-of-ai-agents-on-epigenomics-analysis.md
@@ -0,0 +1,59 @@
+# Paper: EpiBench: Verifiable Evaluation of AI Agents on Epigenomics Analysis
+
+---
+type: paper
+title: "EpiBench: Verifiable Evaluation of AI Agents on Epigenomics Analysis"
+authors: Harihara Muralidharan, Reema Baskar, Soo Hee Lee, Tim Proctor, Kenny Workman
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.13602
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-11
+updated_at: 2026-06-11
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.13602
diff --git a/papers/items/2026-2606-13608-agentbeats-agentifying-agent-assessment-for-openness-standardization-and-reprodu.md b/papers/items/2026-2606-13608-agentbeats-agentifying-agent-assessment-for-openness-standardization-and-reprodu.md
new file mode 100644
index 0000000..44e1e5f
--- /dev/null
+++ b/papers/items/2026-2606-13608-agentbeats-agentifying-agent-assessment-for-openness-standardization-and-reprodu.md
@@ -0,0 +1,63 @@
+# Paper: AgentBeats: Agentifying Agent Assessment for Openness, Standardization, and Reproducibility
+
+---
+type: paper
+title: "AgentBeats: Agentifying Agent Assessment for Openness, Standardization, and Reproducibility"
+authors: Xiaoyuan Liu, Jianhong Tu, Yuqi Chen, Siyuan Xie, Sihan Ren, Tianneng Shi, Gal Gantar, Evan Sandoval, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.13608
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-11
+updated_at: 2026-06-14
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - multi-agent
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, coding-agent, multi-agent, rag, tool-use
+- arXiv categories: cs.AI, cs.LG
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.13608
diff --git a/papers/items/2026-2606-13643-recursive-agent-harnesses.md b/papers/items/2026-2606-13643-recursive-agent-harnesses.md
new file mode 100644
index 0000000..82f1a33
--- /dev/null
+++ b/papers/items/2026-2606-13643-recursive-agent-harnesses.md
@@ -0,0 +1,63 @@
+# Paper: Recursive Agent Harnesses
+
+---
+type: paper
+title: Recursive Agent Harnesses
+authors: Elias Lumer, Sahil Sen, Kevin Paul, Vamse Kumar Subbiah
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.13643
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-11
+updated_at: 2026-06-11
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - planning
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling
+- inferred topics: agent-evaluation, coding-agent, planning, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.CL
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.13643
diff --git a/papers/items/2026-2606-13663-hypertool-beyond-step-wise-tool-calls-for-tool-augmented-agents.md b/papers/items/2026-2606-13663-hypertool-beyond-step-wise-tool-calls-for-tool-augmented-agents.md
new file mode 100644
index 0000000..b151d94
--- /dev/null
+++ b/papers/items/2026-2606-13663-hypertool-beyond-step-wise-tool-calls-for-tool-augmented-agents.md
@@ -0,0 +1,61 @@
+# Paper: HyperTool: Beyond Step-Wise Tool Calls for Tool-Augmented Agents
+
+---
+type: paper
+title: "HyperTool: Beyond Step-Wise Tool Calls for Tool-Augmented Agents"
+authors: Yaxin Du, Yifan Zhou, Yujie Ge, Jiajun Wang, Xianghe Pang, Shuo Tang, Tuney Zheng, Bryan Dai, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.13663
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-11
+updated_at: 2026-06-11
+status: queued
+relevance: high
+topics:
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.CL
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.13663
diff --git a/papers/items/2026-2606-13686-benchmarking-web-agent-safety-under-e-commerce-deceptive-interfaces.md b/papers/items/2026-2606-13686-benchmarking-web-agent-safety-under-e-commerce-deceptive-interfaces.md
new file mode 100644
index 0000000..df750e4
--- /dev/null
+++ b/papers/items/2026-2606-13686-benchmarking-web-agent-safety-under-e-commerce-deceptive-interfaces.md
@@ -0,0 +1,61 @@
+# Paper: Benchmarking Web Agent Safety under E-commerce Deceptive Interfaces
+
+---
+type: paper
+title: Benchmarking Web Agent Safety under E-commerce Deceptive Interfaces
+authors: Zijing Shi, Meng Fang, Ling Chen
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.13686
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-26
+updated_at: 2026-04-26
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.CY
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, computer-use
+- arXiv categories: cs.CL, cs.CY
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.13686
diff --git a/papers/items/2026-2606-13904-sana-what-matters-for-qa-agents-over-massive-data-lakes.md b/papers/items/2026-2606-13904-sana-what-matters-for-qa-agents-over-massive-data-lakes.md
new file mode 100644
index 0000000..94b6935
--- /dev/null
+++ b/papers/items/2026-2606-13904-sana-what-matters-for-qa-agents-over-massive-data-lakes.md
@@ -0,0 +1,64 @@
+# Paper: SANA: What Matters for QA Agents over Massive Data Lakes?
+
+---
+type: paper
+title: "SANA: What Matters for QA Agents over Massive Data Lakes?"
+authors: Austin Senna Wijaya, Jiaxiang Liu, Haonan Wang, Eugene Wu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.13904
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-11
+updated_at: 2026-06-11
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - embodied-agent
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+  - cs.DB
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, computer-use, embodied-agent, planning, tool-use
+- arXiv categories: cs.CL, cs.AI, cs.DB
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.13904
diff --git a/papers/items/2026-2606-13994-hidden-in-plain-sight-benchmarking-agent-safety-against-decomposition-attacks-wi.md b/papers/items/2026-2606-13994-hidden-in-plain-sight-benchmarking-agent-safety-against-decomposition-attacks-wi.md
new file mode 100644
index 0000000..022115a
--- /dev/null
+++ b/papers/items/2026-2606-13994-hidden-in-plain-sight-benchmarking-agent-safety-against-decomposition-attacks-wi.md
@@ -0,0 +1,64 @@
+# Paper: Hidden in Plain Sight: Benchmarking Agent Safety Against Decomposition Attacks with DECOMPBENCH
+
+---
+type: paper
+title: "Hidden in Plain Sight: Benchmarking Agent Safety Against Decomposition Attacks with DECOMPBENCH"
+authors: Vikhyath Kothamasu, Virginia Smith, Chhavi Yadav
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.13994
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-12
+updated_at: 2026-06-12
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 21
+collection_queries: agent-safety, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety, tool-use
+- inferred topics: agent-evaluation, agent-safety, planning, tool-use, workflow-agent
+- arXiv categories: cs.CR, cs.AI, cs.LG
+- collection score: 21
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.13994
diff --git a/papers/items/2026-2606-14106-naive-visual-memory-is-not-enough-a-failure-mode-study-of-gui-agents.md b/papers/items/2026-2606-14106-naive-visual-memory-is-not-enough-a-failure-mode-study-of-gui-agents.md
new file mode 100644
index 0000000..869f5b7
--- /dev/null
+++ b/papers/items/2026-2606-14106-naive-visual-memory-is-not-enough-a-failure-mode-study-of-gui-agents.md
@@ -0,0 +1,63 @@
+# Paper: Naive Visual Memory is Not Enough: A Failure-Mode Study of GUI Agents
+
+---
+type: paper
+title: "Naive Visual Memory is Not Enough: A Failure-Mode Study of GUI Agents"
+authors: Seoyoung Choi, Minseok Ko, Hyunseok Lee, Kunwoong Kim, Woomin Song, Chanseok Jeon, Jinwoo Shin
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.14106
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-12
+updated_at: 2026-06-12
+status: queued
+relevance: high
+topics:
+  - computer-use
+  - memory
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+  - cs.CV
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: computer-use, memory, rag, reasoning, tool-use
+- arXiv categories: cs.MA, cs.CV
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.14106
diff --git a/papers/items/2026-2606-14470-gitofthoughts-version-controlled-reasoning-and-agent-memory-you-can-replay-diff-.md b/papers/items/2026-2606-14470-gitofthoughts-version-controlled-reasoning-and-agent-memory-you-can-replay-diff-.md
new file mode 100644
index 0000000..0cf043b
--- /dev/null
+++ b/papers/items/2026-2606-14470-gitofthoughts-version-controlled-reasoning-and-agent-memory-you-can-replay-diff-.md
@@ -0,0 +1,64 @@
+# Paper: GitOfThoughts: Version-Controlled Reasoning and Agent Memory You Can Replay, Diff, and Merge
+
+---
+type: paper
+title: "GitOfThoughts: Version-Controlled Reasoning and Agent Memory You Can Replay, Diff, and Merge"
+authors: Pavan C Shekar, Abhishek H S, Aswanth Krishnan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.14470
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-12
+updated_at: 2026-06-22
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - memory
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, coding-agent, memory, rag, reasoning
+- arXiv categories: cs.AI, cs.CL, cs.LG
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.14470
diff --git a/papers/items/2026-2606-14502-from-chatbot-to-digital-colleague-the-paradigm-shift-toward-persistent-autonomou.md b/papers/items/2026-2606-14502-from-chatbot-to-digital-colleague-the-paradigm-shift-toward-persistent-autonomou.md
new file mode 100644
index 0000000..d4d3ea3
--- /dev/null
+++ b/papers/items/2026-2606-14502-from-chatbot-to-digital-colleague-the-paradigm-shift-toward-persistent-autonomou.md
@@ -0,0 +1,62 @@
+# Paper: From Chatbot to Digital Colleague: The Paradigm Shift Toward Persistent Autonomous AI
+
+---
+type: paper
+title: "From Chatbot to Digital Colleague: The Paradigm Shift Toward Persistent Autonomous AI"
+authors: Yongheng Zhang, Ziang Liu, Jiaxuan Zhu, Shuai Wang, Xiangqi Chen, Haojing Huang, Jiayi Kuang, Siyu Chen, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.14502
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-12
+updated_at: 2026-06-12
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, memory, rag, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.14502
diff --git a/papers/items/2026-2606-14517-from-shield-to-target-denial-of-service-attacks-on-llm-based-agent-guardrails.md b/papers/items/2026-2606-14517-from-shield-to-target-denial-of-service-attacks-on-llm-based-agent-guardrails.md
new file mode 100644
index 0000000..0ed47d1
--- /dev/null
+++ b/papers/items/2026-2606-14517-from-shield-to-target-denial-of-service-attacks-on-llm-based-agent-guardrails.md
@@ -0,0 +1,63 @@
+# Paper: From Shield to Target: Denial-of-Service Attacks on LLM-Based Agent Guardrails
+
+---
+type: paper
+title: "From Shield to Target: Denial-of-Service Attacks on LLM-Based Agent Guardrails"
+authors: Yuguang Zhou, Xunguang Wang, Pingchuan Ma, Zhantong Xue, Zhaoyu Wang, Shuai Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.14517
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-12
+updated_at: 2026-06-16
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - multi-agent
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-evaluation, autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, autonomous-agent-llm
+- inferred topics: agent-evaluation, agent-safety, computer-use, multi-agent, reasoning
+- arXiv categories: cs.CR, cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.14517
diff --git a/papers/items/2026-2606-14571-streammembench-streaming-evaluation-of-agent-memory-for-future-oriented-assistan.md b/papers/items/2026-2606-14571-streammembench-streaming-evaluation-of-agent-memory-for-future-oriented-assistan.md
new file mode 100644
index 0000000..8d19281
--- /dev/null
+++ b/papers/items/2026-2606-14571-streammembench-streaming-evaluation-of-agent-memory-for-future-oriented-assistan.md
@@ -0,0 +1,60 @@
+# Paper: StreamMemBench: Streaming Evaluation of Agent Memory for Future-Oriented Assistance
+
+---
+type: paper
+title: "StreamMemBench: Streaming Evaluation of Agent Memory for Future-Oriented Assistance"
+authors: Guanming Liu, Yuqi Ren, Hansu Gu, Peng Zhang, Weihang Wang, Jiahao Liu, Ning Gu, Tun Lu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.14571
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-12
+updated_at: 2026-06-12
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, tool-use
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.14571
diff --git a/papers/items/2026-2606-14574-simmer-benchmarking-latent-failures-in-llm-executable-planning-with-a-world-mode.md b/papers/items/2026-2606-14574-simmer-benchmarking-latent-failures-in-llm-executable-planning-with-a-world-mode.md
new file mode 100644
index 0000000..8e1ec38
--- /dev/null
+++ b/papers/items/2026-2606-14574-simmer-benchmarking-latent-failures-in-llm-executable-planning-with-a-world-mode.md
@@ -0,0 +1,64 @@
+# Paper: SIMMER: Benchmarking Latent Failures in LLM Executable Planning with a World Model
+
+---
+type: paper
+title: "SIMMER: Benchmarking Latent Failures in LLM Executable Planning with a World Model"
+authors: Xiaoxin Lu, Ranran Haoran Zhang, Rui Zhang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.14574
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-12
+updated_at: 2026-06-12
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, planning, rag, reasoning, tool-use, world-model
+- arXiv categories: cs.CL, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.14574
diff --git a/papers/items/2026-2606-14790-xflow-an-executable-protocol-programming-system-for-reliable-multi-agent-workflo.md b/papers/items/2026-2606-14790-xflow-an-executable-protocol-programming-system-for-reliable-multi-agent-workflo.md
new file mode 100644
index 0000000..bb51728
--- /dev/null
+++ b/papers/items/2026-2606-14790-xflow-an-executable-protocol-programming-system-for-reliable-multi-agent-workflo.md
@@ -0,0 +1,65 @@
+# Paper: XFlow: An Executable Protocol Programming System for Reliable Multi-Agent Workflows
+
+---
+type: paper
+title: "XFlow: An Executable Protocol Programming System for Reliable Multi-Agent Workflows"
+authors: Hanqi Li, Jing Peng, Zijian Wang, Lu Chen, Kai Yu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.14790
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-11
+updated_at: 2026-06-11
+status: queued
+relevance: high
+topics:
+  - coding-agent
+  - memory
+  - multi-agent
+  - planning
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.PL
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: coding-agent, memory, multi-agent, planning, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.PL, cs.AI
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.14790
diff --git a/papers/items/2026-2606-14805-knowledge-based-zero-replay-debugging-of-multi-agent-llm-traces.md b/papers/items/2026-2606-14805-knowledge-based-zero-replay-debugging-of-multi-agent-llm-traces.md
new file mode 100644
index 0000000..de3fdc7
--- /dev/null
+++ b/papers/items/2026-2606-14805-knowledge-based-zero-replay-debugging-of-multi-agent-llm-traces.md
@@ -0,0 +1,61 @@
+# Paper: Knowledge-Based Zero-Replay Debugging of Multi-Agent LLM Traces
+
+---
+type: paper
+title: Knowledge-Based Zero-Replay Debugging of Multi-Agent LLM Traces
+authors: Dong Ho Kang, Hyeonjeong Cha, Daein Weon
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.14805
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-11
+updated_at: 2026-06-11
+status: queued
+relevance: high
+topics:
+  - memory
+  - multi-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: memory, multi-agent, tool-use
+- arXiv categories: cs.SE, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.14805
diff --git a/papers/items/2026-2606-15017-are-online-skill-and-memory-modules-always-worth-their-tokens-a-budget-constrain.md b/papers/items/2026-2606-15017-are-online-skill-and-memory-modules-always-worth-their-tokens-a-budget-constrain.md
new file mode 100644
index 0000000..5910761
--- /dev/null
+++ b/papers/items/2026-2606-15017-are-online-skill-and-memory-modules-always-worth-their-tokens-a-budget-constrain.md
@@ -0,0 +1,62 @@
+# Paper: Are Online Skill and Memory Modules Always Worth Their Tokens? A Budget-Constrained Study of Web Agents
+
+---
+type: paper
+title: Are Online Skill and Memory Modules Always Worth Their Tokens? A Budget-Constrained Study of Web Agents
+authors: Sina Hajimiri, Masih Aminbeidokhti, Jose Dolz, Ismail Ben Ayed, Issam H. Laradji, Spandana Gella, Nicolas Gontier
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.15017
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-12
+updated_at: 2026-06-12
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - reasoning
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, computer-use, memory, reasoning, workflow-agent
+- arXiv categories: cs.CL
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.15017
diff --git a/papers/items/2026-2606-15034-osguard-a-benchmark-for-safety-in-computer-use-agents.md b/papers/items/2026-2606-15034-osguard-a-benchmark-for-safety-in-computer-use-agents.md
new file mode 100644
index 0000000..22b31ed
--- /dev/null
+++ b/papers/items/2026-2606-15034-osguard-a-benchmark-for-safety-in-computer-use-agents.md
@@ -0,0 +1,61 @@
+# Paper: OSGuard: A Benchmark for Safety in Computer-Use Agents
+
+---
+type: paper
+title: "OSGuard: A Benchmark for Safety in Computer-Use Agents"
+authors: Mina Mohammadmirzaei, Jeffrey Flanigan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.15034
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-13
+updated_at: 2026-06-13
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, agent-safety, computer-use, tool-use
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.15034
diff --git a/papers/items/2026-2606-15079-ling-and-ring-2-6-technical-report-efficient-and-instant-agentic-intelligence-at.md b/papers/items/2026-2606-15079-ling-and-ring-2-6-technical-report-efficient-and-instant-agentic-intelligence-at.md
new file mode 100644
index 0000000..1ebf1d2
--- /dev/null
+++ b/papers/items/2026-2606-15079-ling-and-ring-2-6-technical-report-efficient-and-instant-agentic-intelligence-at.md
@@ -0,0 +1,64 @@
+# Paper: Ling and Ring 2.6 Technical Report: Efficient and Instant Agentic Intelligence at Trillion-Parameter Scale
+
+---
+type: paper
+title: "Ling and Ring 2.6 Technical Report: Efficient and Instant Agentic Intelligence at Trillion-Parameter Scale"
+authors: Ang Li, Ben Liu, Bin Han, Bin Hu, Bin Jing, Binbin Hu, Bing Li, Cai Chen, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.15079
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-13
+updated_at: 2026-06-13
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - coding-agent
+  - computer-use
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-safety, coding-agent, computer-use, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.CL, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.15079
diff --git a/papers/items/2026-2606-15152-can-agents-read-the-room-benchmarking-visual-social-intelligence-in-multimodal-s.md b/papers/items/2026-2606-15152-can-agents-read-the-room-benchmarking-visual-social-intelligence-in-multimodal-s.md
new file mode 100644
index 0000000..adfa810
--- /dev/null
+++ b/papers/items/2026-2606-15152-can-agents-read-the-room-benchmarking-visual-social-intelligence-in-multimodal-s.md
@@ -0,0 +1,61 @@
+# Paper: Can Agents Read the Room? Benchmarking Visual Social Intelligence in Multimodal Simulation
+
+---
+type: paper
+title: Can Agents Read the Room? Benchmarking Visual Social Intelligence in Multimodal Simulation
+authors: Shijun Wan, Xuehai Wu, Jiwen Zhang, Siyuan Wang, Zhongyu Wei
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.15152
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-13
+updated_at: 2026-06-13
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, computer-use, tool-use, world-model
+- arXiv categories: cs.CL
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.15152
diff --git a/papers/items/2026-2606-15242-benign-in-isolation-harmful-in-composition-security-risks-in-agent-skill-ecosyst.md b/papers/items/2026-2606-15242-benign-in-isolation-harmful-in-composition-security-risks-in-agent-skill-ecosyst.md
new file mode 100644
index 0000000..fc1180a
--- /dev/null
+++ b/papers/items/2026-2606-15242-benign-in-isolation-harmful-in-composition-security-risks-in-agent-skill-ecosyst.md
@@ -0,0 +1,62 @@
+# Paper: Benign in Isolation, Harmful in Composition: Security Risks in Agent Skill Ecosystems
+
+---
+type: paper
+title: "Benign in Isolation, Harmful in Composition: Security Risks in Agent Skill Ecosystems"
+authors: Yi Xie, Jiawei Du, Yu Cheng, Jiuan Zhou, Zhaoxia Yin
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.15242
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-13
+updated_at: 2026-06-13
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, agent-safety, planning, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.15242
diff --git a/papers/items/2026-2606-15376-coagent-concurrency-control-for-multi-agent-systems.md b/papers/items/2026-2606-15376-coagent-concurrency-control-for-multi-agent-systems.md
new file mode 100644
index 0000000..ddaf47e
--- /dev/null
+++ b/papers/items/2026-2606-15376-coagent-concurrency-control-for-multi-agent-systems.md
@@ -0,0 +1,63 @@
+# Paper: CoAgent: Concurrency Control for Multi-Agent Systems
+
+---
+type: paper
+title: "CoAgent: Concurrency Control for Multi-Agent Systems"
+authors: Hongtao Lyu, Dingyan Zhang, Mingyu Wu, Xingda Wei, Haibo Chen
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.15376
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-13
+updated_at: 2026-06-13
+status: queued
+relevance: high
+topics:
+  - coding-agent
+  - multi-agent
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.DC
+  - cs.AI
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: coding-agent, multi-agent, planning, tool-use
+- arXiv categories: cs.DC, cs.AI, cs.MA
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.15376
diff --git a/papers/items/2026-2606-15591-agentic-retrieval-and-reinforcement-learned-equation-chains-a-controlled-generat.md b/papers/items/2026-2606-15591-agentic-retrieval-and-reinforcement-learned-equation-chains-a-controlled-generat.md
new file mode 100644
index 0000000..8aabbcb
--- /dev/null
+++ b/papers/items/2026-2606-15591-agentic-retrieval-and-reinforcement-learned-equation-chains-a-controlled-generat.md
@@ -0,0 +1,62 @@
+# Paper: Agentic Retrieval and Reinforcement Learned Equation Chains: A Controlled Generation Framework for Complex and Novel Physics Word Problems
+
+---
+type: paper
+title: "Agentic Retrieval and Reinforcement Learned Equation Chains: A Controlled Generation Framework for Complex and Novel Physics Word Problems"
+authors: Tirthankar Mittra
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.15591
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-14
+updated_at: 2026-06-14
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, computer-use, rag
+- arXiv categories: cs.AI, cs.CL, cs.MA
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.15591
diff --git a/papers/items/2026-2606-15609-fragfuse-bypassing-access-control-of-large-language-model-agents-via-memory-base.md b/papers/items/2026-2606-15609-fragfuse-bypassing-access-control-of-large-language-model-agents-via-memory-base.md
new file mode 100644
index 0000000..e8688cc
--- /dev/null
+++ b/papers/items/2026-2606-15609-fragfuse-bypassing-access-control-of-large-language-model-agents-via-memory-base.md
@@ -0,0 +1,62 @@
+# Paper: FragFuse: Bypassing Access Control of Large Language Model Agents via Memory-Based Query Fragmentation and Fusion
+
+---
+type: paper
+title: "FragFuse: Bypassing Access Control of Large Language Model Agents via Memory-Based Query Fragmentation and Fusion"
+authors: Zixin Rao, Wentian Zhu, Chan Aristella Lu, Zhaorun Chen, Wei Niu, Le Guan, Bo Li, Zhen Xiang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.15609
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-14
+updated_at: 2026-06-14
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, rag, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.15609
diff --git a/papers/items/2026-2606-15684-multi-agent-framework-for-time-sensitive-complementary-collaboration-in-minecraf.md b/papers/items/2026-2606-15684-multi-agent-framework-for-time-sensitive-complementary-collaboration-in-minecraf.md
new file mode 100644
index 0000000..8178d58
--- /dev/null
+++ b/papers/items/2026-2606-15684-multi-agent-framework-for-time-sensitive-complementary-collaboration-in-minecraf.md
@@ -0,0 +1,62 @@
+# Paper: Multi-agent Framework for Time-Sensitive Complementary Collaboration in Minecraft
+
+---
+type: paper
+title: Multi-agent Framework for Time-Sensitive Complementary Collaboration in Minecraft
+authors: Juheon Yi, Jinglu Wang, Xiaoyi Zhang, Yan Lu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.15684
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-14
+updated_at: 2026-06-14
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, agent-safety, multi-agent, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.15684
diff --git a/papers/items/2026-2606-15709-ai-driven-framework-for-adaptive-water-network-management-with-proof-of-concept-.md b/papers/items/2026-2606-15709-ai-driven-framework-for-adaptive-water-network-management-with-proof-of-concept-.md
new file mode 100644
index 0000000..052648f
--- /dev/null
+++ b/papers/items/2026-2606-15709-ai-driven-framework-for-adaptive-water-network-management-with-proof-of-concept-.md
@@ -0,0 +1,64 @@
+# Paper: AI-Driven Framework for Adaptive Water Network Management with Proof-of-Concept Implementation: Addressing Non-Revenue Water in Jordan
+
+---
+type: paper
+title: "AI-Driven Framework for Adaptive Water Network Management with Proof-of-Concept Implementation: Addressing Non-Revenue Water in Jordan"
+authors: Mohammed Fasha, Nahel Al-Maayta, Bilal Sowan, Mohammad Athamneh, Husam Barham
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.15709
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-14
+updated_at: 2026-06-14
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - rag
+  - tool-use
+  - workflow-agent
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: function-calling, rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling, rag-agent
+- inferred topics: agent-evaluation, agent-safety, rag, tool-use, workflow-agent, world-model
+- arXiv categories: cs.AI, cs.MA
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.15709
diff --git a/papers/items/2026-2606-15862-retailbench-benchmarking-long-horizon-reasoning-and-coherent-decision-making-of-.md b/papers/items/2026-2606-15862-retailbench-benchmarking-long-horizon-reasoning-and-coherent-decision-making-of-.md
new file mode 100644
index 0000000..f729ca4
--- /dev/null
+++ b/papers/items/2026-2606-15862-retailbench-benchmarking-long-horizon-reasoning-and-coherent-decision-making-of-.md
@@ -0,0 +1,62 @@
+# Paper: RetailBench: Benchmarking long horizon reasoning and coherent decision making of LLM agents in realistic retail environments
+
+---
+type: paper
+title: "RetailBench: Benchmarking long horizon reasoning and coherent decision making of LLM agents in realistic retail environments"
+authors: Linghua Zhang, Jun Wang, Jingtong Wu, Zhisong Zhang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.15862
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-14
+updated_at: 2026-06-19
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - reasoning
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, planning, reasoning, tool-use, world-model
+- arXiv categories: cs.AI
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.15862
diff --git a/papers/items/2026-2606-15874-llm-as-code-agentic-programming-for-agent-harness.md b/papers/items/2026-2606-15874-llm-as-code-agentic-programming-for-agent-harness.md
new file mode 100644
index 0000000..034458c
--- /dev/null
+++ b/papers/items/2026-2606-15874-llm-as-code-agentic-programming-for-agent-harness.md
@@ -0,0 +1,60 @@
+# Paper: LLM-as-Code: Agentic Programming for Agent Harness
+
+---
+type: paper
+title: "LLM-as-Code: Agentic Programming for Agent Harness"
+authors: Junjia Qi, Zichuan Fu, Jingtong Gao, Wenlin Zhang, Hanyu Yan, Xian Wu, Xiangyu Zhao
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.15874
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-14
+updated_at: 2026-06-22
+status: queued
+relevance: high
+topics:
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: reasoning, tool-use
+- arXiv categories: cs.AI, cs.SE
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.15874
diff --git a/papers/items/2026-2606-15903-control-plane-placement-shapes-forgetting-an-architectural-study-of-agent-memory.md b/papers/items/2026-2606-15903-control-plane-placement-shapes-forgetting-an-architectural-study-of-agent-memory.md
new file mode 100644
index 0000000..6df9d76
--- /dev/null
+++ b/papers/items/2026-2606-15903-control-plane-placement-shapes-forgetting-an-architectural-study-of-agent-memory.md
@@ -0,0 +1,62 @@
+# Paper: Control-Plane Placement Shapes Forgetting: An Architectural Study of Agent Memory Across Thirteen System Configurations
+
+---
+type: paper
+title: "Control-Plane Placement Shapes Forgetting: An Architectural Study of Agent Memory Across Thirteen System Configurations"
+authors: Dongxu Yang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.15903
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-14
+updated_at: 2026-06-16
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, planning, rag
+- arXiv categories: cs.CL, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.15903
diff --git a/papers/items/2026-2606-15906-mage-rag-multigranular-adaptive-graph-evidence-for-agentic-multimodal-rag-in-lon.md b/papers/items/2026-2606-15906-mage-rag-multigranular-adaptive-graph-evidence-for-agentic-multimodal-rag-in-lon.md
new file mode 100644
index 0000000..6ff01d1
--- /dev/null
+++ b/papers/items/2026-2606-15906-mage-rag-multigranular-adaptive-graph-evidence-for-agentic-multimodal-rag-in-lon.md
@@ -0,0 +1,64 @@
+# Paper: MAGE-RAG: Multigranular Adaptive Graph Evidence for Agentic Multimodal RAG in Long-Document QA
+
+---
+type: paper
+title: "MAGE-RAG: Multigranular Adaptive Graph Evidence for Agentic Multimodal RAG in Long-Document QA"
+authors: Yilong Zuo, Xunkai Li, Jing Yuan, Qiangqiang Dai, Hongchao Qin, Ronghua Li
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.15906
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-14
+updated_at: 2026-06-14
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.IR
+  - cs.AI
+  - cs.CL
+  - cs.DB
+  - cs.MM
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, coding-agent, rag
+- arXiv categories: cs.IR, cs.AI, cs.CL, cs.DB, cs.MM
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.15906
diff --git a/papers/items/2026-2606-15931-deeproot-a-kg-coordinated-multi-agent-system-for-therapeutic-reasoning-over-hist.md b/papers/items/2026-2606-15931-deeproot-a-kg-coordinated-multi-agent-system-for-therapeutic-reasoning-over-hist.md
new file mode 100644
index 0000000..2a3a612
--- /dev/null
+++ b/papers/items/2026-2606-15931-deeproot-a-kg-coordinated-multi-agent-system-for-therapeutic-reasoning-over-hist.md
@@ -0,0 +1,63 @@
+# Paper: DeepRoot: A KG-Coordinated Multi-Agent System for Therapeutic Reasoning over Historical Medical Texts
+
+---
+type: paper
+title: "DeepRoot: A KG-Coordinated Multi-Agent System for Therapeutic Reasoning over Historical Medical Texts"
+authors: Zijian Carl Ma, Sean J. Wang, Sijbren Kramer, Li Erran Li
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.15931
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-14
+updated_at: 2026-06-14
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, multi-agent, rag, reasoning, tool-use
+- arXiv categories: cs.MA, cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.15931
diff --git a/papers/items/2026-2606-15994-agentic-framework-for-deep-learning-workload-migration-via-in-context-learning.md b/papers/items/2026-2606-15994-agentic-framework-for-deep-learning-workload-migration-via-in-context-learning.md
new file mode 100644
index 0000000..08cf8f6
--- /dev/null
+++ b/papers/items/2026-2606-15994-agentic-framework-for-deep-learning-workload-migration-via-in-context-learning.md
@@ -0,0 +1,61 @@
+# Paper: Agentic Framework for Deep Learning workload migration via In-Context Learning
+
+---
+type: paper
+title: Agentic Framework for Deep Learning workload migration via In-Context Learning
+authors: Qiyue Liang, Steven Ingram, George Vanica, Andi Gavrilescu, Newfel Harrat, Hassan Sipra, Sethuraman Sankaran
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.15994
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-14
+updated_at: 2026-06-14
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-safety, rag, tool-use
+- arXiv categories: cs.AI, cs.LG
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.15994
diff --git a/papers/items/2026-2606-16111-towards-pareto-optimal-tool-integrated-agents-with-pareto-ranking-policy-optimiz.md b/papers/items/2026-2606-16111-towards-pareto-optimal-tool-integrated-agents-with-pareto-ranking-policy-optimiz.md
new file mode 100644
index 0000000..783849f
--- /dev/null
+++ b/papers/items/2026-2606-16111-towards-pareto-optimal-tool-integrated-agents-with-pareto-ranking-policy-optimiz.md
@@ -0,0 +1,62 @@
+# Paper: Towards Pareto-Optimal Tool-Integrated Agents with Pareto Ranking Policy Optimization
+
+---
+type: paper
+title: Towards Pareto-Optimal Tool-Integrated Agents with Pareto Ranking Policy Optimization
+authors: Junyi Li, Xiaowei Qian, Yingyi Zhang, Wenlin Zhang, Guojing Li, Sheng Zhang, Xiao Han, Yichao Wang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.16111
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-15
+updated_at: 2026-06-15
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - computer-use
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: language-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent, tool-use
+- inferred topics: agent-safety, computer-use, rag, reasoning, tool-use
+- arXiv categories: cs.CL
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.16111
diff --git a/papers/items/2026-2606-16295-visualclaw-a-real-time-personalized-agent-for-the-physical-world.md b/papers/items/2026-2606-16295-visualclaw-a-real-time-personalized-agent-for-the-physical-world.md
new file mode 100644
index 0000000..3fb23b9
--- /dev/null
+++ b/papers/items/2026-2606-16295-visualclaw-a-real-time-personalized-agent-for-the-physical-world.md
@@ -0,0 +1,63 @@
+# Paper: VisualClaw: A Real-Time, Personalized Agent for the Physical World
+
+---
+type: paper
+title: "VisualClaw: A Real-Time, Personalized Agent for the Physical World"
+authors: Haoqin Tu, Jianwen Chen, Zijun Wang, Siwei Han, Juncheng Wu, Hardy Chen, Haonian Ji, Kaiwen Xiong, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.16295
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-15
+updated_at: 2026-06-15
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - computer-use
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-evaluation, tool-use, web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, tool-use, web-gui-agent
+- inferred topics: agent-evaluation, coding-agent, computer-use, rag, tool-use
+- arXiv categories: cs.CV, cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.16295
diff --git a/papers/items/2026-2606-16420-transferable-self-evolving-playbooks-for-agentic-security-auditing.md b/papers/items/2026-2606-16420-transferable-self-evolving-playbooks-for-agentic-security-auditing.md
new file mode 100644
index 0000000..214fa44
--- /dev/null
+++ b/papers/items/2026-2606-16420-transferable-self-evolving-playbooks-for-agentic-security-auditing.md
@@ -0,0 +1,64 @@
+# Paper: Transferable Self-Evolving Playbooks for Agentic Security Auditing
+
+---
+type: paper
+title: Transferable Self-Evolving Playbooks for Agentic Security Auditing
+authors: Ziyue Wang, Cheuk Wang Maurice Ng, Chenchen Yu, Strick Sheng, Kaihua Qin, Liyi Zhou
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.16420
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-15
+updated_at: 2026-06-15
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - embodied-agent
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: agent-safety, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety, tool-use
+- inferred topics: agent-evaluation, agent-safety, computer-use, embodied-agent, rag, tool-use, workflow-agent
+- arXiv categories: cs.CR
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.16420
diff --git a/papers/items/2026-2606-16432-accord-action-conditioned-contextual-grounding-for-language-agents.md b/papers/items/2026-2606-16432-accord-action-conditioned-contextual-grounding-for-language-agents.md
new file mode 100644
index 0000000..e592074
--- /dev/null
+++ b/papers/items/2026-2606-16432-accord-action-conditioned-contextual-grounding-for-language-agents.md
@@ -0,0 +1,62 @@
+# Paper: ACCORD: Action-Conditioned Contextual Grounding for Language Agents
+
+---
+type: paper
+title: "ACCORD: Action-Conditioned Contextual Grounding for Language Agents"
+authors: Lai Jiang, Cheng Qian, Zhenhailong Wang, Pan Lu, Heng Ji, Hao Peng
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.16432
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-15
+updated_at: 2026-06-15
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - embodied-agent
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, embodied-agent, rag, tool-use
+- arXiv categories: cs.CL, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.16432
diff --git a/papers/items/2026-2606-16481-steering-emotional-dynamics-for-art-therapy-controllable-narrative-script-genera.md b/papers/items/2026-2606-16481-steering-emotional-dynamics-for-art-therapy-controllable-narrative-script-genera.md
new file mode 100644
index 0000000..ac9b919
--- /dev/null
+++ b/papers/items/2026-2606-16481-steering-emotional-dynamics-for-art-therapy-controllable-narrative-script-genera.md
@@ -0,0 +1,60 @@
+# Paper: Steering Emotional Dynamics for Art Therapy: Controllable Narrative Script Generation through Hierarchically Guided LLM Agents
+
+---
+type: paper
+title: "Steering Emotional Dynamics for Art Therapy: Controllable Narrative Script Generation through Hierarchically Guided LLM Agents"
+authors: Suqing Wang, Qinghai Miao, Chao Guo, Yisheng Lv
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.16481
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-15
+updated_at: 2026-06-15
+status: queued
+relevance: high
+topics:
+  - computer-use
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: computer-use, planning, tool-use
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.16481
diff --git a/papers/items/2026-2606-16534-generated-parallel-scalable-a-study-of-agentic-ai-generated-julia-code-on-superc.md b/papers/items/2026-2606-16534-generated-parallel-scalable-a-study-of-agentic-ai-generated-julia-code-on-superc.md
new file mode 100644
index 0000000..260bc4a
--- /dev/null
+++ b/papers/items/2026-2606-16534-generated-parallel-scalable-a-study-of-agentic-ai-generated-julia-code-on-superc.md
@@ -0,0 +1,61 @@
+# Paper: Generated, Parallel, Scalable? A Study of Agentic AI-Generated Julia Code on Supercomputers
+
+---
+type: paper
+title: Generated, Parallel, Scalable? A Study of Agentic AI-Generated Julia Code on Supercomputers
+authors: Linus Bantel, Anna-Lena Roth, Jonas Posner, Dirk Pflüger
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.16534
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-15
+updated_at: 2026-06-16
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.DC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, memory, planning, tool-use
+- arXiv categories: cs.DC
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.16534
diff --git a/papers/items/2026-2606-16576-can-llm-agents-infer-world-models-evidence-from-agentic-automata-learning.md b/papers/items/2026-2606-16576-can-llm-agents-infer-world-models-evidence-from-agentic-automata-learning.md
new file mode 100644
index 0000000..49875d4
--- /dev/null
+++ b/papers/items/2026-2606-16576-can-llm-agents-infer-world-models-evidence-from-agentic-automata-learning.md
@@ -0,0 +1,62 @@
+# Paper: Can LLM Agents Infer World Models? Evidence from Agentic Automata Learning
+
+---
+type: paper
+title: Can LLM Agents Infer World Models? Evidence from Agentic Automata Learning
+authors: Reef Menaged, Gili Lior, Shauli Ravfogel, Roee Aharoni, Gabriel Stanovsky
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.16576
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-15
+updated_at: 2026-06-15
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - reasoning
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, planning, reasoning, tool-use, world-model
+- arXiv categories: cs.CL
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.16576
diff --git a/papers/items/2026-2606-16591-sing-synthetic-intention-graph-for-scalable-active-tool-discovery-in-llm-agents.md b/papers/items/2026-2606-16591-sing-synthetic-intention-graph-for-scalable-active-tool-discovery-in-llm-agents.md
new file mode 100644
index 0000000..cfad235
--- /dev/null
+++ b/papers/items/2026-2606-16591-sing-synthetic-intention-graph-for-scalable-active-tool-discovery-in-llm-agents.md
@@ -0,0 +1,62 @@
+# Paper: SING: Synthetic Intention Graph for Scalable Active Tool Discovery in LLM Agents
+
+---
+type: paper
+title: "SING: Synthetic Intention Graph for Scalable Active Tool Discovery in LLM Agents"
+authors: Qiao Xiao, Haochen Shi, Yisen Gao, Wenbin Hu, Huihao Jing, Tianshi Zheng, Baixuan Xu, Ziheng Zhang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.16591
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-15
+updated_at: 2026-06-16
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, multi-agent, planning, rag, tool-use
+- arXiv categories: cs.CL
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.16591
diff --git a/papers/items/2026-2606-16613-coffeebench-benchmarking-long-horizon-llm-agents-in-heterogeneous-multi-agent-ec.md b/papers/items/2026-2606-16613-coffeebench-benchmarking-long-horizon-llm-agents-in-heterogeneous-multi-agent-ec.md
new file mode 100644
index 0000000..89ecf85
--- /dev/null
+++ b/papers/items/2026-2606-16613-coffeebench-benchmarking-long-horizon-llm-agents-in-heterogeneous-multi-agent-ec.md
@@ -0,0 +1,62 @@
+# Paper: CoffeeBench: Benchmarking Long-Horizon LLM Agents in Heterogeneous Multi-Agent Economies
+
+---
+type: paper
+title: "CoffeeBench: Benchmarking Long-Horizon LLM Agents in Heterogeneous Multi-Agent Economies"
+authors: Issa Sugiura, Daichi Hattori, Kazuo Araragi, Keita Ogawa, Shota Onose, Taro Makino, Teppei Usuki, Takashi Ishida
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.16613
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-15
+updated_at: 2026-06-15
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - planning
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 22
+collection_queries: autonomous-agent-llm, planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm, planning-agent
+- inferred topics: agent-evaluation, multi-agent, planning, tool-use, world-model
+- arXiv categories: cs.AI
+- collection score: 22
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.16613
diff --git a/papers/items/2026-2606-16659-fraudsmswalker-benchmarking-agentic-large-language-models-for-sms-to-webpage-fra.md b/papers/items/2026-2606-16659-fraudsmswalker-benchmarking-agentic-large-language-models-for-sms-to-webpage-fra.md
new file mode 100644
index 0000000..1d321fa
--- /dev/null
+++ b/papers/items/2026-2606-16659-fraudsmswalker-benchmarking-agentic-large-language-models-for-sms-to-webpage-fra.md
@@ -0,0 +1,61 @@
+# Paper: FraudSMSWalker: Benchmarking Agentic Large Language Models for SMS-to-Webpage Fraud Detection
+
+---
+type: paper
+title: "FraudSMSWalker: Benchmarking Agentic Large Language Models for SMS-to-Webpage Fraud Detection"
+authors: Y. H. Zhou, Z. M. Ma, Y. J. Zhou, Y. T. Li, H. X. Xiang, Y. M. Cheng, T. L. Chen, K. J. Zhang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.16659
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-15
+updated_at: 2026-06-15
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, agent-safety, computer-use, tool-use
+- arXiv categories: cs.CL
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.16659
diff --git a/papers/items/2026-2606-16748-mypcbench-a-benchmark-for-personally-intelligent-computer-use-agents.md b/papers/items/2026-2606-16748-mypcbench-a-benchmark-for-personally-intelligent-computer-use-agents.md
new file mode 100644
index 0000000..b926176
--- /dev/null
+++ b/papers/items/2026-2606-16748-mypcbench-a-benchmark-for-personally-intelligent-computer-use-agents.md
@@ -0,0 +1,62 @@
+# Paper: MyPCBench: A Benchmark for Personally Intelligent Computer-Use Agents
+
+---
+type: paper
+title: "MyPCBench: A Benchmark for Personally Intelligent Computer-Use Agents"
+authors: Lawrence Keunho Jang, Andrew Keunwoo Jang, Jing Yu Koh, Ruslan Salakhutdinov
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.16748
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-15
+updated_at: 2026-06-15
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-evaluation, web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, web-gui-agent
+- inferred topics: agent-evaluation, computer-use, tool-use, workflow-agent
+- arXiv categories: cs.LG, cs.CL
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.16748
diff --git a/papers/items/2026-2606-16774-openclaw-skill-collective-skill-tree-search-for-agentic-large-language-models.md b/papers/items/2026-2606-16774-openclaw-skill-collective-skill-tree-search-for-agentic-large-language-models.md
new file mode 100644
index 0000000..9555944
--- /dev/null
+++ b/papers/items/2026-2606-16774-openclaw-skill-collective-skill-tree-search-for-agentic-large-language-models.md
@@ -0,0 +1,63 @@
+# Paper: OpenClaw-Skill: Collective Skill Tree Search for Agentic Large Language Models
+
+---
+type: paper
+title: "OpenClaw-Skill: Collective Skill Tree Search for Agentic Large Language Models"
+authors: Tianyi Lin, Chuanyu Sun, Jingyi Zhang, Changxu Wei, Huanjin Yao, Shunyu Liu, Xikun Zhang, Liu Liu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.16774
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-15
+updated_at: 2026-06-15
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: planning-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent, tool-use
+- inferred topics: agent-evaluation, planning, rag, reasoning, tool-use
+- arXiv categories: cs.AI, cs.CL
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.16774
diff --git a/papers/items/2026-2606-16802-labosbench-benchmarking-computer-use-agents-for-scientific-instrument-control.md b/papers/items/2026-2606-16802-labosbench-benchmarking-computer-use-agents-for-scientific-instrument-control.md
new file mode 100644
index 0000000..45b208e
--- /dev/null
+++ b/papers/items/2026-2606-16802-labosbench-benchmarking-computer-use-agents-for-scientific-instrument-control.md
@@ -0,0 +1,62 @@
+# Paper: LabOSBench: Benchmarking Computer Use Agents for Scientific Instrument Control
+
+---
+type: paper
+title: "LabOSBench: Benchmarking Computer Use Agents for Scientific Instrument Control"
+authors: Anqi Zou, Han Deng, Chengyu Zhang, Junquan Hu, Yu Wang, Yuxiang Xing, Aokai Zhang, Hanling Zhang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.16802
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-15
+updated_at: 2026-06-15
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - planning
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 21
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, agent-safety, computer-use, planning, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 21
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.16802
diff --git a/papers/items/2026-2606-16813-gist-cmtf-goal-state-inference-for-causal-minimal-tool-filtering-in-llm-agents.md b/papers/items/2026-2606-16813-gist-cmtf-goal-state-inference-for-causal-minimal-tool-filtering-in-llm-agents.md
new file mode 100644
index 0000000..ae83c32
--- /dev/null
+++ b/papers/items/2026-2606-16813-gist-cmtf-goal-state-inference-for-causal-minimal-tool-filtering-in-llm-agents.md
@@ -0,0 +1,60 @@
+# Paper: GIST-CMTF: Goal-State Inference for Causal Minimal Tool Filtering in LLM Agents
+
+---
+type: paper
+title: "GIST-CMTF: Goal-State Inference for Causal Minimal Tool Filtering in LLM Agents"
+authors: Rahul Suresh Babu, Rohit Shukla
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.16813
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-15
+updated_at: 2026-06-15
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, computer-use, tool-use
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.16813
diff --git a/papers/items/2026-2606-16839-towards-llm-accelerated-rapid-reviews-for-software-tool-discovery-case-for-log-a.md b/papers/items/2026-2606-16839-towards-llm-accelerated-rapid-reviews-for-software-tool-discovery-case-for-log-a.md
new file mode 100644
index 0000000..cdc3916
--- /dev/null
+++ b/papers/items/2026-2606-16839-towards-llm-accelerated-rapid-reviews-for-software-tool-discovery-case-for-log-a.md
@@ -0,0 +1,62 @@
+# Paper: Towards LLM Accelerated Rapid Reviews for Software Tool Discovery -- Case for Log Anomaly Detection
+
+---
+type: paper
+title: Towards LLM Accelerated Rapid Reviews for Software Tool Discovery -- Case for Log Anomaly Detection
+authors: Jesse Nyyssölä, Hamza Bin Mazhar, Alexander Bakhtin, Matteo Esposito, Nana Reinikainen, Yuqing Wang, Ying Song, Davide Taibi, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.16839
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-15
+updated_at: 2026-06-15
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - planning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, coding-agent, planning, tool-use, workflow-agent
+- arXiv categories: cs.SE
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.16839
diff --git a/papers/items/2026-2606-16871-human-on-the-bridge-scalable-evaluation-for-ai-agents.md b/papers/items/2026-2606-16871-human-on-the-bridge-scalable-evaluation-for-ai-agents.md
new file mode 100644
index 0000000..4a9ff10
--- /dev/null
+++ b/papers/items/2026-2606-16871-human-on-the-bridge-scalable-evaluation-for-ai-agents.md
@@ -0,0 +1,62 @@
+# Paper: Human-on-the-Bridge: Scalable Evaluation for AI Agents
+
+---
+type: paper
+title: "Human-on-the-Bridge: Scalable Evaluation for AI Agents"
+authors: Fouad Bousetouane
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.16871
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-15
+updated_at: 2026-06-15
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, computer-use, memory, rag, tool-use
+- arXiv categories: cs.MA
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.16871
diff --git a/papers/items/2026-2606-17041-benchmarking-llm-agents-on-meta-analysis-articles-from-nature-portfolio.md b/papers/items/2026-2606-17041-benchmarking-llm-agents-on-meta-analysis-articles-from-nature-portfolio.md
new file mode 100644
index 0000000..e193ac3
--- /dev/null
+++ b/papers/items/2026-2606-17041-benchmarking-llm-agents-on-meta-analysis-articles-from-nature-portfolio.md
@@ -0,0 +1,63 @@
+# Paper: Benchmarking LLM Agents on Meta-Analysis Articles from Nature Portfolio
+
+---
+type: paper
+title: Benchmarking LLM Agents on Meta-Analysis Articles from Nature Portfolio
+authors: Anzhe Xie, Weihang Su, Yujia Zhou, Yiqun Liu, Qingyao Ai
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.17041
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-15
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - rag
+  - reasoning
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.IR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, computer-use, rag, reasoning, workflow-agent
+- arXiv categories: cs.CL, cs.IR
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.17041
diff --git a/papers/items/2026-2606-17076-cmip-forge-an-agentic-system-that-retrieves-computes-and-self-reviews-climate-sc.md b/papers/items/2026-2606-17076-cmip-forge-an-agentic-system-that-retrieves-computes-and-self-reviews-climate-sc.md
new file mode 100644
index 0000000..88fab2c
--- /dev/null
+++ b/papers/items/2026-2606-17076-cmip-forge-an-agentic-system-that-retrieves-computes-and-self-reviews-climate-sc.md
@@ -0,0 +1,64 @@
+# Paper: CMIP-Forge: An Agentic System that Retrieves, Computes, and Self-Reviews Climate Science
+
+---
+type: paper
+title: "CMIP-Forge: An Agentic System that Retrieves, Computes, and Self-Reviews Climate Science"
+authors: Dmitrii Pantiukhin, Boris Shapkin, Ivan Kuznetsov, Thomas Jung, Nikolay Koldunov
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.17076
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-10
+updated_at: 2026-06-10
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - physics.ao-ph
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, agent-safety, planning, rag, tool-use, workflow-agent
+- arXiv categories: physics.ao-ph, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.17076
diff --git a/papers/items/2026-2606-17114-an-evaluation-of-data-leakage-risks-in-tool-using-llm-agents-in-realistic-scenar.md b/papers/items/2026-2606-17114-an-evaluation-of-data-leakage-risks-in-tool-using-llm-agents-in-realistic-scenar.md
new file mode 100644
index 0000000..d34b019
--- /dev/null
+++ b/papers/items/2026-2606-17114-an-evaluation-of-data-leakage-risks-in-tool-using-llm-agents-in-realistic-scenar.md
@@ -0,0 +1,63 @@
+# Paper: An Evaluation of Data Leakage Risks in Tool-Using LLM Agents in Realistic Scenarios
+
+---
+type: paper
+title: An Evaluation of Data Leakage Risks in Tool-Using LLM Agents in Realistic Scenarios
+authors: Hankyul Baek, Jaewon Noh, Sang Seo, Yongsu Kim, Gabriel Waikin Loh Matienzo, Young Il Kim, Ee Wei Seah, Akriti Vij
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.17114
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-15
+updated_at: 2026-06-15
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - tool-use
+  - workflow-agent
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: agent-safety, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety, tool-use
+- inferred topics: agent-evaluation, agent-safety, tool-use, workflow-agent, world-model
+- arXiv categories: cs.CR, cs.AI
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.17114
diff --git a/papers/items/2026-2606-17246-geodisaster-benchmarking-orchestrated-agents-for-operational-disaster-geo-intell.md b/papers/items/2026-2606-17246-geodisaster-benchmarking-orchestrated-agents-for-operational-disaster-geo-intell.md
new file mode 100644
index 0000000..f3a6f15
--- /dev/null
+++ b/papers/items/2026-2606-17246-geodisaster-benchmarking-orchestrated-agents-for-operational-disaster-geo-intell.md
@@ -0,0 +1,65 @@
+# Paper: GeoDisaster: Benchmarking Orchestrated Agents for Operational Disaster Geo-Intelligence
+
+---
+type: paper
+title: "GeoDisaster: Benchmarking Orchestrated Agents for Operational Disaster Geo-Intelligence"
+authors: Maram Hasan, Aman Verma, Savitra Roy, Hariseetharam Gunduboina, Daksh Jain, Muhammad Haris Khan, Subhasis Chaudhuri, Biplab Banerjee
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.17246
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-15
+updated_at: 2026-06-15
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, agent-safety, multi-agent, rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.CV, cs.MA
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.17246
diff --git a/papers/items/2026-2606-17368-distributed-general-purpose-agent-networks-architecture-key-mechanisms-and-proto.md b/papers/items/2026-2606-17368-distributed-general-purpose-agent-networks-architecture-key-mechanisms-and-proto.md
new file mode 100644
index 0000000..554197a
--- /dev/null
+++ b/papers/items/2026-2606-17368-distributed-general-purpose-agent-networks-architecture-key-mechanisms-and-proto.md
@@ -0,0 +1,63 @@
+# Paper: Distributed General-Purpose Agent Networks: Architecture, Key Mechanisms, and Prototypes
+
+---
+type: paper
+title: "Distributed General-Purpose Agent Networks: Architecture, Key Mechanisms, and Prototypes"
+authors: Shengli Zhang, Deen Ma, Zibin Lin, Taotao Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.17368
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-15
+updated_at: 2026-06-15
+status: queued
+relevance: high
+topics:
+  - computer-use
+  - multi-agent
+  - planning
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.NI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: computer-use, multi-agent, planning, tool-use, world-model
+- arXiv categories: cs.AI, cs.NI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.17368
diff --git a/papers/items/2026-2606-17383-model-validation-of-agentic-ai-systems-a-pomdp-based-framework-for-belief-state-.md b/papers/items/2026-2606-17383-model-validation-of-agentic-ai-systems-a-pomdp-based-framework-for-belief-state-.md
new file mode 100644
index 0000000..78f4ff0
--- /dev/null
+++ b/papers/items/2026-2606-17383-model-validation-of-agentic-ai-systems-a-pomdp-based-framework-for-belief-state-.md
@@ -0,0 +1,63 @@
+# Paper: Model Validation of Agentic AI Systems: A POMDP-Based Framework for Belief-State, Forecast, and Policy Validation
+
+---
+type: paper
+title: "Model Validation of Agentic AI Systems: A POMDP-Based Framework for Belief-State, Forecast, and Policy Validation"
+authors: Matthew Francis Dixon
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.17383
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-16
+updated_at: 2026-06-16
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - q-fin.RM
+  - cs.AI
+  - cs.LG
+  - stat.ML
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-safety, rag, tool-use
+- arXiv categories: q-fin.RM, cs.AI, cs.LG, stat.ML
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.17383
diff --git a/papers/items/2026-2606-17449-mode-rag-manifold-outlier-diagnosis-and-energy-based-retrieval-augmented-generat.md b/papers/items/2026-2606-17449-mode-rag-manifold-outlier-diagnosis-and-energy-based-retrieval-augmented-generat.md
new file mode 100644
index 0000000..ef590d3
--- /dev/null
+++ b/papers/items/2026-2606-17449-mode-rag-manifold-outlier-diagnosis-and-energy-based-retrieval-augmented-generat.md
@@ -0,0 +1,67 @@
+# Paper: MODE-RAG: Manifold Outlier Diagnosis and Energy-based Retrieval-Augmented Generation Evaluation
+
+---
+type: paper
+title: "MODE-RAG: Manifold Outlier Diagnosis and Energy-based Retrieval-Augmented Generation Evaluation"
+authors: Zehang Wei, Jiaxin Dai, Jiamin Yan, Xiang Xiang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.17449
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-16
+updated_at: 2026-06-16
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - multi-agent
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+  - cs.CV
+  - cs.LG
+  - cs.MM
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, agent-safety, computer-use, multi-agent, rag, reasoning
+- arXiv categories: cs.CL, cs.AI, cs.CV, cs.LG, cs.MM
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.17449
diff --git a/papers/items/2026-2606-17453-mapsatisfybench-benchmarking-satisfaction-aware-map-agents-through-behavior-grou.md b/papers/items/2026-2606-17453-mapsatisfybench-benchmarking-satisfaction-aware-map-agents-through-behavior-grou.md
new file mode 100644
index 0000000..58e4745
--- /dev/null
+++ b/papers/items/2026-2606-17453-mapsatisfybench-benchmarking-satisfaction-aware-map-agents-through-behavior-grou.md
@@ -0,0 +1,59 @@
+# Paper: MapSatisfyBench: Benchmarking Satisfaction-Aware Map Agents through Behavior-Grounded Implicit Decision Factors
+
+---
+type: paper
+title: "MapSatisfyBench: Benchmarking Satisfaction-Aware Map Agents through Behavior-Grounded Implicit Decision Factors"
+authors: Lubin Bai, Mengyu Cao, Sixue Wang, Zhongwei Wan, Yue Pan, Jiale Hou, Xiang Li, Xiuyuan Zhang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.17453
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-16
+updated_at: 2026-06-17
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, tool-use
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.17453
diff --git a/papers/items/2026-2606-17459-can-llms-be-ceos-benchmarking-strategic-resource-reallocation-with-multi-role-ag.md b/papers/items/2026-2606-17459-can-llms-be-ceos-benchmarking-strategic-resource-reallocation-with-multi-role-ag.md
new file mode 100644
index 0000000..582477e
--- /dev/null
+++ b/papers/items/2026-2606-17459-can-llms-be-ceos-benchmarking-strategic-resource-reallocation-with-multi-role-ag.md
@@ -0,0 +1,65 @@
+# Paper: Can LLMs Be CEOs? Benchmarking Strategic Resource Reallocation with Multi-Role Agent Simulation
+
+---
+type: paper
+title: Can LLMs Be CEOs? Benchmarking Strategic Resource Reallocation with Multi-Role Agent Simulation
+authors: Yuyang Dai, Xueqing Peng, Lingfei Qian, Zhuohan Xie
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.17459
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-16
+updated_at: 2026-06-16
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - multi-agent
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 22
+collection_queries: agent-evaluation, planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, planning-agent
+- inferred topics: agent-evaluation, computer-use, multi-agent, planning, rag, reasoning, tool-use, world-model
+- arXiv categories: cs.AI
+- collection score: 22
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.17459
diff --git a/papers/items/2026-2606-17573-cordon-semantic-transactions-for-tool-using-llm-agents.md b/papers/items/2026-2606-17573-cordon-semantic-transactions-for-tool-using-llm-agents.md
new file mode 100644
index 0000000..11f573d
--- /dev/null
+++ b/papers/items/2026-2606-17573-cordon-semantic-transactions-for-tool-using-llm-agents.md
@@ -0,0 +1,63 @@
+# Paper: Cordon: Semantic Transactions for Tool-Using LLM Agents
+
+---
+type: paper
+title: "Cordon: Semantic Transactions for Tool-Using LLM Agents"
+authors: Zheng Chen, Hanqing Liu, Duling Xu, Dong Dong, Jialin Li, Bangzheng Pu, Jidong Zhai
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.17573
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-16
+updated_at: 2026-06-16
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.OS
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, agent-safety, memory, tool-use, workflow-agent
+- arXiv categories: cs.OS, cs.CR
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.17573
diff --git a/papers/items/2026-2606-17680-envrl-learn-from-environment-dynamics-in-agentic-reinforcement-learning.md b/papers/items/2026-2606-17680-envrl-learn-from-environment-dynamics-in-agentic-reinforcement-learning.md
new file mode 100644
index 0000000..7c5d6cb
--- /dev/null
+++ b/papers/items/2026-2606-17680-envrl-learn-from-environment-dynamics-in-agentic-reinforcement-learning.md
@@ -0,0 +1,62 @@
+# Paper: EnvRL: Learn from Environment Dynamics in Agentic Reinforcement Learning
+
+---
+type: paper
+title: "EnvRL: Learn from Environment Dynamics in Agentic Reinforcement Learning"
+authors: Zhitong Wang, Songze Li, Hao Peng, Shuzheng Si, Yi Wang, Maosong Sun, Juanzi Li
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.17680
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-16
+updated_at: 2026-06-16
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, planning, rag, tool-use
+- arXiv categories: cs.LG, cs.CL
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.17680
diff --git a/papers/items/2026-2606-18023-loopcoder-v2-only-loop-once-for-efficient-test-time-computation-scaling.md b/papers/items/2026-2606-18023-loopcoder-v2-only-loop-once-for-efficient-test-time-computation-scaling.md
new file mode 100644
index 0000000..b8fc4cb
--- /dev/null
+++ b/papers/items/2026-2606-18023-loopcoder-v2-only-loop-once-for-efficient-test-time-computation-scaling.md
@@ -0,0 +1,63 @@
+# Paper: LoopCoder-v2: Only Loop Once for Efficient Test-Time Computation Scaling
+
+---
+type: paper
+title: "LoopCoder-v2: Only Loop Once for Efficient Test-Time Computation Scaling"
+authors: Jian Yang, Shawn Guo, Wei Zhang, Tianyu Zheng, Yaxin Du, Haau-Sing Li, Jiajun Wu, Yue Song, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.18023
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-16
+updated_at: 2026-06-16
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - memory
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, coding-agent, memory, reasoning, tool-use
+- arXiv categories: cs.LG, cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.18023
diff --git a/papers/items/2026-2606-18037-provenanceguard-source-aware-factuality-verification-for-mcp-based-llm-agents.md b/papers/items/2026-2606-18037-provenanceguard-source-aware-factuality-verification-for-mcp-based-llm-agents.md
new file mode 100644
index 0000000..c6a09c4
--- /dev/null
+++ b/papers/items/2026-2606-18037-provenanceguard-source-aware-factuality-verification-for-mcp-based-llm-agents.md
@@ -0,0 +1,64 @@
+# Paper: ProvenanceGuard: Source-Aware Factuality Verification for MCP-Based LLM Agents
+
+---
+type: paper
+title: "ProvenanceGuard: Source-Aware Factuality Verification for MCP-Based LLM Agents"
+authors: Ander Alvarez, Santhiya Rajan, Samuel Mugel, Román Orús
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.18037
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-16
+updated_at: 2026-06-16
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, agent-safety, rag, reasoning, tool-use
+- arXiv categories: cs.AI, cs.CL, cs.MA
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.18037
diff --git a/papers/items/2026-2606-18051-compositional-skill-routing-for-llm-agents-decompose-retrieve-and-compose.md b/papers/items/2026-2606-18051-compositional-skill-routing-for-llm-agents-decompose-retrieve-and-compose.md
new file mode 100644
index 0000000..52b2ef5
--- /dev/null
+++ b/papers/items/2026-2606-18051-compositional-skill-routing-for-llm-agents-decompose-retrieve-and-compose.md
@@ -0,0 +1,61 @@
+# Paper: Compositional Skill Routing for LLM Agents: Decompose, Retrieve, and Compose
+
+---
+type: paper
+title: "Compositional Skill Routing for LLM Agents: Decompose, Retrieve, and Compose"
+authors: Xueping Gao
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.18051
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-16
+updated_at: 2026-06-16
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, planning, rag, tool-use
+- arXiv categories: cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.18051
diff --git a/papers/items/2026-2606-18068-agentic-ai-based-framework-for-mitigating-premature-diagnostic-handoff-and-silen.md b/papers/items/2026-2606-18068-agentic-ai-based-framework-for-mitigating-premature-diagnostic-handoff-and-silen.md
new file mode 100644
index 0000000..1dc7c01
--- /dev/null
+++ b/papers/items/2026-2606-18068-agentic-ai-based-framework-for-mitigating-premature-diagnostic-handoff-and-silen.md
@@ -0,0 +1,61 @@
+# Paper: Agentic AI-based Framework for Mitigating Premature Diagnostic Handoff and Silent Hallucination in Healthcare Applications
+
+---
+type: paper
+title: Agentic AI-based Framework for Mitigating Premature Diagnostic Handoff and Silent Hallucination in Healthcare Applications
+authors: Divyansh Srivastava, Shreya Ghosh, Anshul Verma, Rajkumar Buyya
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.18068
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-16
+updated_at: 2026-06-16
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-evaluation, agent-safety, multi-agent, reasoning
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.18068
diff --git a/papers/items/2026-2606-18142-your-ai-travel-agent-would-book-you-a-bullfight-an-agentic-benchmark-for-implici.md b/papers/items/2026-2606-18142-your-ai-travel-agent-would-book-you-a-bullfight-an-agentic-benchmark-for-implici.md
new file mode 100644
index 0000000..f15224d
--- /dev/null
+++ b/papers/items/2026-2606-18142-your-ai-travel-agent-would-book-you-a-bullfight-an-agentic-benchmark-for-implici.md
@@ -0,0 +1,61 @@
+# Paper: Your AI Travel Agent Would Book You a Bullfight: An Agentic Benchmark for Implicit Animal Welfare in Frontier AI Models
+
+---
+type: paper
+title: "Your AI Travel Agent Would Book You a Bullfight: An Agentic Benchmark for Implicit Animal Welfare in Frontier AI Models"
+authors: Jasmine Brazilek, Joel Christoph, Maheep Chaudhary, Oliver Tullio, Carol Kline, Miles Tidmarsh, Arturs Kanepajs
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.18142
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-16
+updated_at: 2026-07-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+  - cs.CY
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, agent-safety
+- arXiv categories: cs.AI, cs.CL, cs.CY
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.18142
diff --git a/papers/items/2026-2606-18272-mitigating-anchoring-bias-in-llm-based-agents-for-energy-efficient-6g-autonomous.md b/papers/items/2026-2606-18272-mitigating-anchoring-bias-in-llm-based-agents-for-energy-efficient-6g-autonomous.md
new file mode 100644
index 0000000..38badad
--- /dev/null
+++ b/papers/items/2026-2606-18272-mitigating-anchoring-bias-in-llm-based-agents-for-energy-efficient-6g-autonomous.md
@@ -0,0 +1,63 @@
+# Paper: Mitigating Anchoring Bias in LLM-Based Agents for Energy-Efficient 6G Autonomous Networks
+
+---
+type: paper
+title: Mitigating Anchoring Bias in LLM-Based Agents for Energy-Efficient 6G Autonomous Networks
+authors: Hatim Chergui, Claudia Carballo González, Farhad Rezazadeh, Merouane Debbah
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.18272
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-05
+updated_at: 2026-06-18
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - computer-use
+  - multi-agent
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.NI
+  - cs.AI
+  - eess.SY
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-safety, computer-use, multi-agent, reasoning
+- arXiv categories: cs.NI, cs.AI, eess.SY
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.18272
diff --git a/papers/items/2026-2606-18356-safeclawbench-separating-semantic-audit-evidence-and-sandbox-harm-in-tool-using-.md b/papers/items/2026-2606-18356-safeclawbench-separating-semantic-audit-evidence-and-sandbox-harm-in-tool-using-.md
new file mode 100644
index 0000000..a4ee243
--- /dev/null
+++ b/papers/items/2026-2606-18356-safeclawbench-separating-semantic-audit-evidence-and-sandbox-harm-in-tool-using-.md
@@ -0,0 +1,63 @@
+# Paper: SafeClawBench: Separating Semantic, Audit-Evidence, and Sandbox Harm in Tool-Using LLM Agents
+
+---
+type: paper
+title: "SafeClawBench: Separating Semantic, Audit-Evidence, and Sandbox Harm in Tool-Using LLM Agents"
+authors: Yuchuan Tian, Mengyu Zheng, Haocheng Mei, Ye Yuan, Chao Xu, Xinghao Chen, Hanting Chen, Yu Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.18356
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-16
+updated_at: 2026-06-16
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - memory
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agent-safety, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety, tool-use
+- inferred topics: agent-evaluation, agent-safety, computer-use, memory, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.18356
diff --git a/papers/items/2026-2606-18363-guava-an-effective-and-universal-harness-for-embodied-manipulation.md b/papers/items/2026-2606-18363-guava-an-effective-and-universal-harness-for-embodied-manipulation.md
new file mode 100644
index 0000000..b1d19c9
--- /dev/null
+++ b/papers/items/2026-2606-18363-guava-an-effective-and-universal-harness-for-embodied-manipulation.md
@@ -0,0 +1,64 @@
+# Paper: Guava: An Effective and Universal Harness for Embodied Manipulation
+
+---
+type: paper
+title: "Guava: An Effective and Universal Harness for Embodied Manipulation"
+authors: Haowen Liu, Xirui Li, Shaoxiong Yao, Peng Shi, Tianyi Zhou, Jia-Bin Huang, Furong Huang, Jiayuan Mao
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.18363
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-16
+updated_at: 2026-06-16
+status: queued
+relevance: high
+topics:
+  - embodied-agent
+  - planning
+  - reasoning
+  - tool-use
+  - workflow-agent
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.RO
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agentic-ai, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, tool-use
+- inferred topics: embodied-agent, planning, reasoning, tool-use, workflow-agent, world-model
+- arXiv categories: cs.RO, cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.18363
diff --git a/papers/items/2026-2606-18406-coremem-riemannian-retrieval-and-fisher-guided-distillation-for-long-term-memory.md b/papers/items/2026-2606-18406-coremem-riemannian-retrieval-and-fisher-guided-distillation-for-long-term-memory.md
new file mode 100644
index 0000000..0f7bac4
--- /dev/null
+++ b/papers/items/2026-2606-18406-coremem-riemannian-retrieval-and-fisher-guided-distillation-for-long-term-memory.md
@@ -0,0 +1,63 @@
+# Paper: CoreMem: Riemannian Retrieval and Fisher-Guided Distillation for Long-Term Memory in Dialogue Agents
+
+---
+type: paper
+title: "CoreMem: Riemannian Retrieval and Fisher-Guided Distillation for Long-Term Memory in Dialogue Agents"
+authors: Jiaqi Chen, Yongqin Zeng, Shaoshen Chen, Yijian Zhang, Hai-Tao Zheng, Chunxia Ma, XiuTeng Zhou
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.18406
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-16
+updated_at: 2026-06-16
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, computer-use, memory, rag, reasoning, tool-use
+- arXiv categories: cs.CL
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.18406
diff --git a/papers/items/2026-2606-18467-toolchain-crc-conformal-risk-control-for-agentic-ai-under-retrieval-and-tool-use.md b/papers/items/2026-2606-18467-toolchain-crc-conformal-risk-control-for-agentic-ai-under-retrieval-and-tool-use.md
new file mode 100644
index 0000000..5a2da75
--- /dev/null
+++ b/papers/items/2026-2606-18467-toolchain-crc-conformal-risk-control-for-agentic-ai-under-retrieval-and-tool-use.md
@@ -0,0 +1,62 @@
+# Paper: ToolChain-CRC: Conformal Risk Control for Agentic AI Under Retrieval and Tool-Use Drift
+
+---
+type: paper
+title: "ToolChain-CRC: Conformal Risk Control for Agentic AI Under Retrieval and Tool-Use Drift"
+authors: Jeffery Opoku, David Banahene
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.18467
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-16
+updated_at: 2026-06-16
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - stat.ML
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-evaluation, agentic-ai, ai-agent, rag-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, agentic-ai, ai-agent, rag-agent, tool-use
+- inferred topics: agent-evaluation, agent-safety, rag, tool-use
+- arXiv categories: stat.ML, cs.LG
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.18467
diff --git a/papers/items/2026-2606-18502-towards-scalable-customization-and-deployment-of-multi-agent-systems-for-enterpr.md b/papers/items/2026-2606-18502-towards-scalable-customization-and-deployment-of-multi-agent-systems-for-enterpr.md
new file mode 100644
index 0000000..8feede3
--- /dev/null
+++ b/papers/items/2026-2606-18502-towards-scalable-customization-and-deployment-of-multi-agent-systems-for-enterpr.md
@@ -0,0 +1,62 @@
+# Paper: Towards Scalable Customization and Deployment of Multi-Agent Systems for Enterprise Applications
+
+---
+type: paper
+title: Towards Scalable Customization and Deployment of Multi-Agent Systems for Enterprise Applications
+authors: Paresh Dashore, Shreyas Kulkarni, Uttam Gurram, Nadia Bathaee, Kartik Balasubramaniam, Genta Indra Winata, Sambit Sahu, Shi-Xiong Zhang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.18502
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-16
+updated_at: 2026-06-16
+status: queued
+relevance: high
+topics:
+  - coding-agent
+  - multi-agent
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: coding-agent, multi-agent, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.18502
diff --git a/papers/items/2026-2606-18619-code-augur-agentic-vulnerability-detection-via-specification-inference.md b/papers/items/2026-2606-18619-code-augur-agentic-vulnerability-detection-via-specification-inference.md
new file mode 100644
index 0000000..6f46cae
--- /dev/null
+++ b/papers/items/2026-2606-18619-code-augur-agentic-vulnerability-detection-via-specification-inference.md
@@ -0,0 +1,63 @@
+# Paper: Code-Augur: Agentic Vulnerability Detection via Specification Inference
+
+---
+type: paper
+title: "Code-Augur: Agentic Vulnerability Detection via Specification Inference"
+authors: Zhengxiong Luo, Mehtab Zafar, Dylan Wolff, Abhik Roychoudhury
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.18619
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-17
+updated_at: 2026-06-17
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - computer-use
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent
+- inferred topics: agent-safety, computer-use, rag, reasoning
+- arXiv categories: cs.CR, cs.AI, cs.SE
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.18619
diff --git a/papers/items/2026-2606-18671-hansel-extracting-breadcrumbs-from-web-agent-trajectories-for-interactive-verifi.md b/papers/items/2026-2606-18671-hansel-extracting-breadcrumbs-from-web-agent-trajectories-for-interactive-verifi.md
new file mode 100644
index 0000000..73457d4
--- /dev/null
+++ b/papers/items/2026-2606-18671-hansel-extracting-breadcrumbs-from-web-agent-trajectories-for-interactive-verifi.md
@@ -0,0 +1,61 @@
+# Paper: HANSEL: Extracting Breadcrumbs from Web Agent Trajectories for Interactive Verification
+
+---
+type: paper
+title: "HANSEL: Extracting Breadcrumbs from Web Agent Trajectories for Interactive Verification"
+authors: Yujin Zhang, Daye Nam
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.18671
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-17
+updated_at: 2026-06-17
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - embodied-agent
+  - planning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.HC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: ai-agent, web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent, web-gui-agent
+- inferred topics: agent-evaluation, computer-use, embodied-agent, planning
+- arXiv categories: cs.HC
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.18671
diff --git a/papers/items/2026-2606-18789-poweragentbench-ss-a-benchmark-for-agentic-ai-in-power-system-steady-state-studi.md b/papers/items/2026-2606-18789-poweragentbench-ss-a-benchmark-for-agentic-ai-in-power-system-steady-state-studi.md
new file mode 100644
index 0000000..d6db767
--- /dev/null
+++ b/papers/items/2026-2606-18789-poweragentbench-ss-a-benchmark-for-agentic-ai-in-power-system-steady-state-studi.md
@@ -0,0 +1,64 @@
+# Paper: PowerAgentBench-SS: A Benchmark for Agentic AI in Power System Steady-State Studies
+
+---
+type: paper
+title: "PowerAgentBench-SS: A Benchmark for Agentic AI in Power System Steady-State Studies"
+authors: Costas Mylonas, Magda Foti, Andrea Pomarico, Matheus Duarte, Qian Zhang, Emmanouel Varvarigos
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.18789
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-17
+updated_at: 2026-06-17
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - planning
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - eess.SY
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 24
+collection_queries: agentic-ai, planning-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, planning-agent, tool-use
+- inferred topics: agent-evaluation, agent-safety, computer-use, planning, rag, tool-use, workflow-agent
+- arXiv categories: eess.SY
+- collection score: 24
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.18789
diff --git a/papers/items/2026-2606-18829-gatemem-benchmarking-memory-governance-in-multi-principal-shared-memory-agents.md b/papers/items/2026-2606-18829-gatemem-benchmarking-memory-governance-in-multi-principal-shared-memory-agents.md
new file mode 100644
index 0000000..4d571b4
--- /dev/null
+++ b/papers/items/2026-2606-18829-gatemem-benchmarking-memory-governance-in-multi-principal-shared-memory-agents.md
@@ -0,0 +1,63 @@
+# Paper: GateMem: Benchmarking Memory Governance in Multi-Principal Shared-Memory Agents
+
+---
+type: paper
+title: "GateMem: Benchmarking Memory Governance in Multi-Principal Shared-Memory Agents"
+authors: Zhe Ren, Yibo Yang, Yimeng Chen, Zijun Zhao, Benshuo Fu, Zhihao Shu, Bingjie Zhang, Yangyang Xu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.18829
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-17
+updated_at: 2026-06-17
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+  - rag
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, planning, rag, workflow-agent
+- arXiv categories: cs.LG, cs.CL
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.18829
diff --git a/papers/items/2026-2606-18950-rtsgamebench-an-rts-benchmark-for-strategic-reasoning-by-vision-language-models.md b/papers/items/2026-2606-18950-rtsgamebench-an-rts-benchmark-for-strategic-reasoning-by-vision-language-models.md
new file mode 100644
index 0000000..29c6558
--- /dev/null
+++ b/papers/items/2026-2606-18950-rtsgamebench-an-rts-benchmark-for-strategic-reasoning-by-vision-language-models.md
@@ -0,0 +1,64 @@
+# Paper: RTSGameBench: An RTS Benchmark for Strategic Reasoning by Vision-Language Models
+
+---
+type: paper
+title: "RTSGameBench: An RTS Benchmark for Strategic Reasoning by Vision-Language Models"
+authors: San Kim, Daechul Ahn, Reokyoung Kim, Hyeonbeom Choi, Seungyeon Jwa, Jonghyun Choi
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.18950
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-17
+updated_at: 2026-06-18
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - multi-agent
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, multi-agent, planning, rag, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.18950
diff --git a/papers/items/2026-2606-19063-pypiline-malicious-pypi-package-detection-via-suspicious-api-knowledge-and-agent.md b/papers/items/2026-2606-19063-pypiline-malicious-pypi-package-detection-via-suspicious-api-knowledge-and-agent.md
new file mode 100644
index 0000000..19945cd
--- /dev/null
+++ b/papers/items/2026-2606-19063-pypiline-malicious-pypi-package-detection-via-suspicious-api-knowledge-and-agent.md
@@ -0,0 +1,62 @@
+# Paper: PYPILINE: Malicious PyPI Package Detection via Suspicious API Knowledge and Agent Workflow
+
+---
+type: paper
+title: "PYPILINE: Malicious PyPI Package Detection via Suspicious API Knowledge and Agent Workflow"
+authors: Siyuan Pang, Yepeng Yao, Zhengwei Jiang, Zijing Fan, Haozhe Li, Baoxu Liu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.19063
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-17
+updated_at: 2026-06-26
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agentic-ai, rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, rag-agent
+- inferred topics: agent-evaluation, agent-safety, rag, tool-use, workflow-agent
+- arXiv categories: cs.CR
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.19063
diff --git a/papers/items/2026-2606-19242-runtime-compliance-verification-for-ai-agents.md b/papers/items/2026-2606-19242-runtime-compliance-verification-for-ai-agents.md
new file mode 100644
index 0000000..86760df
--- /dev/null
+++ b/papers/items/2026-2606-19242-runtime-compliance-verification-for-ai-agents.md
@@ -0,0 +1,59 @@
+# Paper: Runtime Compliance Verification for AI Agents
+
+---
+type: paper
+title: Runtime Compliance Verification for AI Agents
+authors: Nafiseh Kahani, Masoud Barati, Diana Addae
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.19242
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-17
+updated_at: 2026-06-17
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: ai-agent, function-calling, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent, function-calling, tool-use
+- inferred topics: agent-evaluation, tool-use
+- arXiv categories: cs.SE
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.19242
diff --git a/papers/items/2026-2606-19245-txbench-pp-analyzing-ai-agent-performance-on-small-molecule-preclinical-pharmaco.md b/papers/items/2026-2606-19245-txbench-pp-analyzing-ai-agent-performance-on-small-molecule-preclinical-pharmaco.md
new file mode 100644
index 0000000..97a7473
--- /dev/null
+++ b/papers/items/2026-2606-19245-txbench-pp-analyzing-ai-agent-performance-on-small-molecule-preclinical-pharmaco.md
@@ -0,0 +1,64 @@
+# Paper: TxBench-PP: Analyzing AI Agent Performance on Small-Molecule Preclinical Pharmacology
+
+---
+type: paper
+title: "TxBench-PP: Analyzing AI Agent Performance on Small-Molecule Preclinical Pharmacology"
+authors: Hannah Le, Ramesh Ramasamy, Alex Urrutia, Mahsa Yazdani, Tim Proctor, Kenny Workman
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.19245
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-17
+updated_at: 2026-06-18
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent
+- inferred topics: agent-evaluation, agent-safety, coding-agent, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI, cs.LG
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.19245
diff --git a/papers/items/2026-2606-19409-openrath-session-centered-runtime-state-for-agent-systems.md b/papers/items/2026-2606-19409-openrath-session-centered-runtime-state-for-agent-systems.md
new file mode 100644
index 0000000..7b7c480
--- /dev/null
+++ b/papers/items/2026-2606-19409-openrath-session-centered-runtime-state-for-agent-systems.md
@@ -0,0 +1,64 @@
+# Paper: OpenRath: Session-Centered Runtime State for Agent Systems
+
+---
+type: paper
+title: "OpenRath: Session-Centered Runtime State for Agent Systems"
+authors: Fukang Wen, Zhijie Wang, Ruilin Xu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.19409
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-17
+updated_at: 2026-06-17
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - multi-agent
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.PL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, multi-agent, rag, tool-use, workflow-agent
+- arXiv categories: cs.SE, cs.PL
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.19409
diff --git a/papers/items/2026-2606-19464-deontic-policies-for-runtime-governance-of-agentic-ai-systems.md b/papers/items/2026-2606-19464-deontic-policies-for-runtime-governance-of-agentic-ai-systems.md
new file mode 100644
index 0000000..d367d35
--- /dev/null
+++ b/papers/items/2026-2606-19464-deontic-policies-for-runtime-governance-of-agentic-ai-systems.md
@@ -0,0 +1,63 @@
+# Paper: Deontic Policies for Runtime Governance of Agentic AI Systems
+
+---
+type: paper
+title: Deontic Policies for Runtime Governance of Agentic AI Systems
+authors: Anupam Joshi, Tim Finin, Karuna Pande Joshi, Lalana Kagal
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.19464
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-17
+updated_at: 2026-06-17
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agentic-ai, autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, autonomous-agent-llm
+- inferred topics: agent-evaluation, agent-safety, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI, cs.MA
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.19464
diff --git a/papers/items/2026-2606-19613-staminabench-stress-testing-coding-agents-over-100-interaction-turns.md b/papers/items/2026-2606-19613-staminabench-stress-testing-coding-agents-over-100-interaction-turns.md
new file mode 100644
index 0000000..541a3cf
--- /dev/null
+++ b/papers/items/2026-2606-19613-staminabench-stress-testing-coding-agents-over-100-interaction-turns.md
@@ -0,0 +1,61 @@
+# Paper: StaminaBench: Stress-Testing Coding Agents over 100 Interaction Turns
+
+---
+type: paper
+title: "StaminaBench: Stress-Testing Coding Agents over 100 Interaction Turns"
+authors: Vlad Sobal, Shuo Yang, Yuting Zhang, Wei Xia, Stefano Soatto
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.19613
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-17
+updated_at: 2026-06-17
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: agent-evaluation, coding-agent, tool-use
+- arXiv categories: cs.SE, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.19613
diff --git a/papers/items/2026-2606-19704-beyond-static-leaderboards-predictive-validity-for-the-evaluation-of-llm-agents.md b/papers/items/2026-2606-19704-beyond-static-leaderboards-predictive-validity-for-the-evaluation-of-llm-agents.md
new file mode 100644
index 0000000..def4fec
--- /dev/null
+++ b/papers/items/2026-2606-19704-beyond-static-leaderboards-predictive-validity-for-the-evaluation-of-llm-agents.md
@@ -0,0 +1,61 @@
+# Paper: Beyond Static Leaderboards: Predictive Validity for the Evaluation of LLM Agents
+
+---
+type: paper
+title: "Beyond Static Leaderboards: Predictive Validity for the Evaluation of LLM Agents"
+authors: Dhaval C. Patel, Kaoutar El Maghraoui, Shuxin Lin, Yusheng Li, Tianjun Feng, Chun-Yi Tsai, Yihan Sun, Wei Alexander Xin, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.19704
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-18
+updated_at: 2026-06-18
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, rag, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.19704
diff --git a/papers/items/2026-2606-19787-oragentbench-can-llm-agents-solve-challenging-operations-research-tasks-end-to-e.md b/papers/items/2026-2606-19787-oragentbench-can-llm-agents-solve-challenging-operations-research-tasks-end-to-e.md
new file mode 100644
index 0000000..eaf4e87
--- /dev/null
+++ b/papers/items/2026-2606-19787-oragentbench-can-llm-agents-solve-challenging-operations-research-tasks-end-to-e.md
@@ -0,0 +1,60 @@
+# Paper: ORAgentBench: Can LLM Agents Solve Challenging Operations Research Tasks End to End?
+
+---
+type: paper
+title: "ORAgentBench: Can LLM Agents Solve Challenging Operations Research Tasks End to End?"
+authors: Jiajun Li, Mingshu Cai, Yixuan Li, Yu Ding, Ran Hou, Guanyu Nie, Xiongwei Han, Wanyuan Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.19787
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-18
+updated_at: 2026-06-18
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, rag, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.19787
diff --git a/papers/items/2026-2606-19812-human-on-the-loop-orchestration-for-ai-assisted-legal-discovery.md b/papers/items/2026-2606-19812-human-on-the-loop-orchestration-for-ai-assisted-legal-discovery.md
new file mode 100644
index 0000000..c38f145
--- /dev/null
+++ b/papers/items/2026-2606-19812-human-on-the-loop-orchestration-for-ai-assisted-legal-discovery.md
@@ -0,0 +1,65 @@
+# Paper: Human-on-the-Loop Orchestration for AI-Assisted Legal Discovery
+
+---
+type: paper
+title: Human-on-the-Loop Orchestration for AI-Assisted Legal Discovery
+authors: Anushree Sinha, Srivaths Ranganathan, Abhishek Dharmaratnakar, Debanshu Das
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.19812
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-18
+updated_at: 2026-06-18
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - rag
+  - reasoning
+  - workflow-agent
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agentic-ai, planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, planning-agent
+- inferred topics: agent-evaluation, agent-safety, planning, rag, reasoning, workflow-agent, world-model
+- arXiv categories: cs.AI, cs.LG
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.19812
diff --git a/papers/items/2026-2606-19852-prompt-plan-extract-zero-shot-agentic-llms-workflows-for-lung-pathology-extracti.md b/papers/items/2026-2606-19852-prompt-plan-extract-zero-shot-agentic-llms-workflows-for-lung-pathology-extracti.md
new file mode 100644
index 0000000..5958c17
--- /dev/null
+++ b/papers/items/2026-2606-19852-prompt-plan-extract-zero-shot-agentic-llms-workflows-for-lung-pathology-extracti.md
@@ -0,0 +1,62 @@
+# Paper: Prompt, Plan, Extract: Zero-Shot Agentic LLMs Workflows for Lung Pathology Extraction from Clinical Narratives
+
+---
+type: paper
+title: "Prompt, Plan, Extract: Zero-Shot Agentic LLMs Workflows for Lung Pathology Extraction from Clinical Narratives"
+authors: Aman Pathak, Cheng Peng, Mengxian Lyu, Ziyi Chen, Reema Solan, Sankalp Talankar, Yasir Khan, Hiren Mehta, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.19852
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-18
+updated_at: 2026-06-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-evaluation, planning, tool-use, workflow-agent
+- arXiv categories: cs.CL, cs.LG
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.19852
diff --git a/papers/items/2026-2606-19899-measuring-biological-capabilities-and-risks-of-ai-agents.md b/papers/items/2026-2606-19899-measuring-biological-capabilities-and-risks-of-ai-agents.md
new file mode 100644
index 0000000..14aba93
--- /dev/null
+++ b/papers/items/2026-2606-19899-measuring-biological-capabilities-and-risks-of-ai-agents.md
@@ -0,0 +1,64 @@
+# Paper: Measuring Biological Capabilities and Risks of AI Agents
+
+---
+type: paper
+title: Measuring Biological Capabilities and Risks of AI Agents
+authors: Patricia Paskov, Jeffrey Lee, Kyle Brady, Alyssa Worland
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.19899
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-18
+updated_at: 2026-06-18
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CY
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: agent-evaluation, agentic-ai, ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, agentic-ai, ai-agent
+- inferred topics: agent-evaluation, agent-safety, computer-use, rag, tool-use, workflow-agent
+- arXiv categories: cs.CY, cs.AI
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.19899
diff --git a/papers/items/2026-2606-19926-memgui-agent-an-end-to-end-long-horizon-mobile-gui-agent-with-proactive-context-.md b/papers/items/2026-2606-19926-memgui-agent-an-end-to-end-long-horizon-mobile-gui-agent-with-proactive-context-.md
new file mode 100644
index 0000000..cc3aad2
--- /dev/null
+++ b/papers/items/2026-2606-19926-memgui-agent-an-end-to-end-long-horizon-mobile-gui-agent-with-proactive-context-.md
@@ -0,0 +1,61 @@
+# Paper: MemGUI-Agent: An End-to-End Long-Horizon Mobile GUI Agent with Proactive Context Management
+
+---
+type: paper
+title: "MemGUI-Agent: An End-to-End Long-Horizon Mobile GUI Agent with Proactive Context Management"
+authors: Guangyi Liu, Gao Wu, Congxiao Liu, Pengxiang Zhao, Liang Liu, Mading Li, Qi Zhang, Mengyan Wang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.19926
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-18
+updated_at: 2026-06-18
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.HC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, computer-use, planning, tool-use
+- arXiv categories: cs.HC
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.19926
diff --git a/papers/items/2026-2606-19930-mobileforge-annotation-free-adaptation-for-mobile-gui-agents-with-hierarchical-f.md b/papers/items/2026-2606-19930-mobileforge-annotation-free-adaptation-for-mobile-gui-agents-with-hierarchical-f.md
new file mode 100644
index 0000000..ce1112d
--- /dev/null
+++ b/papers/items/2026-2606-19930-mobileforge-annotation-free-adaptation-for-mobile-gui-agents-with-hierarchical-f.md
@@ -0,0 +1,60 @@
+# Paper: MobileForge: Annotation-Free Adaptation for Mobile GUI Agents with Hierarchical Feedback-Guided Policy Optimization
+
+---
+type: paper
+title: "MobileForge: Annotation-Free Adaptation for Mobile GUI Agents with Hierarchical Feedback-Guided Policy Optimization"
+authors: Guangyi Liu, Pengxiang Zhao, Gao Wu, Yiwen Yin, Mading Li, Liang Liu, Congxiao Liu, Zhang Qi, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.19930
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-18
+updated_at: 2026-06-18
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.HC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, computer-use, tool-use
+- arXiv categories: cs.HC
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.19930
diff --git a/papers/items/2026-2606-19980-enpire-agentic-robot-policy-self-improvement-in-the-real-world.md b/papers/items/2026-2606-19980-enpire-agentic-robot-policy-self-improvement-in-the-real-world.md
new file mode 100644
index 0000000..53472ae
--- /dev/null
+++ b/papers/items/2026-2606-19980-enpire-agentic-robot-policy-self-improvement-in-the-real-world.md
@@ -0,0 +1,62 @@
+# Paper: ENPIRE: Agentic Robot Policy Self-Improvement in the Real World
+
+---
+type: paper
+title: "ENPIRE: Agentic Robot Policy Self-Improvement in the Real World"
+authors: "Wenli Xiao, Jia Xie, Tonghe Zhang, Haotian Lin, Letian \"Max\" Fu, Haoru Xue, Jalen Lu, Yi Yang, et al."
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.19980
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-18
+updated_at: 2026-06-18
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - embodied-agent
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: coding-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent, tool-use
+- inferred topics: agent-evaluation, coding-agent, embodied-agent, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.19980
diff --git a/papers/items/2026-2606-20023-when-lower-privileges-suffice-investigating-over-privileged-tool-selection-in-ll.md b/papers/items/2026-2606-20023-when-lower-privileges-suffice-investigating-over-privileged-tool-selection-in-ll.md
new file mode 100644
index 0000000..6c0f801
--- /dev/null
+++ b/papers/items/2026-2606-20023-when-lower-privileges-suffice-investigating-over-privileged-tool-selection-in-ll.md
@@ -0,0 +1,62 @@
+# Paper: When Lower Privileges Suffice: Investigating Over-Privileged Tool Selection in LLM Agents
+
+---
+type: paper
+title: "When Lower Privileges Suffice: Investigating Over-Privileged Tool Selection in LLM Agents"
+authors: Kaiyue Yang, Yuyan Bu, Jingwei Yi, Yuchi Wang, Biyu Zhou, Juntao Dai, Songlin Hu, Yaodong Yang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.20023
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-18
+updated_at: 2026-07-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, agent-safety, tool-use
+- arXiv categories: cs.SE, cs.AI, cs.CL
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.20023
diff --git a/papers/items/2026-2606-20041-ai-economist-agent-an-agentic-framework-for-model-grounded-economic-analysis-wit.md b/papers/items/2026-2606-20041-ai-economist-agent-an-agentic-framework-for-model-grounded-economic-analysis-wit.md
new file mode 100644
index 0000000..5f99ce3
--- /dev/null
+++ b/papers/items/2026-2606-20041-ai-economist-agent-an-agentic-framework-for-model-grounded-economic-analysis-wit.md
@@ -0,0 +1,63 @@
+# Paper: AI Economist Agent: An Agentic Framework for Model-Grounded Economic Analysis with RAG, Knowledge Graphs, and Large Language Models
+
+---
+type: paper
+title: "AI Economist Agent: An Agentic Framework for Model-Grounded Economic Analysis with RAG, Knowledge Graphs, and Large Language Models"
+authors: Masahiro Kato
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.20041
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-18
+updated_at: 2026-06-18
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - econ.GN
+  - cs.AI
+  - cs.LG
+  - q-fin.GN
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: ai-agent, rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent, rag-agent
+- inferred topics: agent-evaluation, planning, rag
+- arXiv categories: econ.GN, cs.AI, cs.LG, q-fin.GN
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.20041
diff --git a/papers/items/2026-2606-20047-pacms-submodular-context-selection-as-a-pluggable-engine-for-llm-agents.md b/papers/items/2026-2606-20047-pacms-submodular-context-selection-as-a-pluggable-engine-for-llm-agents.md
new file mode 100644
index 0000000..fa81355
--- /dev/null
+++ b/papers/items/2026-2606-20047-pacms-submodular-context-selection-as-a-pluggable-engine-for-llm-agents.md
@@ -0,0 +1,61 @@
+# Paper: PACMS: Submodular Context Selection as a Pluggable Engine for LLM Agents
+
+---
+type: paper
+title: "PACMS: Submodular Context Selection as a Pluggable Engine for LLM Agents"
+authors: Manu Ghulyani, Arunabh Singh, Karan Bharadwaj, Ankit Nath, Suranjan Goswami
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.20047
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-18
+updated_at: 2026-06-18
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.IR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, memory, rag, tool-use
+- arXiv categories: cs.IR
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.20047
diff --git a/papers/items/2026-2606-20243-phoenix-safe-github-issue-resolution-via-multi-agent-llms.md b/papers/items/2026-2606-20243-phoenix-safe-github-issue-resolution-via-multi-agent-llms.md
new file mode 100644
index 0000000..dea6699
--- /dev/null
+++ b/papers/items/2026-2606-20243-phoenix-safe-github-issue-resolution-via-multi-agent-llms.md
@@ -0,0 +1,64 @@
+# Paper: Phoenix: Safe GitHub Issue Resolution via Multi-Agent LLMs
+
+---
+type: paper
+title: "Phoenix: Safe GitHub Issue Resolution via Multi-Agent LLMs"
+authors: Kipngeno Koech, Muhammad Adam, Baimam Boukar Jean Jacques, Joao Barros
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.20243
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-18
+updated_at: 2026-06-22
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - multi-agent
+  - planning
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: agent-evaluation, agent-safety, coding-agent, multi-agent, planning, rag
+- arXiv categories: cs.SE, cs.MA
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.20243
diff --git a/papers/items/2026-2606-20401-poweragentbench-dyn-a-benchmark-for-agentic-ai-in-power-system-dynamic-studies.md b/papers/items/2026-2606-20401-poweragentbench-dyn-a-benchmark-for-agentic-ai-in-power-system-dynamic-studies.md
new file mode 100644
index 0000000..c3d6168
--- /dev/null
+++ b/papers/items/2026-2606-20401-poweragentbench-dyn-a-benchmark-for-agentic-ai-in-power-system-dynamic-studies.md
@@ -0,0 +1,66 @@
+# Paper: PowerAgentBench-Dyn: A Benchmark for Agentic AI in Power System Dynamic Studies
+
+---
+type: paper
+title: "PowerAgentBench-Dyn: A Benchmark for Agentic AI in Power System Dynamic Studies"
+authors: Qian Zhang, Andrea Pomarico, Costas Mylonas, Magda Foti, Alberto Berizzi, Le Xie
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.20401
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-18
+updated_at: 2026-06-18
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - memory
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - eess.SY
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 24
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-evaluation, agent-safety, coding-agent, memory, planning, rag, reasoning, tool-use, world-model
+- arXiv categories: eess.SY
+- collection score: 24
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.20401
diff --git a/papers/items/2026-2606-20470-analyzing-defensive-misdirection-against-model-guided-automated-attacks-on-agent.md b/papers/items/2026-2606-20470-analyzing-defensive-misdirection-against-model-guided-automated-attacks-on-agent.md
new file mode 100644
index 0000000..a223886
--- /dev/null
+++ b/papers/items/2026-2606-20470-analyzing-defensive-misdirection-against-model-guided-automated-attacks-on-agent.md
@@ -0,0 +1,62 @@
+# Paper: Analyzing Defensive Misdirection Against Model-Guided Automated Attacks on Agentic AI Systems
+
+---
+type: paper
+title: Analyzing Defensive Misdirection Against Model-Guided Automated Attacks on Agentic AI Systems
+authors: Reza Soosahabi, Vivek Namsani
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.20470
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-18
+updated_at: 2026-06-26
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-evaluation, computer-use, tool-use, workflow-agent
+- arXiv categories: cs.CR, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.20470
diff --git a/papers/items/2026-2606-20479-groundcontrol-anticipating-navigation-failures-in-vision-language-agents-via-tra.md b/papers/items/2026-2606-20479-groundcontrol-anticipating-navigation-failures-in-vision-language-agents-via-tra.md
new file mode 100644
index 0000000..f34626a
--- /dev/null
+++ b/papers/items/2026-2606-20479-groundcontrol-anticipating-navigation-failures-in-vision-language-agents-via-tra.md
@@ -0,0 +1,62 @@
+# Paper: GroundControl: Anticipating Navigation Failures in Vision-Language Agents via Trajectory-Consistent Uncertainty Estimates
+
+---
+type: paper
+title: "GroundControl: Anticipating Navigation Failures in Vision-Language Agents via Trajectory-Consistent Uncertainty Estimates"
+authors: Nastaran Darabi, Divake Kumar, Sina Tayebati, Devashri Naik, Amit Ranjan Trivedi
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.20479
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-18
+updated_at: 2026-06-18
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - embodied-agent
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.RO
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, agent-safety, embodied-agent, rag, tool-use
+- arXiv categories: cs.RO
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.20479
diff --git a/papers/items/2026-2606-20510-efficient-and-sound-probabilistic-verification-for-ai-agents.md b/papers/items/2026-2606-20510-efficient-and-sound-probabilistic-verification-for-ai-agents.md
new file mode 100644
index 0000000..426d7bf
--- /dev/null
+++ b/papers/items/2026-2606-20510-efficient-and-sound-probabilistic-verification-for-ai-agents.md
@@ -0,0 +1,62 @@
+# Paper: Efficient and Sound Probabilistic Verification for AI Agents
+
+---
+type: paper
+title: Efficient and Sound Probabilistic Verification for AI Agents
+authors: Alaia Solko-Breslin, Pramod Kaushik Mudrakarta, Mihai Christodorescu, Somesh Jha, Krishnamurthy Dj Dvijotham
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.20510
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-18
+updated_at: 2026-06-18
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent
+- inferred topics: agent-evaluation, agent-safety, computer-use, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.20510
diff --git a/papers/items/2026-2606-20512-probe-and-refine-tuning-of-repository-guidance-for-coding-agents.md b/papers/items/2026-2606-20512-probe-and-refine-tuning-of-repository-guidance-for-coding-agents.md
new file mode 100644
index 0000000..24a1b8b
--- /dev/null
+++ b/papers/items/2026-2606-20512-probe-and-refine-tuning-of-repository-guidance-for-coding-agents.md
@@ -0,0 +1,64 @@
+# Paper: Probe-and-Refine Tuning of Repository Guidance for Coding Agents
+
+---
+type: paper
+title: Probe-and-Refine Tuning of Repository Guidance for Coding Agents
+authors: Asa Shepard, Jeannie Albrecht
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.20512
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-18
+updated_at: 2026-06-19
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - computer-use
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: coding-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent, tool-use
+- inferred topics: agent-evaluation, coding-agent, computer-use, rag, tool-use, workflow-agent
+- arXiv categories: cs.SE, cs.LG
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.20512
diff --git a/papers/items/2026-2606-20515-s-agent-spatial-tool-use-elicits-reasoning-for-spatial-intelligence.md b/papers/items/2026-2606-20515-s-agent-spatial-tool-use-elicits-reasoning-for-spatial-intelligence.md
new file mode 100644
index 0000000..4ea5553
--- /dev/null
+++ b/papers/items/2026-2606-20515-s-agent-spatial-tool-use-elicits-reasoning-for-spatial-intelligence.md
@@ -0,0 +1,62 @@
+# Paper: S-Agent: Spatial Tool-Use Elicits Reasoning for Spatial Intelligence
+
+---
+type: paper
+title: "S-Agent: Spatial Tool-Use Elicits Reasoning for Spatial Intelligence"
+authors: Yalun Dai, Hao Li, Shulin Tian, Runmao Yao, Yuhao Dong, Fangzhou Hong, Zhaoxi Chen, Fangfu Liu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.20515
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-18
+updated_at: 2026-06-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-memory, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory, tool-use
+- inferred topics: agent-evaluation, memory, planning, reasoning, tool-use
+- arXiv categories: cs.CV
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.20515
diff --git a/papers/items/2026-2606-20573-aona-a-comprehensive-architecture-and-workflow-design-for-global-agentic-collabo.md b/papers/items/2026-2606-20573-aona-a-comprehensive-architecture-and-workflow-design-for-global-agentic-collabo.md
new file mode 100644
index 0000000..dc4a64d
--- /dev/null
+++ b/papers/items/2026-2606-20573-aona-a-comprehensive-architecture-and-workflow-design-for-global-agentic-collabo.md
@@ -0,0 +1,61 @@
+# Paper: AONA: A Comprehensive Architecture and Workflow Design for Global Agentic Collaboration
+
+---
+type: paper
+title: "AONA: A Comprehensive Architecture and Workflow Design for Global Agentic Collaboration"
+authors: Jinliang Xu, Runkai Zhu, Bingqi Li, Fanjie Nie, Jin Li, Jiagui Xie
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.20573
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-04-30
+updated_at: 2026-04-30
+status: queued
+relevance: high
+topics:
+  - multi-agent
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.NI
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: multi-agent, tool-use, workflow-agent
+- arXiv categories: cs.NI, cs.MA
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.20573
diff --git a/papers/items/2026-2606-20629-specialize-roles-mix-deployments-pushing-the-cost-accuracy-frontier-of-llm-agent.md b/papers/items/2026-2606-20629-specialize-roles-mix-deployments-pushing-the-cost-accuracy-frontier-of-llm-agent.md
new file mode 100644
index 0000000..f8e54e6
--- /dev/null
+++ b/papers/items/2026-2606-20629-specialize-roles-mix-deployments-pushing-the-cost-accuracy-frontier-of-llm-agent.md
@@ -0,0 +1,65 @@
+# Paper: Specialize Roles, Mix Deployments: Pushing the Cost-Accuracy Frontier of LLM Agent Teams
+
+---
+type: paper
+title: "Specialize Roles, Mix Deployments: Pushing the Cost-Accuracy Frontier of LLM Agent Teams"
+authors: Yinsicheng Jiang, Liang Cheng, Yeqi Huang, Yufan Zhao, Zhan Lu, Li Dong, Wenda Li, Edoardo Ponti, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.20629
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-05-28
+updated_at: 2026-05-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, computer-use, planning, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.MA, cs.AI, cs.LG
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.20629
diff --git a/papers/items/2026-2606-20717-mirage-stealthy-visual-prompt-injection-for-vulnerability-detection-in-web-agent.md b/papers/items/2026-2606-20717-mirage-stealthy-visual-prompt-injection-for-vulnerability-detection-in-web-agent.md
new file mode 100644
index 0000000..509440d
--- /dev/null
+++ b/papers/items/2026-2606-20717-mirage-stealthy-visual-prompt-injection-for-vulnerability-detection-in-web-agent.md
@@ -0,0 +1,65 @@
+# Paper: MIRAGE: Stealthy Visual Prompt Injection for Vulnerability Detection in Web Agents
+
+---
+type: paper
+title: "MIRAGE: Stealthy Visual Prompt Injection for Vulnerability Detection in Web Agents"
+authors: Xuelong Dai, Jianyu Ma, Boyang Ma, Biwei Yan, Yijun Yang, Yue Zhang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.20717
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-16
+updated_at: 2026-06-16
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+  - cs.AI
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, agent-safety, computer-use, rag, tool-use, workflow-agent
+- arXiv categories: cs.CV, cs.AI, cs.CR
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.20717
diff --git a/papers/items/2026-2606-20785-fara-1-5-scalable-learning-environments-for-computer-use-agents.md b/papers/items/2026-2606-20785-fara-1-5-scalable-learning-environments-for-computer-use-agents.md
new file mode 100644
index 0000000..62244dd
--- /dev/null
+++ b/papers/items/2026-2606-20785-fara-1-5-scalable-learning-environments-for-computer-use-agents.md
@@ -0,0 +1,63 @@
+# Paper: Fara-1.5: Scalable Learning Environments for Computer Use Agents
+
+---
+type: paper
+title: "Fara-1.5: Scalable Learning Environments for Computer Use Agents"
+authors: Ahmed Awadallah, Sahil Gupta, Yash Lara, Yadong Lu, Hussein Mozannar, Akshay Nambi, Zach Nussbaum, Yash Pandya, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.20785
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-18
+updated_at: 2026-06-18
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, computer-use, rag, reasoning, tool-use
+- arXiv categories: cs.AI, cs.LG
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.20785
diff --git a/papers/items/2026-2606-20922-think-twice-before-you-act-protecting-llm-agents-against-tool-description-poison.md b/papers/items/2026-2606-20922-think-twice-before-you-act-protecting-llm-agents-against-tool-description-poison.md
new file mode 100644
index 0000000..c73b637
--- /dev/null
+++ b/papers/items/2026-2606-20922-think-twice-before-you-act-protecting-llm-agents-against-tool-description-poison.md
@@ -0,0 +1,61 @@
+# Paper: Think Twice Before You Act: Protecting LLM Agents Against Tool Description Poisoning via Isolated Planning
+
+---
+type: paper
+title: "Think Twice Before You Act: Protecting LLM Agents Against Tool Description Poisoning via Isolated Planning"
+authors: Shanghao Shi, Xiao Wang, Chaoyu Zhang, Hao Li, Wenjing Lou, Thomas Hou, Yevgeniy Vorobeychik, Chongjie Zhang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.20922
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-18
+updated_at: 2026-06-18
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, agent-safety, planning, tool-use
+- arXiv categories: cs.CR
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.20922
diff --git a/papers/items/2026-2606-20950-power-systems-agent-benchmark-executable-evaluation-of-ai-agents-in-electric-pow.md b/papers/items/2026-2606-20950-power-systems-agent-benchmark-executable-evaluation-of-ai-agents-in-electric-pow.md
new file mode 100644
index 0000000..7d3ddba
--- /dev/null
+++ b/papers/items/2026-2606-20950-power-systems-agent-benchmark-executable-evaluation-of-ai-agents-in-electric-pow.md
@@ -0,0 +1,61 @@
+# Paper: Power Systems Agent Benchmark: Executable Evaluation of AI Agents in Electric Power Engineering
+
+---
+type: paper
+title: "Power Systems Agent Benchmark: Executable Evaluation of AI Agents in Electric Power Engineering"
+authors: Sergei Trashchenkov
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.20950
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-18
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - eess.SY
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agent-evaluation, ai-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, ai-agent, tool-use
+- inferred topics: agent-evaluation, rag, tool-use
+- arXiv categories: cs.AI, eess.SY
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.20950
diff --git a/papers/items/2026-2606-20954-learning-what-not-to-forget-long-horizon-agent-memory-from-a-few-kilobytes-of-le.md b/papers/items/2026-2606-20954-learning-what-not-to-forget-long-horizon-agent-memory-from-a-few-kilobytes-of-le.md
new file mode 100644
index 0000000..8bb167f
--- /dev/null
+++ b/papers/items/2026-2606-20954-learning-what-not-to-forget-long-horizon-agent-memory-from-a-few-kilobytes-of-le.md
@@ -0,0 +1,62 @@
+# Paper: Learning What Not to Forget: Long-Horizon Agent Memory from a Few Kilobytes of Learning
+
+---
+type: paper
+title: "Learning What Not to Forget: Long-Horizon Agent Memory from a Few Kilobytes of Learning"
+authors: Nusrat Jahan Lia, Aritra Mazumder
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.20954
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-18
+updated_at: 2026-06-18
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, planning, tool-use
+- arXiv categories: cs.CL, cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.20954
diff --git a/papers/items/2026-2606-21013-agentic-time-machine-as-an-infrastructure-for-future-event-forecasting.md b/papers/items/2026-2606-21013-agentic-time-machine-as-an-infrastructure-for-future-event-forecasting.md
new file mode 100644
index 0000000..c856d51
--- /dev/null
+++ b/papers/items/2026-2606-21013-agentic-time-machine-as-an-infrastructure-for-future-event-forecasting.md
@@ -0,0 +1,63 @@
+# Paper: Agentic Time Machine as an Infrastructure for Future-Event Forecasting
+
+---
+type: paper
+title: Agentic Time Machine as an Infrastructure for Future-Event Forecasting
+authors: Jingyi Chai, Bingyang Zheng, Xiangrui Liu, Hao Lu, Zihang Zhou, Tianchen Wang, Kemeng Zhang, Siheng Chen
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.21013
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-19
+updated_at: 2026-06-19
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, multi-agent, planning, rag, tool-use
+- arXiv categories: cs.AI, cs.LG
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.21013
diff --git a/papers/items/2026-2606-21123-a-multi-agent-audit-framework-for-high-stakes-reasoning-evaluation-and-interpret.md b/papers/items/2026-2606-21123-a-multi-agent-audit-framework-for-high-stakes-reasoning-evaluation-and-interpret.md
new file mode 100644
index 0000000..7241177
--- /dev/null
+++ b/papers/items/2026-2606-21123-a-multi-agent-audit-framework-for-high-stakes-reasoning-evaluation-and-interpret.md
@@ -0,0 +1,62 @@
+# Paper: A Multi-Agent Audit Framework for High-Stakes Reasoning: Evaluation and Interpretability in Clinical Mental Health Screening
+
+---
+type: paper
+title: "A Multi-Agent Audit Framework for High-Stakes Reasoning: Evaluation and Interpretability in Clinical Mental Health Screening"
+authors: Jingchen Ye, Yanpei Yu, Luyao Zhang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.21123
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-19
+updated_at: 2026-06-19
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - rag
+  - reasoning
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, multi-agent, rag, reasoning, workflow-agent
+- arXiv categories: cs.CL
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.21123
diff --git a/papers/items/2026-2606-21129-agenticos-an-intent-oriented-secure-operating-system-architecture-for-autonomous.md b/papers/items/2026-2606-21129-agenticos-an-intent-oriented-secure-operating-system-architecture-for-autonomous.md
new file mode 100644
index 0000000..96a7d13
--- /dev/null
+++ b/papers/items/2026-2606-21129-agenticos-an-intent-oriented-secure-operating-system-architecture-for-autonomous.md
@@ -0,0 +1,61 @@
+# Paper: AgenticOS: An Intent-Oriented Secure Operating System Architecture for Autonomous AI Agents
+
+---
+type: paper
+title: "AgenticOS: An Intent-Oriented Secure Operating System Architecture for Autonomous AI Agents"
+authors: Zhen Zhao, Yu Zhang, Yanpeng Zhu, Jia Wang, Songqiao Tao, Xin Cheng, Jiexin Gao
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.21129
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-19
+updated_at: 2026-06-19
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.OS
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: ai-agent, autonomous-agent-llm, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent, autonomous-agent-llm, tool-use
+- inferred topics: agent-safety, planning, tool-use
+- arXiv categories: cs.CR, cs.OS
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.21129
diff --git a/papers/items/2026-2606-21228-sakana-fugu-technical-report.md b/papers/items/2026-2606-21228-sakana-fugu-technical-report.md
new file mode 100644
index 0000000..31acba7
--- /dev/null
+++ b/papers/items/2026-2606-21228-sakana-fugu-technical-report.md
@@ -0,0 +1,61 @@
+# Paper: Sakana Fugu Technical Report
+
+---
+type: paper
+title: Sakana Fugu Technical Report
+authors: Yujin Tang, Edoardo Cetin, Jinglue Xu, Qi Sun, Stefan Nielsen, Vincent Richard, Haruto Goda, Iaroslav Tymchenko, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.21228
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-19
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - coding-agent
+  - multi-agent
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: coding-agent, multi-agent, rag, reasoning
+- arXiv categories: cs.LG
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.21228
diff --git a/papers/items/2026-2606-21401-swarmx-agentic-scheduling-for-low-latency-agentic-systems.md b/papers/items/2026-2606-21401-swarmx-agentic-scheduling-for-low-latency-agentic-systems.md
new file mode 100644
index 0000000..03de065
--- /dev/null
+++ b/papers/items/2026-2606-21401-swarmx-agentic-scheduling-for-low-latency-agentic-systems.md
@@ -0,0 +1,62 @@
+# Paper: SwarmX: Agentic Scheduling for Low-Latency Agentic Systems
+
+---
+type: paper
+title: "SwarmX: Agentic Scheduling for Low-Latency Agentic Systems"
+authors: Yeqi Huang, Yanwei Ye, Guomin Chen, Wenhao Su, Bin Gong, Jialian Li, Zhan Lu, Yangshen Deng, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.21401
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-19
+updated_at: 2026-06-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.DC
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-evaluation, multi-agent, tool-use, workflow-agent
+- arXiv categories: cs.DC, cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.21401
diff --git a/papers/items/2026-2606-21409-don-t-blindly-trust-it-how-unreliable-feedback-breaks-tool-using-llm-agents.md b/papers/items/2026-2606-21409-don-t-blindly-trust-it-how-unreliable-feedback-breaks-tool-using-llm-agents.md
new file mode 100644
index 0000000..cd843a4
--- /dev/null
+++ b/papers/items/2026-2606-21409-don-t-blindly-trust-it-how-unreliable-feedback-breaks-tool-using-llm-agents.md
@@ -0,0 +1,61 @@
+# Paper: Don't Blindly Trust It: How Unreliable Feedback Breaks Tool-Using LLM Agents
+
+---
+type: paper
+title: "Don't Blindly Trust It: How Unreliable Feedback Breaks Tool-Using LLM Agents"
+authors: Chubin Zhang, Zhenglin Wan, Xingrui Yu, Pengfei Zhou, Wangbo Zhao, Jingxuan Wu, Yaxin Zhou, Ivor Tsang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.21409
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-19
+updated_at: 2026-06-19
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, coding-agent, rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.21409
diff --git a/papers/items/2026-2606-21445-autoras-learning-robust-agentic-systems-with-primitive-representations.md b/papers/items/2026-2606-21445-autoras-learning-robust-agentic-systems-with-primitive-representations.md
new file mode 100644
index 0000000..c7ef1cf
--- /dev/null
+++ b/papers/items/2026-2606-21445-autoras-learning-robust-agentic-systems-with-primitive-representations.md
@@ -0,0 +1,62 @@
+# Paper: AutoRAS: Learning Robust Agentic Systems with Primitive Representations
+
+---
+type: paper
+title: "AutoRAS: Learning Robust Agentic Systems with Primitive Representations"
+authors: Yang Yue, Xuancheng Zhu, Yuyang Ma, Guoshun Nan, Zihan Dou, Jingru Shan, Congyu Guo, Ji Zhang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.21445
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-19
+updated_at: 2026-06-19
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - multi-agent
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-safety, multi-agent, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.21445
diff --git a/papers/items/2026-2606-21553-dissecting-agentic-rag-a-component-ablation-for-multi-hop-qa-with-a-local-7b-mod.md b/papers/items/2026-2606-21553-dissecting-agentic-rag-a-component-ablation-for-multi-hop-qa-with-a-local-7b-mod.md
new file mode 100644
index 0000000..96ff47f
--- /dev/null
+++ b/papers/items/2026-2606-21553-dissecting-agentic-rag-a-component-ablation-for-multi-hop-qa-with-a-local-7b-mod.md
@@ -0,0 +1,62 @@
+# Paper: Dissecting Agentic RAG: A Component Ablation for Multi-Hop QA with a Local 7B Model
+
+---
+type: paper
+title: "Dissecting Agentic RAG: A Component Ablation for Multi-Hop QA with a Local 7B Model"
+authors: Sheroz Shaikh
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.21553
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-19
+updated_at: 2026-06-19
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.IR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, rag, reasoning, tool-use
+- arXiv categories: cs.CL, cs.IR
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.21553
diff --git a/papers/items/2026-2606-21565-composing-verifiable-conceptual-models-via-building-blocks-towards-design-time-v.md b/papers/items/2026-2606-21565-composing-verifiable-conceptual-models-via-building-blocks-towards-design-time-v.md
new file mode 100644
index 0000000..d605684
--- /dev/null
+++ b/papers/items/2026-2606-21565-composing-verifiable-conceptual-models-via-building-blocks-towards-design-time-v.md
@@ -0,0 +1,62 @@
+# Paper: Composing Verifiable Conceptual Models via Building Blocks: Towards Design-Time Verification of Agentic AI Workflows
+
+---
+type: paper
+title: "Composing Verifiable Conceptual Models via Building Blocks: Towards Design-Time Verification of Agentic AI Workflows"
+authors: Noe Y. Flandre, Alexander C. Nwala, Philippe J. Giabbanelli
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.21565
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-19
+updated_at: 2026-06-19
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - reasoning
+  - tool-use
+  - workflow-agent
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-evaluation, reasoning, tool-use, workflow-agent, world-model
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.21565
diff --git a/papers/items/2026-2606-21627-counsel-a-meta-evaluation-dataset-for-agentic-tasks.md b/papers/items/2026-2606-21627-counsel-a-meta-evaluation-dataset-for-agentic-tasks.md
new file mode 100644
index 0000000..4f27e82
--- /dev/null
+++ b/papers/items/2026-2606-21627-counsel-a-meta-evaluation-dataset-for-agentic-tasks.md
@@ -0,0 +1,62 @@
+# Paper: Counsel: A Meta-Evaluation Dataset for Agentic Tasks
+
+---
+type: paper
+title: "Counsel: A Meta-Evaluation Dataset for Agentic Tasks"
+authors: Sashank Pisupati, Henry Broomfield, Eujeong Choi, Antonia Calvi, Charlie Wang, Roman Engeler, Max Bartolo, Patrick Lewis
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.21627
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-19
+updated_at: 2026-06-19
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: agent-evaluation, coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, coding-agent
+- inferred topics: agent-evaluation, agent-safety, coding-agent, reasoning
+- arXiv categories: cs.AI, cs.LG
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.21627
diff --git a/papers/items/2026-2606-21649-evoembedding-evolvable-representations-for-long-context-retrieval-and-agentic-me.md b/papers/items/2026-2606-21649-evoembedding-evolvable-representations-for-long-context-retrieval-and-agentic-me.md
new file mode 100644
index 0000000..a30dede
--- /dev/null
+++ b/papers/items/2026-2606-21649-evoembedding-evolvable-representations-for-long-context-retrieval-and-agentic-me.md
@@ -0,0 +1,62 @@
+# Paper: EvoEmbedding: Evolvable Representations for Long-Context Retrieval and Agentic Memory
+
+---
+type: paper
+title: "EvoEmbedding: Evolvable Representations for Long-Context Retrieval and Agentic Memory"
+authors: Chang Nie, Chaoyou Fu, Junlan Feng, Caifeng Shan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.21649
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-19
+updated_at: 2026-06-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - memory
+  - rag
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agent-memory, agentic-ai, rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory, agentic-ai, rag-agent
+- inferred topics: agent-evaluation, coding-agent, memory, rag, workflow-agent
+- arXiv categories: cs.CL
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.21649
diff --git a/papers/items/2026-2606-21710-privacyalign-contextual-privacy-alignment-for-llm-agents.md b/papers/items/2026-2606-21710-privacyalign-contextual-privacy-alignment-for-llm-agents.md
new file mode 100644
index 0000000..bd7f06e
--- /dev/null
+++ b/papers/items/2026-2606-21710-privacyalign-contextual-privacy-alignment-for-llm-agents.md
@@ -0,0 +1,63 @@
+# Paper: PrivacyAlign: Contextual Privacy Alignment for LLM Agents
+
+---
+type: paper
+title: "PrivacyAlign: Contextual Privacy Alignment for LLM Agents"
+authors: Manveer Singh Tamber, Abhay Puri, Marc-Etienne Brunet, Perouz Taslakian, Jimmy Lin, Spandana Gella
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.21710
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-19
+updated_at: 2026-06-19
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+  - cs.IR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent
+- inferred topics: agent-evaluation, agent-safety, planning, tool-use
+- arXiv categories: cs.CL, cs.AI, cs.IR
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.21710
diff --git a/papers/items/2026-2606-21732-safe-to-check-unsafe-to-use-relinking-at-the-compression-boundary-of-llm-agents.md b/papers/items/2026-2606-21732-safe-to-check-unsafe-to-use-relinking-at-the-compression-boundary-of-llm-agents.md
new file mode 100644
index 0000000..61dd2f8
--- /dev/null
+++ b/papers/items/2026-2606-21732-safe-to-check-unsafe-to-use-relinking-at-the-compression-boundary-of-llm-agents.md
@@ -0,0 +1,62 @@
+# Paper: Safe to Check, Unsafe to Use: Relinking at the Compression Boundary of LLM Agents
+
+---
+type: paper
+title: "Safe to Check, Unsafe to Use: Relinking at the Compression Boundary of LLM Agents"
+authors: Zesen Liu, Zihan Zhang, Dongdong She
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.21732
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-19
+updated_at: 2026-06-19
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, agent-safety, rag, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.21732
diff --git a/papers/items/2026-2606-21740-training-the-orchestrator-a-supervised-approach-to-end-to-end-pddl-planning-with.md b/papers/items/2026-2606-21740-training-the-orchestrator-a-supervised-approach-to-end-to-end-pddl-planning-with.md
new file mode 100644
index 0000000..1bb4457
--- /dev/null
+++ b/papers/items/2026-2606-21740-training-the-orchestrator-a-supervised-approach-to-end-to-end-pddl-planning-with.md
@@ -0,0 +1,62 @@
+# Paper: Training the Orchestrator: A Supervised Approach to End-to-End PDDL Planning with LLM Agents
+
+---
+type: paper
+title: "Training the Orchestrator: A Supervised Approach to End-to-End PDDL Planning with LLM Agents"
+authors: Rajesh Mangannavar, Zachary Coalson, Pranay Dugar, Prasad Tadepalli
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.21740
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-19
+updated_at: 2026-06-19
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, computer-use, planning, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.21740
diff --git a/papers/items/2026-2606-21836-agentdse-reasoning-augmented-architectural-design-space-exploration.md b/papers/items/2026-2606-21836-agentdse-reasoning-augmented-architectural-design-space-exploration.md
new file mode 100644
index 0000000..ee7e384
--- /dev/null
+++ b/papers/items/2026-2606-21836-agentdse-reasoning-augmented-architectural-design-space-exploration.md
@@ -0,0 +1,62 @@
+# Paper: AgentDSE: Reasoning-Augmented Architectural Design Space Exploration
+
+---
+type: paper
+title: "AgentDSE: Reasoning-Augmented Architectural Design Space Exploration"
+authors: Chenyu Wang, Jiahe Caroline Shi, David Kong, Duane S. Boning, Zishen Wan, Yilun Du, Vijay Janapa Reddi
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.21836
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-20
+updated_at: 2026-06-20
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - computer-use
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: agent-evaluation, coding-agent, computer-use, reasoning
+- arXiv categories: cs.AR, cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.21836
diff --git a/papers/items/2026-2606-21842-agent-assisted-side-channel-attacks-on-non-prefix-kv-cache-in-rag.md b/papers/items/2026-2606-21842-agent-assisted-side-channel-attacks-on-non-prefix-kv-cache-in-rag.md
new file mode 100644
index 0000000..15260d0
--- /dev/null
+++ b/papers/items/2026-2606-21842-agent-assisted-side-channel-attacks-on-non-prefix-kv-cache-in-rag.md
@@ -0,0 +1,62 @@
+# Paper: Agent-Assisted Side-Channel Attacks on Non-Prefix KV Cache in RAG
+
+---
+type: paper
+title: Agent-Assisted Side-Channel Attacks on Non-Prefix KV Cache in RAG
+authors: He Sun, Shinan Liu, Siyuan Ma, Junhao Li, Mingjun Xiao, Wenhao Jiang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.21842
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-20
+updated_at: 2026-06-20
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, agent-safety, memory, rag, tool-use
+- arXiv categories: cs.CR
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.21842
diff --git a/papers/items/2026-2606-21877-agentriskbom-a-risk-scoping-security-bill-of-materials-for-agentic-ai-systems.md b/papers/items/2026-2606-21877-agentriskbom-a-risk-scoping-security-bill-of-materials-for-agentic-ai-systems.md
new file mode 100644
index 0000000..c57740b
--- /dev/null
+++ b/papers/items/2026-2606-21877-agentriskbom-a-risk-scoping-security-bill-of-materials-for-agentic-ai-systems.md
@@ -0,0 +1,66 @@
+# Paper: AgentRiskBOM: A Risk-Scoping Security Bill of Materials for Agentic AI Systems
+
+---
+type: paper
+title: "AgentRiskBOM: A Risk-Scoping Security Bill of Materials for Agentic AI Systems"
+authors: Srimonti Dutta, Akshata Kishore Moharir
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.21877
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-20
+updated_at: 2026-06-20
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - memory
+  - multi-agent
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CR
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 21
+collection_queries: agentic-ai, ai-agent, rag-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, ai-agent, rag-agent, tool-use
+- inferred topics: agent-evaluation, agent-safety, coding-agent, memory, multi-agent, rag, tool-use
+- arXiv categories: cs.AI, cs.CR, cs.SE
+- collection score: 21
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.21877
diff --git a/papers/items/2026-2606-21963-holmes-multimodal-agentic-diagnosis-for-mixed-language-mobile-crashes-at-industr.md b/papers/items/2026-2606-21963-holmes-multimodal-agentic-diagnosis-for-mixed-language-mobile-crashes-at-industr.md
new file mode 100644
index 0000000..13b812e
--- /dev/null
+++ b/papers/items/2026-2606-21963-holmes-multimodal-agentic-diagnosis-for-mixed-language-mobile-crashes-at-industr.md
@@ -0,0 +1,63 @@
+# Paper: Holmes: Multimodal Agentic Diagnosis for Mixed-Language Mobile Crashes at Industrial Scale
+
+---
+type: paper
+title: "Holmes: Multimodal Agentic Diagnosis for Mixed-Language Mobile Crashes at Industrial Scale"
+authors: Jia Li, Wenyuan Ma, Ting Peng, Haibin Zheng, Yuetang Deng
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.21963
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-20
+updated_at: 2026-06-20
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - multi-agent
+  - rag
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, computer-use, multi-agent, rag, workflow-agent
+- arXiv categories: cs.AI, cs.SE
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.21963
diff --git a/papers/items/2026-2606-22030-nous-a-predictive-world-model-for-long-term-agent-memory.md b/papers/items/2026-2606-22030-nous-a-predictive-world-model-for-long-term-agent-memory.md
new file mode 100644
index 0000000..b740456
--- /dev/null
+++ b/papers/items/2026-2606-22030-nous-a-predictive-world-model-for-long-term-agent-memory.md
@@ -0,0 +1,64 @@
+# Paper: Nous: A Predictive World Model for Long-Term Agent Memory
+
+---
+type: paper
+title: "Nous: A Predictive World Model for Long-Term Agent Memory"
+authors: Pranav Singh
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.22030
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-20
+updated_at: 2026-06-20
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+  - cs.IR
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, rag, world-model
+- arXiv categories: cs.AI, cs.CL, cs.IR, cs.LG
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.22030
diff --git a/papers/items/2026-2606-22082-codeteam-an-llm-powered-multi-agent-framework-for-repository-level-code-generati.md b/papers/items/2026-2606-22082-codeteam-an-llm-powered-multi-agent-framework-for-repository-level-code-generati.md
new file mode 100644
index 0000000..9015b3d
--- /dev/null
+++ b/papers/items/2026-2606-22082-codeteam-an-llm-powered-multi-agent-framework-for-repository-level-code-generati.md
@@ -0,0 +1,63 @@
+# Paper: CodeTeam: An LLM-Powered Multi-Agent Framework for Repository-Level Code Generation
+
+---
+type: paper
+title: "CodeTeam: An LLM-Powered Multi-Agent Framework for Repository-Level Code Generation"
+authors: Yifei Wang, Ruiyin Li, Peng Liang, Qiong Feng, Zengyang Li, Mojtaba Shahin, Arif Ali Khan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.22082
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-20
+updated_at: 2026-06-20
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - multi-agent
+  - planning
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, coding-agent, multi-agent, planning, rag
+- arXiv categories: cs.SE, cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.22082
diff --git a/papers/items/2026-2606-22110-traceview-interactive-visualization-of-agentic-program-repair-trajectories.md b/papers/items/2026-2606-22110-traceview-interactive-visualization-of-agentic-program-repair-trajectories.md
new file mode 100644
index 0000000..d86416b
--- /dev/null
+++ b/papers/items/2026-2606-22110-traceview-interactive-visualization-of-agentic-program-repair-trajectories.md
@@ -0,0 +1,64 @@
+# Paper: TraceView: Interactive Visualization of Agentic Program Repair Trajectories
+
+---
+type: paper
+title: "TraceView: Interactive Visualization of Agentic Program Repair Trajectories"
+authors: Amirali Sajadi, Tu Nguyen, Kimmie Huynh, Esteban Parra, Preetha Chatterjee
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.22110
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-20
+updated_at: 2026-06-20
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+  - cs.HC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, coding-agent, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.SE, cs.AI, cs.HC
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.22110
diff --git a/papers/items/2026-2606-22151-novelty-aware-agentic-retrieval-comparing-research-contributions-through-structu.md b/papers/items/2026-2606-22151-novelty-aware-agentic-retrieval-comparing-research-contributions-through-structu.md
new file mode 100644
index 0000000..8dca78e
--- /dev/null
+++ b/papers/items/2026-2606-22151-novelty-aware-agentic-retrieval-comparing-research-contributions-through-structu.md
@@ -0,0 +1,62 @@
+# Paper: Novelty-Aware Agentic Retrieval: Comparing Research Contributions Through Structured Multi-Step Reasoning
+
+---
+type: paper
+title: "Novelty-Aware Agentic Retrieval: Comparing Research Contributions Through Structured Multi-Step Reasoning"
+authors: Shou-Tzu Han
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.22151
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-20
+updated_at: 2026-06-20
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.IR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, computer-use, rag, reasoning, tool-use
+- arXiv categories: cs.IR
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.22151
diff --git a/papers/items/2026-2606-22263-revelio-cost-efficient-agentic-memory-safety-vulnerability-detection-for-reposit.md b/papers/items/2026-2606-22263-revelio-cost-efficient-agentic-memory-safety-vulnerability-detection-for-reposit.md
new file mode 100644
index 0000000..f4974b9
--- /dev/null
+++ b/papers/items/2026-2606-22263-revelio-cost-efficient-agentic-memory-safety-vulnerability-detection-for-reposit.md
@@ -0,0 +1,64 @@
+# Paper: Revelio: Cost-Efficient Agentic Memory Safety Vulnerability Detection For Repository-Scale Codebases
+
+---
+type: paper
+title: "Revelio: Cost-Efficient Agentic Memory Safety Vulnerability Detection For Repository-Scale Codebases"
+authors: Yiwei Hou, Hao Wang, Muxi Lyu, Marius Momeu, Eric Nguyen, Taige Yang, Koushik Sen, Dawn Song, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.22263
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-20
+updated_at: 2026-06-20
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - memory
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+  - cs.MA
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: agent-memory, coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory, coding-agent
+- inferred topics: agent-evaluation, agent-safety, coding-agent, memory
+- arXiv categories: cs.CR, cs.AI, cs.MA, cs.SE
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.22263
diff --git a/papers/items/2026-2606-22330-hypothesis-driven-skill-optimization-for-llm-agents.md b/papers/items/2026-2606-22330-hypothesis-driven-skill-optimization-for-llm-agents.md
new file mode 100644
index 0000000..a0dac4d
--- /dev/null
+++ b/papers/items/2026-2606-22330-hypothesis-driven-skill-optimization-for-llm-agents.md
@@ -0,0 +1,64 @@
+# Paper: Hypothesis-Driven Skill Optimization for LLM Agents
+
+---
+type: paper
+title: Hypothesis-Driven Skill Optimization for LLM Agents
+authors: Fangxin Shang, Yehui Yang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.22330
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-21
+updated_at: 2026-06-21
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - coding-agent
+  - memory
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-safety, coding-agent, memory, planning, reasoning, tool-use
+- arXiv categories: cs.AI, cs.SE
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.22330
diff --git a/papers/items/2026-2606-22388-planbench-xl-evaluating-long-horizon-planning-of-llm-tool-use-agents-in-large-sc.md b/papers/items/2026-2606-22388-planbench-xl-evaluating-long-horizon-planning-of-llm-tool-use-agents-in-large-sc.md
new file mode 100644
index 0000000..be7dd7a
--- /dev/null
+++ b/papers/items/2026-2606-22388-planbench-xl-evaluating-long-horizon-planning-of-llm-tool-use-agents-in-large-sc.md
@@ -0,0 +1,62 @@
+# Paper: PlanBench-XL: Evaluating Long-Horizon Planning of LLM Tool-Use Agents in Large-Scale Tool Ecosystems
+
+---
+type: paper
+title: "PlanBench-XL: Evaluating Long-Horizon Planning of LLM Tool-Use Agents in Large-Scale Tool Ecosystems"
+authors: Jiayu Liu, Qihan Lin, Cheng Qian, Rui Wang, Emre Can Acikgoz, Xiaocheng Yang, Jiateng Liu, Zhenhailong Wang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.22388
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-21
+updated_at: 2026-06-21
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: planning-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent, tool-use
+- inferred topics: agent-evaluation, planning, rag, tool-use
+- arXiv categories: cs.AI, cs.CL
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.22388
diff --git a/papers/items/2026-2606-22417-code-isn-t-memory-a-structural-codebase-index-inside-a-coding-agent.md b/papers/items/2026-2606-22417-code-isn-t-memory-a-structural-codebase-index-inside-a-coding-agent.md
new file mode 100644
index 0000000..ce2c7dd
--- /dev/null
+++ b/papers/items/2026-2606-22417-code-isn-t-memory-a-structural-codebase-index-inside-a-coding-agent.md
@@ -0,0 +1,61 @@
+# Paper: Code Isn't Memory: A Structural Codebase Index Inside a Coding Agent
+
+---
+type: paper
+title: "Code Isn't Memory: A Structural Codebase Index Inside a Coding Agent"
+authors: Ishaan Bhola, Adithyan Krishnan, Sravanth Kurmala, Mukunda NS
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.22417
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-21
+updated_at: 2026-06-21
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - memory
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: agent-evaluation, coding-agent, memory, rag
+- arXiv categories: cs.AI
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.22417
diff --git a/papers/items/2026-2606-22484-governed-ai-assisted-engineering-graduated-human-oversight-for-agentic-code-gene.md b/papers/items/2026-2606-22484-governed-ai-assisted-engineering-graduated-human-oversight-for-agentic-code-gene.md
new file mode 100644
index 0000000..be42d35
--- /dev/null
+++ b/papers/items/2026-2606-22484-governed-ai-assisted-engineering-graduated-human-oversight-for-agentic-code-gene.md
@@ -0,0 +1,62 @@
+# Paper: Governed AI-Assisted Engineering: Graduated Human Oversight for Agentic Code Generation in Regulated Domains
+
+---
+type: paper
+title: "Governed AI-Assisted Engineering: Graduated Human Oversight for Agentic Code Generation in Regulated Domains"
+authors: Richard Kang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.22484
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-21
+updated_at: 2026-07-04
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - rag
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.HC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-evaluation, agent-safety, coding-agent, rag, workflow-agent
+- arXiv categories: cs.HC
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.22484
diff --git a/papers/items/2026-2606-22495-grounded-scaling-why-agentic-ai-needs-deterministic-environments.md b/papers/items/2026-2606-22495-grounded-scaling-why-agentic-ai-needs-deterministic-environments.md
new file mode 100644
index 0000000..15f8828
--- /dev/null
+++ b/papers/items/2026-2606-22495-grounded-scaling-why-agentic-ai-needs-deterministic-environments.md
@@ -0,0 +1,62 @@
+# Paper: Grounded Scaling: Why Agentic AI Needs Deterministic Environments
+
+---
+type: paper
+title: "Grounded Scaling: Why Agentic AI Needs Deterministic Environments"
+authors: Liang Ding, Xintong Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.22495
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-21
+updated_at: 2026-06-21
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - embodied-agent
+  - multi-agent
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-safety, embodied-agent, multi-agent, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.22495
diff --git a/papers/items/2026-2606-22557-macagentbench-benchmarking-ai-agents-on-real-world-macos-desktop.md b/papers/items/2026-2606-22557-macagentbench-benchmarking-ai-agents-on-real-world-macos-desktop.md
new file mode 100644
index 0000000..750ceef
--- /dev/null
+++ b/papers/items/2026-2606-22557-macagentbench-benchmarking-ai-agents-on-real-world-macos-desktop.md
@@ -0,0 +1,65 @@
+# Paper: MacAgentBench: Benchmarking AI Agents on Real-World macOS Desktop
+
+---
+type: paper
+title: "MacAgentBench: Benchmarking AI Agents on Real-World macOS Desktop"
+authors: Yikun Fu, Bowen Fu, Zhenyu Wu, Shuang Cheng, Xiaowei Sun, Bowen Yang, Zehao Li, Yibo Zhao, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.22557
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-21
+updated_at: 2026-06-21
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+  - cs.HC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 21
+collection_queries: agent-evaluation, ai-agent, web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, ai-agent, web-gui-agent
+- inferred topics: agent-evaluation, computer-use, planning, rag, tool-use, workflow-agent
+- arXiv categories: cs.AI, cs.CL, cs.HC
+- collection score: 21
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.22557
diff --git a/papers/items/2026-2606-22610-paperclaw-harnessing-agents-for-autonomous-research-and-human-in-the-loop-refine.md b/papers/items/2026-2606-22610-paperclaw-harnessing-agents-for-autonomous-research-and-human-in-the-loop-refine.md
new file mode 100644
index 0000000..34c7d02
--- /dev/null
+++ b/papers/items/2026-2606-22610-paperclaw-harnessing-agents-for-autonomous-research-and-human-in-the-loop-refine.md
@@ -0,0 +1,61 @@
+# Paper: PaperClaw: Harnessing Agents for Autonomous Research and Human-in-the-Loop Refinement
+
+---
+type: paper
+title: "PaperClaw: Harnessing Agents for Autonomous Research and Human-in-the-Loop Refinement"
+authors: Weiwei Ye, Hangchen Liu, Dongyuan Li, Renhe Jiang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.22610
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-21
+updated_at: 2026-06-21
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - multi-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, memory, multi-agent, tool-use
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.22610
diff --git a/papers/items/2026-2606-22647-raven-agentic-rag-for-automated-vulnerability-repair.md b/papers/items/2026-2606-22647-raven-agentic-rag-for-automated-vulnerability-repair.md
new file mode 100644
index 0000000..2851732
--- /dev/null
+++ b/papers/items/2026-2606-22647-raven-agentic-rag-for-automated-vulnerability-repair.md
@@ -0,0 +1,64 @@
+# Paper: RAVEN: Agentic RAG for Automated Vulnerability Repair
+
+---
+type: paper
+title: "RAVEN: Agentic RAG for Automated Vulnerability Repair"
+authors: Varun Gadey, Zijie Liu, Alexandra Dmitrienko
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.22647
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-21
+updated_at: 2026-06-21
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - computer-use
+  - memory
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.LG
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, coding-agent, computer-use, memory, rag
+- arXiv categories: cs.CR, cs.LG, cs.SE
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.22647
diff --git a/papers/items/2026-2606-22673-agentlens-interpretable-safety-steering-via-mechanistic-subspaces-for-multi-turn.md b/papers/items/2026-2606-22673-agentlens-interpretable-safety-steering-via-mechanistic-subspaces-for-multi-turn.md
new file mode 100644
index 0000000..30b8dd4
--- /dev/null
+++ b/papers/items/2026-2606-22673-agentlens-interpretable-safety-steering-via-mechanistic-subspaces-for-multi-turn.md
@@ -0,0 +1,62 @@
+# Paper: AgentLens: Interpretable Safety Steering via Mechanistic Subspaces for Multi-Turn Coding Agent
+
+---
+type: paper
+title: "AgentLens: Interpretable Safety Steering via Mechanistic Subspaces for Multi-Turn Coding Agent"
+authors: Weidi Luo, Qiming Zhang, Yihao Quan, Mingyu Jin, Jie Cai, Chaowei Xiao, Jingcheng Niu, Zhen Xiang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.22673
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-21
+updated_at: 2026-06-21
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: agent-safety, coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety, coding-agent
+- inferred topics: agent-evaluation, agent-safety, coding-agent, tool-use
+- arXiv categories: cs.AI, cs.SE
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.22673
diff --git a/papers/items/2026-2606-22678-rigorbench-benchmarking-engineering-process-discipline-in-autonomous-ai-coding-a.md b/papers/items/2026-2606-22678-rigorbench-benchmarking-engineering-process-discipline-in-autonomous-ai-coding-a.md
new file mode 100644
index 0000000..2b58c07
--- /dev/null
+++ b/papers/items/2026-2606-22678-rigorbench-benchmarking-engineering-process-discipline-in-autonomous-ai-coding-a.md
@@ -0,0 +1,63 @@
+# Paper: RigorBench: Benchmarking Engineering Process Discipline in Autonomous AI Coding Agents
+
+---
+type: paper
+title: "RigorBench: Benchmarking Engineering Process Discipline in Autonomous AI Coding Agents"
+authors: Meher Bhaskar Madiraju, Meher Sai Preetam Madiraju
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.22678
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-21
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: agent-evaluation, coding-agent, planning, rag, tool-use
+- arXiv categories: cs.SE, cs.AI
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.22678
diff --git a/papers/items/2026-2606-22737-groundeval-a-deterministic-replacement-for-llm-as-judge-in-stateful-agent-evalua.md b/papers/items/2026-2606-22737-groundeval-a-deterministic-replacement-for-llm-as-judge-in-stateful-agent-evalua.md
new file mode 100644
index 0000000..454cf04
--- /dev/null
+++ b/papers/items/2026-2606-22737-groundeval-a-deterministic-replacement-for-llm-as-judge-in-stateful-agent-evalua.md
@@ -0,0 +1,62 @@
+# Paper: GroundEval: A Deterministic Replacement for LLM-as-Judge in Stateful Agent Evaluation
+
+---
+type: paper
+title: "GroundEval: A Deterministic Replacement for LLM-as-Judge in Stateful Agent Evaluation"
+authors: Jeffrey Flynt
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.22737
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-22
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, memory, tool-use
+- arXiv categories: cs.AI, cs.CL, cs.SE
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.22737
diff --git a/papers/items/2026-2606-22741-grade-graph-representation-of-llm-agent-dependency-and-execution.md b/papers/items/2026-2606-22741-grade-graph-representation-of-llm-agent-dependency-and-execution.md
new file mode 100644
index 0000000..df23057
--- /dev/null
+++ b/papers/items/2026-2606-22741-grade-graph-representation-of-llm-agent-dependency-and-execution.md
@@ -0,0 +1,60 @@
+# Paper: GRADE: Graph Representation of LLM Agent Dependency and Execution
+
+---
+type: paper
+title: "GRADE: Graph Representation of LLM Agent Dependency and Execution"
+authors: Yue Zhao
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.22741
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-22
+updated_at: 2026-06-22
+status: queued
+relevance: high
+topics:
+  - coding-agent
+  - multi-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: multi-agent-llm, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm, tool-use
+- inferred topics: coding-agent, multi-agent, tool-use
+- arXiv categories: cs.LG
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.22741
diff --git a/papers/items/2026-2606-22844-ramem-contextual-reinstatement-for-long-term-agentic-memory.md b/papers/items/2026-2606-22844-ramem-contextual-reinstatement-for-long-term-agentic-memory.md
new file mode 100644
index 0000000..18ebd27
--- /dev/null
+++ b/papers/items/2026-2606-22844-ramem-contextual-reinstatement-for-long-term-agentic-memory.md
@@ -0,0 +1,62 @@
+# Paper: RaMem: Contextual Reinstatement for Long-term Agentic Memory
+
+---
+type: paper
+title: "RaMem: Contextual Reinstatement for Long-term Agentic Memory"
+authors: Wei Yang, Bryce Kan, Shixuan Li, Li Li, Yuehan Qin, Jiate Li, Paul Bogdan, Jesse Thomason
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.22844
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-22
+updated_at: 2026-06-22
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, rag, tool-use
+- arXiv categories: cs.AI, cs.MA
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.22844
diff --git a/papers/items/2026-2606-22864-when-auc-0-998-is-not-enough-a-candidate-evaluation-protocol-for-hidden-state-pr.md b/papers/items/2026-2606-22864-when-auc-0-998-is-not-enough-a-candidate-evaluation-protocol-for-hidden-state-pr.md
new file mode 100644
index 0000000..0cacb8c
--- /dev/null
+++ b/papers/items/2026-2606-22864-when-auc-0-998-is-not-enough-a-candidate-evaluation-protocol-for-hidden-state-pr.md
@@ -0,0 +1,60 @@
+# Paper: When AUC 0.998 Is Not Enough: A Candidate Evaluation Protocol for Hidden-State Probes of Indirect Prompt Injection in Multimodal Computer-Use Agents
+
+---
+type: paper
+title: "When AUC 0.998 Is Not Enough: A Candidate Evaluation Protocol for Hidden-State Probes of Indirect Prompt Injection in Multimodal Computer-Use Agents"
+authors: Yanhang Li, Zhichao Fan, Zexin Zhuang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.22864
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-22
+updated_at: 2026-06-22
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, agent-safety, tool-use
+- arXiv categories: cs.LG
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.22864
diff --git a/papers/items/2026-2606-22948-envs-environment-native-verified-search-for-long-horizon-gui-agents.md b/papers/items/2026-2606-22948-envs-environment-native-verified-search-for-long-horizon-gui-agents.md
new file mode 100644
index 0000000..928e63c
--- /dev/null
+++ b/papers/items/2026-2606-22948-envs-environment-native-verified-search-for-long-horizon-gui-agents.md
@@ -0,0 +1,63 @@
+# Paper: ENVS: Environment-Native Verified Search for Long-Horizon GUI Agents
+
+---
+type: paper
+title: "ENVS: Environment-Native Verified Search for Long-Horizon GUI Agents"
+authors: Yincheng Zhou, Athena Zhuoming Zhong, Shijie Zhang, Kevin Zhang, Teresa Xiaotao Shang, Shanghang Zhang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.22948
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-22
+updated_at: 2026-06-22
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CV
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, computer-use, planning, reasoning, tool-use
+- arXiv categories: cs.AI, cs.CV
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.22948
diff --git a/papers/items/2026-2606-22953-plans-don-t-persist-why-context-management-is-load-bearing-for-llm-agents.md b/papers/items/2026-2606-22953-plans-don-t-persist-why-context-management-is-load-bearing-for-llm-agents.md
new file mode 100644
index 0000000..158f054
--- /dev/null
+++ b/papers/items/2026-2606-22953-plans-don-t-persist-why-context-management-is-load-bearing-for-llm-agents.md
@@ -0,0 +1,61 @@
+# Paper: Plans Don't Persist: Why Context Management Is Load Bearing for LLM Agents
+
+---
+type: paper
+title: "Plans Don't Persist: Why Context Management Is Load Bearing for LLM Agents"
+authors: Aman Mehta, Anupam Datta
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.22953
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-22
+updated_at: 2026-06-22
+status: queued
+relevance: high
+topics:
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: planning, reasoning, tool-use
+- arXiv categories: cs.AI, cs.CL
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.22953
diff --git a/papers/items/2026-2606-23032-ipo-finance-agent-benchmark-of-llm-financial-analysts-beyond-finance-agent-v2-wi.md b/papers/items/2026-2606-23032-ipo-finance-agent-benchmark-of-llm-financial-analysts-beyond-finance-agent-v2-wi.md
new file mode 100644
index 0000000..1cd9500
--- /dev/null
+++ b/papers/items/2026-2606-23032-ipo-finance-agent-benchmark-of-llm-financial-analysts-beyond-finance-agent-v2-wi.md
@@ -0,0 +1,62 @@
+# Paper: IPO Finance Agent: Benchmark of LLM Financial Analysts Beyond Finance Agent v2, with Automated Rubric Generation, on the SpaceX (SPCX) IPO
+
+---
+type: paper
+title: "IPO Finance Agent: Benchmark of LLM Financial Analysts Beyond Finance Agent v2, with Automated Rubric Generation, on the SpaceX (SPCX) IPO"
+authors: Mostapha Benhenda
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.23032
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-22
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - q-fin.GN
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, agent-safety, rag, tool-use
+- arXiv categories: cs.AI, q-fin.GN
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.23032
diff --git a/papers/items/2026-2606-23130-understanding-the-in-security-of-vibe-coded-applications.md b/papers/items/2026-2606-23130-understanding-the-in-security-of-vibe-coded-applications.md
new file mode 100644
index 0000000..8dcf46f
--- /dev/null
+++ b/papers/items/2026-2606-23130-understanding-the-in-security-of-vibe-coded-applications.md
@@ -0,0 +1,64 @@
+# Paper: Understanding the (In)Security of Vibe-Coded Applications
+
+---
+type: paper
+title: Understanding the (In)Security of Vibe-Coded Applications
+authors: Junquan Deng, Zhiyu Fan, Ruijie Meng
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.23130
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-22
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - memory
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent
+- inferred topics: agent-evaluation, agent-safety, coding-agent, memory, tool-use, workflow-agent
+- arXiv categories: cs.CR, cs.SE
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.23130
diff --git a/papers/items/2026-2606-23195-memory-contagion-cross-temporal-propagation-of-evaluator-bias-via-agent-memory.md b/papers/items/2026-2606-23195-memory-contagion-cross-temporal-propagation-of-evaluator-bias-via-agent-memory.md
new file mode 100644
index 0000000..25c6dd3
--- /dev/null
+++ b/papers/items/2026-2606-23195-memory-contagion-cross-temporal-propagation-of-evaluator-bias-via-agent-memory.md
@@ -0,0 +1,63 @@
+# Paper: Memory Contagion: Cross-Temporal Propagation of Evaluator Bias via Agent Memory
+
+---
+type: paper
+title: "Memory Contagion: Cross-Temporal Propagation of Evaluator Bias via Agent Memory"
+authors: Zewen Liu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.23195
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-22
+updated_at: 2026-06-24
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, computer-use, memory, tool-use
+- arXiv categories: cs.LG, cs.AI, cs.CL
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.23195
diff --git a/papers/items/2026-2606-23283-towards-root-memories-benchmarking-and-enhancing-implicit-logical-memory-retriev.md b/papers/items/2026-2606-23283-towards-root-memories-benchmarking-and-enhancing-implicit-logical-memory-retriev.md
new file mode 100644
index 0000000..35b7bda
--- /dev/null
+++ b/papers/items/2026-2606-23283-towards-root-memories-benchmarking-and-enhancing-implicit-logical-memory-retriev.md
@@ -0,0 +1,60 @@
+# Paper: Towards Root Memories: Benchmarking and Enhancing Implicit Logical Memory Retrieval for Personalized LLMs
+
+---
+type: paper
+title: "Towards Root Memories: Benchmarking and Enhancing Implicit Logical Memory Retrieval for Personalized LLMs"
+authors: Hongxun Ding, Xiang Yu, Chengbing Wang, Jianfei Xiao, Keqin Bao, Wenjie Wang, Xiangnan He
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.23283
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-22
+updated_at: 2026-06-22
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, rag
+- arXiv categories: cs.CL
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.23283
diff --git a/papers/items/2026-2606-23327-videoagent-all-in-one-framework-for-video-understanding-and-editing.md b/papers/items/2026-2606-23327-videoagent-all-in-one-framework-for-video-understanding-and-editing.md
new file mode 100644
index 0000000..deba0b3
--- /dev/null
+++ b/papers/items/2026-2606-23327-videoagent-all-in-one-framework-for-video-understanding-and-editing.md
@@ -0,0 +1,63 @@
+# Paper: VideoAgent: All-in-One Framework for Video Understanding and Editing
+
+---
+type: paper
+title: "VideoAgent: All-in-One Framework for Video Understanding and Editing"
+authors: Hengji Zhou, Lingxuan Huang, Jian Wang, Bing Zhou, Si Wu, Lianghao Xia, Chao Huang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.23327
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-22
+updated_at: 2026-07-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, multi-agent, planning, rag, tool-use
+- arXiv categories: cs.CV, cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.23327
diff --git a/papers/items/2026-2606-23343-group-selection-promotes-prosocial-prompts-in-populations-of-llm-agents.md b/papers/items/2026-2606-23343-group-selection-promotes-prosocial-prompts-in-populations-of-llm-agents.md
new file mode 100644
index 0000000..071a7ea
--- /dev/null
+++ b/papers/items/2026-2606-23343-group-selection-promotes-prosocial-prompts-in-populations-of-llm-agents.md
@@ -0,0 +1,61 @@
+# Paper: Group Selection Promotes Prosocial Prompts in Populations of LLM Agents
+
+---
+type: paper
+title: Group Selection Promotes Prosocial Prompts in Populations of LLM Agents
+authors: Luis Celiktemel, Edward Eichhorn, Levin Brinkmann, Robin Schimmelpfennig, Aron Vallinder, Yaomin Jiang, Edward Hughes, Iyad Rahwan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.23343
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-22
+updated_at: 2026-06-22
+status: queued
+relevance: high
+topics:
+  - coding-agent
+  - computer-use
+  - multi-agent
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CY
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: coding-agent, computer-use, multi-agent, world-model
+- arXiv categories: cs.CY
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.23343
diff --git a/papers/items/2026-2606-23565-holoagent-0-a-unified-embodied-agent-framework-with-3d-spatial-memory.md b/papers/items/2026-2606-23565-holoagent-0-a-unified-embodied-agent-framework-with-3d-spatial-memory.md
new file mode 100644
index 0000000..60cd719
--- /dev/null
+++ b/papers/items/2026-2606-23565-holoagent-0-a-unified-embodied-agent-framework-with-3d-spatial-memory.md
@@ -0,0 +1,66 @@
+# Paper: HoloAgent-0: A Unified Embodied Agent Framework with 3D Spatial Memory
+
+---
+type: paper
+title: "HoloAgent-0: A Unified Embodied Agent Framework with 3D Spatial Memory"
+authors: Xiaolin Zhou, Liu Liu, Tingyang Xiao, Wei Feng, Fa Fu, Xinrui Meng, Xinjie Wang, Jialiang Han, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.23565
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-22
+updated_at: 2026-06-22
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - embodied-agent
+  - memory
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.RO
+  - cs.CV
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, agent-safety, computer-use, embodied-agent, memory, planning, rag, tool-use
+- arXiv categories: cs.RO, cs.CV
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.23565
diff --git a/papers/items/2026-2606-23664-mas-promptbench-when-does-prompt-optimization-improve-multi-agent-llm-systems.md b/papers/items/2026-2606-23664-mas-promptbench-when-does-prompt-optimization-improve-multi-agent-llm-systems.md
new file mode 100644
index 0000000..b643681
--- /dev/null
+++ b/papers/items/2026-2606-23664-mas-promptbench-when-does-prompt-optimization-improve-multi-agent-llm-systems.md
@@ -0,0 +1,61 @@
+# Paper: MAS-PromptBench: When Does Prompt Optimization Improve Multi-Agent LLM Systems?
+
+---
+type: paper
+title: "MAS-PromptBench: When Does Prompt Optimization Improve Multi-Agent LLM Systems?"
+authors: Juyang Bai, Laixi Shi
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.23664
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-22
+updated_at: 2026-06-22
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agentic-ai, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, multi-agent-llm
+- inferred topics: agent-evaluation, multi-agent, workflow-agent
+- arXiv categories: cs.LG, cs.MA
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.23664
diff --git a/papers/items/2026-2606-23752-esaa-conversational-an-event-sourced-memory-layer-for-continuity-handoff-and-cur.md b/papers/items/2026-2606-23752-esaa-conversational-an-event-sourced-memory-layer-for-continuity-handoff-and-cur.md
new file mode 100644
index 0000000..691dd27
--- /dev/null
+++ b/papers/items/2026-2606-23752-esaa-conversational-an-event-sourced-memory-layer-for-continuity-handoff-and-cur.md
@@ -0,0 +1,60 @@
+# Paper: ESAA-Conversational: An Event-Sourced Memory Layer for Continuity, Handoff, and Curation Across Heterogeneous LLM Coding Agents
+
+---
+type: paper
+title: "ESAA-Conversational: An Event-Sourced Memory Layer for Continuity, Handoff, and Curation Across Heterogeneous LLM Coding Agents"
+authors: Elzo Brito dos Santos Filho
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.23752
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-22
+updated_at: 2026-06-22
+status: queued
+relevance: high
+topics:
+  - coding-agent
+  - memory
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: coding-agent, memory, tool-use
+- arXiv categories: cs.SE
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.23752
diff --git a/papers/items/2026-2606-23764-emergent-relational-order-in-llm-agent-societies-from-collective-affect-to-autho.md b/papers/items/2026-2606-23764-emergent-relational-order-in-llm-agent-societies-from-collective-affect-to-autho.md
new file mode 100644
index 0000000..36c0450
--- /dev/null
+++ b/papers/items/2026-2606-23764-emergent-relational-order-in-llm-agent-societies-from-collective-affect-to-autho.md
@@ -0,0 +1,62 @@
+# Paper: Emergent Relational Order in LLM Agent Societies: From Collective Affect to Authority Stratification
+
+---
+type: paper
+title: "Emergent Relational Order in LLM Agent Societies: From Collective Affect to Authority Stratification"
+authors: Zhiyuan Ji, Xinyu Chen, Ziqi Dai, Shiyun Tang, Chunyu Wei, Yueguo Chen
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.23764
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-22
+updated_at: 2026-06-22
+status: queued
+relevance: high
+topics:
+  - multi-agent
+  - planning
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: multi-agent, planning, tool-use, world-model
+- arXiv categories: cs.MA, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.23764
diff --git a/papers/items/2026-2606-23927-rift-bench-dynamic-red-teaming-for-agentic-ai-systems.md b/papers/items/2026-2606-23927-rift-bench-dynamic-red-teaming-for-agentic-ai-systems.md
new file mode 100644
index 0000000..062c0e3
--- /dev/null
+++ b/papers/items/2026-2606-23927-rift-bench-dynamic-red-teaming-for-agentic-ai-systems.md
@@ -0,0 +1,61 @@
+# Paper: RIFT-Bench: Dynamic Red-teaming For Agentic AI Systems
+
+---
+type: paper
+title: "RIFT-Bench: Dynamic Red-teaming For Agentic AI Systems"
+authors: Yarin Yerushalmi Levi, Roy Betser, Amit Giloni, Lidor Erez, Itay Gershon, Oren Rachmil, Sindhu Padakandla, Roman Vainshtein
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.23927
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-22
+updated_at: 2026-06-22
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-evaluation, agent-safety, rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.23927
diff --git a/papers/items/2026-2606-23991-critique-of-agent-model.md b/papers/items/2026-2606-23991-critique-of-agent-model.md
new file mode 100644
index 0000000..e523e10
--- /dev/null
+++ b/papers/items/2026-2606-23991-critique-of-agent-model.md
@@ -0,0 +1,67 @@
+# Paper: Critique of Agent Model
+
+---
+type: paper
+title: Critique of Agent Model
+authors: Eric Xing, Mingkai Deng, Jinyu Hou
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.23991
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-22
+updated_at: 2026-06-22
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - coding-agent
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.LG
+  - cs.MA
+  - cs.RO
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agentic-ai, ai-agent, coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, ai-agent, coding-agent
+- inferred topics: agent-safety, coding-agent, rag, reasoning, tool-use, workflow-agent, world-model
+- arXiv categories: cs.AI, cs.LG, cs.MA, cs.RO
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.23991
diff --git a/papers/items/2026-2606-24193-skychain-intelligence-a-blockchain-secured-multi-agent-drl-framework-for-low-alt.md b/papers/items/2026-2606-24193-skychain-intelligence-a-blockchain-secured-multi-agent-drl-framework-for-low-alt.md
new file mode 100644
index 0000000..3cea856
--- /dev/null
+++ b/papers/items/2026-2606-24193-skychain-intelligence-a-blockchain-secured-multi-agent-drl-framework-for-low-alt.md
@@ -0,0 +1,63 @@
+# Paper: SkyChain Intelligence: A Blockchain-Secured Multi-Agent DRL Framework for Low-Altitude Embodied Artificial Intelligence
+
+---
+type: paper
+title: "SkyChain Intelligence: A Blockchain-Secured Multi-Agent DRL Framework for Low-Altitude Embodied Artificial Intelligence"
+authors: Haoxiang Luo, Tianqi Jiang, Ruichen Zhang, Yinqiu Liu, Gang Sun, Hongfang Yu, Abbas Jamalipour, Dong In Kim
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.24193
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - embodied-agent
+  - multi-agent
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.NI
+  - cs.DC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-safety, embodied-agent, multi-agent, tool-use, world-model
+- arXiv categories: cs.NI, cs.DC
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.24193
diff --git a/papers/items/2026-2606-24235-sp-mind-an-autonomous-reasoning-agent-for-spatial-proteomics-analysis.md b/papers/items/2026-2606-24235-sp-mind-an-autonomous-reasoning-agent-for-spatial-proteomics-analysis.md
new file mode 100644
index 0000000..838cc4a
--- /dev/null
+++ b/papers/items/2026-2606-24235-sp-mind-an-autonomous-reasoning-agent-for-spatial-proteomics-analysis.md
@@ -0,0 +1,63 @@
+# Paper: SP-Mind: An Autonomous Reasoning Agent for Spatial Proteomics Analysis
+
+---
+type: paper
+title: "SP-Mind: An Autonomous Reasoning Agent for Spatial Proteomics Analysis"
+authors: Yucheng Yuan, Yuanfeng Ji, Zhongxiao Li, Ruijiang Li
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.24235
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent
+- inferred topics: agent-evaluation, computer-use, rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.24235
diff --git a/papers/items/2026-2606-24322-securing-llm-agent-long-term-memory-against-poisoning-non-malleable-origin-bound.md b/papers/items/2026-2606-24322-securing-llm-agent-long-term-memory-against-poisoning-non-malleable-origin-bound.md
new file mode 100644
index 0000000..3e5e6c7
--- /dev/null
+++ b/papers/items/2026-2606-24322-securing-llm-agent-long-term-memory-against-poisoning-non-malleable-origin-bound.md
@@ -0,0 +1,60 @@
+# Paper: Securing LLM-Agent Long-Term Memory Against Poisoning: Non-Malleable, Origin-Bound Authority with Machine-Checked Guarantees
+
+---
+type: paper
+title: "Securing LLM-Agent Long-Term Memory Against Poisoning: Non-Malleable, Origin-Bound Authority with Machine-Checked Guarantees"
+authors: Yedidel Louck
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.24322
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, tool-use
+- arXiv categories: cs.CR
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.24322
diff --git a/papers/items/2026-2606-24402-poisoned-playbooks-demystifying-knowledge-poisoning-effects-on-ai-security-agent.md b/papers/items/2026-2606-24402-poisoned-playbooks-demystifying-knowledge-poisoning-effects-on-ai-security-agent.md
new file mode 100644
index 0000000..ad8751f
--- /dev/null
+++ b/papers/items/2026-2606-24402-poisoned-playbooks-demystifying-knowledge-poisoning-effects-on-ai-security-agent.md
@@ -0,0 +1,62 @@
+# Paper: Poisoned Playbooks: Demystifying Knowledge Poisoning Effects on AI Security Agents
+
+---
+type: paper
+title: "Poisoned Playbooks: Demystifying Knowledge Poisoning Effects on AI Security Agents"
+authors: Juho Park, Hyunmin Choi, Kevin Nam
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.24402
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: ai-agent, rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent, rag-agent
+- inferred topics: agent-evaluation, agent-safety, rag, reasoning, tool-use
+- arXiv categories: cs.CR
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.24402
diff --git a/papers/items/2026-2606-24416-agentic-ai-for-bilevel-long-term-optimization-of-policy-driven-physical-layer-sy.md b/papers/items/2026-2606-24416-agentic-ai-for-bilevel-long-term-optimization-of-policy-driven-physical-layer-sy.md
new file mode 100644
index 0000000..0f019aa
--- /dev/null
+++ b/papers/items/2026-2606-24416-agentic-ai-for-bilevel-long-term-optimization-of-policy-driven-physical-layer-sy.md
@@ -0,0 +1,61 @@
+# Paper: Agentic AI for Bilevel Long-Term Optimization of Policy-Driven Physical Layer Systems
+
+---
+type: paper
+title: Agentic AI for Bilevel Long-Term Optimization of Policy-Driven Physical Layer Systems
+authors: Bingnan Xiao, Chenhao Yang, Wei Ni, Xin Wang, Tony Q. S. Quek
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.24416
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - multi-agent
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-evaluation, memory, multi-agent, rag
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.24416
diff --git a/papers/items/2026-2606-24437-rem-moa-reasoning-memory-sustains-mixture-of-agents-scaling.md b/papers/items/2026-2606-24437-rem-moa-reasoning-memory-sustains-mixture-of-agents-scaling.md
new file mode 100644
index 0000000..68fc29a
--- /dev/null
+++ b/papers/items/2026-2606-24437-rem-moa-reasoning-memory-sustains-mixture-of-agents-scaling.md
@@ -0,0 +1,61 @@
+# Paper: ReM-MoA: Reasoning Memory Sustains Mixture-of-Agents Scaling
+
+---
+type: paper
+title: "ReM-MoA: Reasoning Memory Sustains Mixture-of-Agents Scaling"
+authors: Heng Ping, Arijit Bhattacharjee, Peiyu Zhang, Shixuan Li, Wei Yang, Ali Jannesari, Nesreen Ahmed, Paul Bogdan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.24437
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - multi-agent
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, memory, multi-agent, reasoning
+- arXiv categories: cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.24437
diff --git a/papers/items/2026-2606-24453-bayesian-control-for-coding-agents.md b/papers/items/2026-2606-24453-bayesian-control-for-coding-agents.md
new file mode 100644
index 0000000..bebf1c6
--- /dev/null
+++ b/papers/items/2026-2606-24453-bayesian-control-for-coding-agents.md
@@ -0,0 +1,62 @@
+# Paper: Bayesian control for coding agents
+
+---
+type: paper
+title: Bayesian control for coding agents
+authors: Theodore Papamarkou, Vladislav Smirnov, Viktor Mazanov, Artem Vazhentsev, Preslav Nakov, Timothy Baldwin, Artem Shelmanov
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.24453
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: coding-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent, tool-use
+- inferred topics: agent-evaluation, coding-agent, reasoning, tool-use
+- arXiv categories: cs.AI, cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.24453
diff --git a/papers/items/2026-2606-24515-reinforcement-learning-for-computer-use-agents-with-autonomous-evaluation.md b/papers/items/2026-2606-24515-reinforcement-learning-for-computer-use-agents-with-autonomous-evaluation.md
new file mode 100644
index 0000000..9b7345e
--- /dev/null
+++ b/papers/items/2026-2606-24515-reinforcement-learning-for-computer-use-agents-with-autonomous-evaluation.md
@@ -0,0 +1,61 @@
+# Paper: Reinforcement Learning for Computer-Use Agents with Autonomous Evaluation
+
+---
+type: paper
+title: Reinforcement Learning for Computer-Use Agents with Autonomous Evaluation
+authors: Marta Sumyk, Oleksandr Kosovan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.24515
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.HC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, computer-use, rag
+- arXiv categories: cs.AI, cs.HC
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.24515
diff --git a/papers/items/2026-2606-24525-viscritic-visual-state-comparison-as-process-reward-for-gui-agents.md b/papers/items/2026-2606-24525-viscritic-visual-state-comparison-as-process-reward-for-gui-agents.md
new file mode 100644
index 0000000..067a61f
--- /dev/null
+++ b/papers/items/2026-2606-24525-viscritic-visual-state-comparison-as-process-reward-for-gui-agents.md
@@ -0,0 +1,62 @@
+# Paper: VisCritic: Visual State Comparison as Process Reward for GUI Agents
+
+---
+type: paper
+title: "VisCritic: Visual State Comparison as Process Reward for GUI Agents"
+authors: Jiachen Qian
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.24525
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, computer-use, planning, reasoning, tool-use
+- arXiv categories: cs.CV
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.24525
diff --git a/papers/items/2026-2606-24535-governed-shared-memory-for-multi-agent-llm-systems.md b/papers/items/2026-2606-24535-governed-shared-memory-for-multi-agent-llm-systems.md
new file mode 100644
index 0000000..aca6e6c
--- /dev/null
+++ b/papers/items/2026-2606-24535-governed-shared-memory-for-multi-agent-llm-systems.md
@@ -0,0 +1,62 @@
+# Paper: Governed Shared Memory for Multi-Agent LLM Systems
+
+---
+type: paper
+title: Governed Shared Memory for Multi-Agent LLM Systems
+authors: Yanki Margalit, Nurit Cohen-Inger, Erni Avram, Ran Taig, Oded Margalit
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.24535
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - multi-agent
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: agent-memory, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory, multi-agent-llm
+- inferred topics: agent-evaluation, memory, multi-agent, rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.24535
diff --git a/papers/items/2026-2606-24551-gui-vs-cli-execution-bottlenecks-in-screen-only-and-skill-mediated-computer-use-.md b/papers/items/2026-2606-24551-gui-vs-cli-execution-bottlenecks-in-screen-only-and-skill-mediated-computer-use-.md
new file mode 100644
index 0000000..99c6305
--- /dev/null
+++ b/papers/items/2026-2606-24551-gui-vs-cli-execution-bottlenecks-in-screen-only-and-skill-mediated-computer-use-.md
@@ -0,0 +1,64 @@
+# Paper: GUI vs. CLI: Execution Bottlenecks in Screen-Only and Skill-Mediated Computer-Use Agents
+
+---
+type: paper
+title: "GUI vs. CLI: Execution Bottlenecks in Screen-Only and Skill-Mediated Computer-Use Agents"
+authors: Xiao Zhou, Siyue Zhang, Yilun Zhao, Jinbiao Wei, Tingyu Song, Arman Cohan, Chen Zhao
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.24551
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-22
+updated_at: 2026-06-22
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, computer-use, planning, rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.24551
diff --git a/papers/items/2026-2606-24595-memprobe-probing-long-term-agent-memory-via-hidden-user-state-recovery.md b/papers/items/2026-2606-24595-memprobe-probing-long-term-agent-memory-via-hidden-user-state-recovery.md
new file mode 100644
index 0000000..47cc88c
--- /dev/null
+++ b/papers/items/2026-2606-24595-memprobe-probing-long-term-agent-memory-via-hidden-user-state-recovery.md
@@ -0,0 +1,61 @@
+# Paper: MEMPROBE: Probing Long-Term Agent Memory via Hidden User-State Recovery
+
+---
+type: paper
+title: "MEMPROBE: Probing Long-Term Agent Memory via Hidden User-State Recovery"
+authors: Enze Ma, Yufan Zhou, Wei-Chieh Huang, Jie Yang, Huanhuan Ma, Zixuan Wang, Chengze Li, Chunyu Miao, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.24595
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, rag, tool-use
+- arXiv categories: cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.24595
diff --git a/papers/items/2026-2606-24597-qwen-agentworld-language-world-models-for-general-agents.md b/papers/items/2026-2606-24597-qwen-agentworld-language-world-models-for-general-agents.md
new file mode 100644
index 0000000..9ee00fe
--- /dev/null
+++ b/papers/items/2026-2606-24597-qwen-agentworld-language-world-models-for-general-agents.md
@@ -0,0 +1,63 @@
+# Paper: Qwen-AgentWorld: Language World Models for General Agents
+
+---
+type: paper
+title: "Qwen-AgentWorld: Language World Models for General Agents"
+authors: Yuxin Zuo, Zikai Xiao, Li Sheng, Fei Huang, Jianhong Tu, Yuxuan Liu, Tianyi Tang, Xiaomeng Hu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.24597
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, planning, rag, reasoning, tool-use, world-model
+- arXiv categories: cs.CL
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.24597
diff --git a/papers/items/2026-2606-24623-privacy-preserving-rag-via-multi-agent-semantic-rewriting-achieving-confidential.md b/papers/items/2026-2606-24623-privacy-preserving-rag-via-multi-agent-semantic-rewriting-achieving-confidential.md
new file mode 100644
index 0000000..e3f9dec
--- /dev/null
+++ b/papers/items/2026-2606-24623-privacy-preserving-rag-via-multi-agent-semantic-rewriting-achieving-confidential.md
@@ -0,0 +1,63 @@
+# Paper: Privacy-Preserving RAG via Multi-Agent Semantic Rewriting: Achieving Confidentiality Without Compromising Contextual Fidelity
+
+---
+type: paper
+title: "Privacy-Preserving RAG via Multi-Agent Semantic Rewriting: Achieving Confidentiality Without Compromising Contextual Fidelity"
+authors: Yuanhe Zhao, Tianyu Zhang, Huafei Xing, Derek F. Wong, Jianbin Li, Tao Fang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.24623
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: multi-agent-llm, rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm, rag-agent
+- inferred topics: agent-evaluation, agent-safety, multi-agent, rag, tool-use
+- arXiv categories: cs.CL, cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.24623
diff --git a/papers/items/2026-2606-24626-safari-scaling-long-horizon-agentic-fault-attribution-via-active-investigation.md b/papers/items/2026-2606-24626-safari-scaling-long-horizon-agentic-fault-attribution-via-active-investigation.md
new file mode 100644
index 0000000..ecc8034
--- /dev/null
+++ b/papers/items/2026-2606-24626-safari-scaling-long-horizon-agentic-fault-attribution-via-active-investigation.md
@@ -0,0 +1,63 @@
+# Paper: SAFARI: Scaling Long Horizon Agentic Fault Attribution via Active Investigation
+
+---
+type: paper
+title: "SAFARI: Scaling Long Horizon Agentic Fault Attribution via Active Investigation"
+authors: Chenyang Zhu, Jiayu Yao, Kushal Chawla, Youbing Yin, Nathan Wolfe, Pengshan Cai, Jingyu Wu, Spencer Hong, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.24626
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - multi-agent
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: autonomous-agent-llm, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm, multi-agent-llm
+- inferred topics: agent-evaluation, memory, multi-agent, planning, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.24626
diff --git a/papers/items/2026-2606-24649-agentic-collaborative-cognition-for-zero-shot-3d-understanding.md b/papers/items/2026-2606-24649-agentic-collaborative-cognition-for-zero-shot-3d-understanding.md
new file mode 100644
index 0000000..570d919
--- /dev/null
+++ b/papers/items/2026-2606-24649-agentic-collaborative-cognition-for-zero-shot-3d-understanding.md
@@ -0,0 +1,62 @@
+# Paper: Agentic Collaborative Cognition for Zero-Shot 3D Understanding
+
+---
+type: paper
+title: Agentic Collaborative Cognition for Zero-Shot 3D Understanding
+authors: Wenxin Wang, Bo Zhang, Feng Chen, Zixuan Wang, Wen Li, Changsheng Li, Yinjie Lei
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.24649
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-06-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - multi-agent
+  - planning
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, computer-use, multi-agent, planning, rag
+- arXiv categories: cs.CV
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.24649
diff --git a/papers/items/2026-2606-24689-automated-summarization-of-software-documents-an-llm-based-multi-agent-approach.md b/papers/items/2026-2606-24689-automated-summarization-of-software-documents-an-llm-based-multi-agent-approach.md
new file mode 100644
index 0000000..565c014
--- /dev/null
+++ b/papers/items/2026-2606-24689-automated-summarization-of-software-documents-an-llm-based-multi-agent-approach.md
@@ -0,0 +1,61 @@
+# Paper: Automated Summarization of Software Documents: An LLM-based Multi-Agent Approach
+
+---
+type: paper
+title: "Automated Summarization of Software Documents: An LLM-based Multi-Agent Approach"
+authors: Duc S. H. Nguyen, Minh T. Nguyen, Phuong T. Nguyen, Juri Di Rocco, Davide Di Ruscio
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.24689
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - multi-agent
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, coding-agent, multi-agent, workflow-agent
+- arXiv categories: cs.SE
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.24689
diff --git a/papers/items/2026-2606-24694-supplynet-supporting-visual-exploratory-learning-in-supply-chain-via-contextual-.md b/papers/items/2026-2606-24694-supplynet-supporting-visual-exploratory-learning-in-supply-chain-via-contextual-.md
new file mode 100644
index 0000000..f1f5e4d
--- /dev/null
+++ b/papers/items/2026-2606-24694-supplynet-supporting-visual-exploratory-learning-in-supply-chain-via-contextual-.md
@@ -0,0 +1,61 @@
+# Paper: SupplyNet: Supporting Visual Exploratory Learning in Supply Chain via Contextual Multi-Agent Simulation
+
+---
+type: paper
+title: "SupplyNet: Supporting Visual Exploratory Learning in Supply Chain via Contextual Multi-Agent Simulation"
+authors: Yanjia Li, Kelcy Kexin Han, Tianrui Hu, Yi-Fan Cao, Huamin Qu, Sicheng Song
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.24694
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - multi-agent
+  - rag
+  - reasoning
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.HC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: multi-agent, rag, reasoning, world-model
+- arXiv categories: cs.HC
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.24694
diff --git a/papers/items/2026-2606-24775-are-we-ready-for-an-agent-native-memory-system.md b/papers/items/2026-2606-24775-are-we-ready-for-an-agent-native-memory-system.md
new file mode 100644
index 0000000..4905484
--- /dev/null
+++ b/papers/items/2026-2606-24775-are-we-ready-for-an-agent-native-memory-system.md
@@ -0,0 +1,64 @@
+# Paper: Are We Ready For An Agent-Native Memory System?
+
+---
+type: paper
+title: Are We Ready For An Agent-Native Memory System?
+authors: Wei Zhou, Xuanhe Zhou, Shaokun Han, Hongming Xu, Guoliang Li, Zhiyu Li, Feiyu Xiong, Fan Wu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.24775
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.DB
+  - cs.IR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, planning, rag, tool-use
+- arXiv categories: cs.CL, cs.DB, cs.IR
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.24775
diff --git a/papers/items/2026-2606-24779-deepbd-a-grounded-agentic-workflow-for-variant-prioritization-and-diagnosis-of-g.md b/papers/items/2026-2606-24779-deepbd-a-grounded-agentic-workflow-for-variant-prioritization-and-diagnosis-of-g.md
new file mode 100644
index 0000000..91ac291
--- /dev/null
+++ b/papers/items/2026-2606-24779-deepbd-a-grounded-agentic-workflow-for-variant-prioritization-and-diagnosis-of-g.md
@@ -0,0 +1,62 @@
+# Paper: DeepBD: A Grounded Agentic Workflow for Variant Prioritization and Diagnosis of Genetic Birth Defects
+
+---
+type: paper
+title: "DeepBD: A Grounded Agentic Workflow for Variant Prioritization and Diagnosis of Genetic Birth Defects"
+authors: Shiyu Li, Ziqi Yan, Zhihao Wu, Jielong Lu, Weiran Liao, Jiajun Yu, Genjie Li, Zeyu Chu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.24779
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - q-bio.GN
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-evaluation, memory, tool-use, workflow-agent
+- arXiv categories: q-bio.GN, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.24779
diff --git a/papers/items/2026-2606-24820-sherloc-structured-diagnostic-localization-for-code-repair-agents.md b/papers/items/2026-2606-24820-sherloc-structured-diagnostic-localization-for-code-repair-agents.md
new file mode 100644
index 0000000..dfdceb5
--- /dev/null
+++ b/papers/items/2026-2606-24820-sherloc-structured-diagnostic-localization-for-code-repair-agents.md
@@ -0,0 +1,63 @@
+# Paper: SHERLOC: Structured Diagnostic Localization for Code Repair Agents
+
+---
+type: paper
+title: "SHERLOC: Structured Diagnostic Localization for Code Repair Agents"
+authors: Hovhannes Tamoyan, Sean Narenthiran, Erik Arakelyan, Mira Mezini, Boris Ginsburg
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.24820
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - multi-agent
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: coding-agent, multi-agent-llm, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent, multi-agent-llm, tool-use
+- inferred topics: agent-evaluation, coding-agent, multi-agent, rag, reasoning, tool-use
+- arXiv categories: cs.CL
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.24820
diff --git a/papers/items/2026-2606-24839-grading-the-grader-lessons-from-evaluating-an-agentic-data-analysis-system.md b/papers/items/2026-2606-24839-grading-the-grader-lessons-from-evaluating-an-agentic-data-analysis-system.md
new file mode 100644
index 0000000..53dd4d1
--- /dev/null
+++ b/papers/items/2026-2606-24839-grading-the-grader-lessons-from-evaluating-an-agentic-data-analysis-system.md
@@ -0,0 +1,62 @@
+# Paper: Grading the Grader: Lessons from Evaluating an Agentic Data Analysis System
+
+---
+type: paper
+title: "Grading the Grader: Lessons from Evaluating an Agentic Data Analysis System"
+authors: Tian Zheng, Kai-Tai Hsu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.24839
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - multi-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - stat.AP
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, computer-use, multi-agent, tool-use
+- arXiv categories: cs.AI, stat.AP
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.24839
diff --git a/papers/items/2026-2606-24855-openthoughts-agent-data-recipes-for-agentic-models.md b/papers/items/2026-2606-24855-openthoughts-agent-data-recipes-for-agentic-models.md
new file mode 100644
index 0000000..d5d4a23
--- /dev/null
+++ b/papers/items/2026-2606-24855-openthoughts-agent-data-recipes-for-agentic-models.md
@@ -0,0 +1,59 @@
+# Paper: OpenThoughts-Agent: Data Recipes for Agentic Models
+
+---
+type: paper
+title: "OpenThoughts-Agent: Data Recipes for Agentic Models"
+authors: Negin Raoof, Richard Zhuang, Marianna Nezhurina, Etash Guha, Atula Tejaswi, Ryan Marten, Charlie F. Ruan, Tyler Griggs, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.24855
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, rag
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.24855
diff --git a/papers/items/2026-2606-24937-the-hitchhiker-s-guide-to-agentic-ai-from-foundations-to-systems.md b/papers/items/2026-2606-24937-the-hitchhiker-s-guide-to-agentic-ai-from-foundations-to-systems.md
new file mode 100644
index 0000000..f9a8c71
--- /dev/null
+++ b/papers/items/2026-2606-24937-the-hitchhiker-s-guide-to-agentic-ai-from-foundations-to-systems.md
@@ -0,0 +1,68 @@
+# Paper: The Hitchhiker's Guide to Agentic AI: From Foundations to Systems
+
+---
+type: paper
+title: "The Hitchhiker's Guide to Agentic AI: From Foundations to Systems"
+authors: Haggai Roitman
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.24937
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-22
+updated_at: 2026-06-22
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - memory
+  - multi-agent
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+  - cs.IR
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 25
+collection_queries: agentic-ai, multi-agent-llm, rag-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, multi-agent-llm, rag-agent, tool-use
+- inferred topics: agent-evaluation, agent-safety, computer-use, memory, multi-agent, rag, reasoning, tool-use
+- arXiv categories: cs.AI, cs.CL, cs.IR, cs.LG
+- collection score: 25
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.24937
diff --git a/papers/items/2026-2606-24976-diagnosing-and-mitigating-compounding-failures-in-agentic-persuasion-via-taxonom.md b/papers/items/2026-2606-24976-diagnosing-and-mitigating-compounding-failures-in-agentic-persuasion-via-taxonom.md
new file mode 100644
index 0000000..76ca7f5
--- /dev/null
+++ b/papers/items/2026-2606-24976-diagnosing-and-mitigating-compounding-failures-in-agentic-persuasion-via-taxonom.md
@@ -0,0 +1,63 @@
+# Paper: Diagnosing and Mitigating Compounding Failures in Agentic Persuasion via Taxonomic Strategy Retrieval
+
+---
+type: paper
+title: Diagnosing and Mitigating Compounding Failures in Agentic Persuasion via Taxonomic Strategy Retrieval
+authors: Sana Ayromlou, Purvi Sehgal, Pradyumna Narayana
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.24976
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-07-04
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - planning
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, multi-agent, planning, rag
+- arXiv categories: cs.AI, cs.CL, cs.LG
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.24976
diff --git a/papers/items/2026-2606-25115-forget-to-improve-on-device-llm-agent-continual-learning-via-budget-curated-memo.md b/papers/items/2026-2606-25115-forget-to-improve-on-device-llm-agent-continual-learning-via-budget-curated-memo.md
new file mode 100644
index 0000000..542aa1a
--- /dev/null
+++ b/papers/items/2026-2606-25115-forget-to-improve-on-device-llm-agent-continual-learning-via-budget-curated-memo.md
@@ -0,0 +1,61 @@
+# Paper: Forget to Improve: On-Device LLM-Agent Continual Learning via Budget-Curated Memory
+
+---
+type: paper
+title: "Forget to Improve: On-Device LLM-Agent Continual Learning via Budget-Curated Memory"
+authors: Beining Wu, Zihao Ding, Jun Huang, Yanxiao Zhao
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.25115
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - embodied-agent
+  - memory
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.NI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, embodied-agent, memory
+- arXiv categories: cs.LG, cs.NI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.25115
diff --git a/papers/items/2026-2606-25139-buildrix-an-open-platform-for-sharing-and-benchmarking-agentic-ai-skills-in-buil.md b/papers/items/2026-2606-25139-buildrix-an-open-platform-for-sharing-and-benchmarking-agentic-ai-skills-in-buil.md
new file mode 100644
index 0000000..5a55b6d
--- /dev/null
+++ b/papers/items/2026-2606-25139-buildrix-an-open-platform-for-sharing-and-benchmarking-agentic-ai-skills-in-buil.md
@@ -0,0 +1,60 @@
+# Paper: Buildrix: An Open Platform for Sharing and Benchmarking Agentic AI Skills in Building Engineering
+
+---
+type: paper
+title: "Buildrix: An Open Platform for Sharing and Benchmarking Agentic AI Skills in Building Engineering"
+authors: Zixin Jiang, Bing Dong
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.25139
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - eess.SY
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-evaluation, tool-use, workflow-agent
+- arXiv categories: eess.SY
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.25139
diff --git a/papers/items/2026-2606-25161-trustmem-learning-trustworthy-memory-consolidation-for-llm-agents-with-long-term.md b/papers/items/2026-2606-25161-trustmem-learning-trustworthy-memory-consolidation-for-llm-agents-with-long-term.md
new file mode 100644
index 0000000..ac901b8
--- /dev/null
+++ b/papers/items/2026-2606-25161-trustmem-learning-trustworthy-memory-consolidation-for-llm-agents-with-long-term.md
@@ -0,0 +1,63 @@
+# Paper: TRUSTMEM: Learning Trustworthy Memory Consolidation for LLM Agents with Long-Term Memory
+
+---
+type: paper
+title: "TRUSTMEM: Learning Trustworthy Memory Consolidation for LLM Agents with Long-Term Memory"
+authors: Tianyu Yang, Sudipta Paul, Vijay Srinivasan, Vivek Kulkarni, Srinivas Chappidi
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.25161
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, computer-use, memory, rag, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.25161
diff --git a/papers/items/2026-2606-25189-actplane-programmable-os-level-policy-enforcement-for-agent-harnesses.md b/papers/items/2026-2606-25189-actplane-programmable-os-level-policy-enforcement-for-agent-harnesses.md
new file mode 100644
index 0000000..e4a1bcf
--- /dev/null
+++ b/papers/items/2026-2606-25189-actplane-programmable-os-level-policy-enforcement-for-agent-harnesses.md
@@ -0,0 +1,62 @@
+# Paper: ActPlane: Programmable OS-Level Policy Enforcement for Agent Harnesses
+
+---
+type: paper
+title: "ActPlane: Programmable OS-Level Policy Enforcement for Agent Harnesses"
+authors: Yusheng Zheng, Tianyuan Wu, Quanzhi Fu, Tong Yu, Wenan Mao, Tao Ma, Dan Williams, Wei Wang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.25189
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.OS
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent
+- inferred topics: agent-evaluation, agent-safety, coding-agent, planning, tool-use
+- arXiv categories: cs.OS
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.25189
diff --git a/papers/items/2026-2606-25191-to-isolate-or-to-score-model-adaptive-assessment-for-cost-efficient-multi-agent-.md b/papers/items/2026-2606-25191-to-isolate-or-to-score-model-adaptive-assessment-for-cost-efficient-multi-agent-.md
new file mode 100644
index 0000000..5524a93
--- /dev/null
+++ b/papers/items/2026-2606-25191-to-isolate-or-to-score-model-adaptive-assessment-for-cost-efficient-multi-agent-.md
@@ -0,0 +1,62 @@
+# Paper: To Isolate or to Score? Model-Adaptive Assessment for Cost-Efficient Multi-Agent RAG
+
+---
+type: paper
+title: To Isolate or to Score? Model-Adaptive Assessment for Cost-Efficient Multi-Agent RAG
+authors: Jungseob Lee, Chanjun Park, Heuiseok Lim
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.25191
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, multi-agent, rag, reasoning
+- arXiv categories: cs.AI, cs.CL
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.25191
diff --git a/papers/items/2026-2606-25195-sok-ai-secure-code-generation-progress-pitfalls-and-paths-forward.md b/papers/items/2026-2606-25195-sok-ai-secure-code-generation-progress-pitfalls-and-paths-forward.md
new file mode 100644
index 0000000..a28e5fb
--- /dev/null
+++ b/papers/items/2026-2606-25195-sok-ai-secure-code-generation-progress-pitfalls-and-paths-forward.md
@@ -0,0 +1,63 @@
+# Paper: SoK: AI Secure Code Generation: Progress, Pitfalls, and Paths Forward
+
+---
+type: paper
+title: "SoK: AI Secure Code Generation: Progress, Pitfalls, and Paths Forward"
+authors: Rupam Patir, Keyan Guo, Haipeng Cai, Hongxin Hu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.25195
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - computer-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agentic-ai, coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, coding-agent
+- inferred topics: agent-evaluation, agent-safety, coding-agent, computer-use, workflow-agent
+- arXiv categories: cs.CR, cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.25195
diff --git a/papers/items/2026-2606-25206-raven-long-horizon-reasoning-navigation-with-a-visuo-spatio-temporal-memory.md b/papers/items/2026-2606-25206-raven-long-horizon-reasoning-navigation-with-a-visuo-spatio-temporal-memory.md
new file mode 100644
index 0000000..01f441b
--- /dev/null
+++ b/papers/items/2026-2606-25206-raven-long-horizon-reasoning-navigation-with-a-visuo-spatio-temporal-memory.md
@@ -0,0 +1,65 @@
+# Paper: RAVEN: Long-Horizon Reasoning & Navigation with a Visuo-Spatio-Temporal Memory
+
+---
+type: paper
+title: "RAVEN: Long-Horizon Reasoning & Navigation with a Visuo-Spatio-Temporal Memory"
+authors: Yixun Hu, Zhicheng Zheng, Lihan Zha, Chunwei Xing, Rajdeep Singh, Omar Hossain, Antonio Loquercio, Dhruv Shah
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.25206
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-23
+updated_at: 2026-06-23
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - embodied-agent
+  - memory
+  - planning
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.RO
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, embodied-agent, memory, planning, rag, reasoning
+- arXiv categories: cs.RO, cs.AI, cs.CL
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.25206
diff --git a/papers/items/2026-2606-25334-bridging-the-post-discharge-gap-a-traceable-multi-agent-framework-for-safe-and-c.md b/papers/items/2026-2606-25334-bridging-the-post-discharge-gap-a-traceable-multi-agent-framework-for-safe-and-c.md
new file mode 100644
index 0000000..b21dffb
--- /dev/null
+++ b/papers/items/2026-2606-25334-bridging-the-post-discharge-gap-a-traceable-multi-agent-framework-for-safe-and-c.md
@@ -0,0 +1,62 @@
+# Paper: Bridging the Post-discharge Gap: A Traceable Multi-agent Framework for Safe and Continuous Care
+
+---
+type: paper
+title: "Bridging the Post-discharge Gap: A Traceable Multi-agent Framework for Safe and Continuous Care"
+authors: Runwei Guan, Yi Zhou, Heyi Lin, Jinjing Zhu, Mingyuan Hou, Yang Yang, Fang Yuan, Xiaohong Lin, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.25334
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-24
+updated_at: 2026-06-24
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - multi-agent
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, agent-safety, memory, multi-agent, rag
+- arXiv categories: cs.MA
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.25334
diff --git a/papers/items/2026-2606-25358-agentic-knowledge-tracing-a-multi-agent-llm-architecture-for-stealth-assessment-.md b/papers/items/2026-2606-25358-agentic-knowledge-tracing-a-multi-agent-llm-architecture-for-stealth-assessment-.md
new file mode 100644
index 0000000..fb50a54
--- /dev/null
+++ b/papers/items/2026-2606-25358-agentic-knowledge-tracing-a-multi-agent-llm-architecture-for-stealth-assessment-.md
@@ -0,0 +1,63 @@
+# Paper: Agentic Knowledge Tracing: A Multi-Agent LLM Architecture for Stealth Assessment of Financial Literacy in Serious Games
+
+---
+type: paper
+title: "Agentic Knowledge Tracing: A Multi-Agent LLM Architecture for Stealth Assessment of Financial Literacy in Serious Games"
+authors: Gabriel Santos, Rita Julia, Marcelo Nascimento
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.25358
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-24
+updated_at: 2026-06-24
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, agent-safety, multi-agent, reasoning, tool-use
+- arXiv categories: cs.AI, cs.MA
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.25358
diff --git a/papers/items/2026-2606-25361-memory-makes-the-difference-evaluating-how-different-memory-roles-shape-conversa.md b/papers/items/2026-2606-25361-memory-makes-the-difference-evaluating-how-different-memory-roles-shape-conversa.md
new file mode 100644
index 0000000..d934692
--- /dev/null
+++ b/papers/items/2026-2606-25361-memory-makes-the-difference-evaluating-how-different-memory-roles-shape-conversa.md
@@ -0,0 +1,63 @@
+# Paper: Memory Makes the Difference: Evaluating How Different Memory Roles Shape Conversational Agents
+
+---
+type: paper
+title: "Memory Makes the Difference: Evaluating How Different Memory Roles Shape Conversational Agents"
+authors: Yuxin Wang, Paul Thomas, Zhiwei Yu, Yuan Gao, Saeed Hassanpour, Soroush Vosoughi, Robert Sim, Nick Craswell
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.25361
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-24
+updated_at: 2026-06-24
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+  - cs.IR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, memory, rag, tool-use
+- arXiv categories: cs.CL, cs.AI, cs.IR
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.25361
diff --git a/papers/items/2026-2606-25400-brainagent-a-large-language-model-driven-multi-agent-framework-for-autonomous-br.md b/papers/items/2026-2606-25400-brainagent-a-large-language-model-driven-multi-agent-framework-for-autonomous-br.md
new file mode 100644
index 0000000..d8e29c7
--- /dev/null
+++ b/papers/items/2026-2606-25400-brainagent-a-large-language-model-driven-multi-agent-framework-for-autonomous-br.md
@@ -0,0 +1,62 @@
+# Paper: BrainAgent: A Large Language Model-Driven Multi-Agent Framework for Autonomous Brain Signal Understanding
+
+---
+type: paper
+title: "BrainAgent: A Large Language Model-Driven Multi-Agent Framework for Autonomous Brain Signal Understanding"
+authors: Yangxuan Zhou, Sha Zhao, Jiquan Wang, Shijian Li, Gang Pan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.25400
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-24
+updated_at: 2026-06-24
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - planning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, multi-agent, planning, tool-use, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.25400
diff --git a/papers/items/2026-2606-25484-from-causal-discovery-to-implementation-an-agentic-ai-framework-for-e-scooter-mo.md b/papers/items/2026-2606-25484-from-causal-discovery-to-implementation-an-agentic-ai-framework-for-e-scooter-mo.md
new file mode 100644
index 0000000..b36d9bf
--- /dev/null
+++ b/papers/items/2026-2606-25484-from-causal-discovery-to-implementation-an-agentic-ai-framework-for-e-scooter-mo.md
@@ -0,0 +1,63 @@
+# Paper: From Causal Discovery to Implementation: An Agentic AI Framework for E-Scooter Mobility Hub Planning Across 29 German Cities
+
+---
+type: paper
+title: "From Causal Discovery to Implementation: An Agentic AI Framework for E-Scooter Mobility Hub Planning Across 29 German Cities"
+authors: Meng Jin, Melanie Handrich, Simone Martinenz, Nicholas Hoeser, Ziyue Li
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.25484
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-24
+updated_at: 2026-06-24
+status: queued
+relevance: high
+topics:
+  - coding-agent
+  - computer-use
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CY
+  - econ.GN
+  - stat.AP
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: coding-agent, computer-use, planning, tool-use
+- arXiv categories: cs.CY, econ.GN, stat.AP
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.25484
diff --git a/papers/items/2026-2606-25514-unlocking-model-potentials-through-adaptive-multi-agent-scaffolding-for-efficien.md b/papers/items/2026-2606-25514-unlocking-model-potentials-through-adaptive-multi-agent-scaffolding-for-efficien.md
new file mode 100644
index 0000000..2165ef4
--- /dev/null
+++ b/papers/items/2026-2606-25514-unlocking-model-potentials-through-adaptive-multi-agent-scaffolding-for-efficien.md
@@ -0,0 +1,64 @@
+# Paper: Unlocking Model Potentials Through Adaptive Multi-Agent Scaffolding for Efficient Issue Resolution
+
+---
+type: paper
+title: Unlocking Model Potentials Through Adaptive Multi-Agent Scaffolding for Efficient Issue Resolution
+authors: Yang Chen, Aliya Ahmad, Yiheng Zhou, Reyhaneh Jabbarvand
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.25514
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-24
+updated_at: 2026-06-24
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - multi-agent
+  - planning
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: agent-evaluation, coding-agent, multi-agent, planning, rag, tool-use, workflow-agent
+- arXiv categories: cs.SE
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.25514
diff --git a/papers/items/2026-2606-25519-quantization-inflates-reasoning-token-inflation-as-a-hidden-cost-of-low-bit-reas.md b/papers/items/2026-2606-25519-quantization-inflates-reasoning-token-inflation-as-a-hidden-cost-of-low-bit-reas.md
new file mode 100644
index 0000000..d9a21ff
--- /dev/null
+++ b/papers/items/2026-2606-25519-quantization-inflates-reasoning-token-inflation-as-a-hidden-cost-of-low-bit-reas.md
@@ -0,0 +1,63 @@
+# Paper: Quantization Inflates Reasoning: Token Inflation as a Hidden Cost of Low-Bit Reasoning Models
+
+---
+type: paper
+title: "Quantization Inflates Reasoning: Token Inflation as a Hidden Cost of Low-Bit Reasoning Models"
+authors: Xinyu Lian, Walid Krichene, Beichen Huang, Masahiro Tanaka, Olatunji Ruwase, Li Zhang, Minjia Zhang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.25519
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-24
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, coding-agent, rag, reasoning, tool-use
+- arXiv categories: cs.AI, cs.LG
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.25519
diff --git a/papers/items/2026-2606-25588-intenttester-intent-driven-multi-agent-framework-for-cross-library-test-migratio.md b/papers/items/2026-2606-25588-intenttester-intent-driven-multi-agent-framework-for-cross-library-test-migratio.md
new file mode 100644
index 0000000..ee99132
--- /dev/null
+++ b/papers/items/2026-2606-25588-intenttester-intent-driven-multi-agent-framework-for-cross-library-test-migratio.md
@@ -0,0 +1,64 @@
+# Paper: IntentTester: Intent-Driven Multi-agent Framework for Cross-Library Test Migration
+
+---
+type: paper
+title: "IntentTester: Intent-Driven Multi-agent Framework for Cross-Library Test Migration"
+authors: Yi Gao, Ziyuan Zhang, Xing Hu, Xiaohu Yang, Xin Xia
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.25588
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-24
+updated_at: 2026-06-24
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - computer-use
+  - multi-agent
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, agent-safety, coding-agent, computer-use, multi-agent, reasoning, tool-use
+- arXiv categories: cs.SE
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.25588
diff --git a/papers/items/2026-2606-25622-probabilistic-agents-in-deterministic-audits-evaluating-multi-agent-systems-for-.md b/papers/items/2026-2606-25622-probabilistic-agents-in-deterministic-audits-evaluating-multi-agent-systems-for-.md
new file mode 100644
index 0000000..0f8f9f0
--- /dev/null
+++ b/papers/items/2026-2606-25622-probabilistic-agents-in-deterministic-audits-evaluating-multi-agent-systems-for-.md
@@ -0,0 +1,65 @@
+# Paper: Probabilistic Agents in Deterministic Audits: Evaluating Multi-Agent Systems for Automated Audits Based on the German IT-Grundschutz
+
+---
+type: paper
+title: "Probabilistic Agents in Deterministic Audits: Evaluating Multi-Agent Systems for Automated Audits Based on the German IT-Grundschutz"
+authors: Lea Roxanne Muth, Marian Margraf
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.25622
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-24
+updated_at: 2026-06-24
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, agent-safety, multi-agent, rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.CR, cs.AI
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.25622
diff --git a/papers/items/2026-2606-25651-medguards-multi-agent-system-for-reliable-medical-error-detection-and-correction.md b/papers/items/2026-2606-25651-medguards-multi-agent-system-for-reliable-medical-error-detection-and-correction.md
new file mode 100644
index 0000000..5f6b564
--- /dev/null
+++ b/papers/items/2026-2606-25651-medguards-multi-agent-system-for-reliable-medical-error-detection-and-correction.md
@@ -0,0 +1,62 @@
+# Paper: MedGuards: Multi-Agent System for Reliable Medical Error Detection and Correction
+
+---
+type: paper
+title: "MedGuards: Multi-Agent System for Reliable Medical Error Detection and Correction"
+authors: Congbo Ma, Hu Wang, Yichun Zhang, Farah E. Shamout
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.25651
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-24
+updated_at: 2026-06-26
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - multi-agent
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, agent-safety, computer-use, multi-agent, reasoning
+- arXiv categories: cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.25651
diff --git a/papers/items/2026-2606-25656-is-graphrag-needed-from-basic-rag-to-graph-agentic-solutions-with-context-optimi.md b/papers/items/2026-2606-25656-is-graphrag-needed-from-basic-rag-to-graph-agentic-solutions-with-context-optimi.md
new file mode 100644
index 0000000..06150fa
--- /dev/null
+++ b/papers/items/2026-2606-25656-is-graphrag-needed-from-basic-rag-to-graph-agentic-solutions-with-context-optimi.md
@@ -0,0 +1,63 @@
+# Paper: Is GraphRAG Needed? From Basic RAG to Graph-/Agentic Solutions with Context Optimization
+
+---
+type: paper
+title: Is GraphRAG Needed? From Basic RAG to Graph-/Agentic Solutions with Context Optimization
+authors: Long Chen, Ryan Razkenari, Yuxuan Zhou, Yuan Tian, Rahul Ghosh, Venkatesh Pappakrishnan, Disha Ahuja, Vidya Sagar Ravipati
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.25656
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-24
+updated_at: 2026-06-24
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+  - cs.IR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, memory, planning, rag
+- arXiv categories: cs.CL, cs.AI, cs.IR
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.25656
diff --git a/papers/items/2026-2606-25705-gui-agent-guided-exploration-of-user-sensitive-screens.md b/papers/items/2026-2606-25705-gui-agent-guided-exploration-of-user-sensitive-screens.md
new file mode 100644
index 0000000..591f31c
--- /dev/null
+++ b/papers/items/2026-2606-25705-gui-agent-guided-exploration-of-user-sensitive-screens.md
@@ -0,0 +1,60 @@
+# Paper: GUI agent: Guided Exploration of User-Sensitive Screens
+
+---
+type: paper
+title: "GUI agent: Guided Exploration of User-Sensitive Screens"
+authors: Aradhana Nayak, Mussadiq Nazeer, Wang Peng, Feng Liu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.25705
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-24
+updated_at: 2026-06-24
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - computer-use
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-safety, computer-use, tool-use
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.25705
diff --git a/papers/items/2026-2606-25760-uncertainty-quantification-for-computer-use-agents-a-benchmark-across-vision-lan.md b/papers/items/2026-2606-25760-uncertainty-quantification-for-computer-use-agents-a-benchmark-across-vision-lan.md
new file mode 100644
index 0000000..0579093
--- /dev/null
+++ b/papers/items/2026-2606-25760-uncertainty-quantification-for-computer-use-agents-a-benchmark-across-vision-lan.md
@@ -0,0 +1,64 @@
+# Paper: Uncertainty Quantification for Computer-Use Agents: A Benchmark across Vision-Language Models and GUI Grounding Datasets
+
+---
+type: paper
+title: "Uncertainty Quantification for Computer-Use Agents: A Benchmark across Vision-Language Models and GUI Grounding Datasets"
+authors: Divake Kumar, Sina Tayebati, Devashri Naik, Amanda Sofie Rios, Nilesh Ahuja, Omesh Tickoo, Ranganath Krishnan, Amit Ranjan Trivedi
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.25760
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-24
+updated_at: 2026-06-24
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+  - cs.CL
+  - cs.CV
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-evaluation, web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, web-gui-agent
+- inferred topics: agent-evaluation, agent-safety, computer-use, rag
+- arXiv categories: cs.LG, cs.AI, cs.CL, cs.CV
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.25760
diff --git a/papers/items/2026-2606-25819-beyond-function-calling-benchmarking-tool-using-agents-under-tool-environment-un.md b/papers/items/2026-2606-25819-beyond-function-calling-benchmarking-tool-using-agents-under-tool-environment-un.md
new file mode 100644
index 0000000..8dfd06a
--- /dev/null
+++ b/papers/items/2026-2606-25819-beyond-function-calling-benchmarking-tool-using-agents-under-tool-environment-un.md
@@ -0,0 +1,61 @@
+# Paper: Beyond Function Calling: Benchmarking Tool-Using Agents under Tool-Environment Unreliability
+
+---
+type: paper
+title: "Beyond Function Calling: Benchmarking Tool-Using Agents under Tool-Environment Unreliability"
+authors: Yang Tian, Zhengpeng Shi, Yu Zhou, Bo Zhao
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.25819
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-24
+updated_at: 2026-06-27
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: function-calling, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling, tool-use
+- inferred topics: agent-evaluation, tool-use, workflow-agent
+- arXiv categories: cs.CL, cs.SE
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.25819
diff --git a/papers/items/2026-2606-25899-manipulation-is-task-dependent-a-multi-axis-multi-environment-evaluation-of-fron.md b/papers/items/2026-2606-25899-manipulation-is-task-dependent-a-multi-axis-multi-environment-evaluation-of-fron.md
new file mode 100644
index 0000000..84877ae
--- /dev/null
+++ b/papers/items/2026-2606-25899-manipulation-is-task-dependent-a-multi-axis-multi-environment-evaluation-of-fron.md
@@ -0,0 +1,61 @@
+# Paper: Manipulation Is Task-Dependent: A Multi-Axis, Multi-Environment Evaluation of Frontier LLMs
+
+---
+type: paper
+title: "Manipulation Is Task-Dependent: A Multi-Axis, Multi-Environment Evaluation of Frontier LLMs"
+authors: Adeeb Zaman, Erik Nordby, Fred Heiding
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.25899
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-24
+updated_at: 2026-06-24
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-evaluation, rag, tool-use, workflow-agent
+- arXiv categories: cs.MA
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.25899
diff --git a/papers/items/2026-2606-26057-the-unfireable-safety-kernel-execution-time-ai-alignment-for-ai-agents-and-other.md b/papers/items/2026-2606-26057-the-unfireable-safety-kernel-execution-time-ai-alignment-for-ai-agents-and-other.md
new file mode 100644
index 0000000..db11a3c
--- /dev/null
+++ b/papers/items/2026-2606-26057-the-unfireable-safety-kernel-execution-time-ai-alignment-for-ai-agents-and-other.md
@@ -0,0 +1,64 @@
+# Paper: The Unfireable Safety Kernel: Execution-Time AI Alignment for AI Agents and Other Escapable AI Systems
+
+---
+type: paper
+title: "The Unfireable Safety Kernel: Execution-Time AI Alignment for AI Agents and Other Escapable AI Systems"
+authors: Seth Dobrin, Łukasz Chmiel
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.26057
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-24
+updated_at: 2026-06-24
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CR
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent
+- inferred topics: agent-evaluation, agent-safety, planning, tool-use, world-model
+- arXiv categories: cs.AI, cs.CR, cs.LG
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.26057
diff --git a/papers/items/2026-2606-26203-agentic-analysis-for-agentic-infrastructure-an-llm-powered-pipeline-for-comparat.md b/papers/items/2026-2606-26203-agentic-analysis-for-agentic-infrastructure-an-llm-powered-pipeline-for-comparat.md
new file mode 100644
index 0000000..63bb12f
--- /dev/null
+++ b/papers/items/2026-2606-26203-agentic-analysis-for-agentic-infrastructure-an-llm-powered-pipeline-for-comparat.md
@@ -0,0 +1,64 @@
+# Paper: Agentic Analysis for Agentic Infrastructure: An LLM-Powered Pipeline for Comparative Governance of DAO and Corporate AI Protocols
+
+---
+type: paper
+title: "Agentic Analysis for Agentic Infrastructure: An LLM-Powered Pipeline for Comparative Governance of DAO and Corporate AI Protocols"
+authors: Yutian Wang, Luyao Zhang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.26203
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-24
+updated_at: 2026-06-24
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - coding-agent
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CY
+  - cs.MA
+  - cs.SI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agentic-ai, ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, ai-agent
+- inferred topics: agent-safety, coding-agent, rag, tool-use
+- arXiv categories: cs.AI, cs.CY, cs.MA, cs.SI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.26203
diff --git a/papers/items/2026-2606-26205-knowledge-augmented-agentic-ai-for-mental-health-medication-information-seeking.md b/papers/items/2026-2606-26205-knowledge-augmented-agentic-ai-for-mental-health-medication-information-seeking.md
new file mode 100644
index 0000000..3cd9535
--- /dev/null
+++ b/papers/items/2026-2606-26205-knowledge-augmented-agentic-ai-for-mental-health-medication-information-seeking.md
@@ -0,0 +1,60 @@
+# Paper: Knowledge-augmented Agentic AI for Mental Health Medication Information Seeking
+
+---
+type: paper
+title: Knowledge-augmented Agentic AI for Mental Health Medication Information Seeking
+authors: Huizi Yu, Jian Liu, Wenkong Wang, Lingyao Li, Jiayan Zhou, Zhaoqian Xue, Xiang Li, Xinxin Lin, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.26205
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-24
+updated_at: 2026-06-24
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agentic-ai, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, multi-agent-llm
+- inferred topics: agent-evaluation, agent-safety, multi-agent
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.26205
diff --git a/papers/items/2026-2606-26216-cyberchainbench-can-ai-agents-secure-smart-contracts-against-real-world-on-chain.md b/papers/items/2026-2606-26216-cyberchainbench-can-ai-agents-secure-smart-contracts-against-real-world-on-chain.md
new file mode 100644
index 0000000..c4d2978
--- /dev/null
+++ b/papers/items/2026-2606-26216-cyberchainbench-can-ai-agents-secure-smart-contracts-against-real-world-on-chain.md
@@ -0,0 +1,61 @@
+# Paper: CyberChainBench: Can AI Agents Secure Smart Contracts Against Real-World On-Chain Vulnerabilities?
+
+---
+type: paper
+title: "CyberChainBench: Can AI Agents Secure Smart Contracts Against Real-World On-Chain Vulnerabilities?"
+authors: Jintao Huang, Fengqing Jiang, Radha Poovendran, Zhiqiang Lin
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.26216
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-24
+updated_at: 2026-06-24
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-safety, ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety, ai-agent
+- inferred topics: agent-evaluation, agent-safety, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.26216
diff --git a/papers/items/2026-2606-26300-the-verification-horizon-no-silver-bullet-for-coding-agent-rewards.md b/papers/items/2026-2606-26300-the-verification-horizon-no-silver-bullet-for-coding-agent-rewards.md
new file mode 100644
index 0000000..d858dc8
--- /dev/null
+++ b/papers/items/2026-2606-26300-the-verification-horizon-no-silver-bullet-for-coding-agent-rewards.md
@@ -0,0 +1,63 @@
+# Paper: The Verification Horizon: No Silver Bullet for Coding Agent Rewards
+
+---
+type: paper
+title: "The Verification Horizon: No Silver Bullet for Coding Agent Rewards"
+authors: Binghai Wang, Chenlong Zhang, Dayiheng Liu, Jiajun Zhang, Jiawei Chen, Mingze Li, Mouxiang Chen, Rongyao Fang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.26300
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-24
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - planning
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: agent-evaluation, coding-agent, planning, rag, reasoning
+- arXiv categories: cs.AI, cs.CL
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.26300
diff --git a/papers/items/2026-2606-26346-how-do-tool-augmented-llm-agents-perform-on-real-world-energy-analytics-tasks.md b/papers/items/2026-2606-26346-how-do-tool-augmented-llm-agents-perform-on-real-world-energy-analytics-tasks.md
new file mode 100644
index 0000000..ed27959
--- /dev/null
+++ b/papers/items/2026-2606-26346-how-do-tool-augmented-llm-agents-perform-on-real-world-energy-analytics-tasks.md
@@ -0,0 +1,63 @@
+# Paper: How Do Tool-Augmented LLM Agents Perform on Real-World Energy Analytics Tasks?
+
+---
+type: paper
+title: How Do Tool-Augmented LLM Agents Perform on Real-World Energy Analytics Tasks?
+authors: David Akinpelu, Akintonde Abbas, Rereloluwa Alimi, Ayodeji Lana
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.26346
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-24
+updated_at: 2026-06-24
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, agent-safety, coding-agent, rag, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.26346
diff --git a/papers/items/2026-2606-26356-instruction-bleed-cross-module-interference-in-prompt-composed-agentic-systems.md b/papers/items/2026-2606-26356-instruction-bleed-cross-module-interference-in-prompt-composed-agentic-systems.md
new file mode 100644
index 0000000..1242ccb
--- /dev/null
+++ b/papers/items/2026-2606-26356-instruction-bleed-cross-module-interference-in-prompt-composed-agentic-systems.md
@@ -0,0 +1,61 @@
+# Paper: Instruction Bleed: Cross-Module Interference in Prompt-Composed Agentic Systems
+
+---
+type: paper
+title: "Instruction Bleed: Cross-Module Interference in Prompt-Composed Agentic Systems"
+authors: Ching-Yu Lin, Yifan Liu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.26356
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-24
+updated_at: 2026-06-24
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.IR
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-evaluation, agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, agentic-ai
+- inferred topics: agent-evaluation, multi-agent
+- arXiv categories: cs.AI, cs.IR, cs.MA
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.26356
diff --git a/papers/items/2026-2606-26403-profilefoundry-a-synthetic-person-object-substrate-for-privacy-memory-and-tool-u.md b/papers/items/2026-2606-26403-profilefoundry-a-synthetic-person-object-substrate-for-privacy-memory-and-tool-u.md
new file mode 100644
index 0000000..557cab3
--- /dev/null
+++ b/papers/items/2026-2606-26403-profilefoundry-a-synthetic-person-object-substrate-for-privacy-memory-and-tool-u.md
@@ -0,0 +1,60 @@
+# Paper: ProfileFoundry: A Synthetic Person-Object Substrate for Privacy, Memory, and Tool-Use Evaluation in LLM Agent
+
+---
+type: paper
+title: "ProfileFoundry: A Synthetic Person-Object Substrate for Privacy, Memory, and Tool-Use Evaluation in LLM Agent"
+authors: Sriram Selvam, Anneswa Ghosh
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.26403
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-24
+updated_at: 2026-06-24
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, memory, tool-use
+- arXiv categories: cs.CL
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.26403
diff --git a/papers/items/2026-2606-26453-optimizing-cuda-like-a-human-micro-profiling-tools-as-expert-surrogates-for-llm-.md b/papers/items/2026-2606-26453-optimizing-cuda-like-a-human-micro-profiling-tools-as-expert-surrogates-for-llm-.md
new file mode 100644
index 0000000..67a8264
--- /dev/null
+++ b/papers/items/2026-2606-26453-optimizing-cuda-like-a-human-micro-profiling-tools-as-expert-surrogates-for-llm-.md
@@ -0,0 +1,63 @@
+# Paper: Optimizing CUDA like a Human: Micro-Profiling Tools as Expert Surrogates for LLM-Based GPU Kernel Optimization
+
+---
+type: paper
+title: "Optimizing CUDA like a Human: Micro-Profiling Tools as Expert Surrogates for LLM-Based GPU Kernel Optimization"
+authors: Jiading Gai, Shuai Zhang, Kaj Bostrom, Jin Huang, Vihang Patil, Haoyang Fang, Bernie Wang, Huzefa Rangwala, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.26453
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-24
+updated_at: 2026-06-24
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - computer-use
+  - memory
+  - multi-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: coding-agent, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent, multi-agent-llm
+- inferred topics: agent-evaluation, coding-agent, computer-use, memory, multi-agent, tool-use
+- arXiv categories: cs.LG
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.26453
diff --git a/papers/items/2026-2606-26479-adaptive-evaluation-of-out-of-band-defenses-against-prompt-injection-in-llm-agen.md b/papers/items/2026-2606-26479-adaptive-evaluation-of-out-of-band-defenses-against-prompt-injection-in-llm-agen.md
new file mode 100644
index 0000000..2effe82
--- /dev/null
+++ b/papers/items/2026-2606-26479-adaptive-evaluation-of-out-of-band-defenses-against-prompt-injection-in-llm-agen.md
@@ -0,0 +1,64 @@
+# Paper: Adaptive Evaluation of Out-of-Band Defenses Against Prompt Injection in LLM Agents
+
+---
+type: paper
+title: Adaptive Evaluation of Out-of-Band Defenses Against Prompt Injection in LLM Agents
+authors: Praneeth Narisetty, Shiva Nagendra Babu Kore, Uday Kumar Reddy Kattamanchi, Jayaram Kumarapu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.26479
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-25
+updated_at: 2026-06-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+  - cs.CL
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, agent-safety, rag, tool-use
+- arXiv categories: cs.CR, cs.AI, cs.CL, cs.LG
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.26479
diff --git a/papers/items/2026-2606-26511-temporal-validity-in-retrieval-memory-eliminating-stale-fact-errors-for-ai-agent.md b/papers/items/2026-2606-26511-temporal-validity-in-retrieval-memory-eliminating-stale-fact-errors-for-ai-agent.md
new file mode 100644
index 0000000..4e69458
--- /dev/null
+++ b/papers/items/2026-2606-26511-temporal-validity-in-retrieval-memory-eliminating-stale-fact-errors-for-ai-agent.md
@@ -0,0 +1,65 @@
+# Paper: Temporal Validity in Retrieval Memory: Eliminating Stale-Fact Errors for AI Agents over Evolving Knowledge
+
+---
+type: paper
+title: "Temporal Validity in Retrieval Memory: Eliminating Stale-Fact Errors for AI Agents over Evolving Knowledge"
+authors: Neeraj Yadav
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.26511
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-25
+updated_at: 2026-06-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+  - cs.ET
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 21
+collection_queries: ai-agent, rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent, rag-agent
+- inferred topics: agent-evaluation, computer-use, memory, rag, tool-use
+- arXiv categories: cs.CL, cs.AI, cs.ET, cs.LG
+- collection score: 21
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.26511
diff --git a/papers/items/2026-2606-26524-vigil-runtime-enforcement-of-behavioral-specifications-in-ai-agent-skills.md b/papers/items/2026-2606-26524-vigil-runtime-enforcement-of-behavioral-specifications-in-ai-agent-skills.md
new file mode 100644
index 0000000..3cb867c
--- /dev/null
+++ b/papers/items/2026-2606-26524-vigil-runtime-enforcement-of-behavioral-specifications-in-ai-agent-skills.md
@@ -0,0 +1,60 @@
+# Paper: VIGIL: Runtime Enforcement of Behavioral Specifications in AI Agent Skills
+
+---
+type: paper
+title: "VIGIL: Runtime Enforcement of Behavioral Specifications in AI Agent Skills"
+authors: Ying Li, Yanju Chen, Hongbo Wen, Bosi Zhang, Hanzhi Liu, Peiran Wang, Yu Feng, Yuan Tian
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.26524
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-25
+updated_at: 2026-06-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent
+- inferred topics: agent-evaluation, tool-use, workflow-agent
+- arXiv categories: cs.CR
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.26524
diff --git a/papers/items/2026-2606-26614-hilsva-design-and-evaluation-of-a-human-in-the-loop-agentic-system-for-scientifi.md b/papers/items/2026-2606-26614-hilsva-design-and-evaluation-of-a-human-in-the-loop-agentic-system-for-scientifi.md
new file mode 100644
index 0000000..5038903
--- /dev/null
+++ b/papers/items/2026-2606-26614-hilsva-design-and-evaluation-of-a-human-in-the-loop-agentic-system-for-scientifi.md
@@ -0,0 +1,67 @@
+# Paper: HiLSVA: Design and Evaluation of a Human-in-the-Loop Agentic System for Scientific Visualization
+
+---
+type: paper
+title: "HiLSVA: Design and Evaluation of a Human-in-the-Loop Agentic System for Scientific Visualization"
+authors: Kuangshi Ai, Patrick Phuoc Do, Chaoli Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.26614
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-25
+updated_at: 2026-06-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - multi-agent
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.HC
+  - cs.AI
+  - cs.GR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 23
+collection_queries: multi-agent-llm, planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm, planning-agent
+- inferred topics: agent-evaluation, computer-use, multi-agent, planning, rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.HC, cs.AI, cs.GR
+- collection score: 23
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.26614
diff --git a/papers/items/2026-2606-26627-agents-that-know-too-much-a-data-centric-survey-of-privacy-in-llm-agents.md b/papers/items/2026-2606-26627-agents-that-know-too-much-a-data-centric-survey-of-privacy-in-llm-agents.md
new file mode 100644
index 0000000..743f6ba
--- /dev/null
+++ b/papers/items/2026-2606-26627-agents-that-know-too-much-a-data-centric-survey-of-privacy-in-llm-agents.md
@@ -0,0 +1,64 @@
+# Paper: Agents That Know Too Much: A Data-Centric Survey of Privacy in LLM Agents
+
+---
+type: paper
+title: "Agents That Know Too Much: A Data-Centric Survey of Privacy in LLM Agents"
+authors: Nada Lahjouji, Ashwin Gerard Colaco
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.26627
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-25
+updated_at: 2026-06-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, agent-safety, memory, rag, tool-use, workflow-agent
+- arXiv categories: cs.CR, cs.AI
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.26627
diff --git a/papers/items/2026-2606-26649-autoformalization-of-agent-instructions-into-policy-as-code.md b/papers/items/2026-2606-26649-autoformalization-of-agent-instructions-into-policy-as-code.md
new file mode 100644
index 0000000..eb9604d
--- /dev/null
+++ b/papers/items/2026-2606-26649-autoformalization-of-agent-instructions-into-policy-as-code.md
@@ -0,0 +1,62 @@
+# Paper: Autoformalization of Agent Instructions into Policy-as-Code
+
+---
+type: paper
+title: Autoformalization of Agent Instructions into Policy-as-Code
+authors: Adam Mondl, Matthew Maisel, John H. Brock
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.26649
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-25
+updated_at: 2026-06-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, coding-agent, tool-use
+- arXiv categories: cs.AI, cs.CR
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.26649
diff --git a/papers/items/2026-2606-26721-knowledge-based-pull-requests-a-trusted-workflow-for-agent-mediated-knowledge-co.md b/papers/items/2026-2606-26721-knowledge-based-pull-requests-a-trusted-workflow-for-agent-mediated-knowledge-co.md
new file mode 100644
index 0000000..ee821ed
--- /dev/null
+++ b/papers/items/2026-2606-26721-knowledge-based-pull-requests-a-trusted-workflow-for-agent-mediated-knowledge-co.md
@@ -0,0 +1,67 @@
+# Paper: Knowledge-Based Pull Requests: A Trusted Workflow for Agent-Mediated Knowledge Collaboration
+
+---
+type: paper
+title: "Knowledge-Based Pull Requests: A Trusted Workflow for Agent-Mediated Knowledge Collaboration"
+authors: Xinyu Zhang, Weiwei Sun
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.26721
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-25
+updated_at: 2026-06-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - memory
+  - multi-agent
+  - planning
+  - tool-use
+  - workflow-agent
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.HC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: agent-evaluation, agent-safety, coding-agent, memory, multi-agent, planning, tool-use, workflow-agent, world-model
+- arXiv categories: cs.SE, cs.HC
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.26721
diff --git a/papers/items/2026-2606-26758-egg-an-expert-guided-agent-framework-for-kernel-generation.md b/papers/items/2026-2606-26758-egg-an-expert-guided-agent-framework-for-kernel-generation.md
new file mode 100644
index 0000000..95e1588
--- /dev/null
+++ b/papers/items/2026-2606-26758-egg-an-expert-guided-agent-framework-for-kernel-generation.md
@@ -0,0 +1,62 @@
+# Paper: EGG: An Expert-Guided Agent Framework for Kernel Generation
+
+---
+type: paper
+title: "EGG: An Expert-Guided Agent Framework for Kernel Generation"
+authors: Yaochen Han, Ke Fan, Hongxu Jiang, Wanqi Xu, Weiyu Xie, Runhua Zhang, Chenhui Zhu, Yixiang Zhang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.26758
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-25
+updated_at: 2026-06-25
+status: queued
+relevance: high
+topics:
+  - computer-use
+  - memory
+  - multi-agent
+  - rag
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: computer-use, memory, multi-agent, rag, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.26758
diff --git a/papers/items/2026-2606-26793-mirror-novelty-constrained-memory-guided-mcts-red-teaming-for-agentic-rag.md b/papers/items/2026-2606-26793-mirror-novelty-constrained-memory-guided-mcts-red-teaming-for-agentic-rag.md
new file mode 100644
index 0000000..35dfdf8
--- /dev/null
+++ b/papers/items/2026-2606-26793-mirror-novelty-constrained-memory-guided-mcts-red-teaming-for-agentic-rag.md
@@ -0,0 +1,65 @@
+# Paper: MIRROR: Novelty-Constrained Memory-Guided MCTS Red-Teaming for Agentic RAG
+
+---
+type: paper
+title: "MIRROR: Novelty-Constrained Memory-Guided MCTS Red-Teaming for Agentic RAG"
+authors: Inderjeet Singh, Andrés Murillo, Motoyoshi Sekiya, Yuki Unno, Junichi Suga
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.26793
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-25
+updated_at: 2026-06-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, agent-safety, computer-use, memory, rag, tool-use
+- arXiv categories: cs.CR, cs.AI, cs.LG
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.26793
diff --git a/papers/items/2026-2606-26806-memory-depth-not-memory-access-selective-parametric-consolidation-for-long-runni.md b/papers/items/2026-2606-26806-memory-depth-not-memory-access-selective-parametric-consolidation-for-long-runni.md
new file mode 100644
index 0000000..0f316fd
--- /dev/null
+++ b/papers/items/2026-2606-26806-memory-depth-not-memory-access-selective-parametric-consolidation-for-long-runni.md
@@ -0,0 +1,61 @@
+# Paper: Memory Depth, Not Memory Access: Selective Parametric Consolidation for Long-Running Language Agents
+
+---
+type: paper
+title: "Memory Depth, Not Memory Access: Selective Parametric Consolidation for Long-Running Language Agents"
+authors: Haoliang Han
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.26806
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-25
+updated_at: 2026-06-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, memory, rag
+- arXiv categories: cs.AI, cs.LG
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.26806
diff --git a/papers/items/2026-2606-26883-econsimulacra-a-digital-twin-platform-of-socio-economic-systems-powered-by-llm-a.md b/papers/items/2026-2606-26883-econsimulacra-a-digital-twin-platform-of-socio-economic-systems-powered-by-llm-a.md
new file mode 100644
index 0000000..3ca557a
--- /dev/null
+++ b/papers/items/2026-2606-26883-econsimulacra-a-digital-twin-platform-of-socio-economic-systems-powered-by-llm-a.md
@@ -0,0 +1,60 @@
+# Paper: EconSimulacra: A Digital Twin Platform of Socio-Economic Systems Powered by LLM Agents
+
+---
+type: paper
+title: "EconSimulacra: A Digital Twin Platform of Socio-Economic Systems Powered by LLM Agents"
+authors: Ryuji Hashimoto, Masahiro Kaneko, Kentaro Ueda, Takehiro Takayanagi, Kiyoshi Izumi
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.26883
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-25
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - memory
+  - multi-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.DL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: memory, multi-agent, tool-use
+- arXiv categories: cs.DL
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.26883
diff --git a/papers/items/2026-2606-26918-diagnosing-task-insensitivity-in-language-agents.md b/papers/items/2026-2606-26918-diagnosing-task-insensitivity-in-language-agents.md
new file mode 100644
index 0000000..ea68977
--- /dev/null
+++ b/papers/items/2026-2606-26918-diagnosing-task-insensitivity-in-language-agents.md
@@ -0,0 +1,61 @@
+# Paper: Diagnosing Task Insensitivity in Language Agents
+
+---
+type: paper
+title: Diagnosing Task Insensitivity in Language Agents
+authors: Jingyu Liu, Xiaopeng Wu, Kehan Chen, Chuan Yu, Yong Liu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.26918
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-25
+updated_at: 2026-06-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, planning, rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.26918
diff --git a/papers/items/2026-2606-26924-a-deterministic-control-plane-for-llm-coding-agents.md b/papers/items/2026-2606-26924-a-deterministic-control-plane-for-llm-coding-agents.md
new file mode 100644
index 0000000..546ce13
--- /dev/null
+++ b/papers/items/2026-2606-26924-a-deterministic-control-plane-for-llm-coding-agents.md
@@ -0,0 +1,64 @@
+# Paper: A Deterministic Control Plane for LLM Coding Agents
+
+---
+type: paper
+title: A Deterministic Control Plane for LLM Coding Agents
+authors: Padmaraj Madatha
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.26924
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-25
+updated_at: 2026-06-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - planning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: agent-evaluation, coding-agent, planning, tool-use, workflow-agent
+- arXiv categories: cs.SE, cs.AI, cs.CR
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.26924
diff --git a/papers/items/2026-2606-26960-toward-agentic-sysadmin-rethinking-system-administration-with-ai-agents.md b/papers/items/2026-2606-26960-toward-agentic-sysadmin-rethinking-system-administration-with-ai-agents.md
new file mode 100644
index 0000000..b92a620
--- /dev/null
+++ b/papers/items/2026-2606-26960-toward-agentic-sysadmin-rethinking-system-administration-with-ai-agents.md
@@ -0,0 +1,61 @@
+# Paper: Toward Agentic SysAdmin: Rethinking System Administration with AI Agents
+
+---
+type: paper
+title: "Toward Agentic SysAdmin: Rethinking System Administration with AI Agents"
+authors: Gianmaria Frigo, Davide Saladino, Alberto Castagnaro, Francesco Marchiori, Denis Donadel, Luca Pajola, Mauro Conti
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.26960
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-25
+updated_at: 2026-06-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.NI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent
+- inferred topics: agent-evaluation, agent-safety, rag, tool-use
+- arXiv categories: cs.NI
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.26960
diff --git a/papers/items/2026-2606-27009-semantic-early-stopping-for-iterative-llm-agent-loops.md b/papers/items/2026-2606-27009-semantic-early-stopping-for-iterative-llm-agent-loops.md
new file mode 100644
index 0000000..5103faf
--- /dev/null
+++ b/papers/items/2026-2606-27009-semantic-early-stopping-for-iterative-llm-agent-loops.md
@@ -0,0 +1,63 @@
+# Paper: Semantic Early-Stopping for Iterative LLM Agent Loops
+
+---
+type: paper
+title: Semantic Early-Stopping for Iterative LLM Agent Loops
+authors: Sahil Shrivastava
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.27009
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-25
+updated_at: 2026-06-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.LG
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, multi-agent, rag, tool-use
+- arXiv categories: cs.AI, cs.LG, cs.MA
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.27009
diff --git a/papers/items/2026-2606-27154-openrca-2-0-from-outcome-labels-to-causal-process-supervision.md b/papers/items/2026-2606-27154-openrca-2-0-from-outcome-labels-to-causal-process-supervision.md
new file mode 100644
index 0000000..fc7b768
--- /dev/null
+++ b/papers/items/2026-2606-27154-openrca-2-0-from-outcome-labels-to-causal-process-supervision.md
@@ -0,0 +1,61 @@
+# Paper: OpenRCA 2.0: From Outcome Labels to Causal Process Supervision
+
+---
+type: paper
+title: "OpenRCA 2.0: From Outcome Labels to Causal Process Supervision"
+authors: "Aoyang Fang, Yifan Yang, Jin'ao Shang, Qisheng Lu, Junjielung Xu, Rui Wang, Songhan Zhang, Yuzhong Zhang, et al."
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.27154
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-25
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, rag, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.27154
diff --git a/papers/items/2026-2606-27243-nova-a-verification-aware-agent-harness-for-architecture-evolution-in-industrial.md b/papers/items/2026-2606-27243-nova-a-verification-aware-agent-harness-for-architecture-evolution-in-industrial.md
new file mode 100644
index 0000000..55b2d47
--- /dev/null
+++ b/papers/items/2026-2606-27243-nova-a-verification-aware-agent-harness-for-architecture-evolution-in-industrial.md
@@ -0,0 +1,64 @@
+# Paper: NOVA: A Verification-Aware Agent Harness for Architecture Evolution in Industrial Recommender Systems
+
+---
+type: paper
+title: "NOVA: A Verification-Aware Agent Harness for Architecture Evolution in Industrial Recommender Systems"
+authors: Shaohua Liu, Liang Fang, Yilong Sun, Shudong Huang, Qingsong Luo, Shaoxin Liu, Xiaoyang Chen, Dongqiang Liu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.27243
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-25
+updated_at: 2026-06-26
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - computer-use
+  - memory
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.IR
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: agent-evaluation, agent-safety, coding-agent, computer-use, memory, workflow-agent
+- arXiv categories: cs.IR, cs.SE
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.27243
diff --git a/papers/items/2026-2606-27330-empowering-gui-agents-via-autonomous-experience-exploration-and-hindsight-experi.md b/papers/items/2026-2606-27330-empowering-gui-agents-via-autonomous-experience-exploration-and-hindsight-experi.md
new file mode 100644
index 0000000..e769711
--- /dev/null
+++ b/papers/items/2026-2606-27330-empowering-gui-agents-via-autonomous-experience-exploration-and-hindsight-experi.md
@@ -0,0 +1,65 @@
+# Paper: Empowering GUI Agents via Autonomous Experience Exploration and Hindsight Experience Utilization for Task Planning
+
+---
+type: paper
+title: Empowering GUI Agents via Autonomous Experience Exploration and Hindsight Experience Utilization for Task Planning
+authors: Tianyi Men, Zhuoran Jin, Pengfei Cao, Yubo Chen, Kang Liu, Jun Zhao
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.27330
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-25
+updated_at: 2026-06-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+  - cs.CV
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, computer-use, planning, rag, tool-use
+- arXiv categories: cs.CL, cs.AI, cs.CV, cs.LG
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.27330
diff --git a/papers/items/2026-2606-27350-chia-an-open-source-framework-for-principled-agentic-ai-driven-hardware-software.md b/papers/items/2026-2606-27350-chia-an-open-source-framework-for-principled-agentic-ai-driven-hardware-software.md
new file mode 100644
index 0000000..0670d43
--- /dev/null
+++ b/papers/items/2026-2606-27350-chia-an-open-source-framework-for-principled-agentic-ai-driven-hardware-software.md
@@ -0,0 +1,61 @@
+# Paper: CHIA: An open-source framework for principled, agentic AI-driven hardware/software co-design research
+
+---
+type: paper
+title: "CHIA: An open-source framework for principled, agentic AI-driven hardware/software co-design research"
+authors: Angela Cui, Ferran Hermida-Rivera, Jack Toubes, Raghav Gupta, Jim Fang, Chengyi Lux Zhang, Ella Schwarz, Junha Kim, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.27350
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-25
+updated_at: 2026-06-27
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - coding-agent
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agentic-ai, coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, coding-agent
+- inferred topics: agent-safety, coding-agent, tool-use, workflow-agent
+- arXiv categories: cs.AR
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.27350
diff --git a/papers/items/2026-2606-27397-sidconarena-an-environment-evaluating-agents-in-open-ended-positive-sum-bargaini.md b/papers/items/2026-2606-27397-sidconarena-an-environment-evaluating-agents-in-open-ended-positive-sum-bargaini.md
new file mode 100644
index 0000000..3a33d4c
--- /dev/null
+++ b/papers/items/2026-2606-27397-sidconarena-an-environment-evaluating-agents-in-open-ended-positive-sum-bargaini.md
@@ -0,0 +1,64 @@
+# Paper: SidConArena: An Environment Evaluating Agents in Open-Ended,Positive-Sum Bargaining Game
+
+---
+type: paper
+title: "SidConArena: An Environment Evaluating Agents in Open-Ended,Positive-Sum Bargaining Game"
+authors: Yeqi Feng, Yuxin Chen, Tianxing He
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.27397
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-24
+updated_at: 2026-06-24
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+  - cs.AI
+  - cs.GT
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, memory, planning, reasoning, tool-use
+- arXiv categories: cs.MA, cs.AI, cs.GT
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.27397
diff --git a/papers/items/2026-2606-27406-towards-evaluation-of-implicit-software-world-models-in-coding-llms.md b/papers/items/2026-2606-27406-towards-evaluation-of-implicit-software-world-models-in-coding-llms.md
new file mode 100644
index 0000000..206562b
--- /dev/null
+++ b/papers/items/2026-2606-27406-towards-evaluation-of-implicit-software-world-models-in-coding-llms.md
@@ -0,0 +1,63 @@
+# Paper: Towards Evaluation of Implicit Software World Models in Coding LLMs
+
+---
+type: paper
+title: Towards Evaluation of Implicit Software World Models in Coding LLMs
+authors: Egor Bogomolov, Yaroslav Zharov
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.27406
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-25
+updated_at: 2026-06-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - memory
+  - reasoning
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: ai-agent, coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent, coding-agent
+- inferred topics: agent-evaluation, coding-agent, memory, reasoning, world-model
+- arXiv categories: cs.SE, cs.AI
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.27406
diff --git a/papers/items/2026-2606-27416-glite-arf-verifier-driven-research-with-parallel-llm-coding-agents.md b/papers/items/2026-2606-27416-glite-arf-verifier-driven-research-with-parallel-llm-coding-agents.md
new file mode 100644
index 0000000..4a3d259
--- /dev/null
+++ b/papers/items/2026-2606-27416-glite-arf-verifier-driven-research-with-parallel-llm-coding-agents.md
@@ -0,0 +1,61 @@
+# Paper: Glite ARF: Verifier-Driven Research with Parallel LLM Coding Agents
+
+---
+type: paper
+title: "Glite ARF: Verifier-Driven Research with Parallel LLM Coding Agents"
+authors: Vassili Philippov, Pavel Katunin, Dmitry Andreev, Igor Ostanin, Anton Nikolaev
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.27416
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-25
+updated_at: 2026-06-25
+status: queued
+relevance: high
+topics:
+  - coding-agent
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: coding-agent, reasoning, tool-use
+- arXiv categories: cs.MA, cs.SE
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.27416
diff --git a/papers/items/2026-2606-27472-supersede-diagnosing-and-training-the-memory-update-gap-in-llm-agents.md b/papers/items/2026-2606-27472-supersede-diagnosing-and-training-the-memory-update-gap-in-llm-agents.md
new file mode 100644
index 0000000..973034c
--- /dev/null
+++ b/papers/items/2026-2606-27472-supersede-diagnosing-and-training-the-memory-update-gap-in-llm-agents.md
@@ -0,0 +1,64 @@
+# Paper: Supersede: Diagnosing and Training the Memory-Update Gap in LLM Agents
+
+---
+type: paper
+title: "Supersede: Diagnosing and Training the Memory-Update Gap in LLM Agents"
+authors: Vedant Patel
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.27472
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-25
+updated_at: 2026-06-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, memory, planning, reasoning, tool-use
+- arXiv categories: cs.CL, cs.AI, cs.LG
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.27472
diff --git a/papers/items/2026-2606-27483-internalizing-the-future-a-unified-agentic-training-paradigm-for-world-model-pla.md b/papers/items/2026-2606-27483-internalizing-the-future-a-unified-agentic-training-paradigm-for-world-model-pla.md
new file mode 100644
index 0000000..36d2026
--- /dev/null
+++ b/papers/items/2026-2606-27483-internalizing-the-future-a-unified-agentic-training-paradigm-for-world-model-pla.md
@@ -0,0 +1,62 @@
+# Paper: Internalizing the Future: A Unified Agentic Training Paradigm for World Model Planning
+
+---
+type: paper
+title: "Internalizing the Future: A Unified Agentic Training Paradigm for World Model Planning"
+authors: Xuan Zhang, Zhijian Zhou, Lingfeng Qiao, Yulei Qin, Ke Li, Xing Sun, Xiaoyu Tan, Chao Qu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.27483
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-25
+updated_at: 2026-06-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - reasoning
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: planning-agent
+- inferred topics: agent-evaluation, planning, rag, reasoning, world-model
+- arXiv categories: cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.27483
diff --git a/papers/items/2026-2606-27492-queenbee-planner-skill-evolving-communication-topologies-for-token-efficient-llm.md b/papers/items/2026-2606-27492-queenbee-planner-skill-evolving-communication-topologies-for-token-efficient-llm.md
new file mode 100644
index 0000000..bd9fe1a
--- /dev/null
+++ b/papers/items/2026-2606-27492-queenbee-planner-skill-evolving-communication-topologies-for-token-efficient-llm.md
@@ -0,0 +1,61 @@
+# Paper: QueenBee Planner: Skill-Evolving Communication Topologies for Token-Efficient LLM Multi-Agent Systems
+
+---
+type: paper
+title: "QueenBee Planner: Skill-Evolving Communication Topologies for Token-Efficient LLM Multi-Agent Systems"
+authors: Congjia Tian, Yuhang Yao, Jiaming Cui
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.27492
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-25
+updated_at: 2026-06-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, multi-agent, planning, tool-use
+- arXiv categories: cs.MA
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.27492
diff --git a/papers/items/2026-2606-27499-dmv-bench-diagnosing-long-horizon-multimodal-agents-visual-memory-with-incidenta.md b/papers/items/2026-2606-27499-dmv-bench-diagnosing-long-horizon-multimodal-agents-visual-memory-with-incidenta.md
new file mode 100644
index 0000000..4c7c9b4
--- /dev/null
+++ b/papers/items/2026-2606-27499-dmv-bench-diagnosing-long-horizon-multimodal-agents-visual-memory-with-incidenta.md
@@ -0,0 +1,65 @@
+# Paper: DMV-Bench: Diagnosing Long-Horizon Multimodal Agents' Visual Memory with Incidental Cue Injection
+
+---
+type: paper
+title: "DMV-Bench: Diagnosing Long-Horizon Multimodal Agents' Visual Memory with Incidental Cue Injection"
+authors: Yujin Tang, Chenming Shang, Ruize Xu, Nikhil Singh
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.27499
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-25
+updated_at: 2026-06-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - memory
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, coding-agent, memory, planning, rag, tool-use
+- arXiv categories: cs.CV, cs.AI, cs.CL
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.27499
diff --git a/papers/items/2026-2606-27632-yuvion-llm-an-adversarially-aware-large-language-model-for-content-and-ai-safety.md b/papers/items/2026-2606-27632-yuvion-llm-an-adversarially-aware-large-language-model-for-content-and-ai-safety.md
new file mode 100644
index 0000000..31dc712
--- /dev/null
+++ b/papers/items/2026-2606-27632-yuvion-llm-an-adversarially-aware-large-language-model-for-content-and-ai-safety.md
@@ -0,0 +1,62 @@
+# Paper: Yuvion LLM: An Adversarially-Aware Large Language Model for Content And AI Safety
+
+---
+type: paper
+title: "Yuvion LLM: An Adversarially-Aware Large Language Model for Content And AI Safety"
+authors: Ting Ma, Xiufeng Huang, Benlei Cui, Xiaowen Xu, Shikai Qiu, Ruijie Jian, Hongxing Li, Guanghui Wang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.27632
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-26
+updated_at: 2026-06-26
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, agent-safety, planning, reasoning, tool-use
+- arXiv categories: cs.CL
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.27632
diff --git a/papers/items/2026-2606-27806-agent-vs-parametric-world-models-hybrid-planning-for-reliable-language-agents.md b/papers/items/2026-2606-27806-agent-vs-parametric-world-models-hybrid-planning-for-reliable-language-agents.md
new file mode 100644
index 0000000..7064487
--- /dev/null
+++ b/papers/items/2026-2606-27806-agent-vs-parametric-world-models-hybrid-planning-for-reliable-language-agents.md
@@ -0,0 +1,64 @@
+# Paper: Agent vs. Parametric World Models: Hybrid Planning for Reliable Language Agents
+
+---
+type: paper
+title: "Agent vs. Parametric World Models: Hybrid Planning for Reliable Language Agents"
+authors: Xinyuan Song, Zekun Cai
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.27806
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-26
+updated_at: 2026-07-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, agent-safety, planning, rag, reasoning, tool-use, world-model
+- arXiv categories: cs.AI
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.27806
diff --git a/papers/items/2026-2606-27929-when-multi-robot-systems-meet-agentic-ai-towards-embodied-collective-intelligenc.md b/papers/items/2026-2606-27929-when-multi-robot-systems-meet-agentic-ai-towards-embodied-collective-intelligenc.md
new file mode 100644
index 0000000..50c673b
--- /dev/null
+++ b/papers/items/2026-2606-27929-when-multi-robot-systems-meet-agentic-ai-towards-embodied-collective-intelligenc.md
@@ -0,0 +1,62 @@
+# Paper: When Multi-Robot Systems Meet Agentic AI:Towards Embodied Collective Intelligence
+
+---
+type: paper
+title: "When Multi-Robot Systems Meet Agentic AI:Towards Embodied Collective Intelligence"
+authors: Yuxuan Yan, Yuanyuan Jia, Qianqian Yang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.27929
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-26
+updated_at: 2026-06-26
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - embodied-agent
+  - memory
+  - multi-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.RO
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent
+- inferred topics: agent-evaluation, embodied-agent, memory, multi-agent, tool-use
+- arXiv categories: cs.RO
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.27929
diff --git a/papers/items/2026-2606-27990-advancedshellm-a-stateful-multi-agent-llm-honeypot-for-ssh-deception.md b/papers/items/2026-2606-27990-advancedshellm-a-stateful-multi-agent-llm-honeypot-for-ssh-deception.md
new file mode 100644
index 0000000..d61b1ff
--- /dev/null
+++ b/papers/items/2026-2606-27990-advancedshellm-a-stateful-multi-agent-llm-honeypot-for-ssh-deception.md
@@ -0,0 +1,61 @@
+# Paper: AdvancedShelLM: A Stateful Multi-Agent LLM Honeypot for SSH Deception
+
+---
+type: paper
+title: "AdvancedShelLM: A Stateful Multi-Agent LLM Honeypot for SSH Deception"
+authors: Muris Sladić, Eman Alibalić, Veronica Valeros, Carlos Catania, Sebastian Garcia
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.27990
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-26
+updated_at: 2026-06-26
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - multi-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: llm-agent, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent, multi-agent-llm
+- inferred topics: agent-evaluation, memory, multi-agent, tool-use
+- arXiv categories: cs.CR
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.27990
diff --git a/papers/items/2026-2606-28011-from-detection-to-action-using-llm-agents-for-fault-tolerant-control.md b/papers/items/2026-2606-28011-from-detection-to-action-using-llm-agents-for-fault-tolerant-control.md
new file mode 100644
index 0000000..a6f99df
--- /dev/null
+++ b/papers/items/2026-2606-28011-from-detection-to-action-using-llm-agents-for-fault-tolerant-control.md
@@ -0,0 +1,66 @@
+# Paper: From Detection to Action: Using LLM Agents for Fault-Tolerant Control
+
+---
+type: paper
+title: "From Detection to Action: Using LLM Agents for Fault-Tolerant Control"
+authors: Javal Vyas, Milapji Singh Gill, Artan Markaj, Felix Gehlhoff, Mehmet Mercangöz
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28011
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-26
+updated_at: 2026-06-26
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - planning
+  - rag
+  - tool-use
+  - workflow-agent
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - eess.SY
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 24
+collection_queries: agentic-ai, llm-agent, multi-agent-llm, planning-agent, rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, llm-agent, multi-agent-llm, planning-agent, rag-agent
+- inferred topics: agent-evaluation, agent-safety, multi-agent, planning, rag, tool-use, workflow-agent, world-model
+- arXiv categories: eess.SY, cs.LG
+- collection score: 24
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28011
diff --git a/papers/items/2026-2606-28061-toolprivacybench-benchmarking-purpose-bound-privacy-in-tool-using-llm-agents.md b/papers/items/2026-2606-28061-toolprivacybench-benchmarking-purpose-bound-privacy-in-tool-using-llm-agents.md
new file mode 100644
index 0000000..b23bd1a
--- /dev/null
+++ b/papers/items/2026-2606-28061-toolprivacybench-benchmarking-purpose-bound-privacy-in-tool-using-llm-agents.md
@@ -0,0 +1,62 @@
+# Paper: ToolPrivacyBench: Benchmarking Purpose-Bound Privacy in Tool-Using LLM Agents
+
+---
+type: paper
+title: "ToolPrivacyBench: Benchmarking Purpose-Bound Privacy in Tool-Using LLM Agents"
+authors: Shijing Hu, Liang Liu, Zhu Meng, Zhicheng Zhao
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28061
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-26
+updated_at: 2026-06-26
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 22
+collection_queries: agent-evaluation, function-calling, llm-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, function-calling, llm-agent, tool-use
+- inferred topics: agent-evaluation, rag, tool-use, workflow-agent
+- arXiv categories: cs.CR, cs.AI
+- collection score: 22
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28061
diff --git a/papers/items/2026-2606-28182-llawco-learning-laws-of-cooperation-for-modeling-embodied-multi-agent-behavior.md b/papers/items/2026-2606-28182-llawco-learning-laws-of-cooperation-for-modeling-embodied-multi-agent-behavior.md
new file mode 100644
index 0000000..27f5d91
--- /dev/null
+++ b/papers/items/2026-2606-28182-llawco-learning-laws-of-cooperation-for-modeling-embodied-multi-agent-behavior.md
@@ -0,0 +1,66 @@
+# Paper: LLawCo: Learning Laws of Cooperation for Modeling Embodied Multi-Agent Behavior
+
+---
+type: paper
+title: "LLawCo: Learning Laws of Cooperation for Modeling Embodied Multi-Agent Behavior"
+authors: Qinhong Zhou, Chuang Gan, Anoop Cherian
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28182
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-26
+updated_at: 2026-06-26
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - embodied-agent
+  - multi-agent
+  - planning
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+  - cs.CV
+  - cs.RO
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, embodied-agent, multi-agent, planning, rag, reasoning
+- arXiv categories: cs.LG, cs.AI, cs.CV, cs.RO
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28182
diff --git a/papers/items/2026-2606-28187-gbc-gradient-based-connections-for-optimizing-multi-agent-systems.md b/papers/items/2026-2606-28187-gbc-gradient-based-connections-for-optimizing-multi-agent-systems.md
new file mode 100644
index 0000000..e076ff7
--- /dev/null
+++ b/papers/items/2026-2606-28187-gbc-gradient-based-connections-for-optimizing-multi-agent-systems.md
@@ -0,0 +1,60 @@
+# Paper: GBC: Gradient-Based Connections for Optimizing Multi-Agent Systems
+
+---
+type: paper
+title: "GBC: Gradient-Based Connections for Optimizing Multi-Agent Systems"
+authors: Xiaocheng Yang, Abdulrahman Alrabah, Dilek Hakkani-Tür, Gokhan Tur
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28187
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-26
+updated_at: 2026-06-26
+status: queued
+relevance: high
+topics:
+  - multi-agent
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: multi-agent, rag, tool-use
+- arXiv categories: cs.MA
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28187
diff --git a/papers/items/2026-2606-28270-agent-native-immune-system-architecture-taxonomy-and-engineering.md b/papers/items/2026-2606-28270-agent-native-immune-system-architecture-taxonomy-and-engineering.md
new file mode 100644
index 0000000..797ce57
--- /dev/null
+++ b/papers/items/2026-2606-28270-agent-native-immune-system-architecture-taxonomy-and-engineering.md
@@ -0,0 +1,65 @@
+# Paper: Agent-Native Immune System: Architecture, Taxonomy, and Engineering
+
+---
+type: paper
+title: "Agent-Native Immune System: Architecture, Taxonomy, and Engineering"
+authors: Bo Shen, Lifeng Chang, Tianyuan Wei, Yunpeng Li, Feng Shi, Yichen Han, Peijie Gao, Shiyi Kuang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28270
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-26
+updated_at: 2026-06-26
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - multi-agent
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, agent-safety, memory, multi-agent, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI, cs.MA
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28270
diff --git a/papers/items/2026-2606-28279-agentic-hardware-design-as-repository-level-code-evolution.md b/papers/items/2026-2606-28279-agentic-hardware-design-as-repository-level-code-evolution.md
new file mode 100644
index 0000000..c17d48f
--- /dev/null
+++ b/papers/items/2026-2606-28279-agentic-hardware-design-as-repository-level-code-evolution.md
@@ -0,0 +1,60 @@
+# Paper: Agentic Hardware Design as Repository-Level Code Evolution
+
+---
+type: paper
+title: Agentic Hardware Design as Repository-Level Code Evolution
+authors: Cunxi Yu, Chenhui Deng, Nathaniel Pinckney, Brucek Khailany
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28279
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-26
+updated_at: 2026-06-26
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-evaluation, coding-agent
+- arXiv categories: cs.AR, cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28279
diff --git a/papers/items/2026-2606-28349-hmars-a-hierarchical-multi-agent-memory-system-for-long-context-reasoning.md b/papers/items/2026-2606-28349-hmars-a-hierarchical-multi-agent-memory-system-for-long-context-reasoning.md
new file mode 100644
index 0000000..c0c1906
--- /dev/null
+++ b/papers/items/2026-2606-28349-hmars-a-hierarchical-multi-agent-memory-system-for-long-context-reasoning.md
@@ -0,0 +1,64 @@
+# Paper: HMARS: A Hierarchical Multi-Agent Memory System for Long-Context Reasoning
+
+---
+type: paper
+title: "HMARS: A Hierarchical Multi-Agent Memory System for Long-Context Reasoning"
+authors: Zeju Li, Ziyang Zheng, Yizhou Zhou, Qiang Xu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28349
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-03
+updated_at: 2026-06-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - multi-agent
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.IR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, multi-agent, rag, reasoning, tool-use
+- arXiv categories: cs.IR, cs.AI
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28349
diff --git a/papers/items/2026-2606-28360-carolina-guide-a-multi-agent-rag-system-with-institutional-guardrails-for-academ.md b/papers/items/2026-2606-28360-carolina-guide-a-multi-agent-rag-system-with-institutional-guardrails-for-academ.md
new file mode 100644
index 0000000..4a66095
--- /dev/null
+++ b/papers/items/2026-2606-28360-carolina-guide-a-multi-agent-rag-system-with-institutional-guardrails-for-academ.md
@@ -0,0 +1,63 @@
+# Paper: Carolina Guide: A Multi-Agent RAG System with Institutional Guardrails for Academic Policy Assistance
+
+---
+type: paper
+title: "Carolina Guide: A Multi-Agent RAG System with Institutional Guardrails for Academic Policy Assistance"
+authors: Ben Torsion, Jun Zhou
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28360
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-11
+updated_at: 2026-06-11
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - multi-agent
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.IR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, agent-safety, computer-use, multi-agent, rag
+- arXiv categories: cs.IR, cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28360
diff --git a/papers/items/2026-2606-28374-recursive-self-evolving-agents-via-held-out-selection.md b/papers/items/2026-2606-28374-recursive-self-evolving-agents-via-held-out-selection.md
new file mode 100644
index 0000000..d3601ae
--- /dev/null
+++ b/papers/items/2026-2606-28374-recursive-self-evolving-agents-via-held-out-selection.md
@@ -0,0 +1,61 @@
+# Paper: Recursive Self-Evolving Agents via Held-Out Selection
+
+---
+type: paper
+title: Recursive Self-Evolving Agents via Held-Out Selection
+authors: Michael Nguyen, Quoc Nguyen, Paul Vuong
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28374
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-17
+updated_at: 2026-06-17
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28374
diff --git a/papers/items/2026-2606-28409-evidence-driven-llm-agent-for-c-to-synthesizable-c-conversion-and-verification.md b/papers/items/2026-2606-28409-evidence-driven-llm-agent-for-c-to-synthesizable-c-conversion-and-verification.md
new file mode 100644
index 0000000..b2264cc
--- /dev/null
+++ b/papers/items/2026-2606-28409-evidence-driven-llm-agent-for-c-to-synthesizable-c-conversion-and-verification.md
@@ -0,0 +1,63 @@
+# Paper: Evidence-Driven LLM Agent for C-to-Synthesizable-C Conversion and Verification
+
+---
+type: paper
+title: Evidence-Driven LLM Agent for C-to-Synthesizable-C Conversion and Verification
+authors: Zhe Zhao, Hongbing Lang, Zhihan Xiao, Luke Ztz Hu, John Imoleayo Adebisi, Songping Mai
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28409
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-25
+updated_at: 2026-06-25
+status: queued
+relevance: high
+topics:
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: rag, reasoning, tool-use, workflow-agent, world-model
+- arXiv categories: cs.AR, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28409
diff --git a/papers/items/2026-2606-28425-tool-use-enables-undetectable-steganography-in-multi-agent-llm-systems.md b/papers/items/2026-2606-28425-tool-use-enables-undetectable-steganography-in-multi-agent-llm-systems.md
new file mode 100644
index 0000000..9b1bdb7
--- /dev/null
+++ b/papers/items/2026-2606-28425-tool-use-enables-undetectable-steganography-in-multi-agent-llm-systems.md
@@ -0,0 +1,65 @@
+# Paper: Tool Use Enables Undetectable Steganography in Multi-Agent LLM Systems
+
+---
+type: paper
+title: Tool Use Enables Undetectable Steganography in Multi-Agent LLM Systems
+authors: Jimmy Laurence Rippin, Simon C. Marshall, David Demitri Africa, Christian Schroeder de Witt
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28425
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-25
+updated_at: 2026-06-25
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - computer-use
+  - multi-agent
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 25
+collection_queries: agentic-ai, ai-agent, autonomous-agent-llm, multi-agent-llm, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, ai-agent, autonomous-agent-llm, multi-agent-llm, tool-use
+- inferred topics: agent-evaluation, agent-safety, coding-agent, computer-use, multi-agent, rag, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 25
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28425
diff --git a/papers/items/2026-2606-28430-building-to-the-test-coding-agents-deliver-what-you-check-not-what-you-requested.md b/papers/items/2026-2606-28430-building-to-the-test-coding-agents-deliver-what-you-check-not-what-you-requested.md
new file mode 100644
index 0000000..06ab757
--- /dev/null
+++ b/papers/items/2026-2606-28430-building-to-the-test-coding-agents-deliver-what-you-check-not-what-you-requested.md
@@ -0,0 +1,60 @@
+# Paper: Building to the Test: Coding Agents Deliver What You Check, Not What You Requested
+
+---
+type: paper
+title: "Building to the Test: Coding Agents Deliver What You Check, Not What You Requested"
+authors: Yanuo Ma, Ben Kereopa-Yorke, Ben Schultz
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28430
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-26
+updated_at: 2026-06-26
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: agent-evaluation, coding-agent
+- arXiv categories: cs.SE, cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28430
diff --git a/papers/items/2026-2606-28434-swe-mem-learning-adaptive-memory-management-for-long-horizon-coding-agents.md b/papers/items/2026-2606-28434-swe-mem-learning-adaptive-memory-management-for-long-horizon-coding-agents.md
new file mode 100644
index 0000000..9886e34
--- /dev/null
+++ b/papers/items/2026-2606-28434-swe-mem-learning-adaptive-memory-management-for-long-horizon-coding-agents.md
@@ -0,0 +1,63 @@
+# Paper: SWE-MeM: Learning Adaptive Memory Management for Long-Horizon Coding Agents
+
+---
+type: paper
+title: "SWE-MeM: Learning Adaptive Memory Management for Long-Horizon Coding Agents"
+authors: Shuzheng Gao, Wenhao Zeng, Zhaojian Yu, Jianqiao Wangni, Chaozheng Wang, Kai Cai, Shilin He, Michael R. Lyu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28434
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-26
+updated_at: 2026-06-26
+status: queued
+relevance: high
+topics:
+  - coding-agent
+  - memory
+  - planning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: coding-agent, memory, planning, tool-use, workflow-agent
+- arXiv categories: cs.SE, cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28434
diff --git a/papers/items/2026-2606-28436-dockerless-environment-free-program-verifier-for-coding-agents.md b/papers/items/2026-2606-28436-dockerless-environment-free-program-verifier-for-coding-agents.md
new file mode 100644
index 0000000..09453e2
--- /dev/null
+++ b/papers/items/2026-2606-28436-dockerless-environment-free-program-verifier-for-coding-agents.md
@@ -0,0 +1,61 @@
+# Paper: Dockerless: Environment-Free Program Verifier for Coding Agents
+
+---
+type: paper
+title: "Dockerless: Environment-Free Program Verifier for Coding Agents"
+authors: Wenhao Zeng, Yuling Shi, Xiaodong Gu, Chao Hu, Chaofan Wang, Yuhao Cui, Hongting Zhou, Mengnan Qi, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28436
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-26
+updated_at: 2026-06-26
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: agent-evaluation, coding-agent, reasoning
+- arXiv categories: cs.SE, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28436
diff --git a/papers/items/2026-2606-28450-llm-agents-security-duality-a-comprehensive-survey-of-self-security-and-empowere.md b/papers/items/2026-2606-28450-llm-agents-security-duality-a-comprehensive-survey-of-self-security-and-empowere.md
new file mode 100644
index 0000000..f611051
--- /dev/null
+++ b/papers/items/2026-2606-28450-llm-agents-security-duality-a-comprehensive-survey-of-self-security-and-empowere.md
@@ -0,0 +1,61 @@
+# Paper: LLM agents security duality: a comprehensive survey of self-security and empowered cybersecurity
+
+---
+type: paper
+title: "LLM agents security duality: a comprehensive survey of self-security and empowered cybersecurity"
+authors: Yiwei Xu, Yong Zhuang, Xuanming Liu, Tian Zhang, Bowen Xiao, Xiaoyang Xu, Delong Jiang, Juan Wang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28450
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-26
+updated_at: 2026-06-26
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-safety, llm-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety, llm-agent, tool-use
+- inferred topics: agent-evaluation, agent-safety, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28450
diff --git a/papers/items/2026-2606-28456-is-lying-an-emergent-behaviour-in-llms-evidence-from-gaslighting-ai-agents-in-a-.md b/papers/items/2026-2606-28456-is-lying-an-emergent-behaviour-in-llms-evidence-from-gaslighting-ai-agents-in-a-.md
new file mode 100644
index 0000000..2642a06
--- /dev/null
+++ b/papers/items/2026-2606-28456-is-lying-an-emergent-behaviour-in-llms-evidence-from-gaslighting-ai-agents-in-a-.md
@@ -0,0 +1,61 @@
+# Paper: Is Lying an Emergent Behaviour in LLMs? Evidence from Gaslighting AI agents in a Sustainability Game
+
+---
+type: paper
+title: Is Lying an Emergent Behaviour in LLMs? Evidence from Gaslighting AI agents in a Sustainability Game
+authors: Subhendu Bhandary, Federico Carucci, Christos Charalambous, Francesca Dilisante, Ksenia Dvorkina, Anna Garbo, Jiaqi Liang, Riccardo Vasellini, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28456
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-26
+updated_at: 2026-06-26
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - memory
+  - multi-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: ai-agent, llm-agent, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent, llm-agent, multi-agent-llm
+- inferred topics: agent-safety, memory, multi-agent
+- arXiv categories: cs.MA, cs.AI
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28456
diff --git a/papers/items/2026-2606-28467-an-agentic-ai-pipeline-for-appliance-level-energy-anomaly-detection-and-llm-driv.md b/papers/items/2026-2606-28467-an-agentic-ai-pipeline-for-appliance-level-energy-anomaly-detection-and-llm-driv.md
new file mode 100644
index 0000000..b1ac996
--- /dev/null
+++ b/papers/items/2026-2606-28467-an-agentic-ai-pipeline-for-appliance-level-energy-anomaly-detection-and-llm-driv.md
@@ -0,0 +1,65 @@
+# Paper: An Agentic AI Pipeline for Appliance-Level Energy Anomaly Detection and LLM-Driven Recommendations
+
+---
+type: paper
+title: An Agentic AI Pipeline for Appliance-Level Energy Anomaly Detection and LLM-Driven Recommendations
+authors: Dihia Falouz, Aida Douaibia, Amine Bechar, Youssef Elmir, Abbes Amira, Adel Oulefki
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28467
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-26
+updated_at: 2026-06-26
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agentic-ai, rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, rag-agent
+- inferred topics: agent-evaluation, memory, rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.LG, cs.AI, cs.CL
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28467
diff --git a/papers/items/2026-2606-28480-tua-bench-a-benchmark-for-general-purpose-terminal-use-agents.md b/papers/items/2026-2606-28480-tua-bench-a-benchmark-for-general-purpose-terminal-use-agents.md
new file mode 100644
index 0000000..34afe46
--- /dev/null
+++ b/papers/items/2026-2606-28480-tua-bench-a-benchmark-for-general-purpose-terminal-use-agents.md
@@ -0,0 +1,63 @@
+# Paper: TUA-Bench: A Benchmark for General-Purpose Terminal-Use Agents
+
+---
+type: paper
+title: "TUA-Bench: A Benchmark for General-Purpose Terminal-Use Agents"
+authors: Shoufa Chen, Luyuan Wang, Xuan Yang, Zhiheng Liu, Yuren Cong, Yuanfeng Ji, Feiyan Zhou, Xiaohui Zhang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28480
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-26
+updated_at: 2026-06-26
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - computer-use
+  - reasoning
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, coding-agent, computer-use, reasoning, workflow-agent
+- arXiv categories: cs.SE, cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28480
diff --git a/papers/items/2026-2606-28570-digitizing-coaching-intelligence-an-agentic-framework-for-holistic-athlete-profi.md b/papers/items/2026-2606-28570-digitizing-coaching-intelligence-an-agentic-framework-for-holistic-athlete-profi.md
new file mode 100644
index 0000000..56b749f
--- /dev/null
+++ b/papers/items/2026-2606-28570-digitizing-coaching-intelligence-an-agentic-framework-for-holistic-athlete-profi.md
@@ -0,0 +1,64 @@
+# Paper: Digitizing Coaching Intelligence: An Agentic Framework for Holistic Athlete Profiling using VLM and RAG
+
+---
+type: paper
+title: "Digitizing Coaching Intelligence: An Agentic Framework for Holistic Athlete Profiling using VLM and RAG"
+authors: Deep Ghosal, Ishani Sen, Wazib Ansar, Amlan Chakrabarti
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28570
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-26
+updated_at: 2026-06-26
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+  - cs.AI
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: multi-agent-llm, rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm, rag-agent
+- inferred topics: agent-evaluation, multi-agent, rag, reasoning, tool-use
+- arXiv categories: cs.CV, cs.AI, cs.MA
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28570
diff --git a/papers/items/2026-2606-28666-why-trust-your-agent-empirical-security-gains-from-trism-guided-agentic-workflow.md b/papers/items/2026-2606-28666-why-trust-your-agent-empirical-security-gains-from-trism-guided-agentic-workflow.md
new file mode 100644
index 0000000..5435a3c
--- /dev/null
+++ b/papers/items/2026-2606-28666-why-trust-your-agent-empirical-security-gains-from-trism-guided-agentic-workflow.md
@@ -0,0 +1,64 @@
+# Paper: Why Trust Your Agent? Empirical Security Gains from TRiSM-Guided Agentic Workflows in Healthcare
+
+---
+type: paper
+title: Why Trust Your Agent? Empirical Security Gains from TRiSM-Guided Agentic Workflows in Healthcare
+authors: Liam Kearns
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28666
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-27
+updated_at: 2026-06-27
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agentic-ai, rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, rag-agent
+- inferred topics: agent-evaluation, agent-safety, computer-use, rag, tool-use, workflow-agent
+- arXiv categories: cs.CR, cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28666
diff --git a/papers/items/2026-2606-28679-capability-gates-are-not-authorization-confused-deputy-failures-in-llm-agent-fra.md b/papers/items/2026-2606-28679-capability-gates-are-not-authorization-confused-deputy-failures-in-llm-agent-fra.md
new file mode 100644
index 0000000..76eb9de
--- /dev/null
+++ b/papers/items/2026-2606-28679-capability-gates-are-not-authorization-confused-deputy-failures-in-llm-agent-fra.md
@@ -0,0 +1,60 @@
+# Paper: Capability Gates Are Not Authorization: Confused-Deputy Failures in LLM Agent Frameworks
+
+---
+type: paper
+title: "Capability Gates Are Not Authorization: Confused-Deputy Failures in LLM Agent Frameworks"
+authors: David Mellafe Zuvic
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28679
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-27
+updated_at: 2026-06-27
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: llm-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent, tool-use
+- inferred topics: agent-evaluation, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28679
diff --git a/papers/items/2026-2606-28692-an-ai-agent-for-treatment-reasoning-over-a-biomedical-tool-universe.md b/papers/items/2026-2606-28692-an-ai-agent-for-treatment-reasoning-over-a-biomedical-tool-universe.md
new file mode 100644
index 0000000..0059d25
--- /dev/null
+++ b/papers/items/2026-2606-28692-an-ai-agent-for-treatment-reasoning-over-a-biomedical-tool-universe.md
@@ -0,0 +1,61 @@
+# Paper: An AI agent for treatment reasoning over a biomedical tool universe
+
+---
+type: paper
+title: An AI agent for treatment reasoning over a biomedical tool universe
+authors: Shanghua Gao, Ayush Noori, Richard Zhu, Curtis Ginder, Zhenglun Kong, Xiaorui Su, Justin Kauffman, Benjamin S. Glicksberg, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28692
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-27
+updated_at: 2026-06-27
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: ai-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent, tool-use
+- inferred topics: agent-evaluation, multi-agent, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28692
diff --git a/papers/items/2026-2606-28733-agentic-abstention-do-agents-know-when-to-stop-instead-of-act.md b/papers/items/2026-2606-28733-agentic-abstention-do-agents-know-when-to-stop-instead-of-act.md
new file mode 100644
index 0000000..0978965
--- /dev/null
+++ b/papers/items/2026-2606-28733-agentic-abstention-do-agents-know-when-to-stop-instead-of-act.md
@@ -0,0 +1,60 @@
+# Paper: Agentic Abstention: Do Agents Know When to Stop Instead of Act?
+
+---
+type: paper
+title: "Agentic Abstention: Do Agents Know When to Stop Instead of Act?"
+authors: Han Luo, Bingbing Wen, Lucy Lu Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28733
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-27
+updated_at: 2026-06-27
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: agent-evaluation, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28733
diff --git a/papers/items/2026-2606-28739-agent-safety-is-action-alignment.md b/papers/items/2026-2606-28739-agent-safety-is-action-alignment.md
new file mode 100644
index 0000000..e2030ce
--- /dev/null
+++ b/papers/items/2026-2606-28739-agent-safety-is-action-alignment.md
@@ -0,0 +1,60 @@
+# Paper: Agent Safety Is Action Alignment
+
+---
+type: paper
+title: Agent Safety Is Action Alignment
+authors: Shawn Li, Yue Zhao
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28739
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-27
+updated_at: 2026-06-27
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-safety
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety
+- inferred topics: agent-evaluation, agent-safety, tool-use
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28739
diff --git a/papers/items/2026-2606-28781-hyphaedb-a-living-knowledge-topology-for-agent-first-memory.md b/papers/items/2026-2606-28781-hyphaedb-a-living-knowledge-topology-for-agent-first-memory.md
new file mode 100644
index 0000000..ead4de7
--- /dev/null
+++ b/papers/items/2026-2606-28781-hyphaedb-a-living-knowledge-topology-for-agent-first-memory.md
@@ -0,0 +1,63 @@
+# Paper: HyphaeDB: A Living Knowledge Topology for Agent-First Memory
+
+---
+type: paper
+title: "HyphaeDB: A Living Knowledge Topology for Agent-First Memory"
+authors: Krishna Halaharvi
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28781
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-27
+updated_at: 2026-06-27
+status: queued
+relevance: high
+topics:
+  - coding-agent
+  - memory
+  - multi-agent
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-memory, agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory, agentic-ai
+- inferred topics: coding-agent, memory, multi-agent, rag, tool-use
+- arXiv categories: cs.AI, cs.MA
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28781
diff --git a/papers/items/2026-2606-28791-from-determinism-to-delegation-ai-native-software-engineering-and-the-evolution-.md b/papers/items/2026-2606-28791-from-determinism-to-delegation-ai-native-software-engineering-and-the-evolution-.md
new file mode 100644
index 0000000..5175247
--- /dev/null
+++ b/papers/items/2026-2606-28791-from-determinism-to-delegation-ai-native-software-engineering-and-the-evolution-.md
@@ -0,0 +1,65 @@
+# Paper: From Determinism to Delegation: AI-Native Software Engineering and the Evolution of the Agentic Engineer
+
+---
+type: paper
+title: "From Determinism to Delegation: AI-Native Software Engineering and the Evolution of the Agentic Engineer"
+authors: Mamdouh Alenezi
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28791
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-27
+updated_at: 2026-06-27
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - memory
+  - multi-agent
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 23
+collection_queries: agentic-ai, autonomous-agent-llm, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, autonomous-agent-llm, tool-use
+- inferred topics: agent-evaluation, agent-safety, coding-agent, memory, multi-agent, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.SE
+- collection score: 23
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28791
diff --git a/papers/items/2026-2606-28839-the-contagion-tensor-a-framework-for-measuring-output-distribution-coupling-in-m.md b/papers/items/2026-2606-28839-the-contagion-tensor-a-framework-for-measuring-output-distribution-coupling-in-m.md
new file mode 100644
index 0000000..eaae3ff
--- /dev/null
+++ b/papers/items/2026-2606-28839-the-contagion-tensor-a-framework-for-measuring-output-distribution-coupling-in-m.md
@@ -0,0 +1,62 @@
+# Paper: The Contagion Tensor: A Framework for Measuring Output-Distribution Coupling in Multi-Agent LLM Systems -- and Auditing the Claims It Enables
+
+---
+type: paper
+title: "The Contagion Tensor: A Framework for Measuring Output-Distribution Coupling in Multi-Agent LLM Systems -- and Auditing the Claims It Enables"
+authors: Zewen Liu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28839
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-27
+updated_at: 2026-06-27
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - multi-agent
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, computer-use, multi-agent, tool-use, world-model
+- arXiv categories: cs.LG
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28839
diff --git a/papers/items/2026-2606-28841-lamp-lean-based-agentic-framework-with-mcp-and-proof-repair.md b/papers/items/2026-2606-28841-lamp-lean-based-agentic-framework-with-mcp-and-proof-repair.md
new file mode 100644
index 0000000..9db8405
--- /dev/null
+++ b/papers/items/2026-2606-28841-lamp-lean-based-agentic-framework-with-mcp-and-proof-repair.md
@@ -0,0 +1,64 @@
+# Paper: LAMP: Lean-based Agentic framework with MCP and Proof Repair
+
+---
+type: paper
+title: "LAMP: Lean-based Agentic framework with MCP and Proof Repair"
+authors: Santhana Srinivasan R, Maithilee Patawar
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28841
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-27
+updated_at: 2026-06-27
+status: queued
+relevance: high
+topics:
+  - coding-agent
+  - multi-agent
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LO
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: coding-agent, multi-agent, planning, reasoning, tool-use
+- arXiv categories: cs.LO, cs.AI, cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28841
diff --git a/papers/items/2026-2606-28896-a-task-driven-and-quality-assured-agent-framework-for-sar-data-generation.md b/papers/items/2026-2606-28896-a-task-driven-and-quality-assured-agent-framework-for-sar-data-generation.md
new file mode 100644
index 0000000..e4bfa97
--- /dev/null
+++ b/papers/items/2026-2606-28896-a-task-driven-and-quality-assured-agent-framework-for-sar-data-generation.md
@@ -0,0 +1,63 @@
+# Paper: A Task-Driven and Quality-Assured Agent Framework for SAR Data Generation
+
+---
+type: paper
+title: A Task-Driven and Quality-Assured Agent Framework for SAR Data Generation
+authors: Xuanting Wu, Fan Zhanga, Fei Ma, Ling Guan, Guochun Ma, Yongsheng Zhou
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28896
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-27
+updated_at: 2026-06-27
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - eess.IV
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, planning, rag, workflow-agent
+- arXiv categories: eess.IV, cs.AI, cs.LG
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28896
diff --git a/papers/items/2026-2606-28925-multi-agent-routing-as-set-valued-prediction-a-wildchat-benchmark-and-cost-aware.md b/papers/items/2026-2606-28925-multi-agent-routing-as-set-valued-prediction-a-wildchat-benchmark-and-cost-aware.md
new file mode 100644
index 0000000..27c83d5
--- /dev/null
+++ b/papers/items/2026-2606-28925-multi-agent-routing-as-set-valued-prediction-a-wildchat-benchmark-and-cost-aware.md
@@ -0,0 +1,65 @@
+# Paper: Multi-Agent Routing as Set-Valued Prediction: A WildChat Benchmark and Cost-Aware Evaluation
+
+---
+type: paper
+title: "Multi-Agent Routing as Set-Valued Prediction: A WildChat Benchmark and Cost-Aware Evaluation"
+authors: Ananto Nayan Bala, Faisal Muhammad Shah
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28925
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-27
+updated_at: 2026-06-27
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - rag
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+  - cs.IR
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 21
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, multi-agent, rag, tool-use, world-model
+- arXiv categories: cs.LG, cs.AI, cs.IR, cs.MA
+- collection score: 21
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28925
diff --git a/papers/items/2026-2606-28958-when-latent-agents-lie-kv-cache-integrity-in-multi-agent-llm-collaboration.md b/papers/items/2026-2606-28958-when-latent-agents-lie-kv-cache-integrity-in-multi-agent-llm-collaboration.md
new file mode 100644
index 0000000..0f37f1e
--- /dev/null
+++ b/papers/items/2026-2606-28958-when-latent-agents-lie-kv-cache-integrity-in-multi-agent-llm-collaboration.md
@@ -0,0 +1,61 @@
+# Paper: When Latent Agents Lie: KV-Cache Integrity in Multi-Agent LLM Collaboration
+
+---
+type: paper
+title: "When Latent Agents Lie: KV-Cache Integrity in Multi-Agent LLM Collaboration"
+authors: Luís Brito, Carlos Baquero
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.28958
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-27
+updated_at: 2026-06-27
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - memory
+  - multi-agent
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: llm-agent, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent, multi-agent-llm
+- inferred topics: agent-safety, memory, multi-agent, reasoning
+- arXiv categories: cs.MA
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.28958
diff --git a/papers/items/2026-2606-29014-customized-generative-ai-agent-for-transportation-engineering-practice-a-develop.md b/papers/items/2026-2606-29014-customized-generative-ai-agent-for-transportation-engineering-practice-a-develop.md
new file mode 100644
index 0000000..0e50bca
--- /dev/null
+++ b/papers/items/2026-2606-29014-customized-generative-ai-agent-for-transportation-engineering-practice-a-develop.md
@@ -0,0 +1,63 @@
+# Paper: Customized Generative AI Agent for Transportation Engineering Practice: A Development and Continued Pre-training Guideline
+
+---
+type: paper
+title: "Customized Generative AI Agent for Transportation Engineering Practice: A Development and Continued Pre-training Guideline"
+authors: Dianwei Chen, Yuan-Zheng Lei, Zifan Zhang, Yuchen Liu, Xianfeng Yang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29014
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-27
+updated_at: 2026-07-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - planning
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.DL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent
+- inferred topics: agent-evaluation, agent-safety, computer-use, planning, reasoning
+- arXiv categories: cs.AI, cs.DL
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29014
diff --git a/papers/items/2026-2606-29026-preventing-error-propagation-in-multi-agent-ai-through-runtime-monitoring.md b/papers/items/2026-2606-29026-preventing-error-propagation-in-multi-agent-ai-through-runtime-monitoring.md
new file mode 100644
index 0000000..0e216a7
--- /dev/null
+++ b/papers/items/2026-2606-29026-preventing-error-propagation-in-multi-agent-ai-through-runtime-monitoring.md
@@ -0,0 +1,62 @@
+# Paper: Preventing Error Propagation in Multi-Agent AI through Runtime Monitoring
+
+---
+type: paper
+title: Preventing Error Propagation in Multi-Agent AI through Runtime Monitoring
+authors: Shahnewaz Karim Sakib, Anindya Bijoy Das
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29026
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-27
+updated_at: 2026-06-27
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.ET
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-evaluation, agent-safety, multi-agent, reasoning
+- arXiv categories: cs.AI, cs.ET
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29026
diff --git a/papers/items/2026-2606-29030-memory-as-an-attack-surface-in-llm-agents-a-study-on-multiple-choice-question-an.md b/papers/items/2026-2606-29030-memory-as-an-attack-surface-in-llm-agents-a-study-on-multiple-choice-question-an.md
new file mode 100644
index 0000000..9adb5ad
--- /dev/null
+++ b/papers/items/2026-2606-29030-memory-as-an-attack-surface-in-llm-agents-a-study-on-multiple-choice-question-an.md
@@ -0,0 +1,60 @@
+# Paper: Memory as an Attack Surface in LLM Agents: A Study on Multiple-Choice Question Answering
+
+---
+type: paper
+title: "Memory as an Attack Surface in LLM Agents: A Study on Multiple-Choice Question Answering"
+authors: Shahnewaz Karim Sakib, Anindya Bijoy Das
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29030
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-27
+updated_at: 2026-06-27
+status: queued
+relevance: high
+topics:
+  - memory
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.ET
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: ai-agent, llm-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent, llm-agent, tool-use
+- inferred topics: memory, tool-use
+- arXiv categories: cs.AI, cs.ET
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29030
diff --git a/papers/items/2026-2606-29116-characterizing-large-language-model-agentic-workflows-a-study-on-n8n-ecosystem.md b/papers/items/2026-2606-29116-characterizing-large-language-model-agentic-workflows-a-study-on-n8n-ecosystem.md
new file mode 100644
index 0000000..f7b8465
--- /dev/null
+++ b/papers/items/2026-2606-29116-characterizing-large-language-model-agentic-workflows-a-study-on-n8n-ecosystem.md
@@ -0,0 +1,62 @@
+# Paper: Characterizing Large Language Model Agentic Workflows: A Study on N8n Ecosystem
+
+---
+type: paper
+title: "Characterizing Large Language Model Agentic Workflows: A Study on N8n Ecosystem"
+authors: Yutian Tang, Yuming Zhou, Huaming Chen
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29116
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-27
+updated_at: 2026-06-27
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - planning
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: agentic-ai, llm-agent, planning-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, llm-agent, planning-agent, tool-use
+- inferred topics: agent-safety, planning, rag, tool-use, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29116
diff --git a/papers/items/2026-2606-29142-agent-security-meets-regulatory-reality-a-practitioner-systematization-of-autono.md b/papers/items/2026-2606-29142-agent-security-meets-regulatory-reality-a-practitioner-systematization-of-autono.md
new file mode 100644
index 0000000..35a04a0
--- /dev/null
+++ b/papers/items/2026-2606-29142-agent-security-meets-regulatory-reality-a-practitioner-systematization-of-autono.md
@@ -0,0 +1,62 @@
+# Paper: Agent Security Meets Regulatory Reality -- A Practitioner Systematization of Autonomous-Agent Threats and Controls in Regulated Financial Systems
+
+---
+type: paper
+title: Agent Security Meets Regulatory Reality -- A Practitioner Systematization of Autonomous-Agent Threats and Controls in Regulated Financial Systems
+authors: Krishna Mohan, Guda Nagavenkata Srinivasa
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29142
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-28
+updated_at: 2026-06-28
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - computer-use
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CY
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-safety, autonomous-agent-llm, rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety, autonomous-agent-llm, rag-agent
+- inferred topics: agent-safety, computer-use, rag, tool-use
+- arXiv categories: cs.CY, cs.SE
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29142
diff --git a/papers/items/2026-2606-29178-selective-memory-retention-for-long-horizon-llm-agents.md b/papers/items/2026-2606-29178-selective-memory-retention-for-long-horizon-llm-agents.md
new file mode 100644
index 0000000..b54aec2
--- /dev/null
+++ b/papers/items/2026-2606-29178-selective-memory-retention-for-long-horizon-llm-agents.md
@@ -0,0 +1,62 @@
+# Paper: Selective Memory Retention for Long-Horizon LLM Agents
+
+---
+type: paper
+title: Selective Memory Retention for Long-Horizon LLM Agents
+authors: Pranath Reddy
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29178
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-28
+updated_at: 2026-06-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: agent-evaluation, memory, planning
+- arXiv categories: cs.AI, cs.CL, cs.LG
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29178
diff --git a/papers/items/2026-2606-29193-a-multi-dataset-benchmark-for-evaluating-llm-agents-in-microservice-failure-diag.md b/papers/items/2026-2606-29193-a-multi-dataset-benchmark-for-evaluating-llm-agents-in-microservice-failure-diag.md
new file mode 100644
index 0000000..ea0ab1e
--- /dev/null
+++ b/papers/items/2026-2606-29193-a-multi-dataset-benchmark-for-evaluating-llm-agents-in-microservice-failure-diag.md
@@ -0,0 +1,62 @@
+# Paper: A Multi-Dataset Benchmark for Evaluating LLM Agents in Microservice Failure Diagnosis
+
+---
+type: paper
+title: A Multi-Dataset Benchmark for Evaluating LLM Agents in Microservice Failure Diagnosis
+authors: Yuanhong Cai, Xiaohui Nie, Kanglin Yin, Changhua Pei, Yongqian Sun, Shenglin Zhang, Haibin Liu, Guiyang Liu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29193
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-28
+updated_at: 2026-06-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: agent-evaluation, rag, reasoning, tool-use
+- arXiv categories: cs.SE, cs.AI
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29193
diff --git a/papers/items/2026-2606-29225-policyguard-a-dialogue-grounded-sub-agent-verifier-for-policy-adherence-in-llm-a.md b/papers/items/2026-2606-29225-policyguard-a-dialogue-grounded-sub-agent-verifier-for-policy-adherence-in-llm-a.md
new file mode 100644
index 0000000..3ff0837
--- /dev/null
+++ b/papers/items/2026-2606-29225-policyguard-a-dialogue-grounded-sub-agent-verifier-for-policy-adherence-in-llm-a.md
@@ -0,0 +1,62 @@
+# Paper: PolicyGuard: A Dialogue-Grounded Sub-Agent Verifier for Policy Adherence in LLM Agents
+
+---
+type: paper
+title: "PolicyGuard: A Dialogue-Grounded Sub-Agent Verifier for Policy Adherence in LLM Agents"
+authors: Seongjae Kang, Taehyung Yu, Sung Ju Hwang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29225
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-28
+updated_at: 2026-06-28
+status: queued
+relevance: high
+topics:
+  - computer-use
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: computer-use, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI, cs.CL
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29225
diff --git a/papers/items/2026-2606-29270-minority-sentinel-when-to-overturn-majority-voting-in-multi-agent-llm-debates.md b/papers/items/2026-2606-29270-minority-sentinel-when-to-overturn-majority-voting-in-multi-agent-llm-debates.md
new file mode 100644
index 0000000..4de6894
--- /dev/null
+++ b/papers/items/2026-2606-29270-minority-sentinel-when-to-overturn-majority-voting-in-multi-agent-llm-debates.md
@@ -0,0 +1,61 @@
+# Paper: Minority Sentinel: When to Overturn Majority Voting in Multi-Agent LLM Debates
+
+---
+type: paper
+title: "Minority Sentinel: When to Overturn Majority Voting in Multi-Agent LLM Debates"
+authors: Chuan He, Zebin Chen, Zhengyi Yang, Shaobo Qiao, Mingchen Ju, Jiate Liu, Dong Wen, Guanfeng Liu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29270
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-28
+updated_at: 2026-06-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: llm-agent, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent, multi-agent-llm
+- inferred topics: agent-evaluation, agent-safety, multi-agent, reasoning
+- arXiv categories: cs.MA
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29270
diff --git a/papers/items/2026-2606-29315-hierarchical-experimentalist-agents.md b/papers/items/2026-2606-29315-hierarchical-experimentalist-agents.md
new file mode 100644
index 0000000..fa2ff30
--- /dev/null
+++ b/papers/items/2026-2606-29315-hierarchical-experimentalist-agents.md
@@ -0,0 +1,64 @@
+# Paper: Hierarchical Experimentalist Agents
+
+---
+type: paper
+title: Hierarchical Experimentalist Agents
+authors: Abhranil Chandra, Sankaran Vaidyanathan, Utsav Dhanuka, Varun Gandhi, Scott Niekum
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29315
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-28
+updated_at: 2026-06-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: agent-evaluation, planning, rag, reasoning, tool-use, world-model
+- arXiv categories: cs.AI, cs.LG
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29315
diff --git a/papers/items/2026-2606-29354-when-llms-develop-languages-symbolic-communication-for-efficient-multi-agent-rea.md b/papers/items/2026-2606-29354-when-llms-develop-languages-symbolic-communication-for-efficient-multi-agent-rea.md
new file mode 100644
index 0000000..5198728
--- /dev/null
+++ b/papers/items/2026-2606-29354-when-llms-develop-languages-symbolic-communication-for-efficient-multi-agent-rea.md
@@ -0,0 +1,61 @@
+# Paper: When LLMs Develop Languages: Symbolic Communication for Efficient Multi-Agent Reasoning
+
+---
+type: paper
+title: "When LLMs Develop Languages: Symbolic Communication for Efficient Multi-Agent Reasoning"
+authors: Zhengqi Pei, Qingming Huang, Shuhui Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29354
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-28
+updated_at: 2026-06-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.NE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: llm-agent, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent, multi-agent-llm
+- inferred topics: agent-evaluation, multi-agent, reasoning
+- arXiv categories: cs.AI, cs.NE
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29354
diff --git a/papers/items/2026-2606-29445-bridging-videoqa-and-video-guided-agentic-tasks-via-generalized-keyframe-extract.md b/papers/items/2026-2606-29445-bridging-videoqa-and-video-guided-agentic-tasks-via-generalized-keyframe-extract.md
new file mode 100644
index 0000000..e8cc255
--- /dev/null
+++ b/papers/items/2026-2606-29445-bridging-videoqa-and-video-guided-agentic-tasks-via-generalized-keyframe-extract.md
@@ -0,0 +1,62 @@
+# Paper: Bridging VideoQA and Video-Guided Agentic Tasks via Generalized Keyframe Extraction
+
+---
+type: paper
+title: Bridging VideoQA and Video-Guided Agentic Tasks via Generalized Keyframe Extraction
+authors: Sunqi Fan, Qingle Liu, Runqi Yin, Meng-Hao Guo, Shuojin Yang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29445
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-28
+updated_at: 2026-06-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, computer-use, planning, tool-use
+- arXiv categories: cs.CV, cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29445
diff --git a/papers/items/2026-2606-29495-cognitive-world-models-for-process-level-social-influence-evaluation.md b/papers/items/2026-2606-29495-cognitive-world-models-for-process-level-social-influence-evaluation.md
new file mode 100644
index 0000000..4760636
--- /dev/null
+++ b/papers/items/2026-2606-29495-cognitive-world-models-for-process-level-social-influence-evaluation.md
@@ -0,0 +1,61 @@
+# Paper: Cognitive World Models for Process-Level Social Influence Evaluation
+
+---
+type: paper
+title: Cognitive World Models for Process-Level Social Influence Evaluation
+authors: Minghui Ma, Bin Guo, Han Wang, Mengqi Chen, Jingqi Liu, Yan Liu, Zhiwen Yu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29495
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-28
+updated_at: 2026-06-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - multi-agent
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, computer-use, multi-agent, world-model
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29495
diff --git a/papers/items/2026-2606-29537-osworld2-0-benchmarking-computer-use-agents-on-long-horizon-real-world-tasks.md b/papers/items/2026-2606-29537-osworld2-0-benchmarking-computer-use-agents-on-long-horizon-real-world-tasks.md
new file mode 100644
index 0000000..d100c38
--- /dev/null
+++ b/papers/items/2026-2606-29537-osworld2-0-benchmarking-computer-use-agents-on-long-horizon-real-world-tasks.md
@@ -0,0 +1,67 @@
+# Paper: OSWorld2.0: Benchmarking Computer Use Agents on Long-Horizon Real-World Tasks
+
+---
+type: paper
+title: "OSWorld2.0: Benchmarking Computer Use Agents on Long-Horizon Real-World Tasks"
+authors: Mengqi Yuan, Zilong Zhou, Xinzhuang Xiong, Weiming Wu, Jiayang Sun, Jiamin Song, Kaiqian Cui, Bowen Wang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29537
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-28
+updated_at: 2026-06-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - computer-use
+  - memory
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 21
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, agent-safety, coding-agent, computer-use, memory, planning, rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 21
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29537
diff --git a/papers/items/2026-2606-29654-budgeted-act-or-defer-multi-agent-llm-deliberation-with-local-reliability-bounds.md b/papers/items/2026-2606-29654-budgeted-act-or-defer-multi-agent-llm-deliberation-with-local-reliability-bounds.md
new file mode 100644
index 0000000..6b6d306
--- /dev/null
+++ b/papers/items/2026-2606-29654-budgeted-act-or-defer-multi-agent-llm-deliberation-with-local-reliability-bounds.md
@@ -0,0 +1,64 @@
+# Paper: Budgeted Act-or-Defer Multi-Agent LLM Deliberation with Local Reliability Bounds
+
+---
+type: paper
+title: Budgeted Act-or-Defer Multi-Agent LLM Deliberation with Local Reliability Bounds
+authors: Mengdie Flora Wang, Haochen Xie, Guanghui Wang, Devin Zhang, Jae Oh Woo
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29654
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-28
+updated_at: 2026-06-28
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, agent-safety, multi-agent, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI, cs.MA
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29654
diff --git a/papers/items/2026-2606-29719-a-diagnostic-framework-and-multi-evaluator-audit-of-evaluator-driven-preference-.md b/papers/items/2026-2606-29719-a-diagnostic-framework-and-multi-evaluator-audit-of-evaluator-driven-preference-.md
new file mode 100644
index 0000000..8fa118e
--- /dev/null
+++ b/papers/items/2026-2606-29719-a-diagnostic-framework-and-multi-evaluator-audit-of-evaluator-driven-preference-.md
@@ -0,0 +1,60 @@
+# Paper: A Diagnostic Framework and Multi-Evaluator Audit of Evaluator-Driven Preference Dynamics in Self-Adapting LLM Agents
+
+---
+type: paper
+title: A Diagnostic Framework and Multi-Evaluator Audit of Evaluator-Driven Preference Dynamics in Self-Adapting LLM Agents
+authors: Liu Zewen
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29719
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: agent-evaluation, rag
+- arXiv categories: cs.LG, cs.CL
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29719
diff --git a/papers/items/2026-2606-29742-microagent-context-augmented-multi-agent-framework-for-automatic-microservice-de.md b/papers/items/2026-2606-29742-microagent-context-augmented-multi-agent-framework-for-automatic-microservice-de.md
new file mode 100644
index 0000000..d473242
--- /dev/null
+++ b/papers/items/2026-2606-29742-microagent-context-augmented-multi-agent-framework-for-automatic-microservice-de.md
@@ -0,0 +1,63 @@
+# Paper: MicroAgent: Context-Augmented Multi-Agent Framework for Automatic Microservice Decomposition
+
+---
+type: paper
+title: "MicroAgent: Context-Augmented Multi-Agent Framework for Automatic Microservice Decomposition"
+authors: Zishan Su, Junjie Huang, Shiwen Shan, Xingyan Chen, Hui Zeng, Yuxin Su, Yanlin Wang, Michael R. Lyu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29742
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - computer-use
+  - multi-agent
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, coding-agent, computer-use, multi-agent, rag, tool-use
+- arXiv categories: cs.SE
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29742
diff --git a/papers/items/2026-2606-29745-echo-learning-epistemically-adaptive-language-agents-with-turn-level-credit.md b/papers/items/2026-2606-29745-echo-learning-epistemically-adaptive-language-agents-with-turn-level-credit.md
new file mode 100644
index 0000000..a3aed8f
--- /dev/null
+++ b/papers/items/2026-2606-29745-echo-learning-epistemically-adaptive-language-agents-with-turn-level-credit.md
@@ -0,0 +1,61 @@
+# Paper: ECHO: Learning Epistemically Adaptive Language Agents with Turn-Level Credit
+
+---
+type: paper
+title: "ECHO: Learning Epistemically Adaptive Language Agents with Turn-Level Credit"
+authors: Abhijnan Nath, Nikhil Krishnaswamy
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29745
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, rag, reasoning, tool-use
+- arXiv categories: cs.MA
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29745
diff --git a/papers/items/2026-2606-29746-deepmed-search-an-open-source-agentic-platform-for-medical-deep-research-with-in.md b/papers/items/2026-2606-29746-deepmed-search-an-open-source-agentic-platform-for-medical-deep-research-with-in.md
new file mode 100644
index 0000000..156f7ed
--- /dev/null
+++ b/papers/items/2026-2606-29746-deepmed-search-an-open-source-agentic-platform-for-medical-deep-research-with-in.md
@@ -0,0 +1,63 @@
+# Paper: DEEPMED Search: An Open-Source Agentic Platform for Medical Deep Research with Introspective Verification
+
+---
+type: paper
+title: "DEEPMED Search: An Open-Source Agentic Platform for Medical Deep Research with Introspective Verification"
+authors: Maolin Liu, Fanyu Xu, Ruoqing Xu, Jiahang Zhang, Hao Wang, Rui Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29746
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - computer-use
+  - multi-agent
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.HC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: computer-use, multi-agent, rag, reasoning, tool-use
+- arXiv categories: cs.AI, cs.HC
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29746
diff --git a/papers/items/2026-2606-29762-do-recommendation-algorithms-work-when-users-are-llm-agents-a-case-study-on-molt.md b/papers/items/2026-2606-29762-do-recommendation-algorithms-work-when-users-are-llm-agents-a-case-study-on-molt.md
new file mode 100644
index 0000000..c39e53f
--- /dev/null
+++ b/papers/items/2026-2606-29762-do-recommendation-algorithms-work-when-users-are-llm-agents-a-case-study-on-molt.md
@@ -0,0 +1,59 @@
+# Paper: Do Recommendation Algorithms Work When Users Are LLM Agents? A Case Study on Moltbook
+
+---
+type: paper
+title: Do Recommendation Algorithms Work When Users Are LLM Agents? A Case Study on Moltbook
+authors: Daming Li, Simeng Han, Jialu Zhang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29762
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.IR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: ai-agent, llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent, llm-agent
+- inferred topics: agent-evaluation, rag
+- arXiv categories: cs.IR
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29762
diff --git a/papers/items/2026-2606-29771-clqt-a-closed-loop-cost-aware-strategy-consistent-benchmark-for-diagnostic-evalu.md b/papers/items/2026-2606-29771-clqt-a-closed-loop-cost-aware-strategy-consistent-benchmark-for-diagnostic-evalu.md
new file mode 100644
index 0000000..6c3ad95
--- /dev/null
+++ b/papers/items/2026-2606-29771-clqt-a-closed-loop-cost-aware-strategy-consistent-benchmark-for-diagnostic-evalu.md
@@ -0,0 +1,64 @@
+# Paper: CLQT: A Closed-Loop, Cost-Aware, Strategy-Consistent Benchmark for Diagnostic Evaluation of LLM Portfolio-Management Agents
+
+---
+type: paper
+title: "CLQT: A Closed-Loop, Cost-Aware, Strategy-Consistent Benchmark for Diagnostic Evaluation of LLM Portfolio-Management Agents"
+authors: Bo Qu, Mingguang Chen
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29771
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.LG
+  - q-fin.CP
+  - q-fin.PM
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: agent-evaluation, memory, reasoning, tool-use
+- arXiv categories: cs.AI, cs.LG, q-fin.CP, q-fin.PM
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29771
diff --git a/papers/items/2026-2606-29774-analytic-concept-centric-memory-for-agentic-embodied-manipulation.md b/papers/items/2026-2606-29774-analytic-concept-centric-memory-for-agentic-embodied-manipulation.md
new file mode 100644
index 0000000..3fe8066
--- /dev/null
+++ b/papers/items/2026-2606-29774-analytic-concept-centric-memory-for-agentic-embodied-manipulation.md
@@ -0,0 +1,64 @@
+# Paper: Analytic Concept-Centric Memory for Agentic Embodied Manipulation
+
+---
+type: paper
+title: Analytic Concept-Centric Memory for Agentic Embodied Manipulation
+authors: Mingyang Sun, Xiujian Liang, Jiude Wei, Qichen He, Donglin Wang, Cewu Lu, Jianhua Sun
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29774
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - embodied-agent
+  - memory
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.RO
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, embodied-agent, memory, planning, rag, reasoning, tool-use
+- arXiv categories: cs.RO
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29774
diff --git a/papers/items/2026-2606-29778-mandol-an-agglomerative-agent-memory-system-for-long-term-conversations.md b/papers/items/2026-2606-29778-mandol-an-agglomerative-agent-memory-system-for-long-term-conversations.md
new file mode 100644
index 0000000..d740c45
--- /dev/null
+++ b/papers/items/2026-2606-29778-mandol-an-agglomerative-agent-memory-system-for-long-term-conversations.md
@@ -0,0 +1,63 @@
+# Paper: Mandol: An Agglomerative Agent Memory System for Long-Term Conversations
+
+---
+type: paper
+title: "Mandol: An Agglomerative Agent Memory System for Long-Term Conversations"
+authors: Yuhan Zhang, Zhiyuan Guo, Ziheng Zeng, Wei Wang, Wentao Wu, Lijie Xu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29778
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.DB
+  - cs.AI
+  - cs.CL
+  - cs.IR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-memory, rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory, rag-agent
+- inferred topics: agent-evaluation, memory, rag
+- arXiv categories: cs.DB, cs.AI, cs.CL, cs.IR
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29778
diff --git a/papers/items/2026-2606-29788-memleak-diagnosing-information-leaks-in-multimodal-agent-memory.md b/papers/items/2026-2606-29788-memleak-diagnosing-information-leaks-in-multimodal-agent-memory.md
new file mode 100644
index 0000000..3a7ed0c
--- /dev/null
+++ b/papers/items/2026-2606-29788-memleak-diagnosing-information-leaks-in-multimodal-agent-memory.md
@@ -0,0 +1,59 @@
+# Paper: MemLeak: Diagnosing Information Leaks in Multimodal Agent Memory
+
+---
+type: paper
+title: "MemLeak: Diagnosing Information Leaks in Multimodal Agent Memory"
+authors: Kuan Wang, Chao Zhang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29788
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-memory, ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory, ai-agent
+- inferred topics: agent-evaluation, memory
+- arXiv categories: cs.LG
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29788
diff --git a/papers/items/2026-2606-29824-neural-procedural-memory-empowering-llm-agents-with-implicit-activation-steering.md b/papers/items/2026-2606-29824-neural-procedural-memory-empowering-llm-agents-with-implicit-activation-steering.md
new file mode 100644
index 0000000..80ab2bc
--- /dev/null
+++ b/papers/items/2026-2606-29824-neural-procedural-memory-empowering-llm-agents-with-implicit-activation-steering.md
@@ -0,0 +1,64 @@
+# Paper: Neural Procedural Memory: Empowering LLM Agents with Implicit Activation Steering
+
+---
+type: paper
+title: "Neural Procedural Memory: Empowering LLM Agents with Implicit Activation Steering"
+authors: Chengfeng Zhao, Yuqiao Tan, Shizhu He, Yequan Wang, Jun Zhao, Kang Liu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29824
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 24
+collection_queries: agent-evaluation, agent-memory, autonomous-agent-llm, llm-agent, rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, agent-memory, autonomous-agent-llm, llm-agent, rag-agent
+- inferred topics: agent-evaluation, computer-use, memory, rag, tool-use, workflow-agent
+- arXiv categories: cs.CL, cs.AI
+- collection score: 24
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29824
diff --git a/papers/items/2026-2606-29894-saber-math-automated-benchmark-for-information-retrieval-evaluation-in-mathemati.md b/papers/items/2026-2606-29894-saber-math-automated-benchmark-for-information-retrieval-evaluation-in-mathemati.md
new file mode 100644
index 0000000..be18556
--- /dev/null
+++ b/papers/items/2026-2606-29894-saber-math-automated-benchmark-for-information-retrieval-evaluation-in-mathemati.md
@@ -0,0 +1,62 @@
+# Paper: SABER-Math: Automated Benchmark for Information Retrieval Evaluation in Mathematics
+
+---
+type: paper
+title: "SABER-Math: Automated Benchmark for Information Retrieval Evaluation in Mathematics"
+authors: Nikolay Georgiev, Maria Drencheva, Kseniia Ibragimova, Ivo Petrov, Dimitar I. Dimitrov, Martin Vechev
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29894
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.IR
+  - cs.AI
+  - cs.CL
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-evaluation, rag
+- arXiv categories: cs.IR, cs.AI, cs.CL, cs.LG
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29894
diff --git a/papers/items/2026-2606-29914-memdelta-controlled-baselines-and-hidden-confounds-in-agent-memory-evaluation.md b/papers/items/2026-2606-29914-memdelta-controlled-baselines-and-hidden-confounds-in-agent-memory-evaluation.md
new file mode 100644
index 0000000..b8eda4c
--- /dev/null
+++ b/papers/items/2026-2606-29914-memdelta-controlled-baselines-and-hidden-confounds-in-agent-memory-evaluation.md
@@ -0,0 +1,61 @@
+# Paper: MemDelta: Controlled Baselines and Hidden Confounds in Agent Memory Evaluation
+
+---
+type: paper
+title: "MemDelta: Controlled Baselines and Hidden Confounds in Agent Memory Evaluation"
+authors: Kuan Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29914
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-memory, rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory, rag-agent
+- inferred topics: agent-evaluation, memory, rag
+- arXiv categories: cs.CL, cs.LG
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29914
diff --git a/papers/items/2026-2606-29932-saga-scene-aware-goal-evolving-agents-for-long-horizon-civrealm-strategy-plannin.md b/papers/items/2026-2606-29932-saga-scene-aware-goal-evolving-agents-for-long-horizon-civrealm-strategy-plannin.md
new file mode 100644
index 0000000..1427cb2
--- /dev/null
+++ b/papers/items/2026-2606-29932-saga-scene-aware-goal-evolving-agents-for-long-horizon-civrealm-strategy-plannin.md
@@ -0,0 +1,63 @@
+# Paper: SAGA: Scene-Aware, Goal-Evolving Agents for Long-Horizon CivRealm Strategy Planning
+
+---
+type: paper
+title: "SAGA: Scene-Aware, Goal-Evolving Agents for Long-Horizon CivRealm Strategy Planning"
+authors: Tianyu Jin, Shuo Chen, Yida Wang, Liuyu Xiang, Yingzhuo Liu, Zhiyao Jiang, Yexin Li, Zhaofeng He
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29932
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - multi-agent
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, coding-agent, multi-agent, planning, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29932
diff --git a/papers/items/2026-2606-29957-swe-together-evaluating-coding-agents-in-interactive-user-sessions.md b/papers/items/2026-2606-29957-swe-together-evaluating-coding-agents-in-interactive-user-sessions.md
new file mode 100644
index 0000000..f881d33
--- /dev/null
+++ b/papers/items/2026-2606-29957-swe-together-evaluating-coding-agents-in-interactive-user-sessions.md
@@ -0,0 +1,61 @@
+# Paper: SWE-Together: Evaluating Coding Agents in Interactive User Sessions
+
+---
+type: paper
+title: "SWE-Together: Evaluating Coding Agents in Interactive User Sessions"
+authors: Yifan Wu, Zhuokai Zhao, Songlin Li, Ho Hin Lee, Jiacheng Zhu, Shirley Wu, Tianhe Yu, Serena Li, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29957
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-evaluation, coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, coding-agent
+- inferred topics: agent-evaluation, coding-agent, tool-use
+- arXiv categories: cs.SE, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29957
diff --git a/papers/items/2026-2606-29961-duomem-towards-capable-on-device-memory-agents-via-dual-space-distillation.md b/papers/items/2026-2606-29961-duomem-towards-capable-on-device-memory-agents-via-dual-space-distillation.md
new file mode 100644
index 0000000..e50f060
--- /dev/null
+++ b/papers/items/2026-2606-29961-duomem-towards-capable-on-device-memory-agents-via-dual-space-distillation.md
@@ -0,0 +1,61 @@
+# Paper: DuoMem: Towards Capable On-Device Memory Agents via Dual-Space Distillation
+
+---
+type: paper
+title: "DuoMem: Towards Capable On-Device Memory Agents via Dual-Space Distillation"
+authors: Peyman Hosseini, Ondrej Bohdal, Ahmed Alajrami, Andrea Maracani, Ignacio Castro, Matthew Purver, Mete Ozay, Savas Ozkan, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.29961
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - embodied-agent
+  - memory
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, embodied-agent, memory
+- arXiv categories: cs.LG, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.29961
diff --git a/papers/items/2026-2606-30005-llm-agents-are-latent-context-managers-eliciting-self-managed-context-via-a-prop.md b/papers/items/2026-2606-30005-llm-agents-are-latent-context-managers-eliciting-self-managed-context-via-a-prop.md
new file mode 100644
index 0000000..fe76151
--- /dev/null
+++ b/papers/items/2026-2606-30005-llm-agents-are-latent-context-managers-eliciting-self-managed-context-via-a-prop.md
@@ -0,0 +1,60 @@
+# Paper: LLM Agents Are Latent Context Managers: Eliciting Self-Managed Context via a Proprioceptive Dashboard
+
+---
+type: paper
+title: "LLM Agents Are Latent Context Managers: Eliciting Self-Managed Context via a Proprioceptive Dashboard"
+authors: Binyan Xu, Haitao Li, Kehuan Zhang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.30005
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-07-05
+status: queued
+relevance: high
+topics:
+  - memory
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: memory, planning, tool-use
+- arXiv categories: cs.CL
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.30005
diff --git a/papers/items/2026-2606-30111-automating-the-design-of-embodied-agent-architectures.md b/papers/items/2026-2606-30111-automating-the-design-of-embodied-agent-architectures.md
new file mode 100644
index 0000000..10b3b39
--- /dev/null
+++ b/papers/items/2026-2606-30111-automating-the-design-of-embodied-agent-architectures.md
@@ -0,0 +1,66 @@
+# Paper: Automating the Design of Embodied Agent Architectures
+
+---
+type: paper
+title: Automating the Design of Embodied Agent Architectures
+authors: Jian Zhou, Sihao Lin, Jin Li, Shuai Fu, Gengze Zhou, Qi Wu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.30111
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-07-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - embodied-agent
+  - memory
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.RO
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: agent-evaluation, coding-agent, embodied-agent, memory, planning, reasoning, tool-use
+- arXiv categories: cs.RO, cs.AI, cs.LG
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.30111
diff --git a/papers/items/2026-2606-30119-on-the-internet-nobody-knows-you-re-an-llm-bot-unmasking-web-agents-with-multi-l.md b/papers/items/2026-2606-30119-on-the-internet-nobody-knows-you-re-an-llm-bot-unmasking-web-agents-with-multi-l.md
new file mode 100644
index 0000000..cb568f7
--- /dev/null
+++ b/papers/items/2026-2606-30119-on-the-internet-nobody-knows-you-re-an-llm-bot-unmasking-web-agents-with-multi-l.md
@@ -0,0 +1,63 @@
+# Paper: On the Internet, Nobody Knows You're an LLM Bot: Unmasking Web Agents with Multi-Layer Fingerprinting
+
+---
+type: paper
+title: "On the Internet, Nobody Knows You're an LLM Bot: Unmasking Web Agents with Multi-Layer Fingerprinting"
+authors: Iliana Fayolle, Sihem Bouhenniche, Samuel Pélissier, Pierre Laperdrix, Clémentine Maurice, Walter Rudametkin
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.30119
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - embodied-agent
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, computer-use, embodied-agent, rag, tool-use, workflow-agent
+- arXiv categories: cs.CR
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.30119
diff --git a/papers/items/2026-2606-30185-dynamo-dynamic-skill-tool-evolution-for-vision-language-agents.md b/papers/items/2026-2606-30185-dynamo-dynamic-skill-tool-evolution-for-vision-language-agents.md
new file mode 100644
index 0000000..93eac02
--- /dev/null
+++ b/papers/items/2026-2606-30185-dynamo-dynamic-skill-tool-evolution-for-vision-language-agents.md
@@ -0,0 +1,60 @@
+# Paper: Dynamo: Dynamic Skill-Tool Evolution for Vision-Language Agents
+
+---
+type: paper
+title: "Dynamo: Dynamic Skill-Tool Evolution for Vision-Language Agents"
+authors: Yutao Sun, Yanting Miao, Hao-Xuan Ma, Mengyu Zhou, Mingshuai Chen, Tiancheng Zhao, Dexin Wang, Lei Lv, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.30185
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.30185
diff --git a/papers/items/2026-2606-30251-taco-tool-augmented-credit-optimization-for-agentic-tool-use.md b/papers/items/2026-2606-30251-taco-tool-augmented-credit-optimization-for-agentic-tool-use.md
new file mode 100644
index 0000000..418a128
--- /dev/null
+++ b/papers/items/2026-2606-30251-taco-tool-augmented-credit-optimization-for-agentic-tool-use.md
@@ -0,0 +1,61 @@
+# Paper: TACO: Tool-Augmented Credit Optimization for Agentic Tool Use
+
+---
+type: paper
+title: "TACO: Tool-Augmented Credit Optimization for Agentic Tool Use"
+authors: Mingkuan Feng, Jinyang Wu, Hao Gu, Fangrui Lv, Ruihan Jin, Chuyuan Zhang, Zhengqi Wen, Jianhua Tao
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.30251
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, computer-use, reasoning, tool-use
+- arXiv categories: cs.MA
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.30251
diff --git a/papers/items/2026-2606-30259-multi-agentic-system-leveraging-open-source-llms-to-mitigate-disinformation-thre.md b/papers/items/2026-2606-30259-multi-agentic-system-leveraging-open-source-llms-to-mitigate-disinformation-thre.md
new file mode 100644
index 0000000..130a4ef
--- /dev/null
+++ b/papers/items/2026-2606-30259-multi-agentic-system-leveraging-open-source-llms-to-mitigate-disinformation-thre.md
@@ -0,0 +1,60 @@
+# Paper: Multi-Agentic System Leveraging Open-Source LLMs to Mitigate Disinformation Threats
+
+---
+type: paper
+title: Multi-Agentic System Leveraging Open-Source LLMs to Mitigate Disinformation Threats
+authors: Sebastian Kula, Martin Tamajka
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.30259
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, multi-agent, rag
+- arXiv categories: cs.CL
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.30259
diff --git a/papers/items/2026-2606-30266-towards-continual-motion-language-agents-lora-variants-for-incremental-motion-un.md b/papers/items/2026-2606-30266-towards-continual-motion-language-agents-lora-variants-for-incremental-motion-un.md
new file mode 100644
index 0000000..e6651b8
--- /dev/null
+++ b/papers/items/2026-2606-30266-towards-continual-motion-language-agents-lora-variants-for-incremental-motion-un.md
@@ -0,0 +1,59 @@
+# Paper: Towards Continual Motion-Language Agents: LoRA Variants for Incremental Motion Understanding and Generation
+
+---
+type: paper
+title: "Towards Continual Motion-Language Agents: LoRA Variants for Incremental Motion Understanding and Generation"
+authors: Bertram Taetz, Hugo Albuquerque Cosme da Silva, Gabriele Bleser-Taetz
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.30266
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: autonomous-agent-llm, language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm, language-agent
+- inferred topics: agent-evaluation
+- arXiv categories: cs.LG, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.30266
diff --git a/papers/items/2026-2606-30294-rehearsed-multi-agent-live-product-demonstrations-with-real-time-voice-question-.md b/papers/items/2026-2606-30294-rehearsed-multi-agent-live-product-demonstrations-with-real-time-voice-question-.md
new file mode 100644
index 0000000..1a3affc
--- /dev/null
+++ b/papers/items/2026-2606-30294-rehearsed-multi-agent-live-product-demonstrations-with-real-time-voice-question-.md
@@ -0,0 +1,66 @@
+# Paper: Rehearsed Multi-Agent Live Product Demonstrations with Real-Time Voice Question Answering
+
+---
+type: paper
+title: Rehearsed Multi-Agent Live Product Demonstrations with Real-Time Voice Question Answering
+authors: Rahul Khedar, Mayank Malhotra, Avinash Karn, Mouli V, Prakhar Mehrotra
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.30294
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - computer-use
+  - multi-agent
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.HC
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, coding-agent, computer-use, multi-agent, rag, tool-use, workflow-agent
+- arXiv categories: cs.AI, cs.HC, cs.SE
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.30294
diff --git a/papers/items/2026-2606-30383-whose-side-is-your-agent-on-multi-party-principal-loyalty-in-llm-agents.md b/papers/items/2026-2606-30383-whose-side-is-your-agent-on-multi-party-principal-loyalty-in-llm-agents.md
new file mode 100644
index 0000000..33cb107
--- /dev/null
+++ b/papers/items/2026-2606-30383-whose-side-is-your-agent-on-multi-party-principal-loyalty-in-llm-agents.md
@@ -0,0 +1,60 @@
+# Paper: Whose Side Is Your Agent On? Multi-Party Principal Loyalty in LLM Agents
+
+---
+type: paper
+title: Whose Side Is Your Agent On? Multi-Party Principal Loyalty in LLM Agents
+authors: Bojie Li, Noah Shi
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.30383
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: agent-evaluation, agent-safety, tool-use
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.30383
diff --git a/papers/items/2026-2606-30454-collective-cooperation-without-individual-fidelity-in-llm-agents.md b/papers/items/2026-2606-30454-collective-cooperation-without-individual-fidelity-in-llm-agents.md
new file mode 100644
index 0000000..3a6cee2
--- /dev/null
+++ b/papers/items/2026-2606-30454-collective-cooperation-without-individual-fidelity-in-llm-agents.md
@@ -0,0 +1,61 @@
+# Paper: Collective cooperation without individual fidelity in LLM agents
+
+---
+type: paper
+title: Collective cooperation without individual fidelity in LLM agents
+authors: Henrique Ferraz de Arruda, Carlos Gracia Lázaro, Alberto Aleta, Yamir Moreno
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.30454
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - physics.soc-ph
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: agent-evaluation, tool-use, world-model
+- arXiv categories: physics.soc-ph, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.30454
diff --git a/papers/items/2026-2606-30524-the-illusion-of-agentic-complexity-in-readme-md-generation-evaluating-single-age.md b/papers/items/2026-2606-30524-the-illusion-of-agentic-complexity-in-readme-md-generation-evaluating-single-age.md
new file mode 100644
index 0000000..81ba607
--- /dev/null
+++ b/papers/items/2026-2606-30524-the-illusion-of-agentic-complexity-in-readme-md-generation-evaluating-single-age.md
@@ -0,0 +1,63 @@
+# Paper: The Illusion of Agentic Complexity in README.md Generation: Evaluating Single-Agent vs. Multi-Agent RAG Systems
+
+---
+type: paper
+title: "The Illusion of Agentic Complexity in README.md Generation: Evaluating Single-Agent vs. Multi-Agent RAG Systems"
+authors: Abu Saleh, Tesfay Welegebreal Tesfay, Phuong T. Nguyen, Juri Di Rocco, Muhammad Umar Zeshan, Davide Di Ruscio
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.30524
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - computer-use
+  - multi-agent
+  - planning
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 21
+collection_queries: multi-agent-llm, rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm, rag-agent
+- inferred topics: agent-evaluation, coding-agent, computer-use, multi-agent, planning, rag
+- arXiv categories: cs.SE
+- collection score: 21
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.30524
diff --git a/papers/items/2026-2606-30546-mas-lab-a-specification-driven-validation-framework-for-reliable-multi-agent-sys.md b/papers/items/2026-2606-30546-mas-lab-a-specification-driven-validation-framework-for-reliable-multi-agent-sys.md
new file mode 100644
index 0000000..bb7c8a5
--- /dev/null
+++ b/papers/items/2026-2606-30546-mas-lab-a-specification-driven-validation-framework-for-reliable-multi-agent-sys.md
@@ -0,0 +1,62 @@
+# Paper: MAS-Lab: A Specification-Driven Validation Framework for Reliable Multi-Agent Systems
+
+---
+type: paper
+title: "MAS-Lab: A Specification-Driven Validation Framework for Reliable Multi-Agent Systems"
+authors: Jordan Augé, Giovanna Carofiglio, Giulio Grassi, Jacques Samain
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.30546
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - multi-agent
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, memory, multi-agent, tool-use, workflow-agent
+- arXiv categories: cs.MA
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.30546
diff --git a/papers/items/2026-2606-30555-linguistic-firewall-geometry-as-defense-in-multi-agent-systems-routing.md b/papers/items/2026-2606-30555-linguistic-firewall-geometry-as-defense-in-multi-agent-systems-routing.md
new file mode 100644
index 0000000..e822323
--- /dev/null
+++ b/papers/items/2026-2606-30555-linguistic-firewall-geometry-as-defense-in-multi-agent-systems-routing.md
@@ -0,0 +1,64 @@
+# Paper: Linguistic Firewall: Geometry as Defense in Multi-Agent Systems Routing
+
+---
+type: paper
+title: "Linguistic Firewall: Geometry as Defense in Multi-Agent Systems Routing"
+authors: Dvir Alsheich, Adar Peleg, Ben Hagag, Rom Himelstein, Amit Levi, Avi Mendelson
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.30555
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-07-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - multi-agent
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, agent-safety, computer-use, multi-agent, tool-use, workflow-agent
+- arXiv categories: cs.AI, cs.MA
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.30555
diff --git a/papers/items/2026-2606-30560-tracelab-characterizing-coding-agent-workloads-for-llm-serving.md b/papers/items/2026-2606-30560-tracelab-characterizing-coding-agent-workloads-for-llm-serving.md
new file mode 100644
index 0000000..d9f1ee0
--- /dev/null
+++ b/papers/items/2026-2606-30560-tracelab-characterizing-coding-agent-workloads-for-llm-serving.md
@@ -0,0 +1,62 @@
+# Paper: TraceLab: Characterizing Coding Agent Workloads for LLM Serving
+
+---
+type: paper
+title: "TraceLab: Characterizing Coding Agent Workloads for LLM Serving"
+authors: Kan Zhu, Mathew Jacob, Chenxi Ma, Yi Pan, Stephanie Wang, Arvind Krishnamurthy, Baris Kasikci
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.30560
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+  - cs.PF
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: agent-evaluation, coding-agent, tool-use
+- arXiv categories: cs.LG, cs.AI, cs.PF
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.30560
diff --git a/papers/items/2026-2606-30566-forensic-trajectory-signatures-for-agent-memory-poisoning-detection.md b/papers/items/2026-2606-30566-forensic-trajectory-signatures-for-agent-memory-poisoning-detection.md
new file mode 100644
index 0000000..d8e9a61
--- /dev/null
+++ b/papers/items/2026-2606-30566-forensic-trajectory-signatures-for-agent-memory-poisoning-detection.md
@@ -0,0 +1,63 @@
+# Paper: Forensic Trajectory Signatures for Agent Memory Poisoning Detection
+
+---
+type: paper
+title: Forensic Trajectory Signatures for Agent Memory Poisoning Detection
+authors: Jun Wen Leong
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.30566
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-memory, llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory, llm-agent
+- inferred topics: agent-evaluation, computer-use, memory, rag, tool-use
+- arXiv categories: cs.CR, cs.LG
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.30566
diff --git a/papers/items/2026-2606-30573-swe-interact-reimagining-swe-benchmarks-as-user-driven-long-horizon-coding-sessi.md b/papers/items/2026-2606-30573-swe-interact-reimagining-swe-benchmarks-as-user-driven-long-horizon-coding-sessi.md
new file mode 100644
index 0000000..73fa75a
--- /dev/null
+++ b/papers/items/2026-2606-30573-swe-interact-reimagining-swe-benchmarks-as-user-driven-long-horizon-coding-sessi.md
@@ -0,0 +1,63 @@
+# Paper: SWE-INTERACT: Reimagining SWE Benchmarks as User-Driven Long-Horizon Coding Sessions
+
+---
+type: paper
+title: "SWE-INTERACT: Reimagining SWE Benchmarks as User-Driven Long-Horizon Coding Sessions"
+authors: Mohit Raghavendra, Anisha Gunjal, Aakash Sabharwal, Yunzhong He
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.30573
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - computer-use
+  - planning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: agent-evaluation, coding-agent, computer-use, planning, tool-use, workflow-agent
+- arXiv categories: cs.LG
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.30573
diff --git a/papers/items/2026-2606-30602-mesa-prioritizing-vulnerable-communication-channels-for-securing-multi-agent-sys.md b/papers/items/2026-2606-30602-mesa-prioritizing-vulnerable-communication-channels-for-securing-multi-agent-sys.md
new file mode 100644
index 0000000..1d93896
--- /dev/null
+++ b/papers/items/2026-2606-30602-mesa-prioritizing-vulnerable-communication-channels-for-securing-multi-agent-sys.md
@@ -0,0 +1,62 @@
+# Paper: MESA: Prioritizing Vulnerable Communication Channels for Securing Multi-Agent Systems
+
+---
+type: paper
+title: "MESA: Prioritizing Vulnerable Communication Channels for Securing Multi-Agent Systems"
+authors: Kunyang Li, Kyle Domico, Jonathan Gregory, Patrick McDaniel
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.30602
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, agent-safety, multi-agent, workflow-agent
+- arXiv categories: cs.CR, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.30602
diff --git a/papers/items/2026-2606-30616-scaling-the-horizon-not-the-parameters-reaching-trillion-parameter-performance-w.md b/papers/items/2026-2606-30616-scaling-the-horizon-not-the-parameters-reaching-trillion-parameter-performance-w.md
new file mode 100644
index 0000000..3b698a7
--- /dev/null
+++ b/papers/items/2026-2606-30616-scaling-the-horizon-not-the-parameters-reaching-trillion-parameter-performance-w.md
@@ -0,0 +1,63 @@
+# Paper: Scaling the Horizon, Not the Parameters: Reaching Trillion-Parameter Performance with a 35B Agent
+
+---
+type: paper
+title: "Scaling the Horizon, Not the Parameters: Reaching Trillion-Parameter Performance with a 35B Agent"
+authors: Lei Bai, Zongsheng Cao, Yang Chen, Zhiyao Cui, Shangheng Du, Yue Fan, Shiyang Feng, Zijie Guo, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.30616
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, agent-safety, planning, rag, reasoning, tool-use
+- arXiv categories: cs.CL
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.30616
diff --git a/papers/items/2026-2606-30639-self-evolving-world-models-for-llm-agent-planning.md b/papers/items/2026-2606-30639-self-evolving-world-models-for-llm-agent-planning.md
new file mode 100644
index 0000000..72cf0d3
--- /dev/null
+++ b/papers/items/2026-2606-30639-self-evolving-world-models-for-llm-agent-planning.md
@@ -0,0 +1,65 @@
+# Paper: Self-Evolving World Models for LLM Agent Planning
+
+---
+type: paper
+title: Self-Evolving World Models for LLM Agent Planning
+authors: Xuan Zhang, Wenxuan Zhang, See-Kiong Ng, Yang Deng
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.30639
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: llm-agent, planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent, planning-agent
+- inferred topics: agent-evaluation, memory, planning, rag, reasoning, tool-use, world-model
+- arXiv categories: cs.AI, cs.CL
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.30639
diff --git a/papers/items/2026-2606-30697-lumos-a-semantic-operating-system-layer-for-accessibility-grounded-ai-agents.md b/papers/items/2026-2606-30697-lumos-a-semantic-operating-system-layer-for-accessibility-grounded-ai-agents.md
new file mode 100644
index 0000000..87e0f20
--- /dev/null
+++ b/papers/items/2026-2606-30697-lumos-a-semantic-operating-system-layer-for-accessibility-grounded-ai-agents.md
@@ -0,0 +1,63 @@
+# Paper: LUMOS: A Semantic Operating-System Layer for Accessibility-Grounded AI Agents
+
+---
+type: paper
+title: "LUMOS: A Semantic Operating-System Layer for Accessibility-Grounded AI Agents"
+authors: Yogeswar Reddy Thota
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.30697
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - computer-use
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.OS
+  - cs.AI
+  - cs.CV
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: ai-agent, web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent, web-gui-agent
+- inferred topics: computer-use, rag, tool-use, workflow-agent
+- arXiv categories: cs.OS, cs.AI, cs.CV
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.30697
diff --git a/papers/items/2026-2606-30755-understanding-and-evaluating-claw-like-agent-security-through-a-computer-systems.md b/papers/items/2026-2606-30755-understanding-and-evaluating-claw-like-agent-security-through-a-computer-systems.md
new file mode 100644
index 0000000..70b7a53
--- /dev/null
+++ b/papers/items/2026-2606-30755-understanding-and-evaluating-claw-like-agent-security-through-a-computer-systems.md
@@ -0,0 +1,61 @@
+# Paper: Understanding and Evaluating Claw-like Agent Security Through a Computer-Systems Lens
+
+---
+type: paper
+title: Understanding and Evaluating Claw-like Agent Security Through a Computer-Systems Lens
+authors: Peizhi Niu, Wenjie Qu, Shangding Gu, Tianneng Shi, Yuankai Li, Ahmad Tawaha, Hend Alzahrani, Vincent Siu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.30755
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-safety, ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety, ai-agent
+- inferred topics: agent-evaluation, agent-safety, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.30755
diff --git a/papers/items/2026-2606-30840-contrastive-reflection-for-iterative-prompt-optimization.md b/papers/items/2026-2606-30840-contrastive-reflection-for-iterative-prompt-optimization.md
new file mode 100644
index 0000000..85ae508
--- /dev/null
+++ b/papers/items/2026-2606-30840-contrastive-reflection-for-iterative-prompt-optimization.md
@@ -0,0 +1,62 @@
+# Paper: Contrastive Reflection for Iterative Prompt Optimization
+
+---
+type: paper
+title: Contrastive Reflection for Iterative Prompt Optimization
+authors: Derek Koh, Jinghui Mo, Benjamin H. Le, Jiening Zhan, Baofen Zheng, Kevin Bevis, Nathaniel C. Owen, Lauren Elizabeth Charney, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.30840
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - rag
+  - reasoning
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: ai-agent, llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent, llm-agent
+- inferred topics: agent-evaluation, computer-use, rag, reasoning, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.30840
diff --git a/papers/items/2026-2606-30877-a-systematic-approach-to-multi-agent-ai-from-advanced-regulatory-control-theory-.md b/papers/items/2026-2606-30877-a-systematic-approach-to-multi-agent-ai-from-advanced-regulatory-control-theory-.md
new file mode 100644
index 0000000..202ac3d
--- /dev/null
+++ b/papers/items/2026-2606-30877-a-systematic-approach-to-multi-agent-ai-from-advanced-regulatory-control-theory-.md
@@ -0,0 +1,62 @@
+# Paper: A Systematic Approach to Multi-Agent AI from Advanced Regulatory Control Theory: Safe and Auditable LLM Operator Agents for Process Control
+
+---
+type: paper
+title: "A Systematic Approach to Multi-Agent AI from Advanced Regulatory Control Theory: Safe and Auditable LLM Operator Agents for Process Control"
+authors: Idelfonso B. R. Nogueira, Sigurd Skogestad
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.30877
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - eess.SY
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agentic-ai, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, multi-agent-llm
+- inferred topics: agent-evaluation, agent-safety, multi-agent, tool-use
+- arXiv categories: eess.SY, cs.LG
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.30877
diff --git a/papers/items/2026-2606-30887-training-therapeutic-judges-and-multi-agent-systems-for-human-aligned-mental-hea.md b/papers/items/2026-2606-30887-training-therapeutic-judges-and-multi-agent-systems-for-human-aligned-mental-hea.md
new file mode 100644
index 0000000..cd20a93
--- /dev/null
+++ b/papers/items/2026-2606-30887-training-therapeutic-judges-and-multi-agent-systems-for-human-aligned-mental-hea.md
@@ -0,0 +1,63 @@
+# Paper: Training Therapeutic Judges and Multi-Agent Systems for Human-Aligned Mental Health Support
+
+---
+type: paper
+title: Training Therapeutic Judges and Multi-Agent Systems for Human-Aligned Mental Health Support
+authors: Mizanur Rahman, Abeer Badawi, Elahe Rahimi, Laleh Seyyed-Kalantari, Frank Rudzicz, Enamul Hoque, Elham Dolatabadi
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.30887
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, agent-safety, multi-agent, tool-use
+- arXiv categories: cs.CL, cs.AI, cs.MA
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.30887
diff --git a/papers/items/2026-2606-30906-investigating-multi-agent-deliberation-in-law.md b/papers/items/2026-2606-30906-investigating-multi-agent-deliberation-in-law.md
new file mode 100644
index 0000000..9bd1e5b
--- /dev/null
+++ b/papers/items/2026-2606-30906-investigating-multi-agent-deliberation-in-law.md
@@ -0,0 +1,61 @@
+# Paper: Investigating Multi-Agent Deliberation in Law
+
+---
+type: paper
+title: Investigating Multi-Agent Deliberation in Law
+authors: Cor Steging, Ludi van Leeuwen, Tadeusz Zbiegień
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.30906
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: agentic-ai, ai-agent, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, ai-agent, multi-agent-llm
+- inferred topics: agent-evaluation, multi-agent, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.30906
diff --git a/papers/items/2026-2606-30949-agrefactor-self-evolving-agentic-workflow-for-hls-compatibility-and-performance.md b/papers/items/2026-2606-30949-agrefactor-self-evolving-agentic-workflow-for-hls-compatibility-and-performance.md
new file mode 100644
index 0000000..3a2c28b
--- /dev/null
+++ b/papers/items/2026-2606-30949-agrefactor-self-evolving-agentic-workflow-for-hls-compatibility-and-performance.md
@@ -0,0 +1,64 @@
+# Paper: AgRefactor: Self-Evolving Agentic Workflow for HLS Compatibility and Performance
+
+---
+type: paper
+title: "AgRefactor: Self-Evolving Agentic Workflow for HLS Compatibility and Performance"
+authors: Yang Zou, Zijian Ding, Yizhou Sun, Jason Cong
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.30949
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - multi-agent
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.AR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: agentic-ai, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, multi-agent-llm
+- inferred topics: agent-evaluation, memory, multi-agent, rag, tool-use, workflow-agent
+- arXiv categories: cs.AI, cs.AR
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.30949
diff --git a/papers/items/2026-2606-30970-behavioral-governance-for-autonomous-ai-agents-the-agentbound-framework.md b/papers/items/2026-2606-30970-behavioral-governance-for-autonomous-ai-agents-the-agentbound-framework.md
new file mode 100644
index 0000000..e009f87
--- /dev/null
+++ b/papers/items/2026-2606-30970-behavioral-governance-for-autonomous-ai-agents-the-agentbound-framework.md
@@ -0,0 +1,61 @@
+# Paper: Behavioral Governance for Autonomous AI Agents: The AgentBound Framework
+
+---
+type: paper
+title: "Behavioral Governance for Autonomous AI Agents: The AgentBound Framework"
+authors: Anuj Kaul, Qianlong Lan, Pranay Gupta
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.30970
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent
+- inferred topics: agent-evaluation, agent-safety, tool-use, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.30970
diff --git a/papers/items/2026-2606-30986-the-organizational-behavior-of-agentic-ai-collective-intelligence-in-human-agent.md b/papers/items/2026-2606-30986-the-organizational-behavior-of-agentic-ai-collective-intelligence-in-human-agent.md
new file mode 100644
index 0000000..72503bf
--- /dev/null
+++ b/papers/items/2026-2606-30986-the-organizational-behavior-of-agentic-ai-collective-intelligence-in-human-agent.md
@@ -0,0 +1,65 @@
+# Paper: The Organizational Behavior of Agentic AI: Collective Intelligence in Human-Agent Workflows
+
+---
+type: paper
+title: "The Organizational Behavior of Agentic AI: Collective Intelligence in Human-Agent Workflows"
+authors: Canhui Liu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.30986
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - memory
+  - planning
+  - tool-use
+  - workflow-agent
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CY
+  - cs.HC
+  - cs.MA
+  - econ.GN
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agentic-ai, llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, llm-agent
+- inferred topics: memory, planning, tool-use, workflow-agent, world-model
+- arXiv categories: cs.CY, cs.HC, cs.MA, econ.GN
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.30986
diff --git a/papers/items/2026-2606-31046-openlife-toward-open-world-artificial-life-with-autonomous-llm-agents.md b/papers/items/2026-2606-31046-openlife-toward-open-world-artificial-life-with-autonomous-llm-agents.md
new file mode 100644
index 0000000..7d427a8
--- /dev/null
+++ b/papers/items/2026-2606-31046-openlife-toward-open-world-artificial-life-with-autonomous-llm-agents.md
@@ -0,0 +1,60 @@
+# Paper: OpenLife: Toward Open-World Artificial Life with Autonomous LLM Agents
+
+---
+type: paper
+title: "OpenLife: Toward Open-World Artificial Life with Autonomous LLM Agents"
+authors: Atsushi Masumori, Itsuki Doi, Norihiro Maruyama, Ryosuke Takata, Takashi Ikegami
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.31046
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: llm-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent, tool-use
+- inferred topics: agent-evaluation, memory, tool-use
+- arXiv categories: cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.31046
diff --git a/papers/items/2026-2606-31073-multiuav-plat-an-llm-oriented-platform-benchmark-and-framework-for-multi-uav-col.md b/papers/items/2026-2606-31073-multiuav-plat-an-llm-oriented-platform-benchmark-and-framework-for-multi-uav-col.md
new file mode 100644
index 0000000..6038ccc
--- /dev/null
+++ b/papers/items/2026-2606-31073-multiuav-plat-an-llm-oriented-platform-benchmark-and-framework-for-multi-uav-col.md
@@ -0,0 +1,67 @@
+# Paper: MultiUAV-Plat: An LLM-Oriented Platform, Benchmark and Framework for Multi-UAV Collaborative Task Planning
+
+---
+type: paper
+title: "MultiUAV-Plat: An LLM-Oriented Platform, Benchmark and Framework for Multi-UAV Collaborative Task Planning"
+authors: Sheng Zhang, Qinglin Li, Yuechao Zang, Xueqin Huang, Yijia Fu, Cheng Zhu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.31073
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - embodied-agent
+  - memory
+  - multi-agent
+  - planning
+  - rag
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.MA
+  - cs.RO
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: agent-evaluation, llm-agent, planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, llm-agent, planning-agent
+- inferred topics: agent-evaluation, embodied-agent, memory, multi-agent, planning, rag, tool-use, world-model
+- arXiv categories: cs.AI, cs.MA, cs.RO
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.31073
diff --git a/papers/items/2026-2606-31085-ddiagents-mechanism-conditioned-context-flow-for-drug-drug-interaction-predictio.md b/papers/items/2026-2606-31085-ddiagents-mechanism-conditioned-context-flow-for-drug-drug-interaction-predictio.md
new file mode 100644
index 0000000..292cb6d
--- /dev/null
+++ b/papers/items/2026-2606-31085-ddiagents-mechanism-conditioned-context-flow-for-drug-drug-interaction-predictio.md
@@ -0,0 +1,63 @@
+# Paper: DDIAgents: Mechanism-Conditioned Context Flow for Drug-Drug Interaction Prediction
+
+---
+type: paper
+title: "DDIAgents: Mechanism-Conditioned Context Flow for Drug-Drug Interaction Prediction"
+authors: Zhenqian Shen, Yu Liu, Xiaoyi Fu, Quanming Yao
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.31085
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, agent-safety, multi-agent, planning, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.31085
diff --git a/papers/items/2026-2606-31134-beyond-the-library-an-agentic-framework-for-autoformalizing-research-mathematics.md b/papers/items/2026-2606-31134-beyond-the-library-an-agentic-framework-for-autoformalizing-research-mathematics.md
new file mode 100644
index 0000000..8f11d4a
--- /dev/null
+++ b/papers/items/2026-2606-31134-beyond-the-library-an-agentic-framework-for-autoformalizing-research-mathematics.md
@@ -0,0 +1,62 @@
+# Paper: Beyond the Library: An Agentic Framework for Autoformalizing Research Mathematics
+
+---
+type: paper
+title: "Beyond the Library: An Agentic Framework for Autoformalizing Research Mathematics"
+authors: Arshia Soltani Moakhar, Iman Gholami, Max Springer, Mahdi JafariRaviz, MohammadTaghi Hajiaghayi
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.31134
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - multi-agent
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, coding-agent, multi-agent, rag, reasoning
+- arXiv categories: cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.31134
diff --git a/papers/items/2026-2606-31174-clawarena-team-benchmarking-subagent-orchestration-and-dynamic-workflows-in-lang.md b/papers/items/2026-2606-31174-clawarena-team-benchmarking-subagent-orchestration-and-dynamic-workflows-in-lang.md
new file mode 100644
index 0000000..1be872e
--- /dev/null
+++ b/papers/items/2026-2606-31174-clawarena-team-benchmarking-subagent-orchestration-and-dynamic-workflows-in-lang.md
@@ -0,0 +1,61 @@
+# Paper: ClawArena-Team: Benchmarking Subagent Orchestration and Dynamic Workflows in Language-Model Agents
+
+---
+type: paper
+title: "ClawArena-Team: Benchmarking Subagent Orchestration and Dynamic Workflows in Language-Model Agents"
+authors: Kaiwen Xiong, Haonian Ji, Shi Qiu, Zeyu Zheng, Cihang Xie, Xinyu Ye, Huaxiu Yao
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.31174
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: llm-agent, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent, multi-agent-llm
+- inferred topics: agent-evaluation, multi-agent, tool-use, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.31174
diff --git a/papers/items/2026-2606-31179-healthagentbench-a-unified-benchmark-suite-of-realistic-agentic-healthcare-envir.md b/papers/items/2026-2606-31179-healthagentbench-a-unified-benchmark-suite-of-realistic-agentic-healthcare-envir.md
new file mode 100644
index 0000000..c8bdbc9
--- /dev/null
+++ b/papers/items/2026-2606-31179-healthagentbench-a-unified-benchmark-suite-of-realistic-agentic-healthcare-envir.md
@@ -0,0 +1,63 @@
+# Paper: HealthAgentBench: A Unified Benchmark Suite of Realistic Agentic Healthcare Environments for Challenging Frontier AI Agents
+
+---
+type: paper
+title: "HealthAgentBench: A Unified Benchmark Suite of Realistic Agentic Healthcare Environments for Challenging Frontier AI Agents"
+authors: Qianchu Liu, Sheng Zhang, Guanghui Qin, Jeya Maria Jose Valanarasu, Maximilian Rokuss, Mingyu Lu, Timothy Ossowski, Juan Manuel Zambrano Chaves, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.31179
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - reasoning
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+  - cs.CV
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: agent-evaluation, ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, ai-agent
+- inferred topics: agent-evaluation, planning, reasoning, workflow-agent
+- arXiv categories: cs.AI, cs.CL, cs.CV
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.31179
diff --git a/papers/items/2026-2606-31200-agentic-rag-vlm-affordance-aware-retrieval-augmented-generation-with-self-reflec.md b/papers/items/2026-2606-31200-agentic-rag-vlm-affordance-aware-retrieval-augmented-generation-with-self-reflec.md
new file mode 100644
index 0000000..ecad6f7
--- /dev/null
+++ b/papers/items/2026-2606-31200-agentic-rag-vlm-affordance-aware-retrieval-augmented-generation-with-self-reflec.md
@@ -0,0 +1,62 @@
+# Paper: Agentic RAG-VLM: Affordance-Aware Retrieval-Augmented Generation with Self-Reflective Planning for Robotic Grasping
+
+---
+type: paper
+title: "Agentic RAG-VLM: Affordance-Aware Retrieval-Augmented Generation with Self-Reflective Planning for Robotic Grasping"
+authors: Tao Chen, Lizheng Liu, Jiaxu Wang, Ziyue Jiang, Ruiqi Tian, JiGuang Huo, Zhongxue Gan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.31200
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - embodied-agent
+  - planning
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, embodied-agent, planning, rag, reasoning
+- arXiv categories: cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.31200
diff --git a/papers/items/2026-2606-31209-long-term-traffic-simulation-via-structured-autoregressive-modeling.md b/papers/items/2026-2606-31209-long-term-traffic-simulation-via-structured-autoregressive-modeling.md
new file mode 100644
index 0000000..07dec59
--- /dev/null
+++ b/papers/items/2026-2606-31209-long-term-traffic-simulation-via-structured-autoregressive-modeling.md
@@ -0,0 +1,66 @@
+# Paper: Long-term Traffic Simulation via Structured Autoregressive Modeling
+
+---
+type: paper
+title: Long-term Traffic Simulation via Structured Autoregressive Modeling
+authors: Lingyu Xiao, Zexin Feng, Xintao Yan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.31209
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - multi-agent
+  - planning
+  - rag
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.RO
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, agent-safety, memory, multi-agent, planning, rag, tool-use, world-model
+- arXiv categories: cs.AI, cs.RO
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.31209
diff --git a/papers/items/2026-2606-31227-securing-the-ai-agent-a-unified-framework-for-multi-layer-agent-red-teaming.md b/papers/items/2026-2606-31227-securing-the-ai-agent-a-unified-framework-for-multi-layer-agent-red-teaming.md
new file mode 100644
index 0000000..52574b5
--- /dev/null
+++ b/papers/items/2026-2606-31227-securing-the-ai-agent-a-unified-framework-for-multi-layer-agent-red-teaming.md
@@ -0,0 +1,59 @@
+# Paper: Securing the AI Agent: A Unified Framework for Multi-Layer Agent Red Teaming
+
+---
+type: paper
+title: "Securing the AI Agent: A Unified Framework for Multi-Layer Agent Red Teaming"
+authors: Yong Yang, Xing Zheng, Huiyu Wu, Huangsheng Cheng, Xiaorong Shi, Jing Guo, Bo Yang, Yi Zhou, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.31227
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-safety, ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety, ai-agent
+- inferred topics: agent-safety, tool-use
+- arXiv categories: cs.CR
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.31227
diff --git a/papers/items/2026-2606-31229-agentic-ideation-sample-efficient-agentic-trajectories-synthesis-for-scientific-.md b/papers/items/2026-2606-31229-agentic-ideation-sample-efficient-agentic-trajectories-synthesis-for-scientific-.md
new file mode 100644
index 0000000..35409bd
--- /dev/null
+++ b/papers/items/2026-2606-31229-agentic-ideation-sample-efficient-agentic-trajectories-synthesis-for-scientific-.md
@@ -0,0 +1,63 @@
+# Paper: Agentic-Ideation: Sample Efficient Agentic Trajectories Synthesis for Scientific Ideation Agents
+
+---
+type: paper
+title: "Agentic-Ideation: Sample Efficient Agentic Trajectories Synthesis for Scientific Ideation Agents"
+authors: Keyu Zhao, Lingyan Kong, Fengli Xu, Yong Li
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.31229
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - computer-use
+  - multi-agent
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: agentic-ai, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, multi-agent-llm
+- inferred topics: computer-use, multi-agent, rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.31229
diff --git a/papers/items/2026-2606-31252-embodied-cad-solver-grounded-llm-agents-for-parametric-b-rep-assembly-modeling.md b/papers/items/2026-2606-31252-embodied-cad-solver-grounded-llm-agents-for-parametric-b-rep-assembly-modeling.md
new file mode 100644
index 0000000..d6829d1
--- /dev/null
+++ b/papers/items/2026-2606-31252-embodied-cad-solver-grounded-llm-agents-for-parametric-b-rep-assembly-modeling.md
@@ -0,0 +1,62 @@
+# Paper: Embodied CAD: Solver-Grounded LLM Agents for Parametric B-Rep Assembly Modeling
+
+---
+type: paper
+title: "Embodied CAD: Solver-Grounded LLM Agents for Parametric B-Rep Assembly Modeling"
+authors: Fumin Liu, Haoyu Zhou, Fei Hao, Lin Yang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.31252
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - embodied-agent
+  - planning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: llm-agent, planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent, planning-agent
+- inferred topics: agent-evaluation, embodied-agent, planning, tool-use, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.31252
diff --git a/papers/items/2026-2606-31314-a-novel-method-for-differential-algebraic-dynamic-model-discovery-in-power-syste.md b/papers/items/2026-2606-31314-a-novel-method-for-differential-algebraic-dynamic-model-discovery-in-power-syste.md
new file mode 100644
index 0000000..475fbd2
--- /dev/null
+++ b/papers/items/2026-2606-31314-a-novel-method-for-differential-algebraic-dynamic-model-discovery-in-power-syste.md
@@ -0,0 +1,60 @@
+# Paper: A Novel Method for Differential-Algebraic Dynamic Model Discovery in Power Systems: An LLM-Based Multi-Agent Collaborative Framework
+
+---
+type: paper
+title: "A Novel Method for Differential-Algebraic Dynamic Model Discovery in Power Systems: An LLM-Based Multi-Agent Collaborative Framework"
+authors: Xinming Wang, Fan Tang, Yingli Wei, Yakun He, Zhe Liu, Ping Jiang, Haoyu Wu, Zihan Guo, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.31314
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - multi-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - eess.SY
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, computer-use, multi-agent
+- arXiv categories: eess.SY
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.31314
diff --git a/papers/items/2026-2606-31339-verification-gated-agentic-mission-state-governance-for-intelligent-industrial-m.md b/papers/items/2026-2606-31339-verification-gated-agentic-mission-state-governance-for-intelligent-industrial-m.md
new file mode 100644
index 0000000..4d58e2c
--- /dev/null
+++ b/papers/items/2026-2606-31339-verification-gated-agentic-mission-state-governance-for-intelligent-industrial-m.md
@@ -0,0 +1,64 @@
+# Paper: Verification-Gated Agentic Mission-State Governance for Intelligent Industrial Multi-Robot Systems
+
+---
+type: paper
+title: Verification-Gated Agentic Mission-State Governance for Intelligent Industrial Multi-Robot Systems
+authors: Guoqin Tang, Qingxuan Jia, Yichen Tan, Zeyuan Huang, Ning Ji, Gang Chen
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.31339
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - embodied-agent
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.RO
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-evaluation, agent-safety, embodied-agent, planning, rag, reasoning, tool-use
+- arXiv categories: cs.RO
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.31339
diff --git a/papers/items/2026-2606-31410-xiaomi-gui-0-technical-report.md b/papers/items/2026-2606-31410-xiaomi-gui-0-technical-report.md
new file mode 100644
index 0000000..122a286
--- /dev/null
+++ b/papers/items/2026-2606-31410-xiaomi-gui-0-technical-report.md
@@ -0,0 +1,65 @@
+# Paper: Xiaomi-GUI-0 Technical Report
+
+---
+type: paper
+title: Xiaomi-GUI-0 Technical Report
+authors: Wanxia Cao, Chengzhen Duan, Pei Fu, Pengzhi Gao, Niu Lian, Fazhan Liu, Hui Liu, Heng Qu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.31410
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - embodied-agent
+  - memory
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, agent-safety, computer-use, embodied-agent, memory, planning, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.31410
diff --git a/papers/items/2026-2606-31471-think-while-you-map-asynchronous-vision-language-agents-for-incremental-3d-scene.md b/papers/items/2026-2606-31471-think-while-you-map-asynchronous-vision-language-agents-for-incremental-3d-scene.md
new file mode 100644
index 0000000..36838b9
--- /dev/null
+++ b/papers/items/2026-2606-31471-think-while-you-map-asynchronous-vision-language-agents-for-incremental-3d-scene.md
@@ -0,0 +1,59 @@
+# Paper: Think While You Map: Asynchronous Vision-Language Agents for Incremental 3D Scene Graphs
+
+---
+type: paper
+title: "Think While You Map: Asynchronous Vision-Language Agents for Incremental 3D Scene Graphs"
+authors: Deniz Bickici, Michael Pabst, Shohei Mori, Dieter Schmalstieg
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.31471
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, rag
+- arXiv categories: cs.CV
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.31471
diff --git a/papers/items/2026-2606-31612-what-memory-do-gui-agents-really-need-from-passive-records-to-active-task-drivin.md b/papers/items/2026-2606-31612-what-memory-do-gui-agents-really-need-from-passive-records-to-active-task-drivin.md
new file mode 100644
index 0000000..e66f00e
--- /dev/null
+++ b/papers/items/2026-2606-31612-what-memory-do-gui-agents-really-need-from-passive-records-to-active-task-drivin.md
@@ -0,0 +1,64 @@
+# Paper: What Memory Do GUI Agents Really Need? From Passive Records to Active Task-Driving States
+
+---
+type: paper
+title: What Memory Do GUI Agents Really Need? From Passive Records to Active Task-Driving States
+authors: Chen Liu, Ling Chen, Hanzhang Zhou, Xu Zhang, Quyu Kong, Panrong Tong, Wenhao Wang, Xin Yu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.31612
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - planning
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: agent-memory, web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory, web-gui-agent
+- inferred topics: agent-evaluation, computer-use, memory, planning, rag, tool-use, workflow-agent
+- arXiv categories: cs.CV
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.31612
diff --git a/papers/items/2026-2606-31635-a-tutorial-on-autonomous-fault-tolerant-control-using-knowledge-grounded-llm-age.md b/papers/items/2026-2606-31635-a-tutorial-on-autonomous-fault-tolerant-control-using-knowledge-grounded-llm-age.md
new file mode 100644
index 0000000..06fc564
--- /dev/null
+++ b/papers/items/2026-2606-31635-a-tutorial-on-autonomous-fault-tolerant-control-using-knowledge-grounded-llm-age.md
@@ -0,0 +1,63 @@
+# Paper: A Tutorial on Autonomous Fault-Tolerant Control Using Knowledge-Grounded LLM Agents
+
+---
+type: paper
+title: A Tutorial on Autonomous Fault-Tolerant Control Using Knowledge-Grounded LLM Agents
+authors: Javal Vyas, Milapji Singh Gill, Artan Markaj, Felix Gehlhoff, Mehmet Mercangöz
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.31635
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - planning
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - eess.SY
+  - cs.AI
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: agent-safety, planning, tool-use, world-model
+- arXiv categories: eess.SY, cs.AI, cs.MA
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.31635
diff --git a/papers/items/2026-2606-31639-a-lifecycle-and-application-stack-survey-of-large-language-model-vulnerabilities.md b/papers/items/2026-2606-31639-a-lifecycle-and-application-stack-survey-of-large-language-model-vulnerabilities.md
new file mode 100644
index 0000000..3d075d3
--- /dev/null
+++ b/papers/items/2026-2606-31639-a-lifecycle-and-application-stack-survey-of-large-language-model-vulnerabilities.md
@@ -0,0 +1,69 @@
+# Paper: A Lifecycle and Application-Stack Survey of Large Language Model Vulnerabilities: Attacks, Risks, Defenses, and Open Problems
+
+---
+type: paper
+title: "A Lifecycle and Application-Stack Survey of Large Language Model Vulnerabilities: Attacks, Risks, Defenses, and Open Problems"
+authors: Seyed Bagher Hashemi Natanzi, Bo Tang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.31639
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - embodied-agent
+  - memory
+  - planning
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+  - cs.GT
+  - cs.LO
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-evaluation, autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, autonomous-agent-llm
+- inferred topics: agent-evaluation, agent-safety, coding-agent, embodied-agent, memory, planning, rag, tool-use, workflow-agent
+- arXiv categories: cs.CR, cs.AI, cs.GT, cs.LO
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.31639
diff --git a/papers/items/2026-2606-31648-think-in-english-answer-in-korean-efficient-adaptation-of-multilingual-tool-usin.md b/papers/items/2026-2606-31648-think-in-english-answer-in-korean-efficient-adaptation-of-multilingual-tool-usin.md
new file mode 100644
index 0000000..afc4a76
--- /dev/null
+++ b/papers/items/2026-2606-31648-think-in-english-answer-in-korean-efficient-adaptation-of-multilingual-tool-usin.md
@@ -0,0 +1,63 @@
+# Paper: Think in English, Answer in Korean: Efficient Adaptation of Multilingual Tool-Using Agents
+
+---
+type: paper
+title: "Think in English, Answer in Korean: Efficient Adaptation of Multilingual Tool-Using Agents"
+authors: Utsav Garg, Sungjin Hong, Jason Jung, Justin Lee, Shaan Desai, Joon Hee Kim, Anirudh Shrinivason, Edmond Wen, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.31648
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - memory
+  - multi-agent
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agentic-ai, function-calling, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, function-calling, tool-use
+- inferred topics: memory, multi-agent, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI, cs.LG
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.31648
diff --git a/papers/items/2026-2606-31650-echo-prune-to-act-trace-to-learn-with-selective-turn-memory-in-agentic-rl.md b/papers/items/2026-2606-31650-echo-prune-to-act-trace-to-learn-with-selective-turn-memory-in-agentic-rl.md
new file mode 100644
index 0000000..ddf1173
--- /dev/null
+++ b/papers/items/2026-2606-31650-echo-prune-to-act-trace-to-learn-with-selective-turn-memory-in-agentic-rl.md
@@ -0,0 +1,62 @@
+# Paper: ECHO: Prune to act, trace to learn with selective turn memory in agentic RL
+
+---
+type: paper
+title: "ECHO: Prune to act, trace to learn with selective turn memory in agentic RL"
+authors: Zijun Xie, Binbin Zheng, Enlei Gong, Jihua Liu, Yuyang You, Lingfeng Liu, Jiayao Tang, Guanqun Zhao, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.31650
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, memory, planning, tool-use
+- arXiv categories: cs.LG, cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.31650
diff --git a/papers/items/2026-2606-31665-forecastagentsearch-towards-a-multi-expert-agent-search-system-for-geopolitical-.md b/papers/items/2026-2606-31665-forecastagentsearch-towards-a-multi-expert-agent-search-system-for-geopolitical-.md
new file mode 100644
index 0000000..0368d15
--- /dev/null
+++ b/papers/items/2026-2606-31665-forecastagentsearch-towards-a-multi-expert-agent-search-system-for-geopolitical-.md
@@ -0,0 +1,61 @@
+# Paper: ForecastAgentSearch: Towards a Multi-Expert Agent Search System for Geopolitical Event Forecasting
+
+---
+type: paper
+title: "ForecastAgentSearch: Towards a Multi-Expert Agent Search System for Geopolitical Event Forecasting"
+authors: Miaomiao Cai, He Chang, Yunshan Ma, See-kiong Ng
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.31665
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - planning
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, multi-agent, planning, rag
+- arXiv categories: cs.MA
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.31665
diff --git a/papers/items/2026-2606-31693-shopx-a-foundation-model-for-intent-to-item-fulfillment-in-agentic-shopping.md b/papers/items/2026-2606-31693-shopx-a-foundation-model-for-intent-to-item-fulfillment-in-agentic-shopping.md
new file mode 100644
index 0000000..33e48d1
--- /dev/null
+++ b/papers/items/2026-2606-31693-shopx-a-foundation-model-for-intent-to-item-fulfillment-in-agentic-shopping.md
@@ -0,0 +1,64 @@
+# Paper: ShopX: A Foundation Model for Intent-to-Item Fulfillment in Agentic Shopping
+
+---
+type: paper
+title: "ShopX: A Foundation Model for Intent-to-Item Fulfillment in Agentic Shopping"
+authors: Jiacheng Chen, Tao Zhang, Manxi Lin, Dunxian Huang, Teng Shi, Honghao Fu, Mengyan Li, Xinming Zhang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.31693
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.IR
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: llm-agent, planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent, planning-agent
+- inferred topics: agent-evaluation, planning, rag, tool-use, workflow-agent
+- arXiv categories: cs.IR, cs.AI, cs.CL
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.31693
diff --git a/papers/items/2026-2606-31744-a-conversational-agentic-interface-to-physics-based-household-digital-twins-for-.md b/papers/items/2026-2606-31744-a-conversational-agentic-interface-to-physics-based-household-digital-twins-for-.md
new file mode 100644
index 0000000..13be2b8
--- /dev/null
+++ b/papers/items/2026-2606-31744-a-conversational-agentic-interface-to-physics-based-household-digital-twins-for-.md
@@ -0,0 +1,62 @@
+# Paper: A Conversational Agentic Interface to Physics-Based Household Digital Twins for Residential Energy Decision Support
+
+---
+type: paper
+title: A Conversational Agentic Interface to Physics-Based Household Digital Twins for Residential Energy Decision Support
+authors: Costas Mylonas, Titos Georgoulakis, Magda Foti
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.31744
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - eess.SY
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: agent-evaluation, planning, rag, tool-use, world-model
+- arXiv categories: eess.SY
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.31744
diff --git a/papers/items/2026-2606-31767-jeto-bench-a-reproducible-benchmark-for-execution-time-improvement-patches-in-ja.md b/papers/items/2026-2606-31767-jeto-bench-a-reproducible-benchmark-for-execution-time-improvement-patches-in-ja.md
new file mode 100644
index 0000000..1025704
--- /dev/null
+++ b/papers/items/2026-2606-31767-jeto-bench-a-reproducible-benchmark-for-execution-time-improvement-patches-in-ja.md
@@ -0,0 +1,60 @@
+# Paper: JETO-Bench: A Reproducible Benchmark for Execution Time Improvement Patches in Java
+
+---
+type: paper
+title: "JETO-Bench: A Reproducible Benchmark for Execution Time Improvement Patches in Java"
+authors: Khashayar Etemadi, Zhendong Su
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.31767
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: agent-evaluation, coding-agent, tool-use
+- arXiv categories: cs.SE
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.31767
diff --git a/papers/items/2026-2606-31831-an-agentic-ai-framework-to-accelerate-scientific-discovery-in-plant-phenotyping.md b/papers/items/2026-2606-31831-an-agentic-ai-framework-to-accelerate-scientific-discovery-in-plant-phenotyping.md
new file mode 100644
index 0000000..2ea71f5
--- /dev/null
+++ b/papers/items/2026-2606-31831-an-agentic-ai-framework-to-accelerate-scientific-discovery-in-plant-phenotyping.md
@@ -0,0 +1,60 @@
+# Paper: An Agentic AI Framework to Accelerate Scientific Discovery in Plant Phenotyping
+
+---
+type: paper
+title: An Agentic AI Framework to Accelerate Scientific Discovery in Plant Phenotyping
+authors: Renan Souza, Daniel Rosendo, Kelsey Carter, John Lagergren, Frédéric Suter, Shelaine L. Curd, Gerald A. Tuskan, Rafael Ferreira da Silva, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.31831
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agentic-ai, ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, ai-agent
+- inferred topics: agent-safety, planning, tool-use
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.31831
diff --git a/papers/items/2026-2606-31916-theory-of-mind-and-persuasion-beyond-conversation-assessing-the-capacity-of-llms.md b/papers/items/2026-2606-31916-theory-of-mind-and-persuasion-beyond-conversation-assessing-the-capacity-of-llms.md
new file mode 100644
index 0000000..302ea84
--- /dev/null
+++ b/papers/items/2026-2606-31916-theory-of-mind-and-persuasion-beyond-conversation-assessing-the-capacity-of-llms.md
@@ -0,0 +1,62 @@
+# Paper: Theory of Mind and Persuasion Beyond Conversation: Assessing the Capacity of LLMs to Induce Belief States via Planning and Action
+
+---
+type: paper
+title: "Theory of Mind and Persuasion Beyond Conversation: Assessing the Capacity of LLMs to Induce Belief States via Planning and Action"
+authors: Ben Slater, Matteo G. Mecattaf, Lucy G. Cheke, John Burden, Winnie Street
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.31916
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, agent-safety, planning, reasoning, tool-use
+- arXiv categories: cs.CL
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.31916
diff --git a/papers/items/2026-2606-31980-digitalcoach-communication-and-grounding-gaps-in-human-and-agentic-computer-use-.md b/papers/items/2026-2606-31980-digitalcoach-communication-and-grounding-gaps-in-human-and-agentic-computer-use-.md
new file mode 100644
index 0000000..895fe16
--- /dev/null
+++ b/papers/items/2026-2606-31980-digitalcoach-communication-and-grounding-gaps-in-human-and-agentic-computer-use-.md
@@ -0,0 +1,63 @@
+# Paper: DigitalCoach: Communication and Grounding Gaps in Human and Agentic Computer Use Coaching
+
+---
+type: paper
+title: "DigitalCoach: Communication and Grounding Gaps in Human and Agentic Computer Use Coaching"
+authors: Meng Chen, Anya Ji, Tsung-Han Wu, Tobias Maringgele, David M. Chan, Alane Suhr, Amy Pavel
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.31980
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+  - cs.HC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, computer-use, planning, rag
+- arXiv categories: cs.CL, cs.AI, cs.HC
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.31980
diff --git a/papers/items/2026-2606-32025-generative-skill-composition-for-llm-agents.md b/papers/items/2026-2606-32025-generative-skill-composition-for-llm-agents.md
new file mode 100644
index 0000000..44aadc9
--- /dev/null
+++ b/papers/items/2026-2606-32025-generative-skill-composition-for-llm-agents.md
@@ -0,0 +1,62 @@
+# Paper: Generative Skill Composition for LLM Agents
+
+---
+type: paper
+title: Generative Skill Composition for LLM Agents
+authors: Xinyu Zhao, Zhen Tan, Vaishnav Tadiparthi, Nakul Agarwal, Kwonjoon Lee, Ehsan Moradi Pari, Hossein Nourkhiz Mahjoub, Tianlong Chen
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.32025
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - planning
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: coding-agent, llm-agent, planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent, llm-agent, planning-agent
+- inferred topics: agent-evaluation, coding-agent, planning, rag, reasoning
+- arXiv categories: cs.CL
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.32025
diff --git a/papers/items/2026-2606-32034-qval-cheaply-evaluating-dense-supervision-signals-for-long-horizon-llm-agents.md b/papers/items/2026-2606-32034-qval-cheaply-evaluating-dense-supervision-signals-for-long-horizon-llm-agents.md
new file mode 100644
index 0000000..568051c
--- /dev/null
+++ b/papers/items/2026-2606-32034-qval-cheaply-evaluating-dense-supervision-signals-for-long-horizon-llm-agents.md
@@ -0,0 +1,63 @@
+# Paper: QVal: Cheaply Evaluating Dense Supervision Signals for Long-Horizon LLM Agents
+
+---
+type: paper
+title: "QVal: Cheaply Evaluating Dense Supervision Signals for Long-Horizon LLM Agents"
+authors: Sergio Hernández-Gutiérrez, Matteo Merler, Ilze Amanda Auzina, Joschka Strüber, Ameya Prabhu, Matthias Bethge
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2606.32034
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: agent-evaluation, computer-use, planning, tool-use
+- arXiv categories: cs.LG, cs.AI, cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2606.32034
diff --git a/papers/items/2026-2607-00038-stop-hand-holding-your-coding-agent-engineering-the-loops-that-replace-step-by-s.md b/papers/items/2026-2607-00038-stop-hand-holding-your-coding-agent-engineering-the-loops-that-replace-step-by-s.md
new file mode 100644
index 0000000..ca2d9a8
--- /dev/null
+++ b/papers/items/2026-2607-00038-stop-hand-holding-your-coding-agent-engineering-the-loops-that-replace-step-by-s.md
@@ -0,0 +1,63 @@
+# Paper: Stop Hand-Holding Your Coding Agent: Engineering the Loops that Replace Step-by-Step Prompting
+
+---
+type: paper
+title: "Stop Hand-Holding Your Coding Agent: Engineering the Loops that Replace Step-by-Step Prompting"
+authors: Sandeco Macedo
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.00038
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-28
+updated_at: 2026-06-28
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - coding-agent
+  - computer-use
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: agent-safety, coding-agent, computer-use, memory, rag, tool-use
+- arXiv categories: cs.SE
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.00038
diff --git a/papers/items/2026-2607-00041-atm-cid-brokered-pre-write-admission-for-multi-agent-code-co-synthesis.md b/papers/items/2026-2607-00041-atm-cid-brokered-pre-write-admission-for-multi-agent-code-co-synthesis.md
new file mode 100644
index 0000000..966d2b4
--- /dev/null
+++ b/papers/items/2026-2607-00041-atm-cid-brokered-pre-write-admission-for-multi-agent-code-co-synthesis.md
@@ -0,0 +1,64 @@
+# Paper: ATM: CID-Brokered Pre-Write Admission for Multi-Agent Code Co-Synthesis
+
+---
+type: paper
+title: "ATM: CID-Brokered Pre-Write Admission for Multi-Agent Code Co-Synthesis"
+authors: Eagl Huang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.00041
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-29
+updated_at: 2026-06-29
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - computer-use
+  - multi-agent
+  - planning
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agent-evaluation, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, multi-agent-llm
+- inferred topics: agent-evaluation, coding-agent, computer-use, multi-agent, planning, rag
+- arXiv categories: cs.SE, cs.AI
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.00041
diff --git a/papers/items/2026-2607-00233-from-signals-to-structure-how-memory-architecture-drives-language-emergence-in-l.md b/papers/items/2026-2607-00233-from-signals-to-structure-how-memory-architecture-drives-language-emergence-in-l.md
new file mode 100644
index 0000000..99064d3
--- /dev/null
+++ b/papers/items/2026-2607-00233-from-signals-to-structure-how-memory-architecture-drives-language-emergence-in-l.md
@@ -0,0 +1,63 @@
+# Paper: From Signals to Structure: How Memory Architecture Drives Language Emergence in LLM Agents
+
+---
+type: paper
+title: "From Signals to Structure: How Memory Architecture Drives Language Emergence in LLM Agents"
+authors: Yashar Talebirad, Eden Redman, Ali Parsaee, Osmar R. Zaiane
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.00233
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+  - cs.IT
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: memory, rag, tool-use
+- arXiv categories: cs.AI, cs.CL, cs.IT, cs.MA
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.00233
diff --git a/papers/items/2026-2607-00255-slm-llm-or-agentic-ai-toward-intelligent-uav-enabled-wpt-systems-in-low-altitude.md b/papers/items/2026-2607-00255-slm-llm-or-agentic-ai-toward-intelligent-uav-enabled-wpt-systems-in-low-altitude.md
new file mode 100644
index 0000000..d08ab8f
--- /dev/null
+++ b/papers/items/2026-2607-00255-slm-llm-or-agentic-ai-toward-intelligent-uav-enabled-wpt-systems-in-low-altitude.md
@@ -0,0 +1,60 @@
+# Paper: SLM, LLM or Agentic AI? Toward Intelligent UAV-Enabled WPT Systems in Low-Altitude Economy Networks
+
+---
+type: paper
+title: SLM, LLM or Agentic AI? Toward Intelligent UAV-Enabled WPT Systems in Low-Altitude Economy Networks
+authors: Feibo Jiang, Li Dong, Lei Mao, Kezhi Wang, Xianbin Wang, Abbas Jamalipour
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.00255
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - reasoning
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.IT
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-evaluation, reasoning, world-model
+- arXiv categories: cs.IT
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.00255
diff --git a/papers/items/2026-2607-00297-epc-a-standardized-protocol-for-measuring-evaluator-preference-dynamics-in-llm-a.md b/papers/items/2026-2607-00297-epc-a-standardized-protocol-for-measuring-evaluator-preference-dynamics-in-llm-a.md
new file mode 100644
index 0000000..3293abb
--- /dev/null
+++ b/papers/items/2026-2607-00297-epc-a-standardized-protocol-for-measuring-evaluator-preference-dynamics-in-llm-a.md
@@ -0,0 +1,62 @@
+# Paper: EPC: A Standardized Protocol for Measuring Evaluator Preference Dynamics in LLM Agent Systems
+
+---
+type: paper
+title: "EPC: A Standardized Protocol for Measuring Evaluator Preference Dynamics in LLM Agent Systems"
+authors: Zewen Liu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.00297
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - computer-use
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: agent-evaluation, coding-agent, computer-use, tool-use
+- arXiv categories: cs.LG, cs.CL
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.00297
diff --git a/papers/items/2026-2607-00334-managed-autonomy-at-runtime-gear-based-safety-and-governance-for-single-and-mult.md b/papers/items/2026-2607-00334-managed-autonomy-at-runtime-gear-based-safety-and-governance-for-single-and-mult.md
new file mode 100644
index 0000000..ea0f4a8
--- /dev/null
+++ b/papers/items/2026-2607-00334-managed-autonomy-at-runtime-gear-based-safety-and-governance-for-single-and-mult.md
@@ -0,0 +1,63 @@
+# Paper: Managed Autonomy at Runtime: Gear-Based Safety and Governance for Single- and Multi-Agent Cyber-Physical Systems
+
+---
+type: paper
+title: "Managed Autonomy at Runtime: Gear-Based Safety and Governance for Single- and Multi-Agent Cyber-Physical Systems"
+authors: Srini Ramaswamy, Wang Miaosheng
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.00334
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - embodied-agent
+  - multi-agent
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: autonomous-agent-llm, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm, multi-agent-llm
+- inferred topics: agent-evaluation, agent-safety, embodied-agent, multi-agent, planning, tool-use
+- arXiv categories: cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.00334
diff --git a/papers/items/2026-2607-00345-registry-governed-agent-lifecycle-completing-eddops-with-evaluation-drivenregist.md b/papers/items/2026-2607-00345-registry-governed-agent-lifecycle-completing-eddops-with-evaluation-drivenregist.md
new file mode 100644
index 0000000..fca4eff
--- /dev/null
+++ b/papers/items/2026-2607-00345-registry-governed-agent-lifecycle-completing-eddops-with-evaluation-drivenregist.md
@@ -0,0 +1,60 @@
+# Paper: Registry-Governed Agent Lifecycle:Completing EDDOps with Evaluation-DrivenRegistration, Promotion, and Retirement on AWS AgentCore
+
+---
+type: paper
+title: "Registry-Governed Agent Lifecycle:Completing EDDOps with Evaluation-DrivenRegistration, Promotion, and Retirement on AWS AgentCore"
+authors: Richard Kang, Vincent Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.00345
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: agent-evaluation, agent-safety, workflow-agent
+- arXiv categories: cs.SE
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.00345
diff --git a/papers/items/2026-2607-00407-personalization-as-inverse-planning-learning-latent-design-intents-for-agentic-s.md b/papers/items/2026-2607-00407-personalization-as-inverse-planning-learning-latent-design-intents-for-agentic-s.md
new file mode 100644
index 0000000..59c6f48
--- /dev/null
+++ b/papers/items/2026-2607-00407-personalization-as-inverse-planning-learning-latent-design-intents-for-agentic-s.md
@@ -0,0 +1,60 @@
+# Paper: Personalization as Inverse Planning: Learning Latent Design Intents for Agentic Slide Generation via Structural Denoising
+
+---
+type: paper
+title: "Personalization as Inverse Planning: Learning Latent Design Intents for Agentic Slide Generation via Structural Denoising"
+authors: Tianci Liu, Zihan Dong, Linjun Zhang, Haoyu Wang, jing Gao, Emre Kiciman, Ranveer Chandra, Wei-Ting Chen
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.00407
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - multi-agent
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent
+- inferred topics: multi-agent, planning, tool-use
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.00407
diff --git a/papers/items/2026-2607-00422-kidnaprag-a-black-box-attack-for-hijacking-reasoning-in-agentic-retrieval-augmen.md b/papers/items/2026-2607-00422-kidnaprag-a-black-box-attack-for-hijacking-reasoning-in-agentic-retrieval-augmen.md
new file mode 100644
index 0000000..a4dedd0
--- /dev/null
+++ b/papers/items/2026-2607-00422-kidnaprag-a-black-box-attack-for-hijacking-reasoning-in-agentic-retrieval-augmen.md
@@ -0,0 +1,60 @@
+# Paper: KidnapRAG: A Black-Box Attack for Hijacking Reasoning in Agentic Retrieval-Augmented Generation Systems
+
+---
+type: paper
+title: "KidnapRAG: A Black-Box Attack for Hijacking Reasoning in Agentic Retrieval-Augmented Generation Systems"
+authors: Chanwoo Choi, Euntae Kim, Kyuho Lee, Youngsam Chun, Jinhee Jeong, Eunmi Kim, Myunggyo Oh, Junseo Jang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.00422
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, rag, reasoning
+- arXiv categories: cs.CR
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.00422
diff --git a/papers/items/2026-2607-00436-phreeqc-mcq-200-a-diagnostic-benchmark-for-tool-augmented-scientific-simulator-a.md b/papers/items/2026-2607-00436-phreeqc-mcq-200-a-diagnostic-benchmark-for-tool-augmented-scientific-simulator-a.md
new file mode 100644
index 0000000..fc1fa32
--- /dev/null
+++ b/papers/items/2026-2607-00436-phreeqc-mcq-200-a-diagnostic-benchmark-for-tool-augmented-scientific-simulator-a.md
@@ -0,0 +1,61 @@
+# Paper: PHREEQC-MCQ-200: A Diagnostic Benchmark for Tool-Augmented Scientific Simulator Agents
+
+---
+type: paper
+title: "PHREEQC-MCQ-200: A Diagnostic Benchmark for Tool-Augmented Scientific Simulator Agents"
+authors: Ke Zhang, Sahchit Chundur, Mohammad Javad Qomi, Maziar Raissi
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.00436
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, rag, tool-use, world-model
+- arXiv categories: cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.00436
diff --git a/papers/items/2026-2607-00440-minos-a-multi-agent-collaborative-framework-for-provenance-based-backward-tracki.md b/papers/items/2026-2607-00440-minos-a-multi-agent-collaborative-framework-for-provenance-based-backward-tracki.md
new file mode 100644
index 0000000..f5fea34
--- /dev/null
+++ b/papers/items/2026-2607-00440-minos-a-multi-agent-collaborative-framework-for-provenance-based-backward-tracki.md
@@ -0,0 +1,63 @@
+# Paper: Minos: A Multi-Agent Collaborative Framework for Provenance-Based Backward Tracking
+
+---
+type: paper
+title: "Minos: A Multi-Agent Collaborative Framework for Provenance-Based Backward Tracking"
+authors: Jiahui Wang, Zhenyuan Li, Zhengkai Wang, Xiangmin Shen, Fan Zhang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.00440
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - multi-agent
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, agent-safety, computer-use, multi-agent, rag, reasoning
+- arXiv categories: cs.CR
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.00440
diff --git a/papers/items/2026-2607-00454-agri-sage-simulation-grounded-multi-agent-llm-for-context-aware-agricultural-adv.md b/papers/items/2026-2607-00454-agri-sage-simulation-grounded-multi-agent-llm-for-context-aware-agricultural-adv.md
new file mode 100644
index 0000000..753a133
--- /dev/null
+++ b/papers/items/2026-2607-00454-agri-sage-simulation-grounded-multi-agent-llm-for-context-aware-agricultural-adv.md
@@ -0,0 +1,67 @@
+# Paper: Agri-SAGE: Simulation-Grounded Multi-Agent LLM for Context-Aware Agricultural Advisory Generation
+
+---
+type: paper
+title: "Agri-SAGE: Simulation-Grounded Multi-Agent LLM for Context-Aware Agricultural Advisory Generation"
+authors: Vedant Balasubramaniam, Geetha Charan, Manojkumar Patil, Rohit P Suresh, V Priyanka, Kodur Sai Vinay Sathvik, Y. Narahari
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.00454
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - memory
+  - multi-agent
+  - planning
+  - rag
+  - reasoning
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, agent-safety, computer-use, memory, multi-agent, planning, rag, reasoning, world-model
+- arXiv categories: cs.AI, cs.MA
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.00454
diff --git a/papers/items/2026-2607-00502-a-task-state-representation-for-long-horizon-mobile-gui-agents.md b/papers/items/2026-2607-00502-a-task-state-representation-for-long-horizon-mobile-gui-agents.md
new file mode 100644
index 0000000..6602547
--- /dev/null
+++ b/papers/items/2026-2607-00502-a-task-state-representation-for-long-horizon-mobile-gui-agents.md
@@ -0,0 +1,63 @@
+# Paper: A Task-State Representation for Long-Horizon Mobile GUI Agents
+
+---
+type: paper
+title: A Task-State Representation for Long-Horizon Mobile GUI Agents
+authors: Yujie Zheng, Zikang Liu, Xin Zhao, Ji-Rong Wen
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.00502
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, computer-use, memory, planning, reasoning, tool-use
+- arXiv categories: cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.00502
diff --git a/papers/items/2026-2607-00555-rise-from-the-ashes-llm-based-static-analysis-for-deep-learning-framework-bugs.md b/papers/items/2026-2607-00555-rise-from-the-ashes-llm-based-static-analysis-for-deep-learning-framework-bugs.md
new file mode 100644
index 0000000..8672415
--- /dev/null
+++ b/papers/items/2026-2607-00555-rise-from-the-ashes-llm-based-static-analysis-for-deep-learning-framework-bugs.md
@@ -0,0 +1,65 @@
+# Paper: Rise From The Ashes: LLM-based Static Analysis for Deep Learning Framework Bugs
+
+---
+type: paper
+title: "Rise From The Ashes: LLM-based Static Analysis for Deep Learning Framework Bugs"
+authors: Shaoyu Yang, Haifeng Lin, Chunrong Fang, Xiang Chen, Wei Cheng, Jiawei Liu, Yiyu Zhang, Hongyu Liu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.00555
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - computer-use
+  - multi-agent
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agentic-ai, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, multi-agent-llm
+- inferred topics: agent-evaluation, agent-safety, coding-agent, computer-use, multi-agent, rag, tool-use, workflow-agent
+- arXiv categories: cs.SE
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.00555
diff --git a/papers/items/2026-2607-00604-vehicle-routing-problem-meets-large-language-models-an-overview-and-perspectives.md b/papers/items/2026-2607-00604-vehicle-routing-problem-meets-large-language-models-an-overview-and-perspectives.md
new file mode 100644
index 0000000..8dc5e68
--- /dev/null
+++ b/papers/items/2026-2607-00604-vehicle-routing-problem-meets-large-language-models-an-overview-and-perspectives.md
@@ -0,0 +1,63 @@
+# Paper: Vehicle Routing Problem Meets Large Language Models: An Overview and Perspectives
+
+---
+type: paper
+title: "Vehicle Routing Problem Meets Large Language Models: An Overview and Perspectives"
+authors: Xianchao Xiu, Chong Shen, Yanjiao Zhu, Wanquan Liu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.00604
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - planning
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - math.OC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, multi-agent, planning, reasoning, tool-use, workflow-agent
+- arXiv categories: math.OC
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.00604
diff --git a/papers/items/2026-2607-00627-agi-maze-as-a-benchmark-framework-for-world-modeling-agents.md b/papers/items/2026-2607-00627-agi-maze-as-a-benchmark-framework-for-world-modeling-agents.md
new file mode 100644
index 0000000..57588bb
--- /dev/null
+++ b/papers/items/2026-2607-00627-agi-maze-as-a-benchmark-framework-for-world-modeling-agents.md
@@ -0,0 +1,61 @@
+# Paper: AGI Maze as a Benchmark Framework for World-Modeling Agents
+
+---
+type: paper
+title: AGI Maze as a Benchmark Framework for World-Modeling Agents
+authors: Alexey Potapov
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.00627
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: agent-evaluation, memory, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.00627
diff --git a/papers/items/2026-2607-00692-self-gc-self-governing-context-for-long-horizon-llm-agents.md b/papers/items/2026-2607-00692-self-gc-self-governing-context-for-long-horizon-llm-agents.md
new file mode 100644
index 0000000..bac2426
--- /dev/null
+++ b/papers/items/2026-2607-00692-self-gc-self-governing-context-for-long-horizon-llm-agents.md
@@ -0,0 +1,60 @@
+# Paper: Self-GC: Self-Governing Context for Long-Horizon LLM Agents
+
+---
+type: paper
+title: "Self-GC: Self-Governing Context for Long-Horizon LLM Agents"
+authors: Xubin Hao, Hongjin Meng, Xin Yin, Jiawei Zhu, Chenpeng Cao
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.00692
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: llm-agent, planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent, planning-agent
+- inferred topics: planning, rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.00692
diff --git a/papers/items/2026-2607-00911-from-registry-to-repository-how-ai-agent-skills-are-written-adapted-and-maintain.md b/papers/items/2026-2607-00911-from-registry-to-repository-how-ai-agent-skills-are-written-adapted-and-maintain.md
new file mode 100644
index 0000000..f8c3d92
--- /dev/null
+++ b/papers/items/2026-2607-00911-from-registry-to-repository-how-ai-agent-skills-are-written-adapted-and-maintain.md
@@ -0,0 +1,60 @@
+# Paper: From Registry to Repository: How AI Agent Skills Are Written, Adapted, and Maintained
+
+---
+type: paper
+title: "From Registry to Repository: How AI Agent Skills Are Written, Adapted, and Maintained"
+authors: Haoyu Gao, Jai Lal Lulla, Hong Yi Lin, Sebastian Baltes, Christoph Treude, Mansooreh Zahedi
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.00911
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-06
+status: queued
+relevance: high
+topics:
+  - coding-agent
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: ai-agent, coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent, coding-agent
+- inferred topics: coding-agent, tool-use, workflow-agent
+- arXiv categories: cs.SE
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.00911
diff --git a/papers/items/2026-2607-00918-from-personas-to-plot-character-grounded-multi-agent-story-generation-for-long-f.md b/papers/items/2026-2607-00918-from-personas-to-plot-character-grounded-multi-agent-story-generation-for-long-f.md
new file mode 100644
index 0000000..ec470cf
--- /dev/null
+++ b/papers/items/2026-2607-00918-from-personas-to-plot-character-grounded-multi-agent-story-generation-for-long-f.md
@@ -0,0 +1,62 @@
+# Paper: From Personas to Plot: Character-Grounded Multi-Agent Story Generation for Long-Form Narratives
+
+---
+type: paper
+title: "From Personas to Plot: Character-Grounded Multi-Agent Story Generation for Long-Form Narratives"
+authors: Aayush Aluru, Chloe Ho, Muhammad Hammouri, Kerry Luo, Myra Malik, Ryan Lagasse, Arjun Bahuguna, Vasu Sharma
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.00918
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, multi-agent, tool-use
+- arXiv categories: cs.CL, cs.AI, cs.MA
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.00918
diff --git a/papers/items/2026-2607-00939-leveraging-llm-based-agentic-systems-to-generate-quantum-applications-for-test-o.md b/papers/items/2026-2607-00939-leveraging-llm-based-agentic-systems-to-generate-quantum-applications-for-test-o.md
new file mode 100644
index 0000000..b0d8985
--- /dev/null
+++ b/papers/items/2026-2607-00939-leveraging-llm-based-agentic-systems-to-generate-quantum-applications-for-test-o.md
@@ -0,0 +1,63 @@
+# Paper: Leveraging LLM-Based Agentic Systems to Generate Quantum Applications for Test Optimization
+
+---
+type: paper
+title: Leveraging LLM-Based Agentic Systems to Generate Quantum Applications for Test Optimization
+authors: Ming Tao, Yuechen Li, Tao Yue, Man Zhang, Aitor Arrieta Marcos
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.00939
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - multi-agent
+  - rag
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - quant-ph
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, coding-agent, multi-agent, rag, workflow-agent
+- arXiv categories: cs.SE, quant-ph
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.00939
diff --git a/papers/items/2026-2607-00972-bayesian-uncertainty-propagation-for-agentic-rag-pipelines-a-proof-of-concept-st.md b/papers/items/2026-2607-00972-bayesian-uncertainty-propagation-for-agentic-rag-pipelines-a-proof-of-concept-st.md
new file mode 100644
index 0000000..e303ae6
--- /dev/null
+++ b/papers/items/2026-2607-00972-bayesian-uncertainty-propagation-for-agentic-rag-pipelines-a-proof-of-concept-st.md
@@ -0,0 +1,62 @@
+# Paper: Bayesian Uncertainty Propagation for Agentic RAG Pipelines: A Proof-of-Concept Study on Multi-Hop Question Answering
+
+---
+type: paper
+title: "Bayesian Uncertainty Propagation for Agentic RAG Pipelines: A Proof-of-Concept Study on Multi-Hop Question Answering"
+authors: Louis Donaldson, Connor Walker, Koorosh Aslansefat, Yiannis Papadopoulos
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.00972
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - reasoning
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, planning, rag, reasoning, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.00972
diff --git a/papers/items/2026-2607-00990-swe-doctor-guiding-software-engineering-agents-with-runtime-diagnosis-from-multi.md b/papers/items/2026-2607-00990-swe-doctor-guiding-software-engineering-agents-with-runtime-diagnosis-from-multi.md
new file mode 100644
index 0000000..68825dc
--- /dev/null
+++ b/papers/items/2026-2607-00990-swe-doctor-guiding-software-engineering-agents-with-runtime-diagnosis-from-multi.md
@@ -0,0 +1,62 @@
+# Paper: SWE-Doctor: Guiding Software Engineering Agents with Runtime Diagnosis from Multi-Faceted Bug Reproduction Tests
+
+---
+type: paper
+title: "SWE-Doctor: Guiding Software Engineering Agents with Runtime Diagnosis from Multi-Faceted Bug Reproduction Tests"
+authors: Yaoqi Guo, Yang Liu, Jie M. Zhang, Yun Ma, Yiling Lou, Zhenpeng Chen
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.00990
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - computer-use
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: agent-evaluation, coding-agent, computer-use, rag
+- arXiv categories: cs.SE, cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.00990
diff --git a/papers/items/2026-2607-01047-conversable-complexity-agentic-llm-collectives-as-interpretable-substrates.md b/papers/items/2026-2607-01047-conversable-complexity-agentic-llm-collectives-as-interpretable-substrates.md
new file mode 100644
index 0000000..dcd2125
--- /dev/null
+++ b/papers/items/2026-2607-01047-conversable-complexity-agentic-llm-collectives-as-interpretable-substrates.md
@@ -0,0 +1,59 @@
+# Paper: Conversable Complexity: Agentic LLM Collectives as Interpretable Substrates
+
+---
+type: paper
+title: "Conversable Complexity: Agentic LLM Collectives as Interpretable Substrates"
+authors: Elias Najarro, Ane Espeseth, Eleni Nisioti, Sebastian Risi, Stefano Nichele
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.01047
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - memory
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: memory, tool-use
+- arXiv categories: cs.CL
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.01047
diff --git a/papers/items/2026-2607-01061-agentic-generation-of-verifiable-rules-for-deterministic-self-expanding-reaction.md b/papers/items/2026-2607-01061-agentic-generation-of-verifiable-rules-for-deterministic-self-expanding-reaction.md
new file mode 100644
index 0000000..7b8af78
--- /dev/null
+++ b/papers/items/2026-2607-01061-agentic-generation-of-verifiable-rules-for-deterministic-self-expanding-reaction.md
@@ -0,0 +1,62 @@
+# Paper: Agentic generation of verifiable rules for deterministic, self-expanding reaction classification
+
+---
+type: paper
+title: Agentic generation of verifiable rules for deterministic, self-expanding reaction classification
+authors: Daniel Armstrong, Maarten Dobbelaere, Valentas Olikauskas, Helena Avila, Octavian Susanu, Jérôme Waser, Philippe Schwaller
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.01061
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-05
+status: queued
+relevance: high
+topics:
+  - coding-agent
+  - multi-agent
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: coding-agent, multi-agent, planning, tool-use
+- arXiv categories: cs.AI, cs.CL
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.01061
diff --git a/papers/items/2026-2607-01071-memsyco-bench-benchmarking-sycophancy-in-agent-memory.md b/papers/items/2026-2607-01071-memsyco-bench-benchmarking-sycophancy-in-agent-memory.md
new file mode 100644
index 0000000..f7c87c5
--- /dev/null
+++ b/papers/items/2026-2607-01071-memsyco-bench-benchmarking-sycophancy-in-agent-memory.md
@@ -0,0 +1,62 @@
+# Paper: MemSyco-Bench: Benchmarking Sycophancy in Agent Memory
+
+---
+type: paper
+title: "MemSyco-Bench: Benchmarking Sycophancy in Agent Memory"
+authors: Zhishang Xiang, Zerui Chen, Yunbo Tang, Zhimin Wei, Ruqin Ning, Yujie Lin, Qinggang Zhang, Jinsong Su
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.01071
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.IR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, agent-safety, memory, reasoning
+- arXiv categories: cs.IR, cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.01071
diff --git a/papers/items/2026-2607-01084-can-agents-generalize-to-the-open-world-unveiling-the-fragility-of-static-traini.md b/papers/items/2026-2607-01084-can-agents-generalize-to-the-open-world-unveiling-the-fragility-of-static-traini.md
new file mode 100644
index 0000000..0cf7b1f
--- /dev/null
+++ b/papers/items/2026-2607-01084-can-agents-generalize-to-the-open-world-unveiling-the-fragility-of-static-traini.md
@@ -0,0 +1,61 @@
+# Paper: Can Agents Generalize to the Open World? Unveiling the Fragility of Static Training in Tool Use
+
+---
+type: paper
+title: Can Agents Generalize to the Open World? Unveiling the Fragility of Static Training in Tool Use
+authors: Song-Lin Lv, Weiming Wu, Rui Zhu, Zi-Jian Cheng, Lan-Zhe Guo
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.01084
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: llm-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent, tool-use
+- inferred topics: agent-evaluation, rag, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.01084
diff --git a/papers/items/2026-2607-01120-next-generation-agentic-reinforcement-learning-systems-enable-self-evolving-agen.md b/papers/items/2026-2607-01120-next-generation-agentic-reinforcement-learning-systems-enable-self-evolving-agen.md
new file mode 100644
index 0000000..1005c0e
--- /dev/null
+++ b/papers/items/2026-2607-01120-next-generation-agentic-reinforcement-learning-systems-enable-self-evolving-agen.md
@@ -0,0 +1,61 @@
+# Paper: Next-Generation Agentic Reinforcement Learning Systems Enable Self-Evolving Agents
+
+---
+type: paper
+title: Next-Generation Agentic Reinforcement Learning Systems Enable Self-Evolving Agents
+authors: Ran Yan, Wei Fu, Jiale Li, Shusheng Xu, Zhiyu Mei, Jiaxuan Gao, Jiarui Zhang, Wentai Zhang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.01120
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - coding-agent
+  - planning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.DC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: coding-agent, planning, tool-use, workflow-agent
+- arXiv categories: cs.DC
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.01120
diff --git a/papers/items/2026-2607-01211-are-performance-optimization-benchmarks-reliably-measuring-coding-agents.md b/papers/items/2026-2607-01211-are-performance-optimization-benchmarks-reliably-measuring-coding-agents.md
new file mode 100644
index 0000000..b4d67bc
--- /dev/null
+++ b/papers/items/2026-2607-01211-are-performance-optimization-benchmarks-reliably-measuring-coding-agents.md
@@ -0,0 +1,61 @@
+# Paper: Are Performance-Optimization Benchmarks Reliably Measuring Coding Agents?
+
+---
+type: paper
+title: Are Performance-Optimization Benchmarks Reliably Measuring Coding Agents?
+authors: Zhi Chen, Zhensu Sun, Yuling Shi, David Lo, Lingxiao Jiang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.01211
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: agent-evaluation, coding-agent, rag
+- arXiv categories: cs.SE, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.01211
diff --git a/papers/items/2026-2607-01213-reporescue-an-empirical-study-of-llm-agents-on-whole-repository-compatibility-re.md b/papers/items/2026-2607-01213-reporescue-an-empirical-study-of-llm-agents-on-whole-repository-compatibility-re.md
new file mode 100644
index 0000000..a782bdc
--- /dev/null
+++ b/papers/items/2026-2607-01213-reporescue-an-empirical-study-of-llm-agents-on-whole-repository-compatibility-re.md
@@ -0,0 +1,61 @@
+# Paper: RepoRescue: An Empirical Study of LLM Agents on Whole-Repository Compatibility Rescue
+
+---
+type: paper
+title: "RepoRescue: An Empirical Study of LLM Agents on Whole-Repository Compatibility Rescue"
+authors: Zhihao Lin, Mingyi Zhou, Zhensu Sun, Yizhuo Yang, Renyu Yang, David Lo, Li Li
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.01213
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: agent-evaluation, coding-agent, reasoning, tool-use
+- arXiv categories: cs.SE
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.01213
diff --git a/papers/items/2026-2607-01366-auto-fl-research-agentic-search-for-federated-learning-algorithms.md b/papers/items/2026-2607-01366-auto-fl-research-agentic-search-for-federated-learning-algorithms.md
new file mode 100644
index 0000000..b726e3c
--- /dev/null
+++ b/papers/items/2026-2607-01366-auto-fl-research-agentic-search-for-federated-learning-algorithms.md
@@ -0,0 +1,60 @@
+# Paper: Auto-FL-Research: Agentic Search for Federated Learning Algorithms
+
+---
+type: paper
+title: "Auto-FL-Research: Agentic Search for Federated Learning Algorithms"
+authors: Holger R. Roth, Ziyue Xu, Chester Chen, Daguang Xu, Peter Cnudde, Andrew Feng
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.01366
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agentic-ai, coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, coding-agent
+- inferred topics: agent-evaluation, coding-agent, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.01366
diff --git a/papers/items/2026-2607-01421-risk-architecture-for-ai-native-engineering-teams-an-organizational-framework-fo.md b/papers/items/2026-2607-01421-risk-architecture-for-ai-native-engineering-teams-an-organizational-framework-fo.md
new file mode 100644
index 0000000..f365104
--- /dev/null
+++ b/papers/items/2026-2607-01421-risk-architecture-for-ai-native-engineering-teams-an-organizational-framework-fo.md
@@ -0,0 +1,64 @@
+# Paper: Risk Architecture for AI-Native Engineering Teams: An Organizational Framework for Agentic System Governance
+
+---
+type: paper
+title: "Risk Architecture for AI-Native Engineering Teams: An Organizational Framework for Agentic System Governance"
+authors: Laxmipriya Ganesh Iyer
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.01421
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - computer-use
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-evaluation, agent-safety, coding-agent, computer-use, rag, tool-use
+- arXiv categories: cs.SE, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.01421
diff --git a/papers/items/2026-2607-01510-janus-a-playground-for-user-involved-agentic-permission-management.md b/papers/items/2026-2607-01510-janus-a-playground-for-user-involved-agentic-permission-management.md
new file mode 100644
index 0000000..ca5793c
--- /dev/null
+++ b/papers/items/2026-2607-01510-janus-a-playground-for-user-involved-agentic-permission-management.md
@@ -0,0 +1,61 @@
+# Paper: Janus: a Playground for User-Involved Agentic Permission Management
+
+---
+type: paper
+title: "Janus: a Playground for User-Involved Agentic Permission Management"
+authors: Natalie Grace Brigham, Eugene Bagdasarian, Tadayoshi Kohno, Franziska Roesner
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.01510
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent
+- inferred topics: agent-evaluation, agent-safety, tool-use
+- arXiv categories: cs.AI, cs.CR
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.01510
diff --git a/papers/items/2026-2607-01523-multi-head-recurrent-memory-agents.md b/papers/items/2026-2607-01523-multi-head-recurrent-memory-agents.md
new file mode 100644
index 0000000..8d8e6d1
--- /dev/null
+++ b/papers/items/2026-2607-01523-multi-head-recurrent-memory-agents.md
@@ -0,0 +1,62 @@
+# Paper: Multi-Head Recurrent Memory Agents
+
+---
+type: paper
+title: Multi-Head Recurrent Memory Agents
+authors: Jiatong Li, Samuel Yeh, Sharon Li
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.01523
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, agent-safety, memory
+- arXiv categories: cs.LG, cs.AI, cs.CL
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.01523
diff --git a/papers/items/2026-2607-01531-opine-world-programmatic-world-modeling-with-ontology-error-prioritized-interact.md b/papers/items/2026-2607-01531-opine-world-programmatic-world-modeling-with-ontology-error-prioritized-interact.md
new file mode 100644
index 0000000..46304b4
--- /dev/null
+++ b/papers/items/2026-2607-01531-opine-world-programmatic-world-modeling-with-ontology-error-prioritized-interact.md
@@ -0,0 +1,63 @@
+# Paper: OPINE-World: Programmatic World Modeling with Ontology-error-Prioritized Interactive Exploration
+
+---
+type: paper
+title: "OPINE-World: Programmatic World Modeling with Ontology-error-Prioritized Interactive Exploration"
+authors: David Courtis, Wenhao Li, Scott Sanner
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.01531
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: llm-agent, planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent, planning-agent
+- inferred topics: agent-evaluation, computer-use, planning, tool-use, world-model
+- arXiv categories: cs.AI, cs.LG
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.01531
diff --git a/papers/items/2026-2607-01600-boundary-sync-measuring-communication-induced-representational-coupling-in-multi.md b/papers/items/2026-2607-01600-boundary-sync-measuring-communication-induced-representational-coupling-in-multi.md
new file mode 100644
index 0000000..8646e48
--- /dev/null
+++ b/papers/items/2026-2607-01600-boundary-sync-measuring-communication-induced-representational-coupling-in-multi.md
@@ -0,0 +1,60 @@
+# Paper: BOUNDARY_SYNC: Measuring Communication-Induced Representational Coupling in Multi-Agent LLM Systems
+
+---
+type: paper
+title: "BOUNDARY_SYNC: Measuring Communication-Induced Representational Coupling in Multi-Agent LLM Systems"
+authors: Zewen Liu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.01600
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - multi-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: llm-agent, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent, multi-agent-llm
+- inferred topics: multi-agent, tool-use
+- arXiv categories: cs.LG, cs.CL
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.01600
diff --git a/papers/items/2026-2607-01640-agentflow-building-agent-dependency-graphs-for-static-analysis-of-agent-programs.md b/papers/items/2026-2607-01640-agentflow-building-agent-dependency-graphs-for-static-analysis-of-agent-programs.md
new file mode 100644
index 0000000..e5be2b6
--- /dev/null
+++ b/papers/items/2026-2607-01640-agentflow-building-agent-dependency-graphs-for-static-analysis-of-agent-programs.md
@@ -0,0 +1,63 @@
+# Paper: AgentFlow: Building Agent Dependency Graphs for Static Analysis of Agent Programs
+
+---
+type: paper
+title: "AgentFlow: Building Agent Dependency Graphs for Static Analysis of Agent Programs"
+authors: Shenao Wang, Xinyi Hou, Yanjie Zhao, Xiao Cheng, Haoyu Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.01640
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - multi-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: llm-agent, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent, multi-agent-llm
+- inferred topics: agent-evaluation, agent-safety, memory, multi-agent, tool-use
+- arXiv categories: cs.SE, cs.CR
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.01640
diff --git a/papers/items/2026-2607-01641-when-agents-do-not-stop-uncovering-infinite-agentic-loops-in-llm-agents.md b/papers/items/2026-2607-01641-when-agents-do-not-stop-uncovering-infinite-agentic-loops-in-llm-agents.md
new file mode 100644
index 0000000..db02a8a
--- /dev/null
+++ b/papers/items/2026-2607-01641-when-agents-do-not-stop-uncovering-infinite-agentic-loops-in-llm-agents.md
@@ -0,0 +1,62 @@
+# Paper: When Agents Do Not Stop: Uncovering Infinite Agentic Loops in LLM Agents
+
+---
+type: paper
+title: "When Agents Do Not Stop: Uncovering Infinite Agentic Loops in LLM Agents"
+authors: Xinyi Hou, Shenao Wang, Yanjie Zhao, Haoyu Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.01641
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - planning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 21
+collection_queries: llm-agent, planning-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent, planning-agent, tool-use
+- inferred topics: agent-evaluation, multi-agent, planning, tool-use, workflow-agent
+- arXiv categories: cs.SE
+- collection score: 21
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.01641
diff --git a/papers/items/2026-2607-01661-diverse-evidence-better-forecasts-multi-agent-deliberation-under-information-asy.md b/papers/items/2026-2607-01661-diverse-evidence-better-forecasts-multi-agent-deliberation-under-information-asy.md
new file mode 100644
index 0000000..500ae23
--- /dev/null
+++ b/papers/items/2026-2607-01661-diverse-evidence-better-forecasts-multi-agent-deliberation-under-information-asy.md
@@ -0,0 +1,60 @@
+# Paper: Diverse Evidence, Better Forecasts: Multi-Agent Deliberation Under Information Asymmetry
+
+---
+type: paper
+title: "Diverse Evidence, Better Forecasts: Multi-Agent Deliberation Under Information Asymmetry"
+authors: Yuante Li, Yicheng Tao, Kate Zhang, Taozhi Wang, Gefei Gu, Yaxin Zhou
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.01661
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, multi-agent, reasoning
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.01661
diff --git a/papers/items/2026-2607-01668-verichat-an-agentic-conversational-ai-assistant-for-hardware-security-verificati.md b/papers/items/2026-2607-01668-verichat-an-agentic-conversational-ai-assistant-for-hardware-security-verificati.md
new file mode 100644
index 0000000..a098d6b
--- /dev/null
+++ b/papers/items/2026-2607-01668-verichat-an-agentic-conversational-ai-assistant-for-hardware-security-verificati.md
@@ -0,0 +1,65 @@
+# Paper: VeriChat: An Agentic Conversational AI Assistant for Hardware Security Verification
+
+---
+type: paper
+title: "VeriChat: An Agentic Conversational AI Assistant for Hardware Security Verification"
+authors: Dipayan Saha, Khan Thamid Hasan, Shams Tarek, Sujan Kumar Saha, Mark Tehranipoor, Farimah Farahmandi
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.01668
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - multi-agent
+  - rag
+  - tool-use
+  - workflow-agent
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-evaluation, agent-safety, computer-use, multi-agent, rag, tool-use, workflow-agent, world-model
+- arXiv categories: cs.CR
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.01668
diff --git a/papers/items/2026-2607-01709-comfyclaw-self-evolving-skill-harnesses-for-image-generation-workflows.md b/papers/items/2026-2607-01709-comfyclaw-self-evolving-skill-harnesses-for-image-generation-workflows.md
new file mode 100644
index 0000000..d4c11ac
--- /dev/null
+++ b/papers/items/2026-2607-01709-comfyclaw-self-evolving-skill-harnesses-for-image-generation-workflows.md
@@ -0,0 +1,64 @@
+# Paper: COMFYCLAW: Self-Evolving Skill Harnesses for Image Generation Workflows
+
+---
+type: paper
+title: "COMFYCLAW: Self-Evolving Skill Harnesses for Image Generation Workflows"
+authors: Zongxia Li, Dawei Liu, Fuxiao Liu, Yuhang Zhou, Xiyang Wu, Jingxi Chen, Jing Xie, Xiaomin Wu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.01709
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI, cs.LG
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.01709
diff --git a/papers/items/2026-2607-01766-simworlds-a-multi-agent-system-for-dynamic-3d-scene-creation.md b/papers/items/2026-2607-01766-simworlds-a-multi-agent-system-for-dynamic-3d-scene-creation.md
new file mode 100644
index 0000000..56fe884
--- /dev/null
+++ b/papers/items/2026-2607-01766-simworlds-a-multi-agent-system-for-dynamic-3d-scene-creation.md
@@ -0,0 +1,64 @@
+# Paper: SimWorlds: A Multi-Agent System for Dynamic 3D Scene Creation
+
+---
+type: paper
+title: "SimWorlds: A Multi-Agent System for Dynamic 3D Scene Creation"
+authors: Chunjiang Liu, Xiaoyuan Wang, Haoyu Chen, Yizhou Zhao, Ming-Hsuan Yang, László A. Jeni
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.01766
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - embodied-agent
+  - multi-agent
+  - planning
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: llm-agent, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent, multi-agent-llm
+- inferred topics: agent-evaluation, embodied-agent, multi-agent, planning, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.01766
diff --git a/papers/items/2026-2607-01767-repair-the-amplifier-not-the-symptom-stable-world-model-correction-for-agent-rol.md b/papers/items/2026-2607-01767-repair-the-amplifier-not-the-symptom-stable-world-model-correction-for-agent-rol.md
new file mode 100644
index 0000000..09e2cf7
--- /dev/null
+++ b/papers/items/2026-2607-01767-repair-the-amplifier-not-the-symptom-stable-world-model-correction-for-agent-rol.md
@@ -0,0 +1,62 @@
+# Paper: Repair the Amplifier, Not the Symptom: Stable World-Model Correction for Agent Rollouts
+
+---
+type: paper
+title: "Repair the Amplifier, Not the Symptom: Stable World-Model Correction for Agent Rollouts"
+authors: Xinyuan Song, Zekun Cai
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.01767
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, memory, planning, tool-use, world-model
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.01767
diff --git a/papers/items/2026-2607-01788-krca-an-efficient-root-cause-analysis-system-in-hyper-scale-microservice-systems.md b/papers/items/2026-2607-01788-krca-an-efficient-root-cause-analysis-system-in-hyper-scale-microservice-systems.md
new file mode 100644
index 0000000..00f8b2c
--- /dev/null
+++ b/papers/items/2026-2607-01788-krca-an-efficient-root-cause-analysis-system-in-hyper-scale-microservice-systems.md
@@ -0,0 +1,63 @@
+# Paper: KRCA: An Efficient Root Cause Analysis System in Hyper-Scale Microservice Systems via Agentic AI
+
+---
+type: paper
+title: "KRCA: An Efficient Root Cause Analysis System in Hyper-Scale Microservice Systems via Agentic AI"
+authors: Jiamin Jiang, Jingfei Feng, Yu Luo, Qingliang Zhang, Yongqian Su, Wenwei Gu, Shenglin Zhang, Tianyu Cui, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.01788
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - multi-agent
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-evaluation, memory, multi-agent, rag, reasoning, tool-use
+- arXiv categories: cs.SE
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.01788
diff --git a/papers/items/2026-2607-01793-safety-testing-llm-agents-at-scale-from-risk-discovery-to-evidence-grounded-veri.md b/papers/items/2026-2607-01793-safety-testing-llm-agents-at-scale-from-risk-discovery-to-evidence-grounded-veri.md
new file mode 100644
index 0000000..45f28de
--- /dev/null
+++ b/papers/items/2026-2607-01793-safety-testing-llm-agents-at-scale-from-risk-discovery-to-evidence-grounded-veri.md
@@ -0,0 +1,63 @@
+# Paper: Safety Testing LLM Agents at Scale: From Risk Discovery to Evidence-Grounded Verification
+
+---
+type: paper
+title: "Safety Testing LLM Agents at Scale: From Risk Discovery to Evidence-Grounded Verification"
+authors: Yunhao Feng, Ruixiao Lin, Ming Wen, Qinqin He, Yanming Guo, Yifan Ding, Yutao Wu, Jialuo Chen, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.01793
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-04
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: agent-evaluation, agent-safety, coding-agent, rag, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.01793
diff --git a/papers/items/2026-2607-01812-to-master-an-llm-agent-framework-for-automated-topology-optimization.md b/papers/items/2026-2607-01812-to-master-an-llm-agent-framework-for-automated-topology-optimization.md
new file mode 100644
index 0000000..bcfc8c9
--- /dev/null
+++ b/papers/items/2026-2607-01812-to-master-an-llm-agent-framework-for-automated-topology-optimization.md
@@ -0,0 +1,62 @@
+# Paper: TO-Master: an LLM-agent framework for automated topology optimization
+
+---
+type: paper
+title: "TO-Master: an LLM-agent framework for automated topology optimization"
+authors: Haoju Lin, Wenchang Zhang, Weipeng Xu, Xiang Li, Tian Xu, Tianju Xue
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.01812
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: agent-evaluation, computer-use, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.CE
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.01812
diff --git a/papers/items/2026-2607-01874-skillcoach-self-evolving-rubrics-for-evaluating-and-enhancing-agentic-skill-use.md b/papers/items/2026-2607-01874-skillcoach-self-evolving-rubrics-for-evaluating-and-enhancing-agentic-skill-use.md
new file mode 100644
index 0000000..c73adba
--- /dev/null
+++ b/papers/items/2026-2607-01874-skillcoach-self-evolving-rubrics-for-evaluating-and-enhancing-agentic-skill-use.md
@@ -0,0 +1,64 @@
+# Paper: SkillCoach: Self-Evolving Rubrics for Evaluating and Enhancing Agentic Skill-Use
+
+---
+type: paper
+title: "SkillCoach: Self-Evolving Rubrics for Evaluating and Enhancing Agentic Skill-Use"
+authors: Jiayin Zhu, Kelong Mao, Yudong Guo, Dengbo He, Sulong Xu, Simiu Gu, Yutao Yue
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.01874
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - computer-use
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: agent-evaluation, coding-agent, computer-use, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI, cs.CL
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.01874
diff --git a/papers/items/2026-2607-01916-contextsniper-anttrail-s-token-efficient-code-memory-for-repository-level-progra.md b/papers/items/2026-2607-01916-contextsniper-anttrail-s-token-efficient-code-memory-for-repository-level-progra.md
new file mode 100644
index 0000000..8fa37d0
--- /dev/null
+++ b/papers/items/2026-2607-01916-contextsniper-anttrail-s-token-efficient-code-memory-for-repository-level-progra.md
@@ -0,0 +1,62 @@
+# Paper: ContextSniper: AntTrail's Token-Efficient Code Memory for Repository-Level Program Repair
+
+---
+type: paper
+title: "ContextSniper: AntTrail's Token-Efficient Code Memory for Repository-Level Program Repair"
+authors: Chiwang Luk, Matin Mohammad Najafi, Zhifeng Jia, Wei Yang, Xiuchang Li, Jinwei Zhu, Yang Ren, Lei Chen, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.01916
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-memory, coding-agent, rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory, coding-agent, rag-agent
+- inferred topics: agent-evaluation, coding-agent, memory, rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.01916
diff --git a/papers/items/2026-2607-01929-beyond-textual-repository-exploration-dual-modal-structural-reasoning-for-agenti.md b/papers/items/2026-2607-01929-beyond-textual-repository-exploration-dual-modal-structural-reasoning-for-agenti.md
new file mode 100644
index 0000000..573ad61
--- /dev/null
+++ b/papers/items/2026-2607-01929-beyond-textual-repository-exploration-dual-modal-structural-reasoning-for-agenti.md
@@ -0,0 +1,64 @@
+# Paper: Beyond Textual Repository Exploration: Dual-Modal Structural Reasoning for Agentic Issue Resolution
+
+---
+type: paper
+title: "Beyond Textual Repository Exploration: Dual-Modal Structural Reasoning for Agentic Issue Resolution"
+authors: Jiayi Zhang, Kai Huang, Yang Liu, Chunyang Chen
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.01929
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - embodied-agent
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: coding-agent, function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent, function-calling
+- inferred topics: agent-evaluation, coding-agent, embodied-agent, planning, rag, reasoning, tool-use
+- arXiv categories: cs.SE
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.01929
diff --git a/papers/items/2026-2607-01935-a-tma-decoupling-state-aware-memory-failures-in-long-term-agent-memory.md b/papers/items/2026-2607-01935-a-tma-decoupling-state-aware-memory-failures-in-long-term-agent-memory.md
new file mode 100644
index 0000000..7fdd37a
--- /dev/null
+++ b/papers/items/2026-2607-01935-a-tma-decoupling-state-aware-memory-failures-in-long-term-agent-memory.md
@@ -0,0 +1,60 @@
+# Paper: A-TMA: Decoupling State-Aware Memory Failures in Long-Term Agent Memory
+
+---
+type: paper
+title: "A-TMA: Decoupling State-Aware Memory Failures in Long-Term Agent Memory"
+authors: Zitong Shi, Yixuan Tang, Anthony Kum Hoe Tung
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.01935
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-memory, llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory, llm-agent
+- inferred topics: agent-evaluation, memory, rag
+- arXiv categories: cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.01935
diff --git a/papers/items/2026-2607-02032-pace-a-proxy-for-agentic-capability-evaluation.md b/papers/items/2026-2607-02032-pace-a-proxy-for-agentic-capability-evaluation.md
new file mode 100644
index 0000000..2560228
--- /dev/null
+++ b/papers/items/2026-2607-02032-pace-a-proxy-for-agentic-capability-evaluation.md
@@ -0,0 +1,61 @@
+# Paper: PACE: A Proxy for Agentic Capability Evaluation
+
+---
+type: paper
+title: "PACE: A Proxy for Agentic Capability Evaluation"
+authors: Yueqi Song, Lintang Sutawika, Jiarui Liu, Lindia Tjuatja, Jiayi Geng, Yunze Xiao, Daniel Lee, Aditya Bharat Soni, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.02032
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 21
+collection_queries: agent-evaluation, coding-agent, llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, coding-agent, llm-agent
+- inferred topics: agent-evaluation, coding-agent, reasoning
+- arXiv categories: cs.AI, cs.CL
+- collection score: 21
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.02032
diff --git a/papers/items/2026-2607-02186-ua-chatdev-uncertainty-aware-multi-agent-collaboration-for-reliable-software-dev.md b/papers/items/2026-2607-02186-ua-chatdev-uncertainty-aware-multi-agent-collaboration-for-reliable-software-dev.md
new file mode 100644
index 0000000..96a543d
--- /dev/null
+++ b/papers/items/2026-2607-02186-ua-chatdev-uncertainty-aware-multi-agent-collaboration-for-reliable-software-dev.md
@@ -0,0 +1,62 @@
+# Paper: UA-ChatDev: Uncertainty-Aware Multi-Agent Collaboration for Reliable Software Development
+
+---
+type: paper
+title: "UA-ChatDev: Uncertainty-Aware Multi-Agent Collaboration for Reliable Software Development"
+authors: Temitayo Olamilekan Ogunsusi, Lijun Qian, Xishuang Dong
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.02186
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - multi-agent
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, coding-agent, multi-agent, rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.02186
diff --git a/papers/items/2026-2607-02210-criticality-based-guard-rail-validation-for-ai-agent-decisions-in-autonomous-tel.md b/papers/items/2026-2607-02210-criticality-based-guard-rail-validation-for-ai-agent-decisions-in-autonomous-tel.md
new file mode 100644
index 0000000..45af6ce
--- /dev/null
+++ b/papers/items/2026-2607-02210-criticality-based-guard-rail-validation-for-ai-agent-decisions-in-autonomous-tel.md
@@ -0,0 +1,63 @@
+# Paper: Criticality-Based Guard Rail Validation for AI Agent Decisions in Autonomous Telecom Networks
+
+---
+type: paper
+title: Criticality-Based Guard Rail Validation for AI Agent Decisions in Autonomous Telecom Networks
+authors: Ravi Kant Sharma
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.02210
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.NI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent
+- inferred topics: agent-evaluation, agent-safety, multi-agent, rag, tool-use
+- arXiv categories: cs.AI, cs.NI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.02210
diff --git a/papers/items/2026-2607-02245-copewell-a-multi-agent-swarm-architecture-for-equitable-mental-wellness-support.md b/papers/items/2026-2607-02245-copewell-a-multi-agent-swarm-architecture-for-equitable-mental-wellness-support.md
new file mode 100644
index 0000000..e1e2817
--- /dev/null
+++ b/papers/items/2026-2607-02245-copewell-a-multi-agent-swarm-architecture-for-equitable-mental-wellness-support.md
@@ -0,0 +1,62 @@
+# Paper: Copewell: A Multi-Agent Swarm Architecture for Equitable Mental Wellness Support
+
+---
+type: paper
+title: "Copewell: A Multi-Agent Swarm Architecture for Equitable Mental Wellness Support"
+authors: Seren Yenikent, Jack Vinijtrongjit, Katherine Ng
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.02245
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CY
+  - cs.HC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent
+- inferred topics: agent-evaluation, agent-safety, multi-agent
+- arXiv categories: cs.AI, cs.CY, cs.HC
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.02245
diff --git a/papers/items/2026-2607-02255-agenticsts-a-bounded-memory-testbed-for-long-horizon-llm-agents.md b/papers/items/2026-2607-02255-agenticsts-a-bounded-memory-testbed-for-long-horizon-llm-agents.md
new file mode 100644
index 0000000..a4e5ce9
--- /dev/null
+++ b/papers/items/2026-2607-02255-agenticsts-a-bounded-memory-testbed-for-long-horizon-llm-agents.md
@@ -0,0 +1,64 @@
+# Paper: AgenticSTS: A Bounded-Memory Testbed for Long-Horizon LLM Agents
+
+---
+type: paper
+title: "AgenticSTS: A Bounded-Memory Testbed for Long-Horizon LLM Agents"
+authors: Xiangchen Cheng, Yunwei Jiang, Jianwen Sun, Zizhen Li, Chuanhao Li, Xiangcheng Cao, Yihao Liu, Fanrui Zhang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.02255
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 23
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: agent-evaluation, memory, planning, rag, reasoning, tool-use
+- arXiv categories: cs.AI, cs.CL
+- collection score: 23
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.02255
diff --git a/papers/items/2026-2607-02294-coding-agents-are-guessing-measuring-action-boundary-violations-in-underspecifie.md b/papers/items/2026-2607-02294-coding-agents-are-guessing-measuring-action-boundary-violations-in-underspecifie.md
new file mode 100644
index 0000000..2138d3a
--- /dev/null
+++ b/papers/items/2026-2607-02294-coding-agents-are-guessing-measuring-action-boundary-violations-in-underspecifie.md
@@ -0,0 +1,61 @@
+# Paper: Coding Agents Are Guessing: Measuring Action-Boundary Violations in Underspecified DevOps Instructions
+
+---
+type: paper
+title: "Coding Agents Are Guessing: Measuring Action-Boundary Violations in Underspecified DevOps Instructions"
+authors: Zimo Ji, Zekai Zhang, Congying Xu, Zongjie Li, Yudong Gao, Shuai Wang, Shing-Chi Cheung
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.02294
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-evaluation, coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, coding-agent
+- inferred topics: agent-evaluation, agent-safety, coding-agent, tool-use
+- arXiv categories: cs.SE
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.02294
diff --git a/papers/items/2026-2607-02381-hulat2-at-mer-trans-2026-governed-multi-agent-simplification-for-spanish-easy-to.md b/papers/items/2026-2607-02381-hulat2-at-mer-trans-2026-governed-multi-agent-simplification-for-spanish-easy-to.md
new file mode 100644
index 0000000..4037037
--- /dev/null
+++ b/papers/items/2026-2607-02381-hulat2-at-mer-trans-2026-governed-multi-agent-simplification-for-spanish-easy-to.md
@@ -0,0 +1,62 @@
+# Paper: HULAT2 at MER-TRANS 2026: Governed Multi-Agent Simplification for Spanish Easy-to-Read Generation
+
+---
+type: paper
+title: "HULAT2 at MER-TRANS 2026: Governed Multi-Agent Simplification for Spanish Easy-to-Read Generation"
+authors: Lourdes Moreno, Paloma Martínez, Marco Antonio Sanchez-Escudero, Miguel Domínguez-Gómez
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.02381
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - multi-agent
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-evaluation, computer-use, multi-agent, tool-use, workflow-agent
+- arXiv categories: cs.CL
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.02381
diff --git a/papers/items/2026-2607-02448-agentscad-automated-design-for-manufacturing-of-fdm-parts-via-multi-agent-llm-re.md b/papers/items/2026-2607-02448-agentscad-automated-design-for-manufacturing-of-fdm-parts-via-multi-agent-llm-re.md
new file mode 100644
index 0000000..4684a53
--- /dev/null
+++ b/papers/items/2026-2607-02448-agentscad-automated-design-for-manufacturing-of-fdm-parts-via-multi-agent-llm-re.md
@@ -0,0 +1,61 @@
+# Paper: AgentsCAD: Automated Design for Manufacturing of FDM Parts via Multi-Agent LLM Reasoning and Geometric Feature Recognition
+
+---
+type: paper
+title: "AgentsCAD: Automated Design for Manufacturing of FDM Parts via Multi-Agent LLM Reasoning and Geometric Feature Recognition"
+authors: Emmanuel George, Christopher Keefe, Peter Pak, Amir Barati Farimani
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.02448
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - reasoning
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, multi-agent, reasoning, workflow-agent
+- arXiv categories: cs.MA
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.02448
diff --git a/papers/items/2026-2607-02453-adoption-and-ecosystem-health-a-longitudinal-analysis-of-open-source-multi-agent.md b/papers/items/2026-2607-02453-adoption-and-ecosystem-health-a-longitudinal-analysis-of-open-source-multi-agent.md
new file mode 100644
index 0000000..c9c5664
--- /dev/null
+++ b/papers/items/2026-2607-02453-adoption-and-ecosystem-health-a-longitudinal-analysis-of-open-source-multi-agent.md
@@ -0,0 +1,59 @@
+# Paper: Adoption and Ecosystem Health: A Longitudinal Analysis of Open-Source Multi-Agent Frameworks
+
+---
+type: paper
+title: "Adoption and Ecosystem Health: A Longitudinal Analysis of Open-Source Multi-Agent Frameworks"
+authors: Xi Zhang, Papi Menon, Vivian Chu, Koray Cosguner
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.02453
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent
+- inferred topics: agent-evaluation, multi-agent
+- arXiv categories: cs.MA
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.02453
diff --git a/papers/items/2026-2607-02507-what-llm-agents-say-when-no-one-is-watching-social-structure-and-latent-objectiv.md b/papers/items/2026-2607-02507-what-llm-agents-say-when-no-one-is-watching-social-structure-and-latent-objectiv.md
new file mode 100644
index 0000000..b5034d5
--- /dev/null
+++ b/papers/items/2026-2607-02507-what-llm-agents-say-when-no-one-is-watching-social-structure-and-latent-objectiv.md
@@ -0,0 +1,63 @@
+# Paper: What LLM Agents Say When No One Is Watching: Social Structure and Latent Objective Emergence in Multi-Agent Debates
+
+---
+type: paper
+title: "What LLM Agents Say When No One Is Watching: Social Structure and Latent Objective Emergence in Multi-Agent Debates"
+authors: Arman Ghaffarizadeh, Danyal Mohaddes, Aliakbar Izadkhah, Shahriar Noroozizadeh
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.02507
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+  - cs.LG
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agent-evaluation, llm-agent, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, llm-agent, multi-agent-llm
+- inferred topics: agent-evaluation, agent-safety, multi-agent
+- arXiv categories: cs.AI, cs.CL, cs.LG, cs.MA
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.02507
diff --git a/papers/items/2026-2607-02577-benchmarking-the-benchmarks-a-validity-audit-of-tool-calling-evaluation.md b/papers/items/2026-2607-02577-benchmarking-the-benchmarks-a-validity-audit-of-tool-calling-evaluation.md
new file mode 100644
index 0000000..22207ea
--- /dev/null
+++ b/papers/items/2026-2607-02577-benchmarking-the-benchmarks-a-validity-audit-of-tool-calling-evaluation.md
@@ -0,0 +1,60 @@
+# Paper: Benchmarking the Benchmarks: A Validity Audit of Tool-Calling Evaluation
+
+---
+type: paper
+title: "Benchmarking the Benchmarks: A Validity Audit of Tool-Calling Evaluation"
+authors: Vishvesh Bhat, Jay Vaghasiya, Muhammad Ahmed Mohsin, Asad Aali
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.02577
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, agent-safety, tool-use
+- arXiv categories: cs.SE
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.02577
diff --git a/papers/items/2026-2607-02579-when-not-to-write-memory-governing-false-promotion-from-correlated-agent-traces.md b/papers/items/2026-2607-02579-when-not-to-write-memory-governing-false-promotion-from-correlated-agent-traces.md
new file mode 100644
index 0000000..15dc234
--- /dev/null
+++ b/papers/items/2026-2607-02579-when-not-to-write-memory-governing-false-promotion-from-correlated-agent-traces.md
@@ -0,0 +1,63 @@
+# Paper: When Not to Write Memory: Governing False Promotion from Correlated Agent Traces
+
+---
+type: paper
+title: "When Not to Write Memory: Governing False Promotion from Correlated Agent Traces"
+authors: Yijiashun Qi, Xiang Xu, Yuxuan Li
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.02579
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-06-30
+updated_at: 2026-06-30
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-memory, language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory, language-agent
+- inferred topics: agent-evaluation, agent-safety, coding-agent, memory, rag, tool-use
+- arXiv categories: cs.SE
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.02579
diff --git a/papers/items/2026-2607-02599-agentltl-a-trace-verification-framework-for-measuring-enforcing-and-training-pro.md b/papers/items/2026-2607-02599-agentltl-a-trace-verification-framework-for-measuring-enforcing-and-training-pro.md
new file mode 100644
index 0000000..9fe1914
--- /dev/null
+++ b/papers/items/2026-2607-02599-agentltl-a-trace-verification-framework-for-measuring-enforcing-and-training-pro.md
@@ -0,0 +1,63 @@
+# Paper: AgentLTL: A Trace-Verification Framework for Measuring, Enforcing, and Training Procedural Compliance in Tool-Using LLM Agents
+
+---
+type: paper
+title: "AgentLTL: A Trace-Verification Framework for Measuring, Enforcing, and Training Procedural Compliance in Tool-Using LLM Agents"
+authors: Laïla Elkoussy, Julien Perez
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.02599
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+  - cs.LO
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: llm-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent, tool-use
+- inferred topics: agent-evaluation, agent-safety, rag, tool-use
+- arXiv categories: cs.SE, cs.AI, cs.LO
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.02599
diff --git a/papers/items/2026-2607-02606-chainswe-benchmarking-coding-agents-on-multi-bug-software-maintenance.md b/papers/items/2026-2607-02606-chainswe-benchmarking-coding-agents-on-multi-bug-software-maintenance.md
new file mode 100644
index 0000000..4b99c29
--- /dev/null
+++ b/papers/items/2026-2607-02606-chainswe-benchmarking-coding-agents-on-multi-bug-software-maintenance.md
@@ -0,0 +1,60 @@
+# Paper: ChainSWE: Benchmarking Coding Agents on Multi-Bug Software Maintenance
+
+---
+type: paper
+title: "ChainSWE: Benchmarking Coding Agents on Multi-Bug Software Maintenance"
+authors: Qirui Jin, Lingching Tung, Kenan Li, Qiyang Shi, Yushi She, Huanzhong Jia, Harrison Zhao, Kejing Xia, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.02606
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: agent-evaluation, coding-agent, workflow-agent
+- arXiv categories: cs.SE
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.02606
diff --git a/papers/items/2026-2607-02684-can-coding-agents-implement-missed-compiler-optimizations-evaluating-llm-agents-.md b/papers/items/2026-2607-02684-can-coding-agents-implement-missed-compiler-optimizations-evaluating-llm-agents-.md
new file mode 100644
index 0000000..1119220
--- /dev/null
+++ b/papers/items/2026-2607-02684-can-coding-agents-implement-missed-compiler-optimizations-evaluating-llm-agents-.md
@@ -0,0 +1,60 @@
+# Paper: Can Coding Agents Implement Missed Compiler Optimizations? Evaluating LLM Agents on LLVM Peephole Optimizations
+
+---
+type: paper
+title: Can Coding Agents Implement Missed Compiler Optimizations? Evaluating LLM Agents on LLVM Peephole Optimizations
+authors: Hongxu Xu, Chunhao Liao, Xintong Zhou, Chengnian Sun
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.02684
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: coding-agent, llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent, llm-agent
+- inferred topics: agent-evaluation, coding-agent, reasoning
+- arXiv categories: cs.SE
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.02684
diff --git a/papers/items/2026-2607-02689-s-ember-a-large-scale-benchmark-for-streaming-egocentric-memory-retrieval.md b/papers/items/2026-2607-02689-s-ember-a-large-scale-benchmark-for-streaming-egocentric-memory-retrieval.md
new file mode 100644
index 0000000..af80040
--- /dev/null
+++ b/papers/items/2026-2607-02689-s-ember-a-large-scale-benchmark-for-streaming-egocentric-memory-retrieval.md
@@ -0,0 +1,63 @@
+# Paper: S-EMBER: A Large-Scale Benchmark for Streaming Egocentric Memory Retrieval
+
+---
+type: paper
+title: "S-EMBER: A Large-Scale Benchmark for Streaming Egocentric Memory Retrieval"
+authors: Xiaodong Wang, Xuanyi Zhao, Pedro Rodriguez, Devendra Singh Sachan, Barlas Oguz, Seungwhan Moon, Shang-Wen Li, Gargi Ghosh, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.02689
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent
+- inferred topics: agent-evaluation, memory, rag, reasoning, tool-use
+- arXiv categories: cs.CV, cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.02689
diff --git a/papers/items/2026-2607-02703-llmoxie-exploring-agentic-ai-for-scientific-software-development.md b/papers/items/2026-2607-02703-llmoxie-exploring-agentic-ai-for-scientific-software-development.md
new file mode 100644
index 0000000..24bfd41
--- /dev/null
+++ b/papers/items/2026-2607-02703-llmoxie-exploring-agentic-ai-for-scientific-software-development.md
@@ -0,0 +1,65 @@
+# Paper: LLMoxie: Exploring Agentic AI for Scientific Software Development
+
+---
+type: paper
+title: "LLMoxie: Exploring Agentic AI for Scientific Software Development"
+authors: Landung Setiawan, Anant Mittal, Cordero Core, Anshul Tambay, Carlos Garcia Jurado Suarez, David A. C. Beck, Andrew J. Connolly, Vani Mandava
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.02703
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - planning
+  - reasoning
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+  - cs.DC
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: agentic-ai, coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, coding-agent
+- inferred topics: agent-evaluation, coding-agent, planning, reasoning, workflow-agent
+- arXiv categories: cs.SE, cs.AI, cs.DC, cs.MA
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.02703
diff --git a/papers/items/2026-2607-02716-evaluating-large-language-models-for-decision-making-in-agent-based-urban-mobili.md b/papers/items/2026-2607-02716-evaluating-large-language-models-for-decision-making-in-agent-based-urban-mobili.md
new file mode 100644
index 0000000..deb50ad
--- /dev/null
+++ b/papers/items/2026-2607-02716-evaluating-large-language-models-for-decision-making-in-agent-based-urban-mobili.md
@@ -0,0 +1,64 @@
+# Paper: Evaluating Large Language Models for Decision-Making in Agent-Based Urban Mobility Simulations
+
+---
+type: paper
+title: Evaluating Large Language Models for Decision-Making in Agent-Based Urban Mobility Simulations
+authors: Bruno Cascaes Alves, Míriam Blank Born, Ulisses Gilioli Francescatto Júnior, Felipe Moura Goulart, Letícia Brandão Caldas, Marilton Sanchotene de Aguiar
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.02716
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - multi-agent
+  - planning
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, computer-use, memory, multi-agent, planning, tool-use, world-model
+- arXiv categories: cs.MA
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.02716
diff --git a/papers/items/2026-2607-02807-swarmresearch-orchestrating-coding-agents-for-open-ended-discovery.md b/papers/items/2026-2607-02807-swarmresearch-orchestrating-coding-agents-for-open-ended-discovery.md
new file mode 100644
index 0000000..a2e905e
--- /dev/null
+++ b/papers/items/2026-2607-02807-swarmresearch-orchestrating-coding-agents-for-open-ended-discovery.md
@@ -0,0 +1,60 @@
+# Paper: SwarmResearch: Orchestrating Coding Agents for Open-Ended Discovery
+
+---
+type: paper
+title: "SwarmResearch: Orchestrating Coding Agents for Open-Ended Discovery"
+authors: Yuvraj Virk, Zack Edds, Chunqiu Steven Xia, Lingming Zhang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.02807
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-02
+updated_at: 2026-07-02
+status: queued
+relevance: high
+topics:
+  - coding-agent
+  - computer-use
+  - multi-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: coding-agent, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent, multi-agent-llm
+- inferred topics: coding-agent, computer-use, multi-agent
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.02807
diff --git a/papers/items/2026-2607-02846-object-centric-environment-modeling-for-agentic-tasks.md b/papers/items/2026-2607-02846-object-centric-environment-modeling-for-agentic-tasks.md
new file mode 100644
index 0000000..3cdde19
--- /dev/null
+++ b/papers/items/2026-2607-02846-object-centric-environment-modeling-for-agentic-tasks.md
@@ -0,0 +1,61 @@
+# Paper: Object-Centric Environment Modeling for Agentic Tasks
+
+---
+type: paper
+title: Object-Centric Environment Modeling for Agentic Tasks
+authors: Yiyang Li, Tianyi Ma, Zehong Wang, Yijun Ma, Yanfang Ye
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.02846
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-03
+updated_at: 2026-07-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: agent-evaluation, rag, tool-use, world-model
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.02846
diff --git a/papers/items/2026-2607-02857-mosaic-knowledge-guided-cli-command-composition-attack-in-llm-coding-agents.md b/papers/items/2026-2607-02857-mosaic-knowledge-guided-cli-command-composition-attack-in-llm-coding-agents.md
new file mode 100644
index 0000000..42a6658
--- /dev/null
+++ b/papers/items/2026-2607-02857-mosaic-knowledge-guided-cli-command-composition-attack-in-llm-coding-agents.md
@@ -0,0 +1,63 @@
+# Paper: MOSAIC: Knowledge-Guided CLI Command Composition Attack in LLM Coding Agents
+
+---
+type: paper
+title: "MOSAIC: Knowledge-Guided CLI Command Composition Attack in LLM Coding Agents"
+authors: Jiangrong Wu, Huaijin Wang, Yihao Zhang, Yuhong Nan, Shuai Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.02857
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-03
+updated_at: 2026-07-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - computer-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-evaluation, coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, coding-agent
+- inferred topics: agent-evaluation, agent-safety, coding-agent, computer-use, workflow-agent
+- arXiv categories: cs.CR, cs.SE
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.02857
diff --git a/papers/items/2026-2607-02879-medcalc-pro-solving-complex-medical-calculations-with-llm-agents.md b/papers/items/2026-2607-02879-medcalc-pro-solving-complex-medical-calculations-with-llm-agents.md
new file mode 100644
index 0000000..96d9db6
--- /dev/null
+++ b/papers/items/2026-2607-02879-medcalc-pro-solving-complex-medical-calculations-with-llm-agents.md
@@ -0,0 +1,59 @@
+# Paper: MedCalc-Pro: Solving Complex Medical Calculations with LLM Agents
+
+---
+type: paper
+title: "MedCalc-Pro: Solving Complex Medical Calculations with LLM Agents"
+authors: Siran Zhao, Ruihui Hou, Ziyue Huai, Chennuo Zhang, Tong Ruan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.02879
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-03
+updated_at: 2026-07-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: agent-evaluation, tool-use
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.02879
diff --git a/papers/items/2026-2607-02882-diagnosis-driven-automatic-repair-for-agentic-workflow-via-symbolic-inference.md b/papers/items/2026-2607-02882-diagnosis-driven-automatic-repair-for-agentic-workflow-via-symbolic-inference.md
new file mode 100644
index 0000000..c7d5551
--- /dev/null
+++ b/papers/items/2026-2607-02882-diagnosis-driven-automatic-repair-for-agentic-workflow-via-symbolic-inference.md
@@ -0,0 +1,61 @@
+# Paper: Diagnosis-Driven Automatic Repair for Agentic Workflow via Symbolic Inference
+
+---
+type: paper
+title: Diagnosis-Driven Automatic Repair for Agentic Workflow via Symbolic Inference
+authors: Xuyan Ma, Yawen Wang, Junjie Wang, Xiaofei Xie, Boyu Wu, Mingyang Li, Dandan Wang, Qing Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.02882
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-03
+updated_at: 2026-07-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-evaluation, planning, tool-use, workflow-agent
+- arXiv categories: cs.SE
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.02882
diff --git a/papers/items/2026-2607-02911-coact-action-preserving-observation-compression-for-coding-agents.md b/papers/items/2026-2607-02911-coact-action-preserving-observation-compression-for-coding-agents.md
new file mode 100644
index 0000000..57e6907
--- /dev/null
+++ b/papers/items/2026-2607-02911-coact-action-preserving-observation-compression-for-coding-agents.md
@@ -0,0 +1,61 @@
+# Paper: CoACT: Action-Preserving Observation Compression for Coding Agents
+
+---
+type: paper
+title: "CoACT: Action-Preserving Observation Compression for Coding Agents"
+authors: Haorui Chen, Yuancheng Zhu, Yitong Zhang, Jia Li
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.02911
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-03
+updated_at: 2026-07-03
+status: queued
+relevance: high
+topics:
+  - coding-agent
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: coding-agent, rag, tool-use
+- arXiv categories: cs.SE, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.02911
diff --git a/papers/items/2026-2607-02927-videosearcher-empowering-video-deep-research-with-multi-tool-agentic-reasoning-v.md b/papers/items/2026-2607-02927-videosearcher-empowering-video-deep-research-with-multi-tool-agentic-reasoning-v.md
new file mode 100644
index 0000000..da431f8
--- /dev/null
+++ b/papers/items/2026-2607-02927-videosearcher-empowering-video-deep-research-with-multi-tool-agentic-reasoning-v.md
@@ -0,0 +1,62 @@
+# Paper: VideoSearcher: Empowering Video Deep Research with Multi-Tool Agentic Reasoning via Reinforcement Learning
+
+---
+type: paper
+title: "VideoSearcher: Empowering Video Deep Research with Multi-Tool Agentic Reasoning via Reinforcement Learning"
+authors: Zhenkun Gao, Yicheng Bao, Jinlong Peng, Xueheng Li, Theo Huang, Bangwei Liu, Kunquan Li, Zhenye Gan, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.02927
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-03
+updated_at: 2026-07-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, rag, reasoning, tool-use
+- arXiv categories: cs.CV, cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.02927
diff --git a/papers/items/2026-2607-02942-a-workflow-aware-serving-layer-for-agentic-applications.md b/papers/items/2026-2607-02942-a-workflow-aware-serving-layer-for-agentic-applications.md
new file mode 100644
index 0000000..1fb9946
--- /dev/null
+++ b/papers/items/2026-2607-02942-a-workflow-aware-serving-layer-for-agentic-applications.md
@@ -0,0 +1,61 @@
+# Paper: A Workflow-Aware Serving Layer for Agentic Applications
+
+---
+type: paper
+title: A Workflow-Aware Serving Layer for Agentic Applications
+authors: Jiayi Qian, Zishen Wan, Hanchen Yang, Chun Tao, Souvik Kundu, Tushar Krishna
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.02942
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-03
+updated_at: 2026-07-03
+status: queued
+relevance: high
+topics:
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.DC
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: reasoning, tool-use, workflow-agent
+- arXiv categories: cs.DC, cs.MA
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.02942
diff --git a/papers/items/2026-2607-03105-orbit-q-dual-axis-benchmarking-of-autonomous-agents-in-scientific-quantum-progra.md b/papers/items/2026-2607-03105-orbit-q-dual-axis-benchmarking-of-autonomous-agents-in-scientific-quantum-progra.md
new file mode 100644
index 0000000..b1f2e1d
--- /dev/null
+++ b/papers/items/2026-2607-03105-orbit-q-dual-axis-benchmarking-of-autonomous-agents-in-scientific-quantum-progra.md
@@ -0,0 +1,60 @@
+# Paper: ORBIT-Q: Dual-axis benchmarking of autonomous agents in scientific quantum programming
+
+---
+type: paper
+title: "ORBIT-Q: Dual-axis benchmarking of autonomous agents in scientific quantum programming"
+authors: Shi-Xin Zhang, Yu-Qin Chen
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.03105
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-03
+updated_at: 2026-07-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - quant-ph
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: agent-evaluation, coding-agent, workflow-agent
+- arXiv categories: quant-ph
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.03105
diff --git a/papers/items/2026-2607-03162-apeb-benchmarking-personalization-ability-of-large-language-model-agents.md b/papers/items/2026-2607-03162-apeb-benchmarking-personalization-ability-of-large-language-model-agents.md
new file mode 100644
index 0000000..60458e9
--- /dev/null
+++ b/papers/items/2026-2607-03162-apeb-benchmarking-personalization-ability-of-large-language-model-agents.md
@@ -0,0 +1,61 @@
+# Paper: APeB: Benchmarking Personalization Ability of Large Language Model Agents
+
+---
+type: paper
+title: "APeB: Benchmarking Personalization Ability of Large Language Model Agents"
+authors: Garry Yang, Zizhe Chen, Xinru Chen, Yongqiang Chen, Jianxiang Wang, Deyu Zou, Linyi Ding, Jialiang Wu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.03162
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-03
+updated_at: 2026-07-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.HC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-evaluation, tool-use, workflow-agent
+- arXiv categories: cs.AI, cs.HC
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.03162
diff --git a/papers/items/2026-2607-03220-contra-red-teaming-configurations-of-personalizable-agents.md b/papers/items/2026-2607-03220-contra-red-teaming-configurations-of-personalizable-agents.md
new file mode 100644
index 0000000..1ab9d25
--- /dev/null
+++ b/papers/items/2026-2607-03220-contra-red-teaming-configurations-of-personalizable-agents.md
@@ -0,0 +1,64 @@
+# Paper: CONTRA: Red-Teaming Configurations of Personalizable Agents
+
+---
+type: paper
+title: "CONTRA: Red-Teaming Configurations of Personalizable Agents"
+authors: Jonathan Nöther, Adish Singla, Goran Radanovic
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.03220
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-03
+updated_at: 2026-07-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-evaluation, agent-safety, coding-agent, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.CR, cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.03220
diff --git a/papers/items/2026-2607-03233-agentic-and-generative-ai-for-open-source-intelligence-and-cyber-investigations-.md b/papers/items/2026-2607-03233-agentic-and-generative-ai-for-open-source-intelligence-and-cyber-investigations-.md
new file mode 100644
index 0000000..b1f980c
--- /dev/null
+++ b/papers/items/2026-2607-03233-agentic-and-generative-ai-for-open-source-intelligence-and-cyber-investigations-.md
@@ -0,0 +1,65 @@
+# Paper: Agentic and Generative AI for Open-Source Intelligence and Cyber Investigations: Taxonomy, Evaluation, Challenges, and Future Directions
+
+---
+type: paper
+title: "Agentic and Generative AI for Open-Source Intelligence and Cyber Investigations: Taxonomy, Evaluation, Challenges, and Future Directions"
+authors: Eduardo Almeida Palmieri, Mohamed Chahine Ghanem, Dipo Dunsin, Zubair Baig, Ed de Quincey, Kim-Kwang Raymond Choo
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.03233
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-03
+updated_at: 2026-07-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+  - cs.IR
+  - cs.SI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 22
+collection_queries: agentic-ai, rag-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, rag-agent, tool-use
+- inferred topics: agent-evaluation, agent-safety, rag, reasoning, tool-use
+- arXiv categories: cs.CR, cs.AI, cs.IR, cs.SI
+- collection score: 22
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.03233
diff --git a/papers/items/2026-2607-03269-agentic-secpbft-agentic-ai-driven-proactive-security-framework-for-wireless-pbft.md b/papers/items/2026-2607-03269-agentic-secpbft-agentic-ai-driven-proactive-security-framework-for-wireless-pbft.md
new file mode 100644
index 0000000..9058a10
--- /dev/null
+++ b/papers/items/2026-2607-03269-agentic-secpbft-agentic-ai-driven-proactive-security-framework-for-wireless-pbft.md
@@ -0,0 +1,63 @@
+# Paper: Agentic-SecPBFT: Agentic AI-Driven Proactive Security Framework for Wireless PBFT Consensus in Mobile Ad-Hoc Networks
+
+---
+type: paper
+title: "Agentic-SecPBFT: Agentic AI-Driven Proactive Security Framework for Wireless PBFT Consensus in Mobile Ad-Hoc Networks"
+authors: Haoxiang Luo, Yinqiu Liu, Ruichen Zhang, Guangyuan Liu, Gang Sun, Hongfang Yu, Zhu Han, Dong In Kim
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.03269
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-03
+updated_at: 2026-07-03
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - computer-use
+  - multi-agent
+  - rag
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.NI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-safety, computer-use, multi-agent, rag, tool-use, world-model
+- arXiv categories: cs.NI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.03269
diff --git a/papers/items/2026-2607-03316-is-agentic-code-review-helpful-mining-developers-feedback-to-coderabbit-reviews-.md b/papers/items/2026-2607-03316-is-agentic-code-review-helpful-mining-developers-feedback-to-coderabbit-reviews-.md
new file mode 100644
index 0000000..a413d42
--- /dev/null
+++ b/papers/items/2026-2607-03316-is-agentic-code-review-helpful-mining-developers-feedback-to-coderabbit-reviews-.md
@@ -0,0 +1,61 @@
+# Paper: Is Agentic Code Review Helpful? Mining Developers' Feedback to CodeRabbit Reviews in the Wild
+
+---
+type: paper
+title: "Is Agentic Code Review Helpful? Mining Developers' Feedback to CodeRabbit Reviews in the Wild"
+authors: Hong Yi Lin, Mingzhao Liang, Kla Tantithamthavorn, Patanamon Thongtanunam
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.03316
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-03
+updated_at: 2026-07-03
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - coding-agent
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: autonomous-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm
+- inferred topics: agent-safety, coding-agent, workflow-agent
+- arXiv categories: cs.SE, cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.03316
diff --git a/papers/items/2026-2607-03333-spork-self-speculative-forking-to-accelerate-agentic-llm-inference.md b/papers/items/2026-2607-03333-spork-self-speculative-forking-to-accelerate-agentic-llm-inference.md
new file mode 100644
index 0000000..2a1c5b8
--- /dev/null
+++ b/papers/items/2026-2607-03333-spork-self-speculative-forking-to-accelerate-agentic-llm-inference.md
@@ -0,0 +1,64 @@
+# Paper: SPORK: Self-Speculative Forking to Accelerate Agentic LLM Inference
+
+---
+type: paper
+title: "SPORK: Self-Speculative Forking to Accelerate Agentic LLM Inference"
+authors: Huajun Bai, Weiwei Lv, Huichuan Zheng, Youyou Lu, Jiwu Shu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.03333
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-03
+updated_at: 2026-07-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.DC
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: agent-evaluation, coding-agent, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.DC, cs.AI, cs.LG
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.03333
diff --git a/papers/items/2026-2607-03423-securing-multi-tool-ai-agent-chains-with-dynamic-real-time-compositional-policie.md b/papers/items/2026-2607-03423-securing-multi-tool-ai-agent-chains-with-dynamic-real-time-compositional-policie.md
new file mode 100644
index 0000000..414fa93
--- /dev/null
+++ b/papers/items/2026-2607-03423-securing-multi-tool-ai-agent-chains-with-dynamic-real-time-compositional-policie.md
@@ -0,0 +1,62 @@
+# Paper: Securing Multi-Tool AI Agent Chains With Dynamic, Real-Time Compositional Policies
+
+---
+type: paper
+title: Securing Multi-Tool AI Agent Chains With Dynamic, Real-Time Compositional Policies
+authors: Chris Schneider, Kriti Faujdar, Philipp Schoenegger, Ben Bariach
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.03423
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-03
+updated_at: 2026-07-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: ai-agent, coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent, coding-agent
+- inferred topics: agent-evaluation, agent-safety, coding-agent, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.03423
diff --git a/papers/items/2026-2607-03441-no-time-like-the-present-agentic-test-time-training-for-llm-agents.md b/papers/items/2026-2607-03441-no-time-like-the-present-agentic-test-time-training-for-llm-agents.md
new file mode 100644
index 0000000..94b5381
--- /dev/null
+++ b/papers/items/2026-2607-03441-no-time-like-the-present-agentic-test-time-training-for-llm-agents.md
@@ -0,0 +1,61 @@
+# Paper: No Time Like the Present: Agentic Test-Time Training for LLM Agents
+
+---
+type: paper
+title: "No Time Like the Present: Agentic Test-Time Training for LLM Agents"
+authors: Yanbo Wang, Jinhua Hao, Yuze Shi, Kun Yuan, Ming Sun
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.03441
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-03
+updated_at: 2026-07-03
+status: queued
+relevance: high
+topics:
+  - coding-agent
+  - computer-use
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: coding-agent, llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent, llm-agent
+- inferred topics: coding-agent, computer-use, tool-use
+- arXiv categories: cs.LG, cs.AI
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.03441
diff --git a/papers/items/2026-2607-03510-cage-1-control-assurance-and-governance-evaluation-for-enterprise-agentic-ai.md b/papers/items/2026-2607-03510-cage-1-control-assurance-and-governance-evaluation-for-enterprise-agentic-ai.md
new file mode 100644
index 0000000..fe97bae
--- /dev/null
+++ b/papers/items/2026-2607-03510-cage-1-control-assurance-and-governance-evaluation-for-enterprise-agentic-ai.md
@@ -0,0 +1,66 @@
+# Paper: CAGE-1: Control, Assurance, and Governance Evaluation for Enterprise Agentic AI
+
+---
+type: paper
+title: "CAGE-1: Control, Assurance, and Governance Evaluation for Enterprise Agentic AI"
+authors: Roopam W. Sure
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.03510
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-03
+updated_at: 2026-07-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - planning
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+  - cs.CY
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-evaluation, agent-safety, memory, planning, rag, tool-use, workflow-agent
+- arXiv categories: cs.SE, cs.AI, cs.CY
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.03510
diff --git a/papers/items/2026-2607-03525-gameenginebench-evaluating-coding-agents-on-real-c-runtime-environments.md b/papers/items/2026-2607-03525-gameenginebench-evaluating-coding-agents-on-real-c-runtime-environments.md
new file mode 100644
index 0000000..3f65e02
--- /dev/null
+++ b/papers/items/2026-2607-03525-gameenginebench-evaluating-coding-agents-on-real-c-runtime-environments.md
@@ -0,0 +1,64 @@
+# Paper: GameEngineBench: Evaluating Coding Agents on Real C++ Runtime Environments
+
+---
+type: paper
+title: "GameEngineBench: Evaluating Coding Agents on Real C++ Runtime Environments"
+authors: Brian La, Sejoon Chang, Ben Kim, Junyoung Bae, Aamish Ahmad Beg, Sei Chang, Gonzalo Gonzalez-Pumariega
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.03525
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-03
+updated_at: 2026-07-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - embodied-agent
+  - memory
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: agent-evaluation, coding-agent, embodied-agent, memory, tool-use, world-model
+- arXiv categories: cs.SE, cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.03525
diff --git a/papers/items/2026-2607-03601-archeval-measuring-ai-agents-as-computer-architects.md b/papers/items/2026-2607-03601-archeval-measuring-ai-agents-as-computer-architects.md
new file mode 100644
index 0000000..99121c9
--- /dev/null
+++ b/papers/items/2026-2607-03601-archeval-measuring-ai-agents-as-computer-architects.md
@@ -0,0 +1,62 @@
+# Paper: ArchEval: Measuring AI Agents as Computer Architects
+
+---
+type: paper
+title: "ArchEval: Measuring AI Agents as Computer Architects"
+authors: Chenyu Wang, Zishen Wan, Jeffrey Ma, Shvetank Prakash, Zhenting Qi, Haebin Do, Andy Cheng, Arya Tschand, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.03601
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-03
+updated_at: 2026-07-03
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 21
+collection_queries: ai-agent, llm-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent, llm-agent, tool-use
+- inferred topics: agent-evaluation, memory, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AR
+- collection score: 21
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.03601
diff --git a/papers/items/2026-2607-03628-swarm-driven-multi-agent-reasoning-for-smart-city-security.md b/papers/items/2026-2607-03628-swarm-driven-multi-agent-reasoning-for-smart-city-security.md
new file mode 100644
index 0000000..27b7a0f
--- /dev/null
+++ b/papers/items/2026-2607-03628-swarm-driven-multi-agent-reasoning-for-smart-city-security.md
@@ -0,0 +1,62 @@
+# Paper: Swarm-Driven Multi-Agent Reasoning for Smart City Security
+
+---
+type: paper
+title: Swarm-Driven Multi-Agent Reasoning for Smart City Security
+authors: Saeid Jamshidi, Kawser Wazed Nafi, Carol Fung, Foutse Khomh
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.03628
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-03
+updated_at: 2026-07-03
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - multi-agent
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-safety, multi-agent, reasoning, tool-use
+- arXiv categories: cs.CR, cs.MA
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.03628
diff --git a/papers/items/2026-2607-03691-don-t-blame-the-large-language-model-how-scaffolding-evolution-shapes-coding-age.md b/papers/items/2026-2607-03691-don-t-blame-the-large-language-model-how-scaffolding-evolution-shapes-coding-age.md
new file mode 100644
index 0000000..0ade46b
--- /dev/null
+++ b/papers/items/2026-2607-03691-don-t-blame-the-large-language-model-how-scaffolding-evolution-shapes-coding-age.md
@@ -0,0 +1,63 @@
+# Paper: Don't Blame the Large Language Model: How Scaffolding Evolution Shapes Coding Agent Quality
+
+---
+type: paper
+title: "Don't Blame the Large Language Model: How Scaffolding Evolution Shapes Coding Agent Quality"
+authors: Oussama Ben Sghaier, Hao Li, Bram Adams, Ahmed E. Hassan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.03691
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-04
+updated_at: 2026-07-04
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: agent-evaluation, coding-agent, reasoning, tool-use
+- arXiv categories: cs.SE, cs.AI, cs.LG
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.03691
diff --git a/papers/items/2026-2607-03695-social-networks-of-llm-agents.md b/papers/items/2026-2607-03695-social-networks-of-llm-agents.md
new file mode 100644
index 0000000..9efec1c
--- /dev/null
+++ b/papers/items/2026-2607-03695-social-networks-of-llm-agents.md
@@ -0,0 +1,59 @@
+# Paper: Social Networks of LLM Agents
+
+---
+type: paper
+title: Social Networks of LLM Agents
+authors: Kaixuan Liu, Guojun Xiong, Weinan Zhang, Shengpu Tang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.03695
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-04
+updated_at: 2026-07-04
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: llm-agent, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent, multi-agent-llm
+- inferred topics: agent-evaluation, multi-agent
+- arXiv categories: cs.LG
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.03695
diff --git a/papers/items/2026-2607-03702-agent-reinforcement-learning-via-pivotal-aware-self-feedback-retry.md b/papers/items/2026-2607-03702-agent-reinforcement-learning-via-pivotal-aware-self-feedback-retry.md
new file mode 100644
index 0000000..01525b3
--- /dev/null
+++ b/papers/items/2026-2607-03702-agent-reinforcement-learning-via-pivotal-aware-self-feedback-retry.md
@@ -0,0 +1,62 @@
+# Paper: Agent Reinforcement Learning via Pivotal-Aware Self-Feedback Retry
+
+---
+type: paper
+title: Agent Reinforcement Learning via Pivotal-Aware Self-Feedback Retry
+authors: Weiyang Guo, Zesheng Shi, Longhui Zhang, Zeen Zhu, Min Zhang, Jing Li
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.03702
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-04
+updated_at: 2026-07-04
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: agent-evaluation, planning, rag, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.03702
diff --git a/papers/items/2026-2607-03726-selfmem-self-optimizing-memory-for-ai-agents.md b/papers/items/2026-2607-03726-selfmem-self-optimizing-memory-for-ai-agents.md
new file mode 100644
index 0000000..8b91d1b
--- /dev/null
+++ b/papers/items/2026-2607-03726-selfmem-self-optimizing-memory-for-ai-agents.md
@@ -0,0 +1,63 @@
+# Paper: SelfMem: Self-Optimizing Memory for AI Agents
+
+---
+type: paper
+title: "SelfMem: Self-Optimizing Memory for AI Agents"
+authors: Shu Yang, Junchao Wu, Derek F. Wong, Di Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.03726
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-04
+updated_at: 2026-07-04
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: agent-memory, ai-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory, ai-agent, tool-use
+- inferred topics: agent-evaluation, computer-use, memory, planning, rag, tool-use
+- arXiv categories: cs.CL
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.03726
diff --git a/papers/items/2026-2607-03821-dualview-preventing-indirect-prompt-injection-in-personal-ai-agents.md b/papers/items/2026-2607-03821-dualview-preventing-indirect-prompt-injection-in-personal-ai-agents.md
new file mode 100644
index 0000000..6b82d24
--- /dev/null
+++ b/papers/items/2026-2607-03821-dualview-preventing-indirect-prompt-injection-in-personal-ai-agents.md
@@ -0,0 +1,61 @@
+# Paper: DualView: Preventing Indirect Prompt Injection in Personal AI Agents
+
+---
+type: paper
+title: "DualView: Preventing Indirect Prompt Injection in Personal AI Agents"
+authors: Juhee Kim, Woohyuk Choi, Taehyun Kang, Youngmin Kim, Byoungyoung Lee
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.03821
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-04
+updated_at: 2026-07-04
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent
+- inferred topics: agent-evaluation, agent-safety, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.03821
diff --git a/papers/items/2026-2607-03853-cograd-a-cognitively-inspired-multi-agent-framework-for-radiology-report-generat.md b/papers/items/2026-2607-03853-cograd-a-cognitively-inspired-multi-agent-framework-for-radiology-report-generat.md
new file mode 100644
index 0000000..6af619a
--- /dev/null
+++ b/papers/items/2026-2607-03853-cograd-a-cognitively-inspired-multi-agent-framework-for-radiology-report-generat.md
@@ -0,0 +1,61 @@
+# Paper: CogRad: A Cognitively-Inspired Multi-Agent Framework for Radiology Report Generation
+
+---
+type: paper
+title: "CogRad: A Cognitively-Inspired Multi-Agent Framework for Radiology Report Generation"
+authors: Saif Ur Rehman Khan, Hasaan Maqsood, Sebastian Vollmer, Andreas Dengel, Muhammad Nabeel Asim
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.03853
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-04
+updated_at: 2026-07-04
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, multi-agent, rag, reasoning
+- arXiv categories: cs.CV
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.03853
diff --git a/papers/items/2026-2607-03953-the-remarkable-effectiveness-of-providing-ai-agents-with-natural-language-tools-.md b/papers/items/2026-2607-03953-the-remarkable-effectiveness-of-providing-ai-agents-with-natural-language-tools-.md
new file mode 100644
index 0000000..8dd32f2
--- /dev/null
+++ b/papers/items/2026-2607-03953-the-remarkable-effectiveness-of-providing-ai-agents-with-natural-language-tools-.md
@@ -0,0 +1,63 @@
+# Paper: The Remarkable Effectiveness of Providing AI Agents with Natural Language Tools: A Replication Study Validating NLT Performance Across 14 Models
+
+---
+type: paper
+title: "The Remarkable Effectiveness of Providing AI Agents with Natural Language Tools: A Replication Study Validating NLT Performance Across 14 Models"
+authors: Alexander Somma, Isabelle Plante, Fred Premji
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.03953
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-04
+updated_at: 2026-07-04
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: agentic-ai, ai-agent, llm-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, ai-agent, llm-agent, tool-use
+- inferred topics: agent-evaluation, rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.CL, cs.AI
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.03953
diff --git a/papers/items/2026-2607-03968-refused-in-chat-written-in-code-workflow-level-jailbreak-construction-in-ide-cod.md b/papers/items/2026-2607-03968-refused-in-chat-written-in-code-workflow-level-jailbreak-construction-in-ide-cod.md
new file mode 100644
index 0000000..655a6df
--- /dev/null
+++ b/papers/items/2026-2607-03968-refused-in-chat-written-in-code-workflow-level-jailbreak-construction-in-ide-cod.md
@@ -0,0 +1,62 @@
+# Paper: Refused in Chat, Written in Code: Workflow-Level Jailbreak Construction in IDE Coding Agents
+
+---
+type: paper
+title: "Refused in Chat, Written in Code: Workflow-Level Jailbreak Construction in IDE Coding Agents"
+authors: Abhishek Kumar, Carsten Maple
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.03968
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-04
+updated_at: 2026-07-04
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: agent-evaluation, agent-safety, coding-agent, workflow-agent
+- arXiv categories: cs.SE, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.03968
diff --git a/papers/items/2026-2607-04009-physminer-an-agentic-ai-framework-for-discovering-turbulence-physics.md b/papers/items/2026-2607-04009-physminer-an-agentic-ai-framework-for-discovering-turbulence-physics.md
new file mode 100644
index 0000000..cbd02f8
--- /dev/null
+++ b/papers/items/2026-2607-04009-physminer-an-agentic-ai-framework-for-discovering-turbulence-physics.md
@@ -0,0 +1,61 @@
+# Paper: PhysMiner: An Agentic AI Framework for Discovering Turbulence Physics
+
+---
+type: paper
+title: "PhysMiner: An Agentic AI Framework for Discovering Turbulence Physics"
+authors: Jiawei Chen, Han Gao, Ping He
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.04009
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-04
+updated_at: 2026-07-04
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - physics.flu-dyn
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-evaluation, memory, reasoning, tool-use
+- arXiv categories: physics.flu-dyn
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.04009
diff --git a/papers/items/2026-2607-04034-the-i-don-t-know-filter-enhancing-agentic-reliability-in-function-calling.md b/papers/items/2026-2607-04034-the-i-don-t-know-filter-enhancing-agentic-reliability-in-function-calling.md
new file mode 100644
index 0000000..d80a6e7
--- /dev/null
+++ b/papers/items/2026-2607-04034-the-i-don-t-know-filter-enhancing-agentic-reliability-in-function-calling.md
@@ -0,0 +1,61 @@
+# Paper: The "I Don't Know" Filter: Enhancing Agentic Reliability in Function Calling
+
+---
+type: paper
+title: "The \"I Don't Know\" Filter: Enhancing Agentic Reliability in Function Calling"
+authors: Stefan Broecker, Mason del Rosario, Boris Selitser, Thomas Strohmer
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.04034
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-04
+updated_at: 2026-07-04
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-evaluation, function-calling
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, function-calling
+- inferred topics: agent-evaluation, rag, tool-use
+- arXiv categories: cs.SE, cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.04034
diff --git a/papers/items/2026-2607-04089-placemem-toward-a-compute-aware-memory-plane-for-lifelong-agents.md b/papers/items/2026-2607-04089-placemem-toward-a-compute-aware-memory-plane-for-lifelong-agents.md
new file mode 100644
index 0000000..cffa3a0
--- /dev/null
+++ b/papers/items/2026-2607-04089-placemem-toward-a-compute-aware-memory-plane-for-lifelong-agents.md
@@ -0,0 +1,61 @@
+# Paper: PLACEMEM: Toward a Compute-Aware Memory Plane for Lifelong Agents
+
+---
+type: paper
+title: "PLACEMEM: Toward a Compute-Aware Memory Plane for Lifelong Agents"
+authors: Sukanta Ganguly
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.04089
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-05
+updated_at: 2026-07-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - planning
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agent-memory
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory
+- inferred topics: agent-evaluation, memory, planning, rag
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.04089
diff --git a/papers/items/2026-2607-04149-beyond-scene-priors-fine-grained-traffic-scene-reasoning-with-benchmarking-and-q.md b/papers/items/2026-2607-04149-beyond-scene-priors-fine-grained-traffic-scene-reasoning-with-benchmarking-and-q.md
new file mode 100644
index 0000000..8751728
--- /dev/null
+++ b/papers/items/2026-2607-04149-beyond-scene-priors-fine-grained-traffic-scene-reasoning-with-benchmarking-and-q.md
@@ -0,0 +1,63 @@
+# Paper: Beyond Scene Priors: Fine-Grained Traffic Scene Reasoning with Benchmarking and Query-Guided Small-Object Focus
+
+---
+type: paper
+title: "Beyond Scene Priors: Fine-Grained Traffic Scene Reasoning with Benchmarking and Query-Guided Small-Object Focus"
+authors: Waikit Xiu, Qiang Lu, Zian Wang, Xinjie Yang, Zhiwei Chen, Chen Sun, Xiying Li
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.04149
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-05
+updated_at: 2026-07-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - computer-use
+  - multi-agent
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, agent-safety, coding-agent, computer-use, multi-agent, reasoning
+- arXiv categories: cs.CV
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.04149
diff --git a/papers/items/2026-2607-04162-ace-agentic-control-for-embodied-manipulation-via-zero-shot-workflow-reasoning.md b/papers/items/2026-2607-04162-ace-agentic-control-for-embodied-manipulation-via-zero-shot-workflow-reasoning.md
new file mode 100644
index 0000000..4217453
--- /dev/null
+++ b/papers/items/2026-2607-04162-ace-agentic-control-for-embodied-manipulation-via-zero-shot-workflow-reasoning.md
@@ -0,0 +1,66 @@
+# Paper: ACE: Agentic Control for Embodied Manipulation via Zero-shot Workflow Reasoning
+
+---
+type: paper
+title: "ACE: Agentic Control for Embodied Manipulation via Zero-shot Workflow Reasoning"
+authors: Iok Tong Lei, QianZhi Li, Ying Jie Yap, Yujie Zhang, Rui Zhong, Haichao Gui, Xiaolong Liu, Zhidong Deng
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.04162
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-05
+updated_at: 2026-07-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - embodied-agent
+  - memory
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.RO
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-evaluation, embodied-agent, memory, planning, rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.RO, cs.LG
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.04162
diff --git a/papers/items/2026-2607-04212-an-evaluation-of-role-based-multi-agent-code-generation-on-repository-scale-prob.md b/papers/items/2026-2607-04212-an-evaluation-of-role-based-multi-agent-code-generation-on-repository-scale-prob.md
new file mode 100644
index 0000000..251e323
--- /dev/null
+++ b/papers/items/2026-2607-04212-an-evaluation-of-role-based-multi-agent-code-generation-on-repository-scale-prob.md
@@ -0,0 +1,60 @@
+# Paper: An Evaluation of Role-Based Multi-Agent Code Generation on Repository-Scale Problems
+
+---
+type: paper
+title: An Evaluation of Role-Based Multi-Agent Code Generation on Repository-Scale Problems
+authors: Benedetta Donato, Noah Hagar-Dent, Aaron Worsnop, Leonardo Mariani, Valerio Terragni
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.04212
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-05
+updated_at: 2026-07-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - multi-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, coding-agent, multi-agent
+- arXiv categories: cs.SE
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.04212
diff --git a/papers/items/2026-2607-04219-agentic-iot-architectures-applications-and-challenges-toward-the-internet-of-age.md b/papers/items/2026-2607-04219-agentic-iot-architectures-applications-and-challenges-toward-the-internet-of-age.md
new file mode 100644
index 0000000..95999eb
--- /dev/null
+++ b/papers/items/2026-2607-04219-agentic-iot-architectures-applications-and-challenges-toward-the-internet-of-age.md
@@ -0,0 +1,63 @@
+# Paper: Agentic IoT: Architectures, Applications, and Challenges Toward the Internet of Agents
+
+---
+type: paper
+title: "Agentic IoT: Architectures, Applications, and Challenges Toward the Internet of Agents"
+authors: Rümeysa Hilal Sevinç, Bahaeddin Türkoğlu, İbrahim Kök
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.04219
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-05
+updated_at: 2026-07-05
+status: queued
+relevance: high
+topics:
+  - multi-agent
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.MA
+  - cs.NI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: ai-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent, tool-use
+- inferred topics: multi-agent, planning, reasoning, tool-use
+- arXiv categories: cs.AI, cs.MA, cs.NI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.04219
diff --git a/papers/items/2026-2607-04240-biological-motifs-for-agentic-control.md b/papers/items/2026-2607-04240-biological-motifs-for-agentic-control.md
new file mode 100644
index 0000000..4e675b4
--- /dev/null
+++ b/papers/items/2026-2607-04240-biological-motifs-for-agentic-control.md
@@ -0,0 +1,62 @@
+# Paper: Biological Motifs for Agentic Control
+
+---
+type: paper
+title: Biological Motifs for Agentic Control
+authors: Bogdan Banu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.04240
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-05
+updated_at: 2026-07-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - multi-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - q-bio.CB
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: agent-evaluation, autonomous-agent-llm, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, autonomous-agent-llm, multi-agent-llm
+- inferred topics: agent-evaluation, agent-safety, multi-agent, tool-use
+- arXiv categories: cs.AI, q-bio.CB
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.04240
diff --git a/papers/items/2026-2607-04293-causalgame-benchmarking-causal-thinking-of-llm-agents-in-games.md b/papers/items/2026-2607-04293-causalgame-benchmarking-causal-thinking-of-llm-agents-in-games.md
new file mode 100644
index 0000000..36677fe
--- /dev/null
+++ b/papers/items/2026-2607-04293-causalgame-benchmarking-causal-thinking-of-llm-agents-in-games.md
@@ -0,0 +1,64 @@
+# Paper: CausalGame: Benchmarking Causal Thinking of LLM Agents in Games
+
+---
+type: paper
+title: "CausalGame: Benchmarking Causal Thinking of LLM Agents in Games"
+authors: Zhenhao Chen, Yongqiang Chen, Chenxi Liu, Junchi Yu, Xiangchen Song, Zijian Li, Jialin Li, Philip Torr, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.04293
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-05
+updated_at: 2026-07-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+  - cs.LG
+  - stat.ML
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: agent-evaluation, computer-use, planning, reasoning
+- arXiv categories: cs.CL, cs.AI, cs.LG, stat.ML
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.04293
diff --git a/papers/items/2026-2607-04334-do-gui-agents-believe-their-eyes-diagnosing-state-belief-reliance-on-pixels-vers.md b/papers/items/2026-2607-04334-do-gui-agents-believe-their-eyes-diagnosing-state-belief-reliance-on-pixels-vers.md
new file mode 100644
index 0000000..604c506
--- /dev/null
+++ b/papers/items/2026-2607-04334-do-gui-agents-believe-their-eyes-diagnosing-state-belief-reliance-on-pixels-vers.md
@@ -0,0 +1,61 @@
+# Paper: Do GUI Agents Believe Their Eyes? Diagnosing State-Belief Reliance on Pixels versus Structure
+
+---
+type: paper
+title: Do GUI Agents Believe Their Eyes? Diagnosing State-Belief Reliance on Pixels versus Structure
+authors: Guijia Zhang, Harry Yang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.04334
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-05
+updated_at: 2026-07-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: web-gui-agent
+- inferred topics: agent-evaluation, computer-use, rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.04334
diff --git a/papers/items/2026-2607-04391-memory-orchestrated-semantic-system-moss-an-auditable-agentic-memory-architectur.md b/papers/items/2026-2607-04391-memory-orchestrated-semantic-system-moss-an-auditable-agentic-memory-architectur.md
new file mode 100644
index 0000000..22b35da
--- /dev/null
+++ b/papers/items/2026-2607-04391-memory-orchestrated-semantic-system-moss-an-auditable-agentic-memory-architectur.md
@@ -0,0 +1,61 @@
+# Paper: Memory-Orchestrated Semantic System (MOSS): An Auditable Agentic Memory Architecture
+
+---
+type: paper
+title: "Memory-Orchestrated Semantic System (MOSS): An Auditable Agentic Memory Architecture"
+authors: Serge Lacasse, Jérémie Hatier, Alex Baker
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.04391
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-05
+updated_at: 2026-07-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: agent-memory, ai-agent, rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory, ai-agent, rag-agent
+- inferred topics: agent-evaluation, memory, rag, tool-use
+- arXiv categories: cs.CL
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.04391
diff --git a/papers/items/2026-2607-04394-mechmath-agent-team-llm-driven-agents-for-mathematical-research.md b/papers/items/2026-2607-04394-mechmath-agent-team-llm-driven-agents-for-mathematical-research.md
new file mode 100644
index 0000000..cc48000
--- /dev/null
+++ b/papers/items/2026-2607-04394-mechmath-agent-team-llm-driven-agents-for-mathematical-research.md
@@ -0,0 +1,63 @@
+# Paper: MechMath Agent Team: LLM Driven Agents for Mathematical Research
+
+---
+type: paper
+title: "MechMath Agent Team: LLM Driven Agents for Mathematical Research"
+authors: Yichuan Cao, Ruichen Qiu, Junqi Liu, Jiaqi Wang, Dakai Guo, Ruyong Feng, Lihong Zhi, Xiao-Shan Gao
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.04394
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-05
+updated_at: 2026-07-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - planning
+  - rag
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.SC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, multi-agent, planning, rag, reasoning
+- arXiv categories: cs.AI, cs.SC
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.04394
diff --git a/papers/items/2026-2607-04395-nki-agent-domain-specific-fine-tuning-and-agentic-tool-use-for-neuron-kernel-gen.md b/papers/items/2026-2607-04395-nki-agent-domain-specific-fine-tuning-and-agentic-tool-use-for-neuron-kernel-gen.md
new file mode 100644
index 0000000..062cc4b
--- /dev/null
+++ b/papers/items/2026-2607-04395-nki-agent-domain-specific-fine-tuning-and-agentic-tool-use-for-neuron-kernel-gen.md
@@ -0,0 +1,61 @@
+# Paper: NKI-Agent: Domain-Specific Fine-Tuning and Agentic Tool Use for Neuron Kernel Generation
+
+---
+type: paper
+title: "NKI-Agent: Domain-Specific Fine-Tuning and Agentic Tool Use for Neuron Kernel Generation"
+authors: Junjie Tang, Jun Huan, Hao Zhou, Yuhao Zhang, Lin Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.04395
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-05
+updated_at: 2026-07-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, computer-use, memory, tool-use
+- arXiv categories: cs.LG
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.04395
diff --git a/papers/items/2026-2607-04426-ace-brain-0-5-a-unified-embodied-foundational-model-for-physical-agentic-ai.md b/papers/items/2026-2607-04426-ace-brain-0-5-a-unified-embodied-foundational-model-for-physical-agentic-ai.md
new file mode 100644
index 0000000..2acd0ca
--- /dev/null
+++ b/papers/items/2026-2607-04426-ace-brain-0-5-a-unified-embodied-foundational-model-for-physical-agentic-ai.md
@@ -0,0 +1,64 @@
+# Paper: ACE-Brain-0.5: A Unified Embodied Foundational Model for Physical Agentic AI
+
+---
+type: paper
+title: "ACE-Brain-0.5: A Unified Embodied Foundational Model for Physical Agentic AI"
+authors: "ACE-Brain Team, :, Ziyang Gong, Haoming Gu, Zehang Luo, Tianyi Zhang, Tao Tao, Yixiao Chi, et al."
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.04426
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-05
+updated_at: 2026-07-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - embodied-agent
+  - memory
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.RO
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agentic-ai
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai
+- inferred topics: agent-evaluation, embodied-agent, memory, planning, rag, reasoning, tool-use
+- arXiv categories: cs.RO
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.04426
diff --git a/papers/items/2026-2607-04433-autonomous-information-seeking-a-roadmap-for-agentic-recommender-systems.md b/papers/items/2026-2607-04433-autonomous-information-seeking-a-roadmap-for-agentic-recommender-systems.md
new file mode 100644
index 0000000..d3a99c9
--- /dev/null
+++ b/papers/items/2026-2607-04433-autonomous-information-seeking-a-roadmap-for-agentic-recommender-systems.md
@@ -0,0 +1,66 @@
+# Paper: Autonomous Information Seeking: A Roadmap for Agentic Recommender Systems
+
+---
+type: paper
+title: "Autonomous Information Seeking: A Roadmap for Agentic Recommender Systems"
+authors: Xinyu Lin, Yashar Deldjoo, Sunhao Dai, Honghui Bao, Xiaopeng Ye, Fatemeh Nazary, Wenjie Wang, Tommaso Di Noia, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.04433
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-05
+updated_at: 2026-07-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - planning
+  - reasoning
+  - tool-use
+  - workflow-agent
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.IR
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 21
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, agent-safety, memory, planning, reasoning, tool-use, workflow-agent, world-model
+- arXiv categories: cs.IR, cs.CL
+- collection score: 21
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.04433
diff --git a/papers/items/2026-2607-04470-regime-conditional-stabilisation-of-llm-augmented-cooperative-multi-agent-reinfo.md b/papers/items/2026-2607-04470-regime-conditional-stabilisation-of-llm-augmented-cooperative-multi-agent-reinfo.md
new file mode 100644
index 0000000..e0477da
--- /dev/null
+++ b/papers/items/2026-2607-04470-regime-conditional-stabilisation-of-llm-augmented-cooperative-multi-agent-reinfo.md
@@ -0,0 +1,63 @@
+# Paper: Regime-Conditional Stabilisation of LLM-Augmented Cooperative Multi-Agent Reinforcement Learning
+
+---
+type: paper
+title: Regime-Conditional Stabilisation of LLM-Augmented Cooperative Multi-Agent Reinforcement Learning
+authors: Faid Keddouri, Sohaib Houhou, Aissa Boulmerka, Nadir Farhi
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.04470
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-05
+updated_at: 2026-07-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+  - math.OC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, multi-agent, rag, tool-use
+- arXiv categories: cs.LG, cs.AI, math.OC
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.04470
diff --git a/papers/items/2026-2607-04528-measuring-harness-induced-belief-divergence-in-multi-step-llm-agents.md b/papers/items/2026-2607-04528-measuring-harness-induced-belief-divergence-in-multi-step-llm-agents.md
new file mode 100644
index 0000000..58cf0a6
--- /dev/null
+++ b/papers/items/2026-2607-04528-measuring-harness-induced-belief-divergence-in-multi-step-llm-agents.md
@@ -0,0 +1,61 @@
+# Paper: Measuring Harness-Induced Belief Divergence in Multi-Step LLM Agents
+
+---
+type: paper
+title: Measuring Harness-Induced Belief Divergence in Multi-Step LLM Agents
+authors: Haiwen Yi, Xinyuan Song
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.04528
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-05
+updated_at: 2026-07-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-evaluation, llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, llm-agent
+- inferred topics: agent-evaluation, agent-safety, coding-agent, tool-use
+- arXiv categories: cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.04528
diff --git a/papers/items/2026-2607-04569-llms-for-agentic-home-energy-management.md b/papers/items/2026-2607-04569-llms-for-agentic-home-energy-management.md
new file mode 100644
index 0000000..d962db0
--- /dev/null
+++ b/papers/items/2026-2607-04569-llms-for-agentic-home-energy-management.md
@@ -0,0 +1,61 @@
+# Paper: LLMs for Agentic Home Energy Management
+
+---
+type: paper
+title: LLMs for Agentic Home Energy Management
+authors: Sokipriala Jonah
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.04569
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-06
+updated_at: 2026-07-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - eess.SY
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: function-calling, llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling, llm-agent
+- inferred topics: agent-evaluation, agent-safety, rag, tool-use
+- arXiv categories: eess.SY
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.04569
diff --git a/papers/items/2026-2607-04617-mrms-a-multi-resolution-memory-substrate-for-long-lived-ai-agents.md b/papers/items/2026-2607-04617-mrms-a-multi-resolution-memory-substrate-for-long-lived-ai-agents.md
new file mode 100644
index 0000000..cdc2b13
--- /dev/null
+++ b/papers/items/2026-2607-04617-mrms-a-multi-resolution-memory-substrate-for-long-lived-ai-agents.md
@@ -0,0 +1,62 @@
+# Paper: MRMS: A Multi-Resolution Memory Substrate for Long-Lived AI Agents
+
+---
+type: paper
+title: "MRMS: A Multi-Resolution Memory Substrate for Long-Lived AI Agents"
+authors: Jizhizi Li, Amy Shi-Nash
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.04617
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-06
+updated_at: 2026-07-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent
+- inferred topics: agent-evaluation, computer-use, memory, rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.04617
diff --git a/papers/items/2026-2607-04623-can-llms-really-recover-microservice-failures-a-recovery-aware-evaluation-of-dia.md b/papers/items/2026-2607-04623-can-llms-really-recover-microservice-failures-a-recovery-aware-evaluation-of-dia.md
new file mode 100644
index 0000000..40c9bd0
--- /dev/null
+++ b/papers/items/2026-2607-04623-can-llms-really-recover-microservice-failures-a-recovery-aware-evaluation-of-dia.md
@@ -0,0 +1,63 @@
+# Paper: Can LLMs Really Recover Microservice Failures? A Recovery-Aware Evaluation of Diagnosis-to-Action Reasoning
+
+---
+type: paper
+title: Can LLMs Really Recover Microservice Failures? A Recovery-Aware Evaluation of Diagnosis-to-Action Reasoning
+authors: Jiaxing Qi, Zhongzhi Luan, Hongyu Zhang, Shaohan Huang, Carol Fung, Yongxin Tong, Hailong Yang, Depei Qian
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.04623
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-06
+updated_at: 2026-07-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.DC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, planning, rag, reasoning, tool-use
+- arXiv categories: cs.SE, cs.DC
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.04623
diff --git a/papers/items/2026-2607-04686-toolfailbench-diagnosing-tool-use-failures-in-llm-agents.md b/papers/items/2026-2607-04686-toolfailbench-diagnosing-tool-use-failures-in-llm-agents.md
new file mode 100644
index 0000000..222f6c1
--- /dev/null
+++ b/papers/items/2026-2607-04686-toolfailbench-diagnosing-tool-use-failures-in-llm-agents.md
@@ -0,0 +1,62 @@
+# Paper: ToolFailBench: Diagnosing Tool-Use Failures in LLM Agents
+
+---
+type: paper
+title: "ToolFailBench: Diagnosing Tool-Use Failures in LLM Agents"
+authors: Harsh Soni
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.04686
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-06
+updated_at: 2026-07-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: agentic-ai, llm-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, llm-agent, tool-use
+- inferred topics: agent-evaluation, agent-safety, tool-use
+- arXiv categories: cs.CL, cs.AI, cs.SE
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.04686
diff --git a/papers/items/2026-2607-04697-ai-agent-pull-requests-on-github-frequency-structure-and-merge-conflict-rates.md b/papers/items/2026-2607-04697-ai-agent-pull-requests-on-github-frequency-structure-and-merge-conflict-rates.md
new file mode 100644
index 0000000..282983d
--- /dev/null
+++ b/papers/items/2026-2607-04697-ai-agent-pull-requests-on-github-frequency-structure-and-merge-conflict-rates.md
@@ -0,0 +1,60 @@
+# Paper: AI Agent Pull Requests on GitHub: Frequency, Structure, and Merge Conflict Rates
+
+---
+type: paper
+title: "AI Agent Pull Requests on GitHub: Frequency, Structure, and Merge Conflict Rates"
+authors: George Xu, Arjun Subramanian, Nithilan Karthik
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.04697
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-06
+updated_at: 2026-07-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - multi-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: ai-agent, coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent, coding-agent
+- inferred topics: agent-evaluation, coding-agent, multi-agent
+- arXiv categories: cs.SE
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.04697
diff --git a/papers/items/2026-2607-04713-rspo-reward-swap-policy-optimization-for-multi-turn-llm-agents.md b/papers/items/2026-2607-04713-rspo-reward-swap-policy-optimization-for-multi-turn-llm-agents.md
new file mode 100644
index 0000000..4587f2e
--- /dev/null
+++ b/papers/items/2026-2607-04713-rspo-reward-swap-policy-optimization-for-multi-turn-llm-agents.md
@@ -0,0 +1,62 @@
+# Paper: RSPO: Reward-Swap Policy Optimization for Multi-Turn LLM Agents
+
+---
+type: paper
+title: "RSPO: Reward-Swap Policy Optimization for Multi-Turn LLM Agents"
+authors: Qiang Liu, Taian Guo, Ruizhi Qiao, Xing Sun
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.04713
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-06
+updated_at: 2026-07-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-evaluation, llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, llm-agent
+- inferred topics: agent-evaluation, agent-safety, planning, rag
+- arXiv categories: cs.LG, cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.04713
diff --git a/papers/items/2026-2607-04963-stapo-selective-trajectory-aware-policy-optimization-for-llm-agent-training.md b/papers/items/2026-2607-04963-stapo-selective-trajectory-aware-policy-optimization-for-llm-agent-training.md
new file mode 100644
index 0000000..0e084fb
--- /dev/null
+++ b/papers/items/2026-2607-04963-stapo-selective-trajectory-aware-policy-optimization-for-llm-agent-training.md
@@ -0,0 +1,60 @@
+# Paper: STAPO: Selective Trajectory-Aware Policy Optimization for LLM Agent Training
+
+---
+type: paper
+title: "STAPO: Selective Trajectory-Aware Policy Optimization for LLM Agent Training"
+authors: Qiuyi Qi, Tian Liang, Mutian Bao, Jinjian Zhang, Dongnan Liu, Wei Zhou, Linjian Mo, Ming Kong, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.04963
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-06
+updated_at: 2026-07-06
+status: queued
+relevance: high
+topics:
+  - planning
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: planning, rag, tool-use
+- arXiv categories: cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.04963
diff --git a/papers/items/2026-2607-05001-tactic-kg-toward-small-agent-teams-for-cyber-threat-intelligence-knowledge-graph.md b/papers/items/2026-2607-05001-tactic-kg-toward-small-agent-teams-for-cyber-threat-intelligence-knowledge-graph.md
new file mode 100644
index 0000000..5e8bc27
--- /dev/null
+++ b/papers/items/2026-2607-05001-tactic-kg-toward-small-agent-teams-for-cyber-threat-intelligence-knowledge-graph.md
@@ -0,0 +1,64 @@
+# Paper: TACTIC-KG: Toward Small Agent Teams for Cyber Threat Intelligence Knowledge Graph Construction
+
+---
+type: paper
+title: "TACTIC-KG: Toward Small Agent Teams for Cyber Threat Intelligence Knowledge Graph Construction"
+authors: Mouhamed Amine Bouchiha, Gregory Blanc
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.05001
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-06
+updated_at: 2026-07-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+  - cs.LG
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: agent-evaluation, agent-safety, reasoning, tool-use
+- arXiv categories: cs.CR, cs.AI, cs.LG, cs.MA
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.05001
diff --git a/papers/items/2026-2607-05029-your-agent-s-memories-are-not-its-own-forged-reasoning-attacks-on-llm-agent-memo.md b/papers/items/2026-2607-05029-your-agent-s-memories-are-not-its-own-forged-reasoning-attacks-on-llm-agent-memo.md
new file mode 100644
index 0000000..8bae70b
--- /dev/null
+++ b/papers/items/2026-2607-05029-your-agent-s-memories-are-not-its-own-forged-reasoning-attacks-on-llm-agent-memo.md
@@ -0,0 +1,62 @@
+# Paper: Your Agent's Memories Are Not Its Own: Forged Reasoning Attacks on LLM Agent Memory and Defenses
+
+---
+type: paper
+title: "Your Agent's Memories Are Not Its Own: Forged Reasoning Attacks on LLM Agent Memory and Defenses"
+authors: Neeraj Karamchandani, Piyush Nagasubramaniam, Sencun Zhu, Dinghao Wu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.05029
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-06
+updated_at: 2026-07-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agent-memory, llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-memory, llm-agent
+- inferred topics: agent-evaluation, memory, reasoning, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.05029
diff --git a/papers/items/2026-2607-05055-toward-trustworthy-large-language-model-agents-in-healthcare.md b/papers/items/2026-2607-05055-toward-trustworthy-large-language-model-agents-in-healthcare.md
new file mode 100644
index 0000000..e862959
--- /dev/null
+++ b/papers/items/2026-2607-05055-toward-trustworthy-large-language-model-agents-in-healthcare.md
@@ -0,0 +1,62 @@
+# Paper: Toward Trustworthy Large Language Model Agents in Healthcare
+
+---
+type: paper
+title: Toward Trustworthy Large Language Model Agents in Healthcare
+authors: Hadi Hasan, Safaa Salman, Adam Tai Abou Dargham, Ammar Mohanna, Ali Chehab
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.05055
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-06
+updated_at: 2026-07-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: function-calling, rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: function-calling, rag-agent
+- inferred topics: agent-evaluation, agent-safety, rag, tool-use, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.05055
diff --git a/papers/items/2026-2607-05120-agent-data-injection-attacks-are-realistic-threats-to-ai-agents.md b/papers/items/2026-2607-05120-agent-data-injection-attacks-are-realistic-threats-to-ai-agents.md
new file mode 100644
index 0000000..b6809c0
--- /dev/null
+++ b/papers/items/2026-2607-05120-agent-data-injection-attacks-are-realistic-threats-to-ai-agents.md
@@ -0,0 +1,63 @@
+# Paper: Agent Data Injection Attacks are Realistic Threats to AI Agents
+
+---
+type: paper
+title: Agent Data Injection Attacks are Realistic Threats to AI Agents
+authors: Woohyuk Choi, Juhee Kim, Taehyun Kang, Jihyeon Jeong, Luyi Xing, Byoungyoung Lee
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.05120
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-06
+updated_at: 2026-07-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - computer-use
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agent-safety, ai-agent, coding-agent, web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-safety, ai-agent, coding-agent, web-gui-agent
+- inferred topics: agent-evaluation, agent-safety, coding-agent, computer-use, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.05120
diff --git a/papers/items/2026-2607-05132-when-agents-lie-premeditation-persistence-and-exploitation-in-repeated-games.md b/papers/items/2026-2607-05132-when-agents-lie-premeditation-persistence-and-exploitation-in-repeated-games.md
new file mode 100644
index 0000000..f2766d5
--- /dev/null
+++ b/papers/items/2026-2607-05132-when-agents-lie-premeditation-persistence-and-exploitation-in-repeated-games.md
@@ -0,0 +1,62 @@
+# Paper: When Agents Lie: Premeditation, Persistence, and Exploitation in Repeated Games
+
+---
+type: paper
+title: "When Agents Lie: Premeditation, Persistence, and Exploitation in Repeated Games"
+authors: Jerick Shi, Terry Jingcheng Zhang, Bernhard Schölkopf, Vincent Conitzer, Zhijing Jin
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.05132
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-06
+updated_at: 2026-07-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CY
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: autonomous-agent-llm, llm-agent, planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm, llm-agent, planning-agent
+- inferred topics: agent-evaluation, agent-safety, planning, tool-use
+- arXiv categories: cs.CY, cs.CL
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.05132
diff --git a/papers/items/2026-2607-05174-agentgym2-benchmarking-large-language-model-agents-in-de-idealized-real-world-en.md b/papers/items/2026-2607-05174-agentgym2-benchmarking-large-language-model-agents-in-de-idealized-real-world-en.md
new file mode 100644
index 0000000..78b10e4
--- /dev/null
+++ b/papers/items/2026-2607-05174-agentgym2-benchmarking-large-language-model-agents-in-de-idealized-real-world-en.md
@@ -0,0 +1,61 @@
+# Paper: AgentGym2: Benchmarking Large Language Model Agents in De-Idealized Real-World Environments
+
+---
+type: paper
+title: "AgentGym2: Benchmarking Large Language Model Agents in De-Idealized Real-World Environments"
+authors: Zhiheng Xi, Dingwen Yang, Jiaqi Liu, Jixuan Huang, Honglin Guo, Baodai Huang, Tinggang Chen, Qi Zhang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.05174
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-06
+updated_at: 2026-07-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: language-agent, llm-agent, planning-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent, llm-agent, planning-agent
+- inferred topics: agent-evaluation, planning, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.05174
diff --git a/papers/items/2026-2607-05188-latent-programming-horizons-in-coding-agents.md b/papers/items/2026-2607-05188-latent-programming-horizons-in-coding-agents.md
new file mode 100644
index 0000000..a6f9128
--- /dev/null
+++ b/papers/items/2026-2607-05188-latent-programming-horizons-in-coding-agents.md
@@ -0,0 +1,61 @@
+# Paper: Latent Programming Horizons in Coding Agents
+
+---
+type: paper
+title: Latent Programming Horizons in Coding Agents
+authors: André Silva, Han Tu, Martin Monperrus
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.05188
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-06
+updated_at: 2026-07-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: agent-evaluation, coding-agent, reasoning
+- arXiv categories: cs.LG, cs.SE
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.05188
diff --git a/papers/items/2026-2607-05202-evoagentbench-benchmarking-agent-self-evolution-via-ability-transfer.md b/papers/items/2026-2607-05202-evoagentbench-benchmarking-agent-self-evolution-via-ability-transfer.md
new file mode 100644
index 0000000..8aa9b68
--- /dev/null
+++ b/papers/items/2026-2607-05202-evoagentbench-benchmarking-agent-self-evolution-via-ability-transfer.md
@@ -0,0 +1,63 @@
+# Paper: EvoAgentBench: Benchmarking Agent Self-Evolution via Ability Transfer
+
+---
+type: paper
+title: "EvoAgentBench: Benchmarking Agent Self-Evolution via Ability Transfer"
+authors: Xingze Gao, Chuanrui Hu, Hongda Chen, Pengfei Yao, Zhao Wang, Yi Bai, Zhengwei Wu, Yunyun Han, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.05202
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-06
+updated_at: 2026-07-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - computer-use
+  - memory
+  - planning
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 19
+collection_queries: agent-evaluation
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation
+- inferred topics: agent-evaluation, coding-agent, computer-use, memory, planning, reasoning
+- arXiv categories: cs.AI
+- collection score: 19
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.05202
diff --git a/papers/items/2026-2607-05297-metaskill-evolve-recursive-self-improvement-of-llm-agents-via-two-timescale-meta.md b/papers/items/2026-2607-05297-metaskill-evolve-recursive-self-improvement-of-llm-agents-via-two-timescale-meta.md
new file mode 100644
index 0000000..49f4be4
--- /dev/null
+++ b/papers/items/2026-2607-05297-metaskill-evolve-recursive-self-improvement-of-llm-agents-via-two-timescale-meta.md
@@ -0,0 +1,61 @@
+# Paper: MetaSkill-Evolve: Recursive Self-Improvement of LLM Agents via Two-Timescale Meta-Skill Evolution
+
+---
+type: paper
+title: "MetaSkill-Evolve: Recursive Self-Improvement of LLM Agents via Two-Timescale Meta-Skill Evolution"
+authors: Zefeng Wang, Minxi Yan, Jinhe Bi, Sikuan Yan, Volker Tresp, Yunpu Ma
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.05297
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-06
+updated_at: 2026-07-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - planning
+  - reasoning
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: agent-evaluation, llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, llm-agent
+- inferred topics: agent-evaluation, planning, reasoning, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.05297
diff --git a/papers/items/2026-2607-05318-pisas-benchmarking-contextual-integrity-in-multi-user-agentic-systems.md b/papers/items/2026-2607-05318-pisas-benchmarking-contextual-integrity-in-multi-user-agentic-systems.md
new file mode 100644
index 0000000..71ecade
--- /dev/null
+++ b/papers/items/2026-2607-05318-pisas-benchmarking-contextual-integrity-in-multi-user-agentic-systems.md
@@ -0,0 +1,62 @@
+# Paper: PiSAs: Benchmarking Contextual Integrity in Multi-User Agentic Systems
+
+---
+type: paper
+title: "PiSAs: Benchmarking Contextual Integrity in Multi-User Agentic Systems"
+authors: Shubham Gupta, Nazanin Mohammadi Sepahvand, Abhinav Kumar, Cem Subakan, Spandana Gella, Pierre-André Noël, Perouz Taslakian, Eugene Bagdasarian, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.05318
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-06
+updated_at: 2026-07-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - memory
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.MA
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 20
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: agent-evaluation, agent-safety, memory, tool-use
+- arXiv categories: cs.MA, cs.CR
+- collection score: 20
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.05318
diff --git a/papers/items/2026-2607-05363-sovereignpa-bench-evaluating-user-owned-personal-agents-under-evolving-intent-pl.md b/papers/items/2026-2607-05363-sovereignpa-bench-evaluating-user-owned-personal-agents-under-evolving-intent-pl.md
new file mode 100644
index 0000000..8c637d3
--- /dev/null
+++ b/papers/items/2026-2607-05363-sovereignpa-bench-evaluating-user-owned-personal-agents-under-evolving-intent-pl.md
@@ -0,0 +1,63 @@
+# Paper: SovereignPA-Bench: Evaluating User-Owned Personal Agents under Evolving Intent, Platform Mediation, and Consent Constraints
+
+---
+type: paper
+title: "SovereignPA-Bench: Evaluating User-Owned Personal Agents under Evolving Intent, Platform Mediation, and Consent Constraints"
+authors: Dylan Zongmin Liu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.05363
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-06
+updated_at: 2026-07-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - embodied-agent
+  - memory
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-evaluation, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, tool-use
+- inferred topics: agent-evaluation, agent-safety, computer-use, embodied-agent, memory, tool-use
+- arXiv categories: cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.05363
diff --git a/papers/items/2026-2607-05378-compactionrl-reinforcement-learning-with-context-compaction-for-long-horizon-age.md b/papers/items/2026-2607-05378-compactionrl-reinforcement-learning-with-context-compaction-for-long-horizon-age.md
new file mode 100644
index 0000000..46b5067
--- /dev/null
+++ b/papers/items/2026-2607-05378-compactionrl-reinforcement-learning-with-context-compaction-for-long-horizon-age.md
@@ -0,0 +1,60 @@
+# Paper: CompactionRL: Reinforcement Learning with Context Compaction for Long-Horizon Agents
+
+---
+type: paper
+title: "CompactionRL: Reinforcement Learning with Context Compaction for Long-Horizon Agents"
+authors: Yujiang Li, Zhenyu Hou, Yi Jing, Jie Tang, Yuxiao Dong
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.05378
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-06
+updated_at: 2026-07-06
+status: queued
+relevance: high
+topics:
+  - coding-agent
+  - planning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: coding-agent, llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent, llm-agent
+- inferred topics: coding-agent, planning, tool-use
+- arXiv categories: cs.LG
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.05378
diff --git a/papers/items/2026-2607-05391-llm-as-a-verifier-a-general-purpose-verification-framework.md b/papers/items/2026-2607-05391-llm-as-a-verifier-a-general-purpose-verification-framework.md
new file mode 100644
index 0000000..fc7fc6a
--- /dev/null
+++ b/papers/items/2026-2607-05391-llm-as-a-verifier-a-general-purpose-verification-framework.md
@@ -0,0 +1,65 @@
+# Paper: LLM-as-a-Verifier: A General-Purpose Verification Framework
+
+---
+type: paper
+title: "LLM-as-a-Verifier: A General-Purpose Verification Framework"
+authors: Jacky Kwok, Shulu Li, Pranav Atreya, Yuejiang Liu, Yixing Jiang, Chelsea Finn, Marco Pavone, Ion Stoica, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.05391
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-06
+updated_at: 2026-07-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - embodied-agent
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+  - cs.LG
+  - cs.MA
+  - cs.RO
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: agent-evaluation, coding-agent, embodied-agent, reasoning
+- arXiv categories: cs.AI, cs.CL, cs.LG, cs.MA, cs.RO
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.05391
diff --git a/papers/items/2026-2607-05428-charlie-an-on-premise-multi-agent-retrieval-augmented-generation-system-for-evid.md b/papers/items/2026-2607-05428-charlie-an-on-premise-multi-agent-retrieval-augmented-generation-system-for-evid.md
new file mode 100644
index 0000000..ba55ef9
--- /dev/null
+++ b/papers/items/2026-2607-05428-charlie-an-on-premise-multi-agent-retrieval-augmented-generation-system-for-evid.md
@@ -0,0 +1,66 @@
+# Paper: CHARLIE: An On-Premise Multi-Agent Retrieval-Augmented Generation System for Evidential Reasoning in Forensic Science
+
+---
+type: paper
+title: "CHARLIE: An On-Premise Multi-Agent Retrieval-Augmented Generation System for Evidential Reasoning in Forensic Science"
+authors: Leandro D. Carneiro, Andre L. S. Meirelles, Juliano de A. Gomes, Rafael C. A. Cabral
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.05428
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-01
+updated_at: 2026-07-01
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - multi-agent
+  - planning
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.DL
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: rag-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: rag-agent
+- inferred topics: agent-evaluation, memory, multi-agent, planning, rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.DL, cs.AI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.05428
diff --git a/papers/items/2026-2607-05456-prompt-to-paper-agentic-ai-system-for-bioinformatics.md b/papers/items/2026-2607-05456-prompt-to-paper-agentic-ai-system-for-bioinformatics.md
new file mode 100644
index 0000000..3695c17
--- /dev/null
+++ b/papers/items/2026-2607-05456-prompt-to-paper-agentic-ai-system-for-bioinformatics.md
@@ -0,0 +1,64 @@
+# Paper: Prompt-to-Paper: Agentic AI System for Bioinformatics
+
+---
+type: paper
+title: "Prompt-to-Paper: Agentic AI System for Bioinformatics"
+authors: Ramsha Kamran, Maheera Amjad, Zartasha Mustansar, Arsalan Shaukat, Salma Sherbaz, Muhammad U. S. Khan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.05456
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-05
+updated_at: 2026-07-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - multi-agent
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+  - q-bio.QM
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 21
+collection_queries: agentic-ai, coding-agent, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, coding-agent, multi-agent-llm
+- inferred topics: agent-evaluation, coding-agent, multi-agent, rag, tool-use
+- arXiv categories: cs.AI, cs.CL, q-bio.QM
+- collection score: 21
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.05456
diff --git a/papers/items/2026-2607-05458-learning-to-control-llm-agent-harnesses-with-offline-reinforcement-learning.md b/papers/items/2026-2607-05458-learning-to-control-llm-agent-harnesses-with-offline-reinforcement-learning.md
new file mode 100644
index 0000000..6e66037
--- /dev/null
+++ b/papers/items/2026-2607-05458-learning-to-control-llm-agent-harnesses-with-offline-reinforcement-learning.md
@@ -0,0 +1,63 @@
+# Paper: Learning to Control LLM Agent Harnesses with Offline Reinforcement Learning
+
+---
+type: paper
+title: Learning to Control LLM Agent Harnesses with Offline Reinforcement Learning
+authors: Haiwen Yi, Xinyuan Song
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.05458
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-05
+updated_at: 2026-07-05
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.LG
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: agent-evaluation, coding-agent, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.LG, cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.05458
diff --git a/papers/items/2026-2607-05518-aiauthz-off-host-identity-bound-authorization-for-ai-agents.md b/papers/items/2026-2607-05518-aiauthz-off-host-identity-bound-authorization-for-ai-agents.md
new file mode 100644
index 0000000..255c81e
--- /dev/null
+++ b/papers/items/2026-2607-05518-aiauthz-off-host-identity-bound-authorization-for-ai-agents.md
@@ -0,0 +1,61 @@
+# Paper: aiAuthZ: Off-Host, Identity-Bound Authorization for AI Agents
+
+---
+type: paper
+title: "aiAuthZ: Off-Host, Identity-Bound Authorization for AI Agents"
+authors: Sai Varun Kodathala
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.05518
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-06
+updated_at: 2026-07-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: ai-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent
+- inferred topics: agent-evaluation, agent-safety, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.05518
diff --git a/papers/items/2026-2607-05659-agents-with-feelings-personality-and-emotion-in-multi-agent-software-teams.md b/papers/items/2026-2607-05659-agents-with-feelings-personality-and-emotion-in-multi-agent-software-teams.md
new file mode 100644
index 0000000..172ca12
--- /dev/null
+++ b/papers/items/2026-2607-05659-agents-with-feelings-personality-and-emotion-in-multi-agent-software-teams.md
@@ -0,0 +1,61 @@
+# Paper: Agents with Feelings? Personality and Emotion in Multi-Agent Software Teams
+
+---
+type: paper
+title: Agents with Feelings? Personality and Emotion in Multi-Agent Software Teams
+authors: Yunyan Ding, Thomas Zimmermann, Iftekhar Ahmed
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.05659
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-06
+updated_at: 2026-07-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - multi-agent
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: llm-agent, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent, multi-agent-llm
+- inferred topics: agent-evaluation, coding-agent, multi-agent, workflow-agent
+- arXiv categories: cs.SE
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.05659
diff --git a/papers/items/2026-2607-05666-what-do-ai-agents-actually-change-an-empirical-taxonomy-of-mutation-patterns-in-.md b/papers/items/2026-2607-05666-what-do-ai-agents-actually-change-an-empirical-taxonomy-of-mutation-patterns-in-.md
new file mode 100644
index 0000000..fa057ed
--- /dev/null
+++ b/papers/items/2026-2607-05666-what-do-ai-agents-actually-change-an-empirical-taxonomy-of-mutation-patterns-in-.md
@@ -0,0 +1,59 @@
+# Paper: What Do AI Agents Actually Change? An Empirical Taxonomy of Mutation Patterns in Performance-Improving Pull Requests
+
+---
+type: paper
+title: What Do AI Agents Actually Change? An Empirical Taxonomy of Mutation Patterns in Performance-Improving Pull Requests
+authors: Illia Dovhoshliubnyi, Nima Soroush, Ashkan Sami, Alexander Brownlee
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.05666
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-06
+updated_at: 2026-07-06
+status: queued
+relevance: high
+topics:
+  - coding-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: ai-agent, coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: ai-agent, coding-agent
+- inferred topics: coding-agent
+- arXiv categories: cs.SE, cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.05666
diff --git a/papers/items/2026-2607-05677-from-conversation-to-contribution-characterizing-coding-agent-in-open-source-sof.md b/papers/items/2026-2607-05677-from-conversation-to-contribution-characterizing-coding-agent-in-open-source-sof.md
new file mode 100644
index 0000000..163e50b
--- /dev/null
+++ b/papers/items/2026-2607-05677-from-conversation-to-contribution-characterizing-coding-agent-in-open-source-sof.md
@@ -0,0 +1,64 @@
+# Paper: From Conversation to Contribution: Characterizing Coding Agent in Open-Source Software
+
+---
+type: paper
+title: "From Conversation to Contribution: Characterizing Coding Agent in Open-Source Software"
+authors: Zihan Fang, Yueke Zhang, Ningzhi Tang, Collin McMillan, Toby Jia-Jun Li, Yu Huang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.05677
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-06
+updated_at: 2026-07-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - computer-use
+  - multi-agent
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.HC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: agent-evaluation, coding-agent, computer-use, multi-agent, tool-use, workflow-agent
+- arXiv categories: cs.SE, cs.HC
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.05677
diff --git a/papers/items/2026-2607-05690-memory-in-the-loop-in-process-retrieval-as-extendedworking-memory-for-language-a.md b/papers/items/2026-2607-05690-memory-in-the-loop-in-process-retrieval-as-extendedworking-memory-for-language-a.md
new file mode 100644
index 0000000..0fe57af
--- /dev/null
+++ b/papers/items/2026-2607-05690-memory-in-the-loop-in-process-retrieval-as-extendedworking-memory-for-language-a.md
@@ -0,0 +1,62 @@
+# Paper: Memory in the Loop: In-Process Retrieval as ExtendedWorking Memory for Language Agents
+
+---
+type: paper
+title: "Memory in the Loop: In-Process Retrieval as ExtendedWorking Memory for Language Agents"
+authors: Yusuf Khan, Carlo Lipizzi
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.05690
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-06
+updated_at: 2026-07-06
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: language-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: language-agent
+- inferred topics: agent-evaluation, memory, rag, tool-use
+- arXiv categories: cs.AI, cs.CL
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.05690
diff --git a/papers/items/2026-2607-05743-the-balkanization-of-execution-security-research-for-ai-coding-agents-isolation-.md b/papers/items/2026-2607-05743-the-balkanization-of-execution-security-research-for-ai-coding-agents-isolation-.md
new file mode 100644
index 0000000..4f59bc7
--- /dev/null
+++ b/papers/items/2026-2607-05743-the-balkanization-of-execution-security-research-for-ai-coding-agents-isolation-.md
@@ -0,0 +1,62 @@
+# Paper: The Balkanization of Execution-Security Research for AI Coding Agents: Isolation, Access Control, and Time-of-Check-to-Time-of-Use Vulnerabilities
+
+---
+type: paper
+title: "The Balkanization of Execution-Security Research for AI Coding Agents: Isolation, Access Control, and Time-of-Check-to-Time-of-Use Vulnerabilities"
+authors: Mohammadreza Rashidi
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.05743
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-07
+updated_at: 2026-07-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agentic-ai, coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, coding-agent
+- inferred topics: agent-evaluation, agent-safety, coding-agent, tool-use
+- arXiv categories: cs.CR, cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.05743
diff --git a/papers/items/2026-2607-05772-detecting-vulnerability-inducing-commits-via-multi-stage-reasoning-with-llm-base.md b/papers/items/2026-2607-05772-detecting-vulnerability-inducing-commits-via-multi-stage-reasoning-with-llm-base.md
new file mode 100644
index 0000000..9896f02
--- /dev/null
+++ b/papers/items/2026-2607-05772-detecting-vulnerability-inducing-commits-via-multi-stage-reasoning-with-llm-base.md
@@ -0,0 +1,63 @@
+# Paper: Detecting Vulnerability-Inducing Commits via Multi-Stage Reasoning with LLM-Based Agents
+
+---
+type: paper
+title: Detecting Vulnerability-Inducing Commits via Multi-Stage Reasoning with LLM-Based Agents
+authors: Liyou Chen, Hailong Sun, Xiang Gao, Yue Pan
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.05772
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-07
+updated_at: 2026-07-07
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - multi-agent
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-safety, multi-agent, rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.SE
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.05772
diff --git a/papers/items/2026-2607-05773-beyond-static-evaluation-building-simulation-environments-for-scalable-agentic-r.md b/papers/items/2026-2607-05773-beyond-static-evaluation-building-simulation-environments-for-scalable-agentic-r.md
new file mode 100644
index 0000000..0ffa378
--- /dev/null
+++ b/papers/items/2026-2607-05773-beyond-static-evaluation-building-simulation-environments-for-scalable-agentic-r.md
@@ -0,0 +1,61 @@
+# Paper: Beyond Static Evaluation: Building Simulation Environments for Scalable Agentic Reinforcement Learning
+
+---
+type: paper
+title: "Beyond Static Evaluation: Building Simulation Environments for Scalable Agentic Reinforcement Learning"
+authors: Akshay Arora, Ishan Nigam, Ashutosh Aggarwal, Shefali Bansal, Krishna Singh, Sweta Kumari, Nikhil Mittal, Shariq Farhan, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.05773
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-07
+updated_at: 2026-07-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 21
+collection_queries: autonomous-agent-llm, tool-use, web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: autonomous-agent-llm, tool-use, web-gui-agent
+- inferred topics: agent-evaluation, computer-use, tool-use, world-model
+- arXiv categories: cs.AI
+- collection score: 21
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.05773
diff --git a/papers/items/2026-2607-05775-beyond-the-leaderboard-a-synthesis-of-tool-use-planning-and-reasoning-failures-i.md b/papers/items/2026-2607-05775-beyond-the-leaderboard-a-synthesis-of-tool-use-planning-and-reasoning-failures-i.md
new file mode 100644
index 0000000..930596d
--- /dev/null
+++ b/papers/items/2026-2607-05775-beyond-the-leaderboard-a-synthesis-of-tool-use-planning-and-reasoning-failures-i.md
@@ -0,0 +1,65 @@
+# Paper: Beyond the Leaderboard: A Synthesis of Tool-Use, Planning, and Reasoning Failures in Large Language Model Agents
+
+---
+type: paper
+title: "Beyond the Leaderboard: A Synthesis of Tool-Use, Planning, and Reasoning Failures in Large Language Model Agents"
+authors: Wael Albayaydh, Rui Zhao, Ivan Flechais
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.05775
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-07
+updated_at: 2026-07-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - coding-agent
+  - embodied-agent
+  - multi-agent
+  - planning
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 23
+collection_queries: llm-agent, multi-agent-llm, planning-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent, multi-agent-llm, planning-agent, tool-use
+- inferred topics: agent-evaluation, agent-safety, coding-agent, embodied-agent, multi-agent, planning, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 23
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.05775
diff --git a/papers/items/2026-2607-05794-from-passive-retrieval-to-active-memory-navigation-learning-to-use-memory-as-a-s.md b/papers/items/2026-2607-05794-from-passive-retrieval-to-active-memory-navigation-learning-to-use-memory-as-a-s.md
new file mode 100644
index 0000000..625166d
--- /dev/null
+++ b/papers/items/2026-2607-05794-from-passive-retrieval-to-active-memory-navigation-learning-to-use-memory-as-a-s.md
@@ -0,0 +1,63 @@
+# Paper: From Passive Retrieval to Active Memory Navigation: Learning to Use Memory as a Structured Action Space
+
+---
+type: paper
+title: "From Passive Retrieval to Active Memory Navigation: Learning to Use Memory as a Structured Action Space"
+authors: Yue Xu, Yutao Sun, Yihao Liu, Mengyu Zhou, Jiayi Qiao, Lu Ma, Kai Tang, Wenjie Wang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.05794
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-07
+updated_at: 2026-07-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - embodied-agent
+  - memory
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: tool-use
+- inferred topics: agent-evaluation, embodied-agent, memory, rag, reasoning, tool-use
+- arXiv categories: cs.AI
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.05794
diff --git a/papers/items/2026-2607-05805-onnes-a-physics-grounded-multi-agent-llm-simulator-for-cryogenic-fault-diagnosis.md b/papers/items/2026-2607-05805-onnes-a-physics-grounded-multi-agent-llm-simulator-for-cryogenic-fault-diagnosis.md
new file mode 100644
index 0000000..a33917f
--- /dev/null
+++ b/papers/items/2026-2607-05805-onnes-a-physics-grounded-multi-agent-llm-simulator-for-cryogenic-fault-diagnosis.md
@@ -0,0 +1,61 @@
+# Paper: Onnes: A Physics-Grounded Multi-Agent LLM Simulator for Cryogenic Fault Diagnosis in Quantum Computing Infrastructure
+
+---
+type: paper
+title: "Onnes: A Physics-Grounded Multi-Agent LLM Simulator for Cryogenic Fault Diagnosis in Quantum Computing Infrastructure"
+authors: Praneeth Narisetty, Uday Kumar Reddy Kattamanchi, Shiva Nagendra Babu Kore
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.05805
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-07
+updated_at: 2026-07-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.LG
+  - quant-ph
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: llm-agent, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent, multi-agent-llm
+- inferred topics: agent-evaluation, multi-agent
+- arXiv categories: cs.AI, cs.LG, quant-ph
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.05805
diff --git a/papers/items/2026-2607-05915-pcbworld-a-benchmark-environment-for-engine-grounded-pcb-design-automation.md b/papers/items/2026-2607-05915-pcbworld-a-benchmark-environment-for-engine-grounded-pcb-design-automation.md
new file mode 100644
index 0000000..678969c
--- /dev/null
+++ b/papers/items/2026-2607-05915-pcbworld-a-benchmark-environment-for-engine-grounded-pcb-design-automation.md
@@ -0,0 +1,60 @@
+# Paper: PCBWorld: A Benchmark Environment for Engine-Grounded PCB Design Automation
+
+---
+type: paper
+title: "PCBWorld: A Benchmark Environment for Engine-Grounded PCB Design Automation"
+authors: Hyungseok Song, Junseok Park, Won-Seok Choi, Seohui Bae, Han-Seul Jeong, Youngjoon Park, Soonyoung Lee
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.05915
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-07
+updated_at: 2026-07-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: agentic-ai, llm-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, llm-agent, tool-use
+- inferred topics: agent-evaluation, tool-use, workflow-agent
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.05915
diff --git a/papers/items/2026-2607-06000-context-to-execution-integrity-for-llm-agents.md b/papers/items/2026-2607-06000-context-to-execution-integrity-for-llm-agents.md
new file mode 100644
index 0000000..9ce7fb2
--- /dev/null
+++ b/papers/items/2026-2607-06000-context-to-execution-integrity-for-llm-agents.md
@@ -0,0 +1,60 @@
+# Paper: Context-to-Execution Integrity for LLM Agents
+
+---
+type: paper
+title: Context-to-Execution Integrity for LLM Agents
+authors: Igor Santos-Grueiro
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.06000
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-07
+updated_at: 2026-07-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CR
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-evaluation, coding-agent, llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, coding-agent, llm-agent
+- inferred topics: agent-evaluation, coding-agent, tool-use
+- arXiv categories: cs.CR
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.06000
diff --git a/papers/items/2026-2607-06001-information-limits-and-attractor-dynamics-in-economies-of-frontier-llm-agents-a-.md b/papers/items/2026-2607-06001-information-limits-and-attractor-dynamics-in-economies-of-frontier-llm-agents-a-.md
new file mode 100644
index 0000000..e609671
--- /dev/null
+++ b/papers/items/2026-2607-06001-information-limits-and-attractor-dynamics-in-economies-of-frontier-llm-agents-a-.md
@@ -0,0 +1,62 @@
+# Paper: Information Limits and Attractor Dynamics in Economies of Frontier LLM Agents: A Pre-Registered Test
+
+---
+type: paper
+title: "Information Limits and Attractor Dynamics in Economies of Frontier LLM Agents: A Pre-Registered Test"
+authors: Cheng Qian
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.06001
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-07
+updated_at: 2026-07-07
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - multi-agent
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.MA
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: llm-agent, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent, multi-agent-llm
+- inferred topics: agent-safety, multi-agent, reasoning, tool-use
+- arXiv categories: cs.AI, cs.MA
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.06001
diff --git a/papers/items/2026-2607-06008-polyworkbench-benchmarking-multilingual-long-horizon-llm-agents.md b/papers/items/2026-2607-06008-polyworkbench-benchmarking-multilingual-long-horizon-llm-agents.md
new file mode 100644
index 0000000..be00e09
--- /dev/null
+++ b/papers/items/2026-2607-06008-polyworkbench-benchmarking-multilingual-long-horizon-llm-agents.md
@@ -0,0 +1,64 @@
+# Paper: PolyWorkBench: Benchmarking Multilingual Long-Horizon LLM Agents
+
+---
+type: paper
+title: "PolyWorkBench: Benchmarking Multilingual Long-Horizon LLM Agents"
+authors: Hongliang Li, Yijin Liu, Zhiwei Zhang, Zihe Liu, Xinyue Lou, Jinan Xu, Fandong Meng, Kaiyu Huang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.06008
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-07
+updated_at: 2026-07-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 25
+collection_queries: agent-evaluation, llm-agent, planning-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, llm-agent, planning-agent, tool-use
+- inferred topics: agent-evaluation, computer-use, planning, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.AI, cs.CL
+- collection score: 25
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.06008
diff --git a/papers/items/2026-2607-06080-from-blueprint-to-reality-modeling-and-applying-putnam-s-social-capital-theory-w.md b/papers/items/2026-2607-06080-from-blueprint-to-reality-modeling-and-applying-putnam-s-social-capital-theory-w.md
new file mode 100644
index 0000000..54488db
--- /dev/null
+++ b/papers/items/2026-2607-06080-from-blueprint-to-reality-modeling-and-applying-putnam-s-social-capital-theory-w.md
@@ -0,0 +1,64 @@
+# Paper: From Blueprint to Reality: Modeling and Applying Putnam's Social Capital Theory with LLM-based Multi-agent Simulations
+
+---
+type: paper
+title: "From Blueprint to Reality: Modeling and Applying Putnam's Social Capital Theory with LLM-based Multi-agent Simulations"
+authors: Shiyi Ling, Zhi Zheng, Hui Zheng, Wenjun Xue, Feng Ye, Tong Xu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.06080
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-07
+updated_at: 2026-07-07
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - multi-agent
+  - rag
+  - tool-use
+  - world-model
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+  - cs.SI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: llm-agent, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent, multi-agent-llm
+- inferred topics: agent-safety, multi-agent, rag, tool-use, world-model
+- arXiv categories: cs.CL, cs.AI, cs.SI
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.06080
diff --git a/papers/items/2026-2607-06101-agents-that-teach-towards-designing-incidental-learning-back-into-ai-assisted-so.md b/papers/items/2026-2607-06101-agents-that-teach-towards-designing-incidental-learning-back-into-ai-assisted-so.md
new file mode 100644
index 0000000..86fdc04
--- /dev/null
+++ b/papers/items/2026-2607-06101-agents-that-teach-towards-designing-incidental-learning-back-into-ai-assisted-so.md
@@ -0,0 +1,68 @@
+# Paper: Agents That Teach: Towards Designing Incidental Learning Back into AI-Assisted Software Development
+
+---
+type: paper
+title: "Agents That Teach: Towards Designing Incidental Learning Back into AI-Assisted Software Development"
+authors: Rohit Mehra, Samdyuti Suri, Prithviraj K Tagadinamani, Kapil Singi, Vikrant Kaulgud, Adam P. Burden
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.06101
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-07
+updated_at: 2026-07-07
+status: queued
+relevance: high
+topics:
+  - agent-safety
+  - coding-agent
+  - computer-use
+  - multi-agent
+  - rag
+  - reasoning
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+  - cs.CY
+  - cs.HC
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: agent-safety, coding-agent, computer-use, multi-agent, rag, reasoning, tool-use, workflow-agent
+- arXiv categories: cs.SE, cs.AI, cs.CY, cs.HC
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.06101
diff --git a/papers/items/2026-2607-06118-webretriever-a-large-scale-comprehensive-benchmark-for-efficient-web-agent-evalu.md b/papers/items/2026-2607-06118-webretriever-a-large-scale-comprehensive-benchmark-for-efficient-web-agent-evalu.md
new file mode 100644
index 0000000..c89f63f
--- /dev/null
+++ b/papers/items/2026-2607-06118-webretriever-a-large-scale-comprehensive-benchmark-for-efficient-web-agent-evalu.md
@@ -0,0 +1,65 @@
+# Paper: WebRetriever: A Large-Scale Comprehensive Benchmark for Efficient Web Agent Evaluation
+
+---
+type: paper
+title: "WebRetriever: A Large-Scale Comprehensive Benchmark for Efficient Web Agent Evaluation"
+authors: Wei Dong, Tianyu Fu, Zhe Yu, Hanning Wang, Anyang Su, Zhizhou Fang, Yuyang Chen, Shuo Wang, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.06118
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-07
+updated_at: 2026-07-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - agent-safety
+  - computer-use
+  - embodied-agent
+  - rag
+  - tool-use
+  - workflow-agent
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CV
+  - cs.MM
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 21
+collection_queries: agent-evaluation, web-gui-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, web-gui-agent
+- inferred topics: agent-evaluation, agent-safety, computer-use, embodied-agent, rag, tool-use, workflow-agent
+- arXiv categories: cs.CV, cs.MM
+- collection score: 21
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.06118
diff --git a/papers/items/2026-2607-06140-curateevo-data-curation-evolving-for-agentic-post-training.md b/papers/items/2026-2607-06140-curateevo-data-curation-evolving-for-agentic-post-training.md
new file mode 100644
index 0000000..d6094d5
--- /dev/null
+++ b/papers/items/2026-2607-06140-curateevo-data-curation-evolving-for-agentic-post-training.md
@@ -0,0 +1,60 @@
+# Paper: CurateEvo: Data-Curation Evolving for Agentic Post-Training
+
+---
+type: paper
+title: "CurateEvo: Data-Curation Evolving for Agentic Post-Training"
+authors: Dingzirui Wang, Xuanliang Zhang, Keyan Xu, Qingfu Zhu, Wanxiang Che
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.06140
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-07
+updated_at: 2026-07-07
+status: queued
+relevance: high
+topics:
+  - memory
+  - planning
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: memory, planning, rag
+- arXiv categories: cs.CL
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.06140
diff --git a/papers/items/2026-2607-06157-llm-agents-for-deliberative-collaboration-a-study-on-joint-decision-making-under.md b/papers/items/2026-2607-06157-llm-agents-for-deliberative-collaboration-a-study-on-joint-decision-making-under.md
new file mode 100644
index 0000000..b8bf994
--- /dev/null
+++ b/papers/items/2026-2607-06157-llm-agents-for-deliberative-collaboration-a-study-on-joint-decision-making-under.md
@@ -0,0 +1,62 @@
+# Paper: LLM Agents for Deliberative Collaboration: A Study on Joint Decision Making Under Partial Observability
+
+---
+type: paper
+title: "LLM Agents for Deliberative Collaboration: A Study on Joint Decision Making Under Partial Observability"
+authors: Chenxu Wang, Yongkun Yang, Boyuan Du, Shiwei Lin, Huaping Liu
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.06157
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-07
+updated_at: 2026-07-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 18
+collection_queries: llm-agent, multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent, multi-agent-llm
+- inferred topics: agent-evaluation, multi-agent, reasoning, tool-use
+- arXiv categories: cs.CL, cs.AI
+- collection score: 18
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.06157
diff --git a/papers/items/2026-2607-06195-logichunter-testing-llm-agent-frameworks-with-an-agentic-oracle.md b/papers/items/2026-2607-06195-logichunter-testing-llm-agent-frameworks-with-an-agentic-oracle.md
new file mode 100644
index 0000000..986e544
--- /dev/null
+++ b/papers/items/2026-2607-06195-logichunter-testing-llm-agent-frameworks-with-an-agentic-oracle.md
@@ -0,0 +1,60 @@
+# Paper: LogicHunter: Testing LLM Agent Frameworks with an Agentic Oracle
+
+---
+type: paper
+title: "LogicHunter: Testing LLM Agent Frameworks with an Agentic Oracle"
+authors: Minghui Long, Yanjie Zhao, Haoyu Wang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.06195
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-07
+updated_at: 2026-07-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: agent-evaluation, computer-use, memory
+- arXiv categories: cs.SE
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.06195
diff --git a/papers/items/2026-2607-06223-information-gain-based-rollout-policy-optimization-an-adaptive-tree-structured-r.md b/papers/items/2026-2607-06223-information-gain-based-rollout-policy-optimization-an-adaptive-tree-structured-r.md
new file mode 100644
index 0000000..6515f0b
--- /dev/null
+++ b/papers/items/2026-2607-06223-information-gain-based-rollout-policy-optimization-an-adaptive-tree-structured-r.md
@@ -0,0 +1,61 @@
+# Paper: Information Gain-based Rollout Policy Optimization: An Adaptive Tree-Structured Rollout Approach for Multi-Turn LLM Agents
+
+---
+type: paper
+title: "Information Gain-based Rollout Policy Optimization: An Adaptive Tree-Structured Rollout Approach for Multi-Turn LLM Agents"
+authors: Yijun Zhang, Fan Xu, Jiaxin Ding, Yule Xie, Shiqing Gao, Xin Ding, Haoxiang Zhang, Luoyi Fu, et al.
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.06223
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-07
+updated_at: 2026-07-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - planning
+  - rag
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 15
+collection_queries: llm-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent
+- inferred topics: agent-evaluation, computer-use, planning, rag
+- arXiv categories: cs.AI
+- collection score: 15
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.06223
diff --git a/papers/items/2026-2607-06273-agenttether-graph-guided-diagnosis-and-runtime-intervention-for-reliable-llm-age.md b/papers/items/2026-2607-06273-agenttether-graph-guided-diagnosis-and-runtime-intervention-for-reliable-llm-age.md
new file mode 100644
index 0000000..b2cfee5
--- /dev/null
+++ b/papers/items/2026-2607-06273-agenttether-graph-guided-diagnosis-and-runtime-intervention-for-reliable-llm-age.md
@@ -0,0 +1,62 @@
+# Paper: AgentTether: Graph-Guided Diagnosis and Runtime Intervention for Reliable LLM Agent Operation
+
+---
+type: paper
+title: "AgentTether: Graph-Guided Diagnosis and Runtime Intervention for Reliable LLM Agent Operation"
+authors: Chenyu Zhao, Shenglin Zhang, Wenwei Gu, Yongqian Sun, Dan Pei, Chetan Bansal, Saravan Rajmohan, Minghua Ma
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.06273
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-07
+updated_at: 2026-07-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - computer-use
+  - memory
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 17
+collection_queries: llm-agent, tool-use
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: llm-agent, tool-use
+- inferred topics: agent-evaluation, computer-use, memory, reasoning, tool-use
+- arXiv categories: cs.SE
+- collection score: 17
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.06273
diff --git a/papers/items/2026-2607-06341-harnessing-code-agents-for-automatic-software-verification.md b/papers/items/2026-2607-06341-harnessing-code-agents-for-automatic-software-verification.md
new file mode 100644
index 0000000..3ed1aff
--- /dev/null
+++ b/papers/items/2026-2607-06341-harnessing-code-agents-for-automatic-software-verification.md
@@ -0,0 +1,64 @@
+# Paper: Harnessing Code Agents for Automatic Software Verification
+
+---
+type: paper
+title: Harnessing Code Agents for Automatic Software Verification
+authors: Shuangxiang Kan, Shuanglong Kan, Sebastian Ertel
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.06341
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-07
+updated_at: 2026-07-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - memory
+  - rag
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.FL
+  - cs.AI
+  - cs.SE
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 13
+collection_queries: coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: coding-agent
+- inferred topics: agent-evaluation, coding-agent, memory, rag, tool-use
+- arXiv categories: cs.FL, cs.AI, cs.SE
+- collection score: 13
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.06341
diff --git a/papers/items/2026-2607-06411-rubench-a-repository-level-agentic-coding-benchmark-with-natively-authored-russi.md b/papers/items/2026-2607-06411-rubench-a-repository-level-agentic-coding-benchmark-with-natively-authored-russi.md
new file mode 100644
index 0000000..250cd26
--- /dev/null
+++ b/papers/items/2026-2607-06411-rubench-a-repository-level-agentic-coding-benchmark-with-natively-authored-russi.md
@@ -0,0 +1,62 @@
+# Paper: RuBench: A Repository-Level Agentic Coding Benchmark with Natively Authored Russian Task Specifications
+
+---
+type: paper
+title: "RuBench: A Repository-Level Agentic Coding Benchmark with Natively Authored Russian Task Specifications"
+authors: Evgeny Shilov
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.06411
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-07
+updated_at: 2026-07-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.SE
+  - cs.AI
+  - cs.CL
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agent-evaluation, coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agent-evaluation, coding-agent
+- inferred topics: agent-evaluation, coding-agent, reasoning
+- arXiv categories: cs.SE, cs.AI, cs.CL
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.06411
diff --git a/papers/items/2026-2607-06413-an-experimental-design-approach-to-evaluating-agentic-ai-s-autonomous-model-disc.md b/papers/items/2026-2607-06413-an-experimental-design-approach-to-evaluating-agentic-ai-s-autonomous-model-disc.md
new file mode 100644
index 0000000..aede1e9
--- /dev/null
+++ b/papers/items/2026-2607-06413-an-experimental-design-approach-to-evaluating-agentic-ai-s-autonomous-model-disc.md
@@ -0,0 +1,61 @@
+# Paper: An Experimental Design Approach to Evaluating Agentic AI's Autonomous Model Discovery
+
+---
+type: paper
+title: "An Experimental Design Approach to Evaluating Agentic AI's Autonomous Model Discovery"
+authors: Hao He, Xueying Liu, Chris J. Kuhlman, Xinwei Deng
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.06413
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-07
+updated_at: 2026-07-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - coding-agent
+  - reasoning
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - stat.ME
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 16
+collection_queries: agentic-ai, coding-agent
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: agentic-ai, coding-agent
+- inferred topics: agent-evaluation, coding-agent, reasoning
+- arXiv categories: stat.ME, cs.AI
+- collection score: 16
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.06413
diff --git a/papers/items/2026-2607-06452-from-voting-to-agent-collaboration-answer-type-aware-llm-pipelines-for-bioasq-14.md b/papers/items/2026-2607-06452-from-voting-to-agent-collaboration-answer-type-aware-llm-pipelines-for-bioasq-14.md
new file mode 100644
index 0000000..768a8a8
--- /dev/null
+++ b/papers/items/2026-2607-06452-from-voting-to-agent-collaboration-answer-type-aware-llm-pipelines-for-bioasq-14.md
@@ -0,0 +1,63 @@
+# Paper: From Voting to Agent Collaboration: Answer-Type-Aware LLM Pipelines for BioASQ 14b
+
+---
+type: paper
+title: "From Voting to Agent Collaboration: Answer-Type-Aware LLM Pipelines for BioASQ 14b"
+authors: Taeyun Roh, Eunha Lee, Wonjune Jang, Sohyun Chung, Junha Jung, Jaewoo Kang
+year: 2026
+venue: arXiv
+url: https://arxiv.org/abs/2607.06452
+code_url:
+source: arxiv
+collected_at: 2026-07-08
+published_at: 2026-07-07
+updated_at: 2026-07-07
+status: queued
+relevance: high
+topics:
+  - agent-evaluation
+  - multi-agent
+  - rag
+  - reasoning
+  - tool-use
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+  - cs.CL
+  - cs.AI
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: 14
+collection_queries: multi-agent-llm
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: multi-agent-llm
+- inferred topics: agent-evaluation, multi-agent, rag, reasoning, tool-use
+- arXiv categories: cs.CL, cs.AI
+- collection score: 14
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: https://arxiv.org/abs/2607.06452
diff --git a/papers/paper-insights.md b/papers/paper-insights.md
index d6e0ae9..3ad63e2 100644
--- a/papers/paper-insights.md
+++ b/papers/paper-insights.md
@@ -7,9 +7,13 @@ status: seed
 ## Current Snapshot
 
 - date: 2026-07-08
-- scope: Agent evaluation, memory, coding-agent benchmarks
-- papers reviewed: 6 starter items
-- caveat: sources are provisional; source approval workflow still needs user confirmation.
+- scope: arXiv Agent-related papers from 2025-07-08 to 2026-07-08
+- candidates seen: 1506
+- high-relevance candidates selected into paper items: 970
+- reserve candidates retained in manifest: 536
+- total paper items: 974
+- detailed summary: [Paper Corpus Summary](corpus-summary-2026-07-08.md)
+- caveat: high-recall automated sweep; most items are `queued` and need human skim.
 
 ## Strong Signals
 
@@ -17,12 +21,15 @@ status: seed
 - Memory 不能只看事实召回；更重要的是是否改善流程执行和跨任务决策。
 - Coding-agent benchmark 正在从单 issue patch 转向长程软件演进和人机对话。
 - JD 高频词和论文趋势高度重合：memory、eval、planning/tool use、coding agent。
+- 最近一年 Agent 论文的重心明显落在 evaluation、tool use、memory、safety、coding agent、computer-use/GUI agent。
+- 2026-05 到 2026-07 的论文密度明显上升，说明 Agent research 正在快速扩张。
 
 ## Weak or Unproven Claims
 
 - “加 memory 就更强”不成立；EvoMemBench 显示长上下文 baseline 很强，memory 价值依赖任务。
 - 单一 safety benchmark 不足以说明 Agent 安全；不同 benchmark 的方法差异可能改变结论。
 - SWE-Bench 类单点任务不足以代表真实 coding agent 生产力。
+- 自动标签是召回导向，不能直接当作精确分类；后续需要人工把高价值论文升到 `skimmed`。
 
 ## Mature Directions
 
@@ -30,6 +37,8 @@ status: seed
 - Memory benchmark and memory architecture comparison
 - Coding agent benchmark beyond SWE-Bench
 - Agent failure diagnosis and trajectory analysis
+- Runtime governance and agent security
+- Computer-use / GUI agent benchmark
 
 ## Emerging Directions
 
@@ -37,6 +46,10 @@ status: seed
 - Stateful enterprise task benchmarks
 - Agent safety benchmark taxonomy
 - Skill/prompt optimization as trainable artifacts
+- Agent memory poisoning and memory provenance
+- Workflow-agent and enterprise automation benchmarks
+- Multi-agent shared memory and governance
+- Agentic RAG for technical/scientific literature
 
 ## Experiment Candidates
 
@@ -46,10 +59,12 @@ status: seed
 | Memory vs Long Context | Memory survey, EvoMemBench, STATE-Bench | success rate, token cost, latency, contradiction rate | 必须包含 long-context baseline |
 | Long-horizon Coding Eval | SWE-EVO, Dialogue-SWEBench | task success, fix rate, clarification quality | 从真实 repo release notes 构造小任务 |
 | Failure Step Diagnosis | AgentRx, OpenAgentSafety | critical failure step accuracy | 需要统一 trace schema |
+| GUI Agent Safety | OSWorld2.0, MacAgentBench, computer-use papers | irreversible action rate, prompt-injection catch rate | 需要 sandbox 和人工确认策略 |
+| Agent Memory Security | memory poisoning / forged reasoning papers | poisoned recall rate, provenance coverage | 和长期记忆系统强相关 |
 
 ## Links Back to Knowledge Base
 
 - docs to update: evaluation, memory, coding-agent, safety
-- experiments to create: memory comparison, mini safety suite, coding-agent eval
+- experiments to create: memory comparison, mini safety suite, coding-agent eval, GUI-agent safety eval
 - projects affected: future front-end/indexing project, coding-agent experiments
 - job skills connected: agent-evaluation, memory, coding-agent, tool-use, planning
diff --git a/papers/source-registry.md b/papers/source-registry.md
index 3c776b7..9b2ec04 100644
--- a/papers/source-registry.md
+++ b/papers/source-registry.md
@@ -8,7 +8,7 @@
 
 | Source | Status | Notes |
 | --- | --- | --- |
-| arXiv | provisional-used | 新论文发现和预印本；first sweep 已使用，待用户正式确认 |
+| arXiv | provisional-used | 新论文发现和预印本；2026-07-08 扩召回看到 1506 个候选，970 个 high-relevance 入库，536 个留作 reserve |
 | OpenReview | candidate | ICLR、NeurIPS 等投稿和评审线索 |
 | ACL Anthology | candidate | NLP、RAG、tool use、evaluation 相关论文 |
 | Conference proceedings | candidate | NeurIPS、ICLR、ICML、ACL、EMNLP、KDD、WWW 等 |
diff --git a/tools/collection/README.md b/tools/collection/README.md
index a7cc60f..b5efdcb 100644
--- a/tools/collection/README.md
+++ b/tools/collection/README.md
@@ -43,6 +43,26 @@ python3 tools/collection/check_urls.py
 
 可加 `--json` 输出机器可读结果。
 
+### `collect_arxiv.py`
+
+用 arXiv API 批量收集最近一年 Agent 相关论文，生成 `papers/items/` 条目、`data/arxiv-agent-papers-*.json` 入库清单和 `data/arxiv-agent-candidates-*.json` 全量候选清单。
+
+```bash
+python3 tools/collection/collect_arxiv.py \
+  --from-date 2025-07-08 \
+  --to-date 2026-07-08 \
+  --max-items 180 \
+  --sleep 1.0
+```
+
+默认策略：
+
+- 多组关键词召回：LLM agent、language agent、tool use、memory、evaluation、coding agent、web/GUI agent、multi-agent、safety、RAG 等。
+- 自动去重：按 arXiv ID 去重。
+- 自动打标签：根据标题和摘要推断 topics。
+- 自动分层：生成 `queued` 条目，后续人工精读后再升级状态。
+- API 退避：遇到 arXiv 429 或临时网络错误时按 `--retries` 和 `--retry-sleep` 重试。
+
 ## Method
 
 1. 搜索或导入资料。
@@ -51,3 +71,22 @@ python3 tools/collection/check_urls.py
 4. 运行 `build_index.py` 更新结构化数据。
 5. 运行 `check_urls.py` 检查来源可达性。
 6. 更新对应 insight 文件。
+
+大规模论文扩展时，先运行 `collect_arxiv.py`，再运行 `build_index.py`。
+
+### `promote_arxiv_manifest.py`
+
+从 `collect_arxiv.py` 生成的候选池里，把达到阈值的记录批量提升为 `papers/items/` 条目。适合在 arXiv API 限流后继续本地调阈值。
+
+```bash
+python3 tools/collection/promote_arxiv_manifest.py \
+  data/arxiv-agent-candidates-2025-07-08-to-2026-07-08.json \
+  --min-score 13 \
+  --relevance high \
+  --output-manifest data/arxiv-agent-papers-2025-07-08-to-2026-07-08.json
+```
+
+默认用途：
+
+- `score >= 13` 且 `relevance=high` 的论文进入主阅读队列。
+- 低分候选仍留在 candidate manifest，后续可以人工抽查或降低阈值再提升。
diff --git a/tools/collection/collect_arxiv.py b/tools/collection/collect_arxiv.py
new file mode 100644
index 0000000..9f41649
--- /dev/null
+++ b/tools/collection/collect_arxiv.py
@@ -0,0 +1,425 @@
+#!/usr/bin/env python3
+"""Collect recent arXiv papers related to LLM/AI agents."""
+
+from __future__ import annotations
+
+import argparse
+import json
+import re
+import time
+import urllib.parse
+import urllib.request
+import urllib.error
+import xml.etree.ElementTree as ET
+from collections import defaultdict
+from datetime import date
+from pathlib import Path
+
+
+ROOT = Path(__file__).resolve().parents[2]
+API_URL = "https://export.arxiv.org/api/query"
+NS = {
+    "atom": "http://www.w3.org/2005/Atom",
+    "arxiv": "http://arxiv.org/schemas/atom",
+}
+
+
+QUERIES = [
+    ("llm-agent", 'all:"LLM agent" OR all:"LLM agents"'),
+    ("language-agent", 'all:"language agent" OR all:"language agents"'),
+    ("ai-agent", 'all:"AI agent" OR all:"AI agents"'),
+    ("agentic-ai", 'all:"agentic AI" OR all:"agentic workflow"'),
+    ("agent-evaluation", 'all:"agent evaluation" OR all:"agent benchmark" OR all:"agentic benchmark"'),
+    ("agent-memory", 'all:"agent memory" OR all:"memory agent" OR all:"memory agents"'),
+    ("tool-use", 'all:"tool use" AND all:"agent"'),
+    ("function-calling", 'all:"function calling" AND all:"agent"'),
+    ("coding-agent", 'all:"coding agent" OR all:"software engineering agent" OR all:"SWE-bench"'),
+    ("web-gui-agent", 'all:"web agent" OR all:"browser agent" OR all:"GUI agent" OR all:"computer use"'),
+    ("multi-agent-llm", 'all:"multi-agent" AND (all:"LLM" OR all:"large language model")'),
+    ("agent-safety", 'all:"agent safety" OR all:"AI agent safety" OR all:"agent security"'),
+    ("rag-agent", 'all:"RAG" AND all:"agent"'),
+    ("planning-agent", 'all:"planning" AND all:"LLM agent"'),
+    ("autonomous-agent-llm", 'all:"autonomous agent" AND (all:"LLM" OR all:"large language model")'),
+]
+
+
+TOPIC_RULES = [
+    ("agent-evaluation", ("evaluation", "benchmark", "eval", "metric", "leaderboard", "assessment")),
+    ("memory", ("memory", "remember", "stateful", "long-term", "episodic")),
+    ("tool-use", ("tool", "function calling", "api", "mcp", "action")),
+    ("coding-agent", ("coding", "software engineering", "swe-bench", "repository", "program repair", "code agent")),
+    ("computer-use", ("computer use", "gui", "browser", "web agent", "mobile", "desktop")),
+    ("multi-agent", ("multi-agent", "multiagent", "agent society", "collaboration", "debate")),
+    ("agent-safety", ("safety", "security", "risk", "prompt injection", "alignment", "guardrail")),
+    ("rag", ("retrieval", "rag", "knowledge base", "grounding")),
+    ("planning", ("planning", "planner", "plan", "long-horizon", "task decomposition")),
+    ("reasoning", ("reasoning", "reflection", "self-improvement", "verifier")),
+    ("workflow-agent", ("workflow", "enterprise", "office", "productivity", "automation")),
+    ("embodied-agent", ("embodied", "robot", "robotic", "navigation")),
+    ("world-model", ("world model", "simulation", "environment model")),
+]
+
+
+HIGH_SIGNAL_TERMS = (
+    "llm agent",
+    "language agent",
+    "ai agent",
+    "agentic",
+    "tool use",
+    "memory",
+    "benchmark",
+    "evaluation",
+    "coding agent",
+    "swe-bench",
+    "computer use",
+    "gui agent",
+    "multi-agent",
+    "agent safety",
+)
+
+
+def normalize_space(value: str) -> str:
+    return re.sub(r"\s+", " ", value).strip()
+
+
+def slugify(text: str) -> str:
+    text = text.lower()
+    text = re.sub(r"[^a-z0-9]+", "-", text)
+    return re.sub(r"-+", "-", text).strip("-")[:80] or "untitled"
+
+
+def yaml_value(value: str | int | None) -> str:
+    if value is None:
+        return ""
+    text = str(value).replace("\n", " ").strip()
+    if not text:
+        return ""
+    if any(char in text for char in [":", "#", "[", "]", "{", "}", "\"", "'"]):
+        return json.dumps(text, ensure_ascii=False)
+    return text
+
+
+def arxiv_id(entry_id: str) -> str:
+    raw = entry_id.rsplit("/", 1)[-1]
+    return raw.split("v", 1)[0]
+
+
+def date_window_query(raw_query: str, from_date: str, to_date: str) -> str:
+    start = from_date.replace("-", "") + "0000"
+    end = to_date.replace("-", "") + "2359"
+    return f"({raw_query}) AND submittedDate:[{start} TO {end}]"
+
+
+def fetch_query(
+    raw_query: str,
+    from_date: str,
+    to_date: str,
+    start: int,
+    max_results: int,
+    retries: int,
+    retry_sleep: float,
+) -> str:
+    params = {
+        "search_query": date_window_query(raw_query, from_date, to_date),
+        "start": start,
+        "max_results": max_results,
+        "sortBy": "submittedDate",
+        "sortOrder": "descending",
+    }
+    url = f"{API_URL}?{urllib.parse.urlencode(params)}"
+    request = urllib.request.Request(url, headers={"User-Agent": "agent-kb-arxiv-collector/0.1"})
+    for attempt in range(retries + 1):
+        try:
+            with urllib.request.urlopen(request, timeout=30) as response:
+                return response.read().decode("utf-8", "replace")
+        except urllib.error.HTTPError as exc:
+            if exc.code != 429 or attempt >= retries:
+                raise
+            time.sleep(retry_sleep * (attempt + 1))
+        except urllib.error.URLError:
+            if attempt >= retries:
+                raise
+            time.sleep(retry_sleep * (attempt + 1))
+    raise RuntimeError("unreachable fetch retry state")
+
+
+def manifest_record(record: dict, topics: list[str], score: int, relevance: str, matched_queries: set[str]) -> dict:
+    sorted_queries = sorted(matched_queries)
+    return {
+        "id": record["id"],
+        "arxiv_id": record["id"],
+        "source": "arxiv",
+        "source_id": f"arxiv:{record['id']}",
+        "title": record["title"],
+        "url": record["url"],
+        "pdf_url": f"https://arxiv.org/pdf/{record['id']}",
+        "published": record["published"],
+        "updated": record["updated"],
+        "authors": record["authors"],
+        "categories": record["categories"],
+        "topics": topics,
+        "score": score,
+        "relevance": relevance,
+        "primary_query": sorted_queries[0] if sorted_queries else "",
+        "matched_queries": sorted_queries,
+    }
+
+
+def parse_feed(xml_text: str) -> list[dict]:
+    root = ET.fromstring(xml_text)
+    records = []
+    for entry in root.findall("atom:entry", NS):
+        entry_id = entry.findtext("atom:id", default="", namespaces=NS)
+        title = normalize_space(entry.findtext("atom:title", default="", namespaces=NS))
+        summary = normalize_space(entry.findtext("atom:summary", default="", namespaces=NS))
+        published = entry.findtext("atom:published", default="", namespaces=NS)[:10]
+        updated = entry.findtext("atom:updated", default="", namespaces=NS)[:10]
+        authors = [
+            normalize_space(author.findtext("atom:name", default="", namespaces=NS))
+            for author in entry.findall("atom:author", NS)
+        ]
+        categories = [
+            category.attrib.get("term", "")
+            for category in entry.findall("atom:category", NS)
+            if category.attrib.get("term")
+        ]
+        records.append(
+            {
+                "id": arxiv_id(entry_id),
+                "url": f"https://arxiv.org/abs/{arxiv_id(entry_id)}",
+                "title": title,
+                "summary": summary,
+                "published": published,
+                "updated": updated,
+                "authors": authors,
+                "categories": categories,
+            }
+        )
+    return records
+
+
+def classify(record: dict, matched_queries: set[str]) -> tuple[list[str], int, str]:
+    text = f"{record['title']} {record['summary']}".lower()
+    title = record["title"].lower()
+    topics = []
+    score = 0
+
+    if "agent" in title or "agentic" in title:
+        score += 4
+    elif "agent" in text or "agentic" in text:
+        score += 2
+
+    if "llm" in text or "large language model" in text or "language model" in text:
+        score += 2
+
+    for term in HIGH_SIGNAL_TERMS:
+        if term in title:
+            score += 3
+        elif term in text:
+            score += 1
+
+    for topic, keywords in TOPIC_RULES:
+        if any(keyword in text for keyword in keywords):
+            topics.append(topic)
+            score += 1
+
+    if matched_queries:
+        score += min(len(matched_queries), 4)
+
+    if not topics and ("agent" in text or "agentic" in text):
+        topics.append("agent")
+
+    if score >= 13:
+        relevance = "high"
+    elif score >= 7:
+        relevance = "medium"
+    else:
+        relevance = "low"
+
+    return sorted(set(topics)), score, relevance
+
+
+def existing_arxiv_ids() -> set[str]:
+    ids = set()
+    for path in (ROOT / "papers" / "items").glob("*.md"):
+        text = path.read_text(encoding="utf-8")
+        for match in re.findall(r"https://arxiv\.org/(?:abs|html)/([0-9]+\.[0-9]+)", text):
+            ids.add(match)
+    return ids
+
+
+def item_path(record: dict) -> Path:
+    year = (record.get("published") or "0000")[:4]
+    title_slug = slugify(record["title"])
+    id_slug = record["id"].replace(".", "-")
+    return ROOT / "papers" / "items" / f"{year}-{id_slug}-{title_slug}.md"
+
+
+def write_item(record: dict, topics: list[str], score: int, relevance: str, matched_queries: set[str], today: str) -> Path:
+    path = item_path(record)
+    authors = ", ".join(record["authors"][:8])
+    if len(record["authors"]) > 8:
+        authors += ", et al."
+    categories = ", ".join(record["categories"])
+    query_list = ", ".join(sorted(matched_queries))
+    topic_block = "\n".join(f"  - {topic}" for topic in topics) or "  - agent"
+    category_block = "\n".join(f"  - {category}" for category in record["categories"]) or "  -"
+    content = f"""# Paper: {record['title']}
+
+---
+type: paper
+title: {yaml_value(record['title'])}
+authors: {yaml_value(authors)}
+year: {yaml_value((record.get('published') or '')[:4])}
+venue: arXiv
+url: {record['url']}
+code_url:
+source: arxiv
+collected_at: {today}
+published_at: {record.get('published') or ''}
+updated_at: {record.get('updated') or ''}
+status: queued
+relevance: {relevance}
+topics:
+{topic_block}
+methods:
+  -
+benchmarks:
+  -
+models:
+  -
+datasets:
+{category_block}
+related_concepts:
+  -
+related_jobs:
+  -
+related_experiments:
+  -
+related_projects:
+  -
+collection_score: {score}
+collection_queries: {yaml_value(query_list)}
+---
+
+## One-line Takeaway
+
+Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.
+
+## Why Collected
+
+- matched queries: {query_list or 'agent-related arXiv sweep'}
+- inferred topics: {', '.join(topics) or 'agent'}
+- arXiv categories: {categories or 'unknown'}
+- collection score: {score}
+
+## Review Checklist
+
+- Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
+- Does it include a benchmark, dataset, code, or reproducible experimental setup?
+- Should it be promoted from `queued` to `skimmed` or `summarized`?
+
+## Links
+
+- arXiv: {record['url']}
+"""
+    path.write_text(content, encoding="utf-8")
+    return path
+
+
+def main() -> int:
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--from-date", default="2025-07-08")
+    parser.add_argument("--to-date", default=date.today().isoformat())
+    parser.add_argument("--per-query", type=int, default=120)
+    parser.add_argument("--page-size", type=int, default=60)
+    parser.add_argument("--max-items", type=int, default=180)
+    parser.add_argument("--min-score", type=int, default=6)
+    parser.add_argument("--sleep", type=float, default=1.0)
+    parser.add_argument("--retries", type=int, default=4)
+    parser.add_argument("--retry-sleep", type=float, default=10.0)
+    parser.add_argument("--dry-run", action="store_true")
+    args = parser.parse_args()
+
+    by_id: dict[str, dict] = {}
+    matches: dict[str, set[str]] = defaultdict(set)
+
+    for label, raw_query in QUERIES:
+        fetched = 0
+        start = 0
+        while fetched < args.per_query:
+            page_size = min(args.page_size, args.per_query - fetched)
+            xml_text = fetch_query(
+                raw_query,
+                args.from_date,
+                args.to_date,
+                start,
+                page_size,
+                args.retries,
+                args.retry_sleep,
+            )
+            records = parse_feed(xml_text)
+            if not records:
+                break
+            for record in records:
+                by_id.setdefault(record["id"], record)
+                matches[record["id"]].add(label)
+            fetched += len(records)
+            start += len(records)
+            if len(records) < page_size:
+                break
+            time.sleep(args.sleep)
+
+    existing = existing_arxiv_ids()
+    scored = []
+    all_candidates = []
+    for record_id, record in by_id.items():
+        topics, score, relevance = classify(record, matches[record_id])
+        all_candidates.append(manifest_record(record, topics, score, relevance, matches[record["id"]]))
+        if score < args.min_score:
+            continue
+        scored.append((score, relevance, topics, record))
+
+    scored.sort(key=lambda item: (item[0], item[3].get("published") or ""), reverse=True)
+    selected = scored[: args.max_items]
+
+    data_dir = ROOT / "data"
+    data_dir.mkdir(exist_ok=True)
+    manifest = []
+    written = []
+    skipped_existing = 0
+    today = date.today().isoformat()
+    for score, relevance, topics, record in selected:
+        manifest.append(manifest_record(record, topics, score, relevance, matches[record["id"]]))
+        if record["id"] in existing:
+            skipped_existing += 1
+            continue
+        if not args.dry_run:
+            path = write_item(record, topics, score, relevance, matches[record["id"]], today)
+            written.append(str(path.relative_to(ROOT)))
+
+    all_candidates.sort(key=lambda item: (item["score"], item["published"]), reverse=True)
+    candidates_path = data_dir / f"arxiv-agent-candidates-{args.from_date}-to-{args.to_date}.json"
+    candidates_path.write_text(json.dumps(all_candidates, ensure_ascii=False, indent=2) + "\n", encoding="utf-8")
+
+    manifest_path = data_dir / f"arxiv-agent-papers-{args.from_date}-to-{args.to_date}.json"
+    manifest_path.write_text(json.dumps(manifest, ensure_ascii=False, indent=2) + "\n", encoding="utf-8")
+
+    summary = {
+        "from_date": args.from_date,
+        "to_date": args.to_date,
+        "queries": [label for label, _ in QUERIES],
+        "unique_seen": len(by_id),
+        "selected": len(selected),
+        "written": len(written),
+        "skipped_existing": skipped_existing,
+        "candidates_manifest": str(candidates_path.relative_to(ROOT)),
+        "manifest": str(manifest_path.relative_to(ROOT)),
+        "written_paths": written,
+    }
+    print(json.dumps(summary, ensure_ascii=False, indent=2))
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())
diff --git a/tools/collection/promote_arxiv_manifest.py b/tools/collection/promote_arxiv_manifest.py
new file mode 100644
index 0000000..5a4730f
--- /dev/null
+++ b/tools/collection/promote_arxiv_manifest.py
@@ -0,0 +1,101 @@
+#!/usr/bin/env python3
+"""Promote arXiv manifest records into paper item notes."""
+
+from __future__ import annotations
+
+import argparse
+import json
+from datetime import date
+from pathlib import Path
+
+from collect_arxiv import ROOT, existing_arxiv_ids, write_item
+
+
+def parse_args() -> argparse.Namespace:
+    parser = argparse.ArgumentParser()
+    parser.add_argument("manifest", help="Path to an arXiv candidate manifest JSON file.")
+    parser.add_argument("--min-score", type=int, default=13)
+    parser.add_argument("--relevance", default="high")
+    parser.add_argument("--max-items", type=int, default=0, help="0 means no cap.")
+    parser.add_argument("--output-manifest", help="Optional JSON file for promoted records.")
+    parser.add_argument("--dry-run", action="store_true")
+    return parser.parse_args()
+
+
+def normalize_record(record: dict) -> dict:
+    arxiv_id = record.get("arxiv_id") or record["id"]
+    return {
+        "id": arxiv_id,
+        "url": record.get("url") or f"https://arxiv.org/abs/{arxiv_id}",
+        "title": record["title"],
+        "published": record.get("published") or "",
+        "updated": record.get("updated") or "",
+        "authors": record.get("authors") or [],
+        "categories": record.get("categories") or [],
+    }
+
+
+def main() -> int:
+    args = parse_args()
+    manifest_path = Path(args.manifest)
+    if not manifest_path.is_absolute():
+        manifest_path = ROOT / manifest_path
+
+    records = json.loads(manifest_path.read_text(encoding="utf-8"))
+    selected = [
+        record
+        for record in records
+        if int(record.get("score") or 0) >= args.min_score
+        and (not args.relevance or record.get("relevance") == args.relevance)
+    ]
+    selected.sort(key=lambda item: (int(item.get("score") or 0), item.get("published") or ""), reverse=True)
+    if args.max_items:
+        selected = selected[: args.max_items]
+
+    existing = existing_arxiv_ids()
+    today = date.today().isoformat()
+    written = []
+    skipped_existing = 0
+    for record in selected:
+        arxiv_id = record.get("arxiv_id") or record["id"]
+        if arxiv_id in existing:
+            skipped_existing += 1
+            continue
+        if not args.dry_run:
+            path = write_item(
+                normalize_record(record),
+                record.get("topics") or ["agent"],
+                int(record.get("score") or 0),
+                record.get("relevance") or "unknown",
+                set(record.get("matched_queries") or []),
+                today,
+            )
+            written.append(str(path.relative_to(ROOT)))
+            existing.add(arxiv_id)
+
+    if args.output_manifest and not args.dry_run:
+        output_path = Path(args.output_manifest)
+        if not output_path.is_absolute():
+            output_path = ROOT / output_path
+        output_path.parent.mkdir(parents=True, exist_ok=True)
+        output_path.write_text(json.dumps(selected, ensure_ascii=False, indent=2) + "\n", encoding="utf-8")
+
+    print(
+        json.dumps(
+            {
+                "manifest": str(manifest_path.relative_to(ROOT)),
+                "selected": len(selected),
+                "written": len(written),
+                "skipped_existing": skipped_existing,
+                "output_manifest": args.output_manifest or "",
+                "written_paths": written,
+            },
+            ensure_ascii=False,
+            indent=2,
+        )
+    )
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())