wuyang/agent
2
2
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
Files
main
agent/data/research/memory-corpus.json
T
wuyang ab758e774e Audit recent Agent memory research
2026-07-10 18:52:00 +08:00

9768 lines
737 KiB
JSON
Raw Permalink Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
{
"topic": "memory",
"generated_at": "2026-07-10",
"selected_from_index": 293,
"resolved": 293,
"missing_arxiv_id": [],
"unresolved": [],
"papers": [
{
"arxiv_id": "2509.02444",
"title": "AppCopilot: Toward General, Accurate, Long-Horizon, and Efficient Mobile Agent",
"abstract": "With the raid evolution of large language models and multimodal models, the mobile-agent landscape has proliferated without converging on the fundamental challenges. This paper identifies four core problems that should be solved for mobile agents to deliver practical, scalable impact: (1) generalization across tasks, APPs, and devices; (2) accuracy, specifically precise on-screen interaction and click targeting; (3) long-horizon capability for sustained, multi-step goals; and (4) efficiency, specifically high-performance runtime on resource-constrained devices. We present AppCopilot, a multimodal, multi-agent, general-purpose mobile agent that operates across applications. AppCopilot operationalizes this position through an end-to-end pipeline spanning data collection, training, finetuning, efficient inference, and PC/mobile application. At the model layer, it integrates multimodal foundation models with robust Chinese-English support. At the reasoning and control layer, it combines chain-of-thought reasoning, hierarchical task planning and decomposition, and multi-agent collaboration. At the execution layer, it enables experiential adaptation, voice interaction, function calling, cross-APP and cross-device orchestration, and comprehensive mobile APP support. The system design incorporates profiling-driven optimization for latency and memory across heterogeneous hardware. Empirically, AppCopilot achieves significant improvements on four dimensions: stronger generalization, higher precision of on screen actions, more reliable long horizon task completion, and faster, more resource efficient runtime. By articulating a cohesive position and a reference architecture that closes the loop from data collection, training to finetuning and efficient inference, this paper offers a concrete roadmap for general purpose mobile agent and provides actionable guidance.",
"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"
],
"url": "https://arxiv.org/abs/2509.02444",
"pdf_url": "https://arxiv.org/pdf/2509.02444",
"path": "papers/items/2025-2509-02444-appcopilot-toward-general-accurate-long-horizon-and-efficient-mobile-agent.md",
"local_title": "\"AppCopilot: Toward General, Accurate, Long-Horizon, and Efficient Mobile Agent\"",
"local_status": "queued",
"local_topics": [
"computer-use",
"memory",
"multi-agent",
"planning",
"reasoning",
"tool-use"
],
"collection_queries": "function-calling",
"collection_score": "15"
},
{
"arxiv_id": "2509.10769",
"title": "AgentArch: A Comprehensive Benchmark to Evaluate Agent Architectures in Enterprise",
"abstract": "While individual components of agentic architectures have been studied in isolation, there remains limited empirical understanding of how different design dimensions interact within complex multi-agent systems. This study aims to address these gaps by providing a comprehensive enterprise-specific benchmark evaluating 18 distinct agentic configurations across state-of-the-art large language models. We examine four critical agentic system dimensions: orchestration strategy, agent prompt implementation (ReAct versus function calling), memory architecture, and thinking tool integration. Our benchmark reveals significant model-specific architectural preferences that challenge the prevalent one-size-fits-all paradigm in agentic AI systems. It also reveals significant weaknesses in overall agentic performance on enterprise tasks with the highest scoring models achieving a maximum of only 35.3\\% success on the more complex task and 70.8\\% on the simpler task. We hope these findings inform the design of future agentic systems by enabling more empirically backed decisions regarding architectural components and model selection.",
"published": "2025-09-13",
"updated": "2026-01-06",
"authors": [
"Tara Bogavelli",
"Roshnee Sharma",
"Hari Subramani"
],
"categories": [
"cs.AI",
"cs.CL",
"cs.MA"
],
"url": "https://arxiv.org/abs/2509.10769",
"pdf_url": "https://arxiv.org/pdf/2509.10769",
"path": "papers/items/2025-2509-10769-agentarch-a-comprehensive-benchmark-to-evaluate-agent-architectures-in-enterpris.md",
"local_title": "\"AgentArch: A Comprehensive Benchmark to Evaluate Agent Architectures in Enterprise\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"multi-agent",
"tool-use",
"workflow-agent"
],
"collection_queries": "function-calling",
"collection_score": "18"
},
{
"arxiv_id": "2510.14548",
"title": "LLM Agents Beyond Utility: An Open-Ended Perspective",
"abstract": "Recent LLM agents have made great use of chain of thought reasoning and function calling. As their capabilities grow, an important question arises: can this software represent not only a smart problem-solving tool, but an entity in its own right, that can plan, design immediate tasks, and reason toward broader, more ambiguous goals? To study this question, we adopt an open-ended experimental setting where we augment a pretrained LLM agent with the ability to generate its own tasks, accumulate knowledge, and interact extensively with its environment. We study the resulting open-ended agent qualitatively. It can reliably follow complex multi-step instructions, store and reuse information across runs, and propose and solve its own tasks, though it remains sensitive to prompt design, prone to repetitive task generation, and unable to form self-representations. These findings illustrate both the promise and current limits of adapting pretrained LLMs toward open-endedness, and point to future directions for training agents to manage memory, explore productively, and pursue abstract long-term goals.",
"published": "2025-10-16",
"updated": "2025-10-16",
"authors": [
"Asen Nachkov",
"Xi Wang",
"Luc Van Gool"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2510.14548",
"pdf_url": "https://arxiv.org/pdf/2510.14548",
"path": "papers/items/2025-2510-14548-llm-agents-beyond-utility-an-open-ended-perspective.md",
"local_title": "\"LLM Agents Beyond Utility: An Open-Ended Perspective\"",
"local_status": "queued",
"local_topics": [
"memory",
"planning",
"reasoning",
"tool-use"
],
"collection_queries": "function-calling",
"collection_score": "15"
},
{
"arxiv_id": "2510.18586",
"title": "TokenCake: A KV-Cache-centric Serving Framework for LLM-based Multi-Agent Applications",
"abstract": "Large Language Models (LLMs) are increasingly deployed in complex multi-agent applications that rely on external function calls. This workload creates severe performance challenges for the KV Cache: spatial contention leads to the eviction of critical agents' caches and temporal underutilization leaves the cache of agents stalled on long-running function calls idling in GPU memory. We present TokenCake, a KV-Cache-centric serving framework that bridges this gap by co-optimizing scheduling and memory management through an agent-aware design. TokenCake's Temporal Scheduler employs an event-driven, opportunistic policy to proactively offload idle KV Caches during function calls and uses predictive uploading to hide data transfer latency. TokenCake's Spatial Scheduler uses dynamic memory partitioning, guided by a hybrid priority metric combining graph structure and runtime state, to reserve GPU memory for critical-path agents. Our evaluation on representative multi-agent benchmarks shows that TokenCake reduces end-to-end latency by over 47.06% and improves effective GPU memory utilization by up to 16.9% compared to vLLM.",
"published": "2025-10-21",
"updated": "2026-05-20",
"authors": [
"Zhuohang Bian",
"Feiyang Wu",
"Zhuoran Li",
"Teng Ma",
"Youwei Zhuo"
],
"categories": [
"cs.DC"
],
"url": "https://arxiv.org/abs/2510.18586",
"pdf_url": "https://arxiv.org/pdf/2510.18586",
"path": "papers/items/2025-2510-18586-tokencake-a-kv-cache-centric-serving-framework-for-llm-based-multi-agent-applica.md",
"local_title": "\"TokenCake: A KV-Cache-centric Serving Framework for LLM-based Multi-Agent Applications\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"multi-agent"
],
"collection_queries": "function-calling",
"collection_score": "17"
},
{
"arxiv_id": "2512.02605",
"title": "IACT: A Self-Organizing Recursive Model for General AI Agents: A Technical White Paper on the Architecture Behind kragent.ai",
"abstract": "This technical white paper introduces the Interactive Agents Call Tree (IACT), a computational model designed to address the limitations of static, hard-coded agent workflows. Unlike traditional systems that require pre-defined graphs or specialized programming, IACT operates as a general-purpose autonomous system driven purely by user dialogue. Given a high-level objective, the system autonomously grows a dynamic, recursive agent topology incrementally tailored to the problem's structure. This allows it to scale its organizational complexity to match open-ended tasks. To mitigate the error propagation inherent in unidirectional function calls, IACT introduces interactional redundancy by replacing rigid invocations with bidirectional, stateful dialogues. This mechanism enables runtime error correction and ambiguity resolution. We describe the architecture, design principles, and practical lessons behind the production deployment of this model in the kragent.ai system, presenting qualitative evidence from real-world workflows rather than exhaustive benchmark results.",
"published": "2025-12-02",
"updated": "2025-12-02",
"authors": [
"Pengju Lu"
],
"categories": [
"cs.AI",
"cs.MA",
"cs.SE"
],
"url": "https://arxiv.org/abs/2512.02605",
"pdf_url": "https://arxiv.org/pdf/2512.02605",
"path": "papers/items/2025-2512-02605-iact-a-self-organizing-recursive-model-for-general-ai-agents-a-technical-white-p.md",
"local_title": "\"IACT: A Self-Organizing Recursive Model for General AI Agents: A Technical White Paper on the Architecture Behind kragent.ai\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"rag",
"tool-use",
"workflow-agent"
],
"collection_queries": "function-calling",
"collection_score": "15"
},
{
"arxiv_id": "2602.03224",
"title": "TAME: A Trustworthy Test-Time Evolution of Agent Memory with Systematic Benchmarking",
"abstract": "Test-time evolution of agent memory represents a pivotal paradigm for advancing AGI, as it strengthens complex reasoning through experience accumulation without requiring parameter updates. However, even during benign task evolution, agent safety alignment remains vulnerable, a phenomenon known as Agent Memory Misevolution. To evaluate this phenomenon, we construct the Trust-Memevo benchmark and find that agents exhibit an overall decline in trustworthiness across multiple tasks during benign task evolution. To address this issue, we propose TAME, a trust-aware memory evolution framework in which a shared memory bank is jointly governed by an Executor and an Evaluator. The Executor retrieves and applies transferable experiences to support task solving, while the Evaluator assesses the contribution of each utilized experience to the outcome and produces trust-aware feedback to guide subsequent memory use. This executor-evaluator loop enables memory to be selectively reinforced, cautiously reused, and continuously expanded over time. Experiments show that TAME mitigates memory misevolution while achieving strong task performance. In particular, on the GPT-5.2 AIME benchmark, TAME improves accuracy by 14.6 percentage points over the strongest existing method and maintains competitive trustworthiness.",
"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"
],
"url": "https://arxiv.org/abs/2602.03224",
"pdf_url": "https://arxiv.org/pdf/2602.03224",
"path": "papers/items/2026-2602-03224-tame-a-trustworthy-test-time-evolution-of-agent-memory-with-systematic-benchmark.md",
"local_title": "\"TAME: A Trustworthy Test-Time Evolution of Agent Memory with Systematic Benchmarking\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"computer-use",
"memory",
"reasoning"
],
"collection_queries": "agent-safety",
"collection_score": "17"
},
{
"arxiv_id": "2602.07652",
"title": "Agent-Fence: Mapping Security Vulnerabilities Across Deep Research Agents",
"abstract": "Large language models are increasingly deployed as *deep agents* that plan, maintain persistent state, and invoke external tools, shifting safety failures from unsafe text to unsafe *trajectories*. We introduce **AgentFence**, an architecture-centric security evaluation that defines 14 trust-boundary attack classes spanning planning, memory, retrieval, tool use, and delegation, and detects failures via *trace-auditable conversation breaks* (unauthorized or unsafe tool use, wrong-principal actions, state/objective integrity violations, and attack-linked deviations). Holding the base model fixed, we evaluate eight agent archetypes under persistent multi-turn interaction and observe substantial architectural variation in mean security break rate (MSBR), ranging from $0.29 \\pm 0.04$ (LangGraph) to $0.51 \\pm 0.07$ (AutoGPT). The highest-risk classes are operational: Denial-of-Wallet ($0.62 \\pm 0.08$), Authorization Confusion ($0.54 \\pm 0.10$), Retrieval Poisoning ($0.47 \\pm 0.09$), and Planning Manipulation ($0.44 \\pm 0.11$), while prompt-centric classes remain below $0.20$ under standard settings. Breaks are dominated by boundary violations (SIV 31%, WPA 27%, UTI+UTA 24%, ATD 18%), and authorization confusion correlates with objective and tool hijacking ($ρ\\approx 0.63$ and $ρ\\approx 0.58$). AgentFence reframes agent security around what matters operationally: whether an agent stays within its goal and authority envelope over time.",
"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"
],
"url": "https://arxiv.org/abs/2602.07652",
"pdf_url": "https://arxiv.org/pdf/2602.07652",
"path": "papers/items/2026-2602-07652-agent-fence-mapping-security-vulnerabilities-across-deep-research-agents.md",
"local_title": "\"Agent-Fence: Mapping Security Vulnerabilities Across Deep Research Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"planning",
"rag",
"tool-use"
],
"collection_queries": "agent-safety",
"collection_score": "16"
},
{
"arxiv_id": "2602.08412",
"title": "From Assistant to Double Agent: Formalizing and Benchmarking Attacks on OpenClaw for Personalized Local AI Agent",
"abstract": "Although large language model (LLM)-based agents, exemplified by OpenClaw, are increasingly evolving from task-oriented systems into personalized AI assistants for solving complex real-world tasks, their practical deployment also introduces severe security risks. However, existing agent security research and evaluation frameworks primarily focus on synthetic or task-centric settings, and thus fail to accurately capture the attack surface and risk propagation mechanisms of personalized agents in real-world deployments. To address this gap, we propose Personalized Agent Security Bench (PASB), an end-to-end security evaluation framework tailored for real-world personalized agents. Building upon existing agent attack paradigms, PASB incorporates personalized usage scenarios, realistic toolchains, and long-horizon interactions, enabling black-box, end-to-end security evaluation on real systems. Using OpenClaw as a representative case study, we systematically evaluate its security across multiple personalized scenarios, tool capabilities, and attack types. Our results indicate that OpenClaw exhibits critical vulnerabilities at different execution stages, including user prompt processing, tool usage, and memory retrieval, highlighting substantial security risks in personalized agent deployments. The code for the proposed PASB framework is available at https://github.com/AstorYH/PASB.",
"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"
],
"url": "https://arxiv.org/abs/2602.08412",
"pdf_url": "https://arxiv.org/pdf/2602.08412",
"path": "papers/items/2026-2602-08412-from-assistant-to-double-agent-formalizing-and-benchmarking-attacks-on-openclaw-.md",
"local_title": "\"From Assistant to Double Agent: Formalizing and Benchmarking Attacks on OpenClaw for Personalized Local AI Agent\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"planning",
"rag",
"tool-use"
],
"collection_queries": "agent-safety",
"collection_score": "21"
},
{
"arxiv_id": "2602.13530",
"title": "REMem: Reasoning with Episodic Memory in Language Agent",
"abstract": "Humans excel at remembering concrete experiences along spatiotemporal contexts and performing reasoning across those events, i.e., the capacity for episodic memory. In contrast, memory in language agents remains mainly semantic, and current agents are not yet capable of effectively recollecting and reasoning over interaction histories. We identify and formalize the core challenges of episodic recollection and reasoning from this gap, and observe that existing work often overlooks episodicity, lacks explicit event modeling, or overemphasizes simple retrieval rather than complex reasoning. We present REMem, a two-phase framework for constructing and reasoning with episodic memory: 1) Offline indexing, where REMem converts experiences into a hybrid memory graph that flexibly links time-aware gists and facts. 2) Online inference, where REMem employs an agentic retriever with carefully curated tools for iterative retrieval over the memory graph. Comprehensive evaluation across four episodic memory benchmarks shows that REMem substantially outperforms state-of-the-art memory systems such as Mem0 and HippoRAG 2, showing 3.4% and 13.4% absolute improvements on episodic recollection and reasoning tasks, respectively. Moreover, REMem also demonstrates more robust refusal behavior for unanswerable questions.",
"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"
],
"url": "https://arxiv.org/abs/2602.13530",
"pdf_url": "https://arxiv.org/pdf/2602.13530",
"path": "papers/items/2026-2602-13530-remem-reasoning-with-episodic-memory-in-language-agent.md",
"local_title": "\"REMem: Reasoning with Episodic Memory in Language Agent\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "language-agent",
"collection_score": "19"
},
{
"arxiv_id": "2602.23320",
"title": "ParamMem: Augmenting Language Agents with Parametric Reflective Memory",
"abstract": "Self-reflection enables language agents to iteratively refine solutions, yet often produces repetitive outputs that limit reasoning performance. Recent studies have attempted to address this limitation through various approaches, among which increasing reflective diversity has shown promise. Our empirical analysis reveals a strong positive correlation between reflective diversity and task success, further motivating the need for diverse reflection signals. We introduce ParamMem, a parametric memory module that encodes cross-sample reflection patterns into model parameters, enabling diverse reflection generation through temperature-controlled sampling. Building on this module, we propose ParamAgent, a reflection-based agent framework that integrates parametric memory with episodic and cross-sample memory. Extensive experiments on code generation, mathematical reasoning, and multi-hop question answering demonstrate consistent improvements over state-of-the-art baselines. Further analysis reveals that ParamMem is sample-efficient, enables weak-to-strong transfer across model scales, and supports self-improvement without reliance on stronger external model, highlighting the potential of ParamMem as an effective component for enhancing language agents.",
"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"
],
"url": "https://arxiv.org/abs/2602.23320",
"pdf_url": "https://arxiv.org/pdf/2602.23320",
"path": "papers/items/2026-2602-23320-parammem-augmenting-language-agents-with-parametric-reflective-memory.md",
"local_title": "\"ParamMem: Augmenting Language Agents with Parametric Reflective Memory\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"reasoning"
],
"collection_queries": "language-agent",
"collection_score": "14"
},
{
"arxiv_id": "2603.03680",
"title": "MAGE: Meta-Reinforcement Learning for Language Agents toward Strategic Exploration and Exploitation",
"abstract": "Large Language Model (LLM) agents have demonstrated remarkable proficiency in learned tasks, yet they often struggle to adapt to non-stationary environments with feedback. While In-Context Learning and external memory offer some flexibility, they fail to internalize the adaptive ability required for long-term improvement. Meta-Reinforcement Learning (meta-RL) provides an alternative by embedding the learning process directly within the model. However, existing meta-RL approaches for LLMs focus primarily on exploration in single-agent settings, neglecting the strategic exploitation necessary for multi-agent environments. We propose MAGE, a meta-RL framework that empowers LLM agents for strategic exploration and exploitation. MAGE utilizes a multi-episode training regime where interaction histories and reflections are integrated into the context window. By using the final episode reward as the objective, MAGE incentivizes the agent to refine its strategy based on past experiences. We further combine population-based training with an agent-specific advantage normalization technique to enrich agent diversity and ensure stable learning. Experiment results show that MAGE outperforms existing baselines in both exploration and exploitation tasks. Furthermore, MAGE exhibits strong generalization to unseen opponents, suggesting it has internalized the ability for strategic exploration and exploitation. Code is available at https://github.com/Lu-Yang666/MAGE.",
"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"
],
"url": "https://arxiv.org/abs/2603.03680",
"pdf_url": "https://arxiv.org/pdf/2603.03680",
"path": "papers/items/2026-2603-03680-mage-meta-reinforcement-learning-for-language-agents-toward-strategic-exploratio.md",
"local_title": "\"MAGE: Meta-Reinforcement Learning for Language Agents toward Strategic Exploration and Exploitation\"",
"local_status": "queued",
"local_topics": [
"memory",
"multi-agent",
"reasoning",
"tool-use"
],
"collection_queries": "language-agent",
"collection_score": "17"
},
{
"arxiv_id": "2603.07670",
"title": "Memory for Autonomous LLM Agents:Mechanisms, Evaluation, and Emerging Frontiers",
"abstract": "Large language model (LLM) agents increasingly operate in settings where a single context window is far too small to capture what has happened, what was learned, and what should not be repeated. Memory -- the ability to persist, organize, and selectively recall information across interactions -- is what turns a stateless text generator into a genuinely adaptive agent. This survey offers a structured account of how memory is designed, implemented, and evaluated in modern LLM-based agents, covering work from 2022 through early 2026. We formalize agent memory as a \\emph{write--manage--read} loop tightly coupled with perception and action, then introduce a three-dimensional taxonomy spanning temporal scope, representational substrate, and control policy. Five mechanism families are examined in depth: context-resident compression, retrieval-augmented stores, reflective self-improvement, hierarchical virtual context, and policy-learned management. On the evaluation side, we trace the shift from static recall benchmarks to multi-session agentic tests that interleave memory with decision-making, analyzing four recent benchmarks that expose stubborn gaps in current systems. We also survey applications where memory is the differentiating factor -- personal assistants, coding agents, open-world games, scientific reasoning, and multi-agent teamwork -- and address the engineering realities of write-path filtering, contradiction handling, latency budgets, and privacy governance. The paper closes with open challenges: continual consolidation, causally grounded retrieval, trustworthy reflection, learned forgetting, and multimodal embodied memory.",
"published": "2026-03-08",
"updated": "2026-03-08",
"authors": [
"Pengfei Du"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2603.07670",
"pdf_url": "https://arxiv.org/pdf/2603.07670",
"path": "papers/items/2026-memory-agent-survey.md",
"local_title": "\"Memory for Autonomous LLM Agents: Mechanisms, Evaluation, and Emerging Frontiers\"",
"local_status": "skimmed",
"local_topics": [
"memory",
"agent-architecture",
"agent-evaluation"
],
"collection_queries": null,
"collection_score": null
},
{
"arxiv_id": "2603.09002",
"title": "Security Considerations for Multi-agent Systems",
"abstract": "Multi-agent artificial intelligence systems or MAS are systems of autonomous agents that exercise delegated tool authority, share persistent memory, and coordinate via inter-agent communication. MAS introduces qualitatively distinct security vulnerabilities from those documented for singular AI models. Existing security and governance frameworks were not designed for these emerging attack surfaces. This study systematically characterizes the threat landscape of MAS and quantitatively evaluates 16 security frameworks for AI against it. A four-phase methodology is proposed: constructing a deep technical knowledge base of production multi-agent architectures; conducting generative AI-assisted threat modeling scoped to MAS cybersecurity risks and validated by domain experts; structuring survey plans at individual-threat granularity; and scoring each framework on a three-point scale against the cybersecurity risks. The risks were organized into 193 distinct main threat items across nine risk categories. The expected minimal average score is 2. No reviewed framework achieves majority coverage of any single category. Non-Determinism (mean score 1.231 across all 16 frameworks) and Data Leakage (1.340) are the most under-addressed domains. The OWASP Agentic Security Initiative leads overall at 65.3\\% coverage and in the design phase; the CDAO Generative AI Responsible AI Toolkit leads in development and operational coverage. These results provide the first empirical cross-framework comparison for MAS security and offer evidence-based guidance for framework selection. Please check back for information on the published journal version.",
"published": "2026-03-09",
"updated": "2026-04-26",
"authors": [
"Tam Nguyen",
"Moses Ndebugre",
"Dheeraj Arremsetty"
],
"categories": [
"cs.CR",
"cs.AI"
],
"url": "https://arxiv.org/abs/2603.09002",
"pdf_url": "https://arxiv.org/pdf/2603.09002",
"path": "papers/items/2026-2603-09002-security-considerations-for-multi-agent-systems.md",
"local_title": "Security Considerations for Multi-agent Systems",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"computer-use",
"memory",
"multi-agent",
"planning",
"rag",
"tool-use"
],
"collection_queries": "agent-safety",
"collection_score": "18"
},
{
"arxiv_id": "2603.15666",
"title": "Compiled Memory: Not More Information, but More Precise Instructions for Language Agents",
"abstract": "Existing memory systems for language agents address memory management: how to retrieve and page more information within a context budget. We address a complementary problem -- memory utility: what experience is worth keeping, and how it should change agent behavior. We present Atlas, a memory kernel that compiles accumulated task experience into an agent's instruction structure -- without fine-tuning, RAG, or human intervention. Memory is distillation, not storage; delivery is instruction rewriting, not context injection. Facts extracted from agent failures and successes are verified through a three-step promotion gate and delivered by rewriting the agent's system prompt with learned sub-bullets. On CUAD contract analysis, the evolved prompt improves GPT-4o token-level F1 by $+8.7$pp and precision by $+12.5$pp. On HotpotQA multi-hop QA, joint F1 improves $+3.16$pp. An ablation isolates the mechanism's defining property -- the training signal constraint: the evolved prompt learns exactly what it is taught, and nothing more. Applied to Claude Sonnet~4.5 using the same evolved prompt -- compiled from GPT-4o errors, unchanged -- joint F1 improves $+2.31$pp, with gains concentrating where Claude's stronger baseline leaves the most room -- confirming that the compiled knowledge is task-shaped, not model-shaped.",
"published": "2026-03-12",
"updated": "2026-03-12",
"authors": [
"James Rhodes",
"George Kang"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2603.15666",
"pdf_url": "https://arxiv.org/pdf/2603.15666",
"path": "papers/items/2026-2603-15666-compiled-memory-not-more-information-but-more-precise-instructions-for-language-.md",
"local_title": "\"Compiled Memory: Not More Information, but More Precise Instructions for Language Agents\"",
"local_status": "skimmed",
"local_topics": [
"memory",
"rag"
],
"collection_queries": "language-agent",
"collection_score": "13"
},
{
"arxiv_id": "2603.16734",
"title": "Differential Harm Propensity in Personalized LLM Agents: The Curious Case of Mental Health Disclosure",
"abstract": "Large language models (LLMs) are increasingly deployed as tool-using agents, shifting safety concerns from harmful text generation to harmful task completion. Deployed systems often condition on user profiles or persistent memory, yet agent safety evaluations typically ignore personalization signals. To address this gap, we investigated how mental health disclosure, a sensitive and realistic user-context cue, affects harmful behavior in agentic settings. Building on the AgentHarm benchmark, we evaluated frontier and open-source LLMs on multi-step malicious tasks (and their benign counterparts) under controlled prompt conditions that vary user-context personalization (no bio, bio-only, bio+mental health disclosure) and include a lightweight jailbreak injection. Our results reveal that harmful task completion is non-trivial across models: frontier lab models (e.g., GPT 5.2, Claude Sonnet 4.5, Gemini 3-Pro) still complete a measurable fraction of harmful tasks, while an open model (DeepSeek 3.2) exhibits substantially higher harmful completion. Adding a bio-only context generally reduces harm scores and increases refusals. Adding an explicit mental health disclosure often shifts outcomes further in the same direction, though effects are modest and not uniformly reliable after multiple-testing correction. Importantly, the refusal increase also appears on benign tasks, indicating a safety--utility trade-off via over-refusal. Finally, jailbreak prompting sharply elevates harm relative to benign conditions and can weaken or override the protective shift induced by personalization. Taken together, our results indicate that personalization can act as a weak protective factor in agentic misuse settings, but it is fragile under minimal adversarial pressure, highlighting the need for personalization-aware evaluations and safeguards that remain robust across user-context conditions.",
"published": "2026-03-17",
"updated": "2026-03-17",
"authors": [
"Caglar Yildirim"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2603.16734",
"pdf_url": "https://arxiv.org/pdf/2603.16734",
"path": "papers/items/2026-2603-16734-differential-harm-propensity-in-personalized-llm-agents-the-curious-case-of-ment.md",
"local_title": "\"Differential Harm Propensity in Personalized LLM Agents: The Curious Case of Mental Health Disclosure\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"rag",
"tool-use"
],
"collection_queries": "agent-safety",
"collection_score": "20"
},
{
"arxiv_id": "2603.19469",
"title": "A Framework for Formalizing LLM Agent Security",
"abstract": "Security in LLM agents is inherently contextual. For example, the same action taken by an agent may represent legitimate behavior or a security violation depending on whose instruction led to the action, what objective is being pursued, and whether the action serves that objective. However, existing definitions of security attacks against LLM agents often fail to capture this contextual nature. As a result, defenses face a fundamental utility-security tradeoff: applying defenses uniformly across all contexts can lead to significant utility loss, while applying defenses in insufficient or inappropriate contexts can result in security vulnerabilities. In this work, we present a framework that systematizes existing attacks and defenses from the perspective of contextual security. To this end, we propose four security properties that capture contextual security for LLM agents: task alignment (pursuing authorized objectives), action alignment (individual actions serving those objectives), source authorization (executing commands from authenticated sources), and data isolation (ensuring information flows respect privilege boundaries). We further introduce a set of oracle functions that enable verification of whether these security properties are violated as an agent executes a user task. Using this framework, we reformalize existing attacks, such as indirect prompt injection, direct prompt injection, jailbreak, task drift, and memory poisoning, as violations of one or more security properties, thereby providing precise and contextual definitions of these attacks. Similarly, we reformalize defenses as mechanisms that strengthen oracle functions or perform security property checks. Finally, we discuss several important future research directions enabled by our framework.",
"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"
],
"url": "https://arxiv.org/abs/2603.19469",
"pdf_url": "https://arxiv.org/pdf/2603.19469",
"path": "papers/items/2026-2603-19469-a-framework-for-formalizing-llm-agent-security.md",
"local_title": "A Framework for Formalizing LLM Agent Security",
"local_status": "queued",
"local_topics": [
"agent-safety",
"memory",
"tool-use"
],
"collection_queries": "agent-safety",
"collection_score": "14"
},
{
"arxiv_id": "2603.21357",
"title": "AgentHER: Hindsight Experience Replay for LLM Agent Trajectory Relabeling",
"abstract": "LLM-agent training pipelines routinely discard failed trajectories even though GPT-4o achieves only 14-20% on WebArena and below 55% pass@1 on ToolBench; even specialised systems at 50-65% leave the majority of trajectories unused. We introduce AgentHER, which recovers this lost signal by adapting Hindsight Experience Replay (HER) to natural-language agent trajectories: a trajectory that fails goal A is often a correct demonstration for an achievable alternative goal B. AgentHER realises this through a four-stage pipeline (failure classification, outcome extraction, LLM-guided relabeling with confidence gating, and data packaging) that converts discarded failures into SFT, DPO, and ShareGPT training data. On WebArena and ToolBench under a strict task-disjoint held-out protocol, AgentHER improves over success-only SFT by +7.6-11.4% across four model families (GPT-4o, Qwen2.5-72B/7B, LLaMA-3.1-8B), achieves 2x sample efficiency, and beats the strongest experience-centric baseline (Agent Workflow Memory) by +3.0-6.2%. Two robustness mechanisms, failure-severity weighting and cross-model multi-judge verification (gpt-4o-mini paired with Qwen2.5-72B-Instruct), reduce label noise from 5.9% to 2.9% and raise human-rated relabeling precision to 97.1% on WebArena and 96.0% on ToolBench. A full system-cost audit shows the entire relabeling pipeline costs 2.98 and 26 wall-clock minutes for 3,000 trajectories, i.e. 1.4 x 10^-3 per accepted pair. Code: https://github.com/alphadl/AgentHER",
"published": "2026-03-22",
"updated": "2026-05-10",
"authors": [
"Liang Ding"
],
"categories": [
"cs.AI",
"cs.CL"
],
"url": "https://arxiv.org/abs/2603.21357",
"pdf_url": "https://arxiv.org/pdf/2603.21357",
"path": "papers/items/2026-2603-21357-agenther-hindsight-experience-replay-for-llm-agent-trajectory-relabeling.md",
"local_title": "\"AgentHER: Hindsight Experience Replay for LLM Agent Trajectory Relabeling\"",
"local_status": "queued",
"local_topics": [
"computer-use",
"memory",
"tool-use",
"workflow-agent"
],
"collection_queries": "language-agent",
"collection_score": "16"
},
{
"arxiv_id": "2603.21564",
"title": "Toward a Theory of Hierarchical Memory for Language Agents",
"abstract": "Many recent long-context and agentic systems address context-length limitations by adding hierarchical memory: they extract atomic units from raw data, build multi-level representatives by grouping and compression, and traverse this structure to retrieve content under a token budget. Despite recurring implementations, there is no shared formalism for comparing design choices. We propose a unifying theory in terms of three operators. Extraction ($α$) maps raw data to atomic information units; coarsening ($C = (π, ρ)$) partitions units and assigns a representative to each group; and traversal ($τ$) selects which units to include in context given a query and budget. We identify a self-sufficiency spectrum for the representative function $ρ$ and show how it constrains viable retrieval strategies (a coarsening-traversal coupling). Finally, we instantiate the decomposition on eleven existing systems spanning document hierarchies, conversational memory, and agent execution traces, showcasing its generality.",
"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"
],
"url": "https://arxiv.org/abs/2603.21564",
"pdf_url": "https://arxiv.org/pdf/2603.21564",
"path": "papers/items/2026-2603-21564-toward-a-theory-of-hierarchical-memory-for-language-agents.md",
"local_title": "Toward a Theory of Hierarchical Memory for Language Agents",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"tool-use"
],
"collection_queries": "language-agent",
"collection_score": "16"
},
{
"arxiv_id": "2603.24257",
"title": "Memory-Augmented Vision-Language Agents for Persistent and Semantically Consistent Object Captioning",
"abstract": "Vision-Language Models (VLMs) often yield inconsistent descriptions of the same object across viewpoints, hindering the ability of embodied agents to construct consistent semantic representations over time. Previous methods resolved inconsistencies using offline multi-view aggregation or multi-stage pipelines that decouple exploration, data association, and caption learning, with limited capacity to reason over previously observed objects. In this paper, we introduce a unified, memory-augmented Vision-Language agent that simultaneously handles data association, object captioning, and exploration policy within a single autoregressive framework. The model processes the current RGB observation, a top-down explored map, and an object-level episodic memory serialized into object-level tokens, ensuring persistent object identity and semantic consistency across extended sequences. To train the model in a self-supervised manner, we collect a dataset in photorealistic 3D environments using a disagreement-based policy and a pseudo-captioning model that enforces consistency across multi-view caption histories. Extensive evaluation on a manually annotated object-level test set, demonstrate improvements of up to +11.86% in standard captioning scores and +7.39% in caption self-similarity over baseline models, while enabling scalable performance through a compact scene representation. Code, model weights, and data are available at https://hsp-iit.github.io/epos-vlm/.",
"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"
],
"url": "https://arxiv.org/abs/2603.24257",
"pdf_url": "https://arxiv.org/pdf/2603.24257",
"path": "papers/items/2026-2603-24257-memory-augmented-vision-language-agents-for-persistent-and-semantically-consiste.md",
"local_title": "Memory-Augmented Vision-Language Agents for Persistent and Semantically Consistent Object Captioning",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"embodied-agent",
"memory"
],
"collection_queries": "language-agent",
"collection_score": "17"
},
{
"arxiv_id": "2603.28428",
"title": "Synergy: A Next-Generation General-Purpose Agent for Open Agentic Web",
"abstract": "AI agents are rapidly expanding in both capability and population: they now write code, operate computers across platforms, manage cloud infrastructure, and make purchasing decisions, while open-source frameworks such as OpenClaw are putting personal agents in the hands of millions and embodied agents are spreading across smartphones, vehicles, and robots. As the internet prepares to host billions of such entities, it is shifting toward what we call Open Agentic Web, a decentralized digital ecosystem in which agents from different users, organizations, and runtimes can discover one another, negotiate task boundaries, and delegate work across open technical and social surfaces at scale. Yet most of today's agents remain isolated tools or closed-ecosystem orchestrators rather than socially integrated participants in open networks. We argue that the next generation of agents must become Agentic Citizens, defined by three requirements: Agentic-Web-Native Collaboration, participation in open collaboration networks rather than only closed internal orchestration; Agent Identity and Personhood, continuity as a social entity rather than a resettable function call; and Lifelong Evolution, improvement across task performance, communication, and collaboration over time. We present Synergy, a general-purpose agent architecture and runtime harness for persistent, collaborative, and evolving agents on Open Agentic Web, grounding collaboration in session-native orchestration, repository-backed workspaces, and social communication; identity in typed memory, notes, agenda, skills, and persistent social relationships; and evolution in an experience-centered learning mechanism that proactively recalls rewarded trajectories at inference time.",
"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"
],
"url": "https://arxiv.org/abs/2603.28428",
"pdf_url": "https://arxiv.org/pdf/2603.28428",
"path": "papers/items/2026-2603-28428-synergy-a-next-generation-general-purpose-agent-for-open-agentic-web.md",
"local_title": "\"Synergy: A Next-Generation General-Purpose Agent for Open Agentic Web\"",
"local_status": "queued",
"local_topics": [
"coding-agent",
"embodied-agent",
"memory",
"multi-agent",
"rag",
"tool-use"
],
"collection_queries": "function-calling",
"collection_score": "16"
},
{
"arxiv_id": "2604.17562",
"title": "SafeAgent: A Runtime Protection Architecture for Agentic Systems",
"abstract": "Large language model (LLM) agents are vulnerable to prompt-injection attacks that propagate through multi-step workflows, tool interactions, and persistent context, making input-output filtering alone insufficient for reliable protection. This paper presents SafeAgent, a runtime security architecture that treats agent safety as a stateful decision problem over evolving interaction trajectories. The proposed design separates execution governance from semantic risk reasoning through two coordinated components: a runtime controller that mediates actions around the agent loop and a context-aware decision core that operates over persistent session state. The core is formalized as a context-aware advanced machine intelligence and instantiated through operators for risk encoding, utility-cost evaluation, consequence modeling, policy arbitration, and state synchronization. Experiments on Agent Security Bench (ASB) and InjecAgent show that SafeAgent consistently improves robustness over baseline and text-level guardrail methods while maintaining competitive benign-task performance. Ablation studies further show that recovery confidence and policy weighting determine distinct safety-utility operating points.",
"published": "2026-04-19",
"updated": "2026-04-19",
"authors": [
"Hailin Liu",
"Eugene Ilyushin",
"Jie Ni",
"Min Zhu"
],
"categories": [
"cs.AI",
"cs.MA"
],
"url": "https://arxiv.org/abs/2604.17562",
"pdf_url": "https://arxiv.org/pdf/2604.17562",
"path": "papers/items/2026-2604-17562-safeagent-a-runtime-protection-architecture-for-agentic-systems.md",
"local_title": "\"SafeAgent: A Runtime Protection Architecture for Agentic Systems\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"coding-agent",
"memory",
"reasoning",
"tool-use",
"workflow-agent"
],
"collection_queries": "agent-safety",
"collection_score": "19"
},
{
"arxiv_id": "2604.18847",
"title": "Human-Guided Harm Recovery for Computer Use Agents",
"abstract": "As LM agents gain the ability to execute actions on real computer systems, we need ways to not only prevent harmful actions at scale but also effectively remediate harm when prevention fails. We formalize a solution to this neglected challenge in post-execution safeguards as harm recovery: the problem of optimally steering an agent from a harmful state back to a safe one in alignment with human preferences. We ground preference-aligned recovery through a formative user study that identifies valued recovery dimensions and produces a natural language rubric. Our dataset of 1,130 pairwise judgments reveals context-dependent shifts in attribute importance, such as preferences for pragmatic, targeted strategies over comprehensive long-term approaches. We operationalize these learned insights in a reward model, re-ranking multiple candidate recovery plans generated by an agent scaffold at test time. To evaluate recovery capabilities systematically, we introduce BackBench, a benchmark of 50 computer-use tasks that test an agent's ability to recover from harmful states. Human evaluation shows our reward model scaffold yields higher-quality recovery trajectories than base agents and rubric-based scaffolds. Together, these contributions lay the foundation for a new class of agent safety methods -- ones that confront harm not only by preventing it, but by navigating its aftermath with alignment and intent.",
"published": "2026-04-20",
"updated": "2026-05-28",
"authors": [
"Christy Li",
"Sky CH-Wang",
"Andi Peng",
"Andreea Bobu"
],
"categories": [
"cs.AI",
"cs.CL"
],
"url": "https://arxiv.org/abs/2604.18847",
"pdf_url": "https://arxiv.org/pdf/2604.18847",
"path": "papers/items/2026-2604-18847-human-guided-harm-recovery-for-computer-use-agents.md",
"local_title": "Human-Guided Harm Recovery for Computer Use Agents",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"computer-use",
"memory",
"planning",
"rag",
"tool-use"
],
"collection_queries": "agent-safety",
"collection_score": "18"
},
{
"arxiv_id": "2604.23374",
"title": "Ghost in the Agent: Redefining Information Flow Tracking for LLM Agents",
"abstract": "Autonomous Large Language Model (LLM) agents are increasingly deployed to conduct complex tasks by interacting with external tools, APIs, and memory stores. However, processing untrusted external data exposes these agents to severe security threats, such as indirect prompt injection and unauthorized tool execution. Securing these systems requires effective information flow tracking. Yet, traditional taint analysis that is designed for program memory states fundamentally fails when applied to LLMs, where data propagation is governed by probabilistic natural language reasoning. In this paper, we present NeuroTaint, the first comprehensive taint tracking framework tailored for the unique information flow characteristics of LLM agents. Our key insight is that taint propagation in LLM agents must be understood not only as explicit content transfer, but also as semantic transformation, causal influence on decisions, and cross-session persistence through memory. NeuroTaint therefore audits execution traces offline to reconstruct provenance from untrusted sources to privileged sinks using semantic evidence, causal reasoning, and persistent context tracking, rather than relying on exact string matches or pre-defined source-sink paths alone. Extensive evaluation using TaintBench, our 400-scenario benchmark spanning 20 real-world agent frameworks, shows that NeuroTaint substantially outperforms FIDES, an information-flow-control (IFC)-style baseline for LLM agents, in source-sink propagation detection. We further show that NeuroTaint remains effective on established agent-security benchmarks, including InjecAgent and ToolEmu, while operating offline with modest additional auditing cost.",
"published": "2026-04-25",
"updated": "2026-04-25",
"authors": [
"Yuandao Cai",
"Wensheng Tang",
"Cheng Wen",
"Shengchao Qin"
],
"categories": [
"cs.CR"
],
"url": "https://arxiv.org/abs/2604.23374",
"pdf_url": "https://arxiv.org/pdf/2604.23374",
"path": "papers/items/2026-2604-23374-ghost-in-the-agent-redefining-information-flow-tracking-for-llm-agents.md",
"local_title": "\"Ghost in the Agent: Redefining Information Flow Tracking for LLM Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"reasoning",
"tool-use"
],
"collection_queries": "agent-safety",
"collection_score": "18"
},
{
"arxiv_id": "2604.23459",
"title": "Architecture Matters for Multi-Agent Security",
"abstract": "Multi-agent systems (MAS), composed of networks of two or more autonomous AI agents, have become increasingly popular in production deployments, yet introduce security risks that do not arise in single-agent settings. Even if individual agents exhibit robust security, architectural decisions governing their coordination can create attack surfaces that have not been systematically characterized. In this work, we present an empirical study of how MAS design decisions shape the tradeoff between task performance and attack resistance. Across three agentic environments (browser, desktop, and code) and 13 architectural configurations, we use stagewise evaluations that distinguish planning refusal, execution-stage interception, partial harmful execution, and successful attack completion to study three key design choices: (i) agent roles, which determine how authority and responsibility are allocated; (ii) communication topology, which shapes how and when agents interact; and (iii) memory, which determines the context and state visibility accessible to each agent. We find that multi-agent architectures are more vulnerable than standalone agents in the majority of configurations, with attack success rates varying by up to 3.8x at comparable or higher benign accuracy, and that no single design is universally safer. These results motivate the development of further evaluations that move beyond the security properties of a single agent.",
"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"
],
"url": "https://arxiv.org/abs/2604.23459",
"pdf_url": "https://arxiv.org/pdf/2604.23459",
"path": "papers/items/2026-2604-23459-architecture-matters-for-multi-agent-security.md",
"local_title": "Architecture Matters for Multi-Agent Security",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"computer-use",
"memory",
"multi-agent",
"planning"
],
"collection_queries": "agent-safety",
"collection_score": "18"
},
{
"arxiv_id": "2604.24212",
"title": "Empowering Autonomous Debugging Agents with Efficient Dynamic Analysis",
"abstract": "Autonomous agents for automated program repair represent a promising frontier in software engineering, yet their effectiveness is often hindered by reliance on post-mortem, coarse-grained execution feedback. While integrating traditional interactive debuggers seems a natural solution, their low-level, line-by-line interaction paradigm turns out to be cost-inefficient for LLM-based agents, leading to exhausted budgets and unproductive loops. To mitigate this, we introduce Agent-centric Debugging Interface (ADI), a novel agent-centric debugging interface designed for cost-efficient, end-to-end autonomous interaction. Specifically, Agent-centric Debugging Interface realizes a function-level interaction paradigm, powered by our Frame Lifetime Trace, a comprehensive data structure encapsulating a function's stateful execution trace, and a set of high-level navigational commands. Our extensive evaluation on the SWE-bench benchmark demonstrates the effectiveness and efficiency of ADI. By simply equipping a basic agent with ADI, it successfully resolves 63.8\\% of the tasks on the SWE-bench Verified set, even slightly outperforming the highly optimized and high-investment Claude-Tools agent, at an average cost of USD 1.28 per task with Claude-Sonnet-3.7. Furthermore, we demonstrate ADI's generality by integrating it as a plug-and-play component into existing SOTA agents, delivering consistent gains ranging from 6.2\\% to 18.5\\% on the resolved tasks. These results indicate that Agent-centric Debugging Interface can provide a general and efficient enhancement for existing autonomous agents.",
"published": "2026-04-27",
"updated": "2026-04-27",
"authors": [
"Jiahong Xiang",
"Xiaoyang Xu",
"Xiaopan Chu",
"Hongliang Tian",
"Yuqun Zhang"
],
"categories": [
"cs.SE"
],
"url": "https://arxiv.org/abs/2604.24212",
"pdf_url": "https://arxiv.org/pdf/2604.24212",
"path": "papers/items/2026-2604-24212-empowering-autonomous-debugging-agents-with-efficient-dynamic-analysis.md",
"local_title": "Empowering Autonomous Debugging Agents with Efficient Dynamic Analysis",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"coding-agent",
"embodied-agent",
"memory",
"rag",
"tool-use"
],
"collection_queries": "autonomous-agent-llm",
"collection_score": "16"
},
{
"arxiv_id": "2604.27092",
"title": "End-to-end autonomous scientific discovery on a real optical platform",
"abstract": "Scientific research has long been human-led, driving new knowledge and transformative technologies through the continual revision of questions, methods and claims as evidence accumulates. Although large language model (LLM)-based agents are beginning to move beyond assisting predefined research workflows, none has yet demonstrated end-to-end autonomous discovery in a real physical system that produces a nontrivial result supported by experimental evidence. Here we introduce Qiushi Discovery Engine, an LLM-based agentic system for end-to-end autonomous scientific discovery on a real optical platform. Qiushi Engine combines nonlinear research phases, Meta-Trace memory and a dual-layer architecture to maintain adaptive and stable research trajectories across long-horizon investigations involving thousands of LLM-mediated reasoning, measurement and revision actions. It autonomously reproduces a published transmission-matrix experiment on a non-original platform and converts an abstract coherence-order theory into experimental observables, providing, to our knowledge, the first observation of this class of coherence-order structure. More importantly, in an open-ended study involving 145.9 million tokens, 3,242 LLM calls, 1,242 tool calls, 163 research notes and 44 scripts, Qiushi Engine proposes and experimentally validates optical bilinear interaction, a physical mechanism structurally analogous to a core operation in Transformer attention. This AI-discovered mechanism suggests a route towards high-speed, energy-efficient optical hardware for pairwise computation. To our knowledge, this is the first demonstration of an AI agentic system autonomously identifying and experimentally validating a nontrivial, previously unreported physical mechanism, marking a milestone for research-level autonomous agents.",
"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"
],
"url": "https://arxiv.org/abs/2604.27092",
"pdf_url": "https://arxiv.org/pdf/2604.27092",
"path": "papers/items/2026-2604-27092-end-to-end-autonomous-scientific-discovery-on-a-real-optical-platform.md",
"local_title": "End-to-end autonomous scientific discovery on a real optical platform",
"local_status": "queued",
"local_topics": [
"memory",
"planning",
"reasoning",
"tool-use",
"workflow-agent"
],
"collection_queries": "autonomous-agent-llm",
"collection_score": "13"
},
{
"arxiv_id": "2604.27699",
"title": "Bridging Values and Behavior: A Hierarchical Framework for Proactive Embodied Agents",
"abstract": "Current embodied agents are often limited to passive instruction-following or reactive need-satisfaction, lacking a stable, high-order value framework essential for long-term, self-directed behavior and resolving motivational conflicts. We introduce \\textit{ValuePlanner}, a hierarchical cognitive architecture that decouples high-level value scheduling from low-level action execution. \\textit{ValuePlanner} employs an LLM-based cognitive module to generate symbolic subgoals by reasoning through abstract value trade-offs, which are then translated into executable action plans by a classical PDDL planner. This process is refined via a closed-loop feedback mechanism. Evaluating such autonomy requires methods beyond task-success rates, and we therefore propose a value-centric evaluation suite measuring cumulative value gain, preference alignment, and behavioral diversity. Experiments in the TongSim household environment demonstrate that \\textit{ValuePlanner} arbitrates competing values to generate coherent, long-horizon, self-directed behavior absent from instruction-following and needs-driven baselines. Our work offers a structured approach to bridging intrinsic values and grounded behavior for autonomous agents.",
"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"
],
"url": "https://arxiv.org/abs/2604.27699",
"pdf_url": "https://arxiv.org/pdf/2604.27699",
"path": "papers/items/2026-2604-27699-bridging-values-and-behavior-a-hierarchical-framework-for-proactive-embodied-age.md",
"local_title": "\"Bridging Values and Behavior: A Hierarchical Framework for Proactive Embodied Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"embodied-agent",
"memory",
"planning",
"reasoning",
"tool-use"
],
"collection_queries": "autonomous-agent-llm",
"collection_score": "15"
},
{
"arxiv_id": "2604.27859",
"title": "Rethinking Agentic Reinforcement Learning In Large Language Models",
"abstract": "Reinforcement Learning (RL) has traditionally focused on training specialized agents to optimize predefined reward functions within narrowly defined environments. However, the advent of powerful Large Language Models (LLMs) and increasingly complex, open-ended tasks has catalyzed a paradigm shift towards agentic paradigms within RL. This emerging framework extends beyond traditional RL by emphasizing the development of autonomous agents capable of goal-setting, long-term planning, dynamic strategy adaptation, and interactive reasoning in uncertain, real-world environments. Unlike conventional approaches that rely heavily on static objectives and episodic interactions, LLM-based Agentic RL incorporates cognitive-like capabilities such as meta-reasoning, self-reflection, and multi-step decision-making directly into the learning loop. In this paper, we provide a deep insight for looking the conceptual foundations, methodological innovations, and effective designs underlying this trend. Furthermore, we identify critical challenges and outline promising future directions for building LLM-based Agentic RL.",
"published": "2026-04-30",
"updated": "2026-05-15",
"authors": [
"Fangming Cui",
"Ruixiao Zhu",
"Cheng Fang",
"Sunan Li",
"Jiahong Li"
],
"categories": [
"cs.AI",
"cs.ET"
],
"url": "https://arxiv.org/abs/2604.27859",
"pdf_url": "https://arxiv.org/pdf/2604.27859",
"path": "papers/items/2026-2604-27859-rethinking-agentic-reinforcement-learning-in-large-language-models.md",
"local_title": "Rethinking Agentic Reinforcement Learning In Large Language Models",
"local_status": "queued",
"local_topics": [
"memory",
"planning",
"reasoning",
"tool-use"
],
"collection_queries": "autonomous-agent-llm",
"collection_score": "14"
},
{
"arxiv_id": "2604.28157",
"title": "FlashRT: Towards Computationally and Memory Efficient Red-Teaming for Prompt Injection and Knowledge Corruption",
"abstract": "Long-context large language models (LLMs)-for example, Gemini-3.1-Pro and Qwen-3.5-are widely used to empower many real-world applications, such as retrieval-augmented generation, autonomous agents, and AI assistants. However, security remains a major concern for their widespread deployment, with threats such as prompt injection and knowledge corruption. To quantify the security risks faced by LLMs under these threats, the research community has developed heuristic-based and optimization-based red-teaming methods. Optimization-based methods generally produce stronger attacks than heuristic attacks and thus provide a more rigorous assessment of LLM security risks. However, they are often resource-intensive, requiring significant computation and GPU memory, especially for long context scenarios. The resource-intensive nature poses a major obstacle for the community (especially academic researchers) to systematically evaluate the security risks of long-context LLMs and assess the effectiveness of defense strategies at scale. In this work, we propose FlashRT, the first framework to improve the efficiency (in terms of both computation and memory) for optimization-based prompt injection and knowledge corruption attacks under long-context LLMs. Through extensive evaluations, we find that FlashRT consistently delivers a 2x-7x speedup (e.g., reducing runtime from one hour to less than ten minutes) and a 2x-4x reduction in GPU memory consumption (e.g., reducing from 264.1 GB to 65.7 GB GPU memory for a 32K token context) compared to state-of-the-art baseline nanoGCG. FlashRT can be broadly applied to black-box optimization methods, such as TAP and AutoDAN. We hope FlashRT can serve as a red-teaming tool to enable systematic evaluation of long-context LLM security. The code is available at: https://github.com/Wang-Yanting/FlashRT",
"published": "2026-04-30",
"updated": "2026-04-30",
"authors": [
"Yanting Wang",
"Chenlong Yin",
"Ying Chen",
"Jinyuan Jia"
],
"categories": [
"cs.CR"
],
"url": "https://arxiv.org/abs/2604.28157",
"pdf_url": "https://arxiv.org/pdf/2604.28157",
"path": "papers/items/2026-2604-28157-flashrt-towards-computationally-and-memory-efficient-red-teaming-for-prompt-inje.md",
"local_title": "\"FlashRT: Towards Computationally and Memory Efficient Red-Teaming for Prompt Injection and Knowledge Corruption\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"rag",
"tool-use"
],
"collection_queries": "autonomous-agent-llm",
"collection_score": "14"
},
{
"arxiv_id": "2605.03312",
"title": "MemFlow: Intent-Driven Memory Orchestration for Small Language Model Agents",
"abstract": "Modern language agents must operate over long-horizon, multi-turn histories, yet deploying such agents with Small Language Models (SLMs) remains fundamentally difficult. Full-context prompting causes context overflow, flat retrieval exposes the model to noisy evidence, and open-ended agentic loops are unreliable under limited reasoning capacity. We argue that a substantial portion of SLM memory failure arises from mismatched memory operations: different query types demand categorically different retrieval strategies, evidence transformations, and context budgets that SLMs cannot reliably self-orchestrate through open-ended reasoning. We introduce MemFlow, a training-free memory orchestration framework that externalizes memory planning from the SLM. A Router Agent classifies each query by intent and dispatches it to the Memory Agent, which executes one of three specialized tiers (Profile Lookup, Targeted Retrieval, or Deep Reasoning) and assembles the resulting evidence under a dynamic, tier-aware token budget. An Answer Agent then generates a response from this compact context, and a Validator Agent optionally retries with a heavier memory tier when the response is not supported by the provided evidence. This route-then-compile design avoids tool-selection hallucination and reasoning loops while keeping the answer context compact. Evaluated on a frozen Qwen3-1.7B backbone across long-horizon memory benchmarks - LongMemEval, LoCoMo, and LongBench - MemFlow improves accuracy by nearly 2x over full-context SLM baselines. These results suggest that structured intent routing and deterministic evidence preparation can make limited-capacity models substantially more effective in resource-constrained long-horizon agents.",
"published": "2026-05-05",
"updated": "2026-05-05",
"authors": [
"Jiayi Chen",
"Yingcong Li",
"Guiling Wang"
],
"categories": [
"cs.MA"
],
"url": "https://arxiv.org/abs/2605.03312",
"pdf_url": "https://arxiv.org/pdf/2605.03312",
"path": "papers/items/2026-2605-03312-memflow-intent-driven-memory-orchestration-for-small-language-model-agents.md",
"local_title": "\"MemFlow: Intent-Driven Memory Orchestration for Small Language Model Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"planning",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "language-agent",
"collection_score": "19"
},
{
"arxiv_id": "2605.05704",
"title": "SafeHarbor: Hierarchical Memory-Augmented Guardrail for LLM Agent Safety",
"abstract": "Recent advances in foundation models have transformed LLMs from passive conversational systems into autonomous agents capable of reasoning and tool execution. While these capabilities unlock substantial practical value, they also introduce new security risks, as adversaries can manipulate agents into performing harmful actions in real-world environments. Existing defense strategies mitigate such threats but frequently struggle to balance safety and utility, resulting in over-refusal of benign user requests. To mitigate this trade-off, we propose SafeHarbor, a novel framework designed to establish precise decision boundaries for LLM agents. Unlike static guidelines, SafeHarbor extracts context-aware defense rules through enhanced adversarial generation. We design a local hierarchical memory system for dynamic rule injection, offering a training-free, efficient, and plug-and-play solution. Furthermore, we introduce an information entropy-based self-evolution mechanism that continuously optimizes the memory structure through dynamic node splitting and merging. Extensive experiments demonstrate that SafeHarbor achieves state-of-the-art performance on both ambiguous benign tasks and explicit malicious attacks, notably attaining a peak benign utility of 63.6\\% on GPT-4o while maintaining a robust refusal rate exceeding 93\\% against harmful requests. The source code is publicly available at https://github.com/ljj-cyber/SafeHarbor.",
"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"
],
"url": "https://arxiv.org/abs/2605.05704",
"pdf_url": "https://arxiv.org/pdf/2605.05704",
"path": "papers/items/2026-2605-05704-safeharbor-hierarchical-memory-augmented-guardrail-for-llm-agent-safety.md",
"local_title": "\"SafeHarbor: Hierarchical Memory-Augmented Guardrail for LLM Agent Safety\"",
"local_status": "queued",
"local_topics": [
"agent-safety",
"computer-use",
"memory",
"reasoning",
"tool-use"
],
"collection_queries": "agent-safety, autonomous-agent-llm",
"collection_score": "22"
},
{
"arxiv_id": "2605.05716",
"title": "More Is Not Always Better: Cross-Component Interference in LLM Agent Scaffolding",
"abstract": "LLM agent systems are built by stacking scaffolding components (planning, tools, memory, self-reflection, retrieval) assuming more is better. We study cross-component interference (CCI): degradation when components interact destructively. We run a full factorial experiment over all 2^5=32 subsets of five components on HotpotQA and GSM8K with Llama-3.1-8B/70B (96 conditions, up to 10 seeds). The All-In system is consistently suboptimal: on HotpotQA, a single-tool agent surpasses All-In by 32% (F1 0.233 vs 0.177, p=0.023); on GSM8K, a 3-component subset beats All-In by 79% (0.43 vs 0.24, p=0.010). Optimal component count is task-dependent (k*=1-4) and scale-sensitive: at 70B, combinations that hurt at 8B provide gains, though All-In still trails the best subset. We fit a main-effects regression (R^2=0.916, adj-R^2=0.899, LOOCV=0.872), compute exact Shapley values, and find 183/325 submodularity violations (56.3%), showing greedy selection is unreliable. A three-body synergy among Tool Use, Self-Reflection, and Retrieval (INT_3=+0.175, 95% CI [+0.003,+0.351]) is reported as exploratory. CCI replicates across model families (Qwen2.5) and is robust to prompt paraphrasing. Our findings suggest maximally-equipped agent defaults should be replaced by task-specific subset selection via interaction-aware analysis.",
"published": "2026-05-07",
"updated": "2026-05-07",
"authors": [
"Ming Liu"
],
"categories": [
"cs.AI",
"cs.CL"
],
"url": "https://arxiv.org/abs/2605.05716",
"pdf_url": "https://arxiv.org/pdf/2605.05716",
"path": "papers/items/2026-2605-05716-more-is-not-always-better-cross-component-interference-in-llm-agent-scaffolding.md",
"local_title": "\"More Is Not Always Better: Cross-Component Interference in LLM Agent Scaffolding\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"planning",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "planning-agent",
"collection_score": "18"
},
{
"arxiv_id": "2605.06716",
"title": "From Storage to Experience: A Survey on the Evolution of LLM Agent Memory Mechanisms",
"abstract": "Large Language Model (LLM)-based agents have fundamentally reshaped artificial intelligence by integrating external tools and planning capabilities. While memory mechanisms have emerged as the architectural cornerstone of these systems, current research remains fragmented, oscillating between operating system engineering and cognitive science. This theoretical divide prevents a unified view of technological synthesis and a coherent evolutionary perspective. To bridge this gap, this survey proposes a novel evolutionary framework for LLM agent memory mechanisms, formalizing the development process into three stages: Storage (trajectory preservation), Reflection (trajectory refinement), and Experience (trajectory abstraction). We first formally define these three stages before analyzing the three core drivers of this evolution: the necessity for long-range consistency, the challenges in dynamic environments, and the ultimate goal of continual learning. Furthermore, we specifically explore two transformative mechanisms in the frontier Experience stage: proactive exploration and cross-trajectory abstraction. By synthesizing these disparate views, this work offers robust design principles and a clear roadmap for the development of next-generation LLM agents.",
"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"
],
"url": "https://arxiv.org/abs/2605.06716",
"pdf_url": "https://arxiv.org/pdf/2605.06716",
"path": "papers/items/2026-2605-06716-from-storage-to-experience-a-survey-on-the-evolution-of-llm-agent-memory-mechani.md",
"local_title": "\"From Storage to Experience: A Survey on the Evolution of LLM Agent Memory Mechanisms\"",
"local_status": "queued",
"local_topics": [
"memory",
"planning",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "planning-agent",
"collection_score": "18"
},
{
"arxiv_id": "2605.06812",
"title": "Towards Security-Auditable LLM Agents: A Unified Graph Representation",
"abstract": "LLM-based agentic systems are rapidly evolving to perform complex autonomous tasks through dynamic tool invocation, stateful memory management, and multi-agent collaboration. However, this semantics-driven execution paradigm creates a severe semantic gap between low-level physical events and high-level execution intent, making post-hoc security auditing fundamentally difficult. Existing representation mechanisms, including static SBOMs and runtime logs, provide only fragmented evidence and fail to capture cognitive-state evolution, capability bindings, persistent memory contamination, and cascading risk propagation across interacting agents. To bridge this gap, we propose Agent-BOM, a unified structural representation for agent security auditing. Agent-BOM models an agentic system as a hierarchical attributed directed graph that separates static capability bases, such as models, tools, and long-term memory, from dynamic runtime semantic states, such as goals, reasoning trajectories, and actions. These layers are connected through semantic edges and security attributes, transforming fragmented execution traces into queryable audit paths. Building on Agent-BOM, we develop a graph-query-based paradigm for path-level risk assessment and instantiate it with the OWASP Agentic Top 10. We further implement an auditing plugin in the OpenClaw environment to construct Agent-BOM from live executions. Evaluation on representative real-world agentic attack scenarios shows that Agent-BOM can reconstruct stealthy attack chains, including cross-session memory poisoning and tool misuse, capability supply-chain hijacking and unexpected code execution, multi-agent ecosystem hijacking, and privilege and trust abuse. These results demonstrate that Agent-BOM provides a unified and auditable foundation for root-cause analysis and security adjudication in complex agentic ecosystems.",
"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"
],
"url": "https://arxiv.org/abs/2605.06812",
"pdf_url": "https://arxiv.org/pdf/2605.06812",
"path": "papers/items/2026-2605-06812-towards-security-auditable-llm-agents-a-unified-graph-representation.md",
"local_title": "\"Towards Security-Auditable LLM Agents: A Unified Graph Representation\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"multi-agent",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "agent-safety",
"collection_score": "21"
},
{
"arxiv_id": "2605.08374",
"title": "MemQ: Integrating Q-Learning into Self-Evolving Memory Agents over Provenance DAGs",
"abstract": "Episodic memory allows LLM agents to accumulate and retrieve experience, but current methods treat each memory independently, i.e., evaluating retrieval quality in isolation without accounting for the dependency chains through which memories enable the creation of future memories. We introduce MemQ, which applies TD($λ$) eligibility traces to memory Q-values, propagating credit backward through a provenance DAG that records which memories were retrieved when each new memory was created. Credit weight decays as $(γλ)^d$ with DAG depth $d$, replacing temporal distance with structural proximity. We formalize the setting as an Exogenous-Context MDP, whose factored transition decouples the exogenous task stream from the endogenous memory store. Across six benchmarks, spanning OS interaction, function calling, code generation, multimodal reasoning, embodied reasoning, and expert-level QA, MemQ achieves the highest success rate on all six in generalization evaluation and runtime learning, with gains largest on multi-step tasks that produce deep and relevant provenance chains (up to +5.7~pp) and smallest on single-step classification (+0.77~pp) where single-step updates already suffice. We further study how $γ$ and $λ$ interact with the EC-MDP structure, providing principled guidance for parameter selection and future research. Code is available at https://github.com/jwliao-ai/MemQ.",
"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"
],
"url": "https://arxiv.org/abs/2605.08374",
"pdf_url": "https://arxiv.org/pdf/2605.08374",
"path": "papers/items/2026-2605-08374-memq-integrating-q-learning-into-self-evolving-memory-agents-over-provenance-dag.md",
"local_title": "\"MemQ: Integrating Q-Learning into Self-Evolving Memory Agents over Provenance DAGs\"",
"local_status": "skimmed",
"local_topics": [
"agent-evaluation",
"computer-use",
"embodied-agent",
"memory",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "function-calling",
"collection_score": "20"
},
{
"arxiv_id": "2605.08442",
"title": "Defense effectiveness across architectural layers: a mechanistic evaluation of persistent memory attacks on stateful LLM agents",
"abstract": "Persistent memory in LLM agents creates an attack surface that production safety classifiers do not observe: the payload enters via RAG retrieval and persists across sessions via tool-mediated memory. We evaluate six defenses across four architectural layers against delayed-trigger attacks on nine open-source models (5,040 runs, N=40 per condition). Five of six defenses fail: input-level filters never see the payload (it enters via RAG, not user input); retrieval-level classifiers observe it but cannot distinguish compliance-framed injection from legitimate policy; instruction-level hardening is overridden by the stored rule's compliance framing. Only tool-gating at the memory layer (Memory Sandbox) reduces ASR to 0% for eight of nine models, with zero utility cost. A reasoning model inverts this defense via goal-directed RAG fallback, a mechanism that replicates cross-family on Bedrock. A reasoning-mode ablation reveals a double dissociation: no single sandbox implementation is safe across both reasoning and non-reasoning model classes. We resolve this with a content-layer proof-of-concept (RATG), validated on non-reasoning models. A loaded-corpus frontier evaluation (21 models, 3 providers, N=40) overturns an initial empty-corpus screen showing 0/210 exfiltrations: that was a threat-model artifact, not model safety. Under realistic conditions, Gemini 3.1 Pro Preview exfiltrates at 95% ASR, GPT-5.1 regresses to 22.5% relative to GPT-5 (5%), and Anthropic blocks at the injection layer (0-17.5% storage, 0% ASR). Nearly all OpenAI and Gemini models store the rule at 100% regardless of execution resistance, creating supply-chain risk in shared-memory deployments. Defense effectiveness is determined by architectural layer and reasoning capability, not classifier quality.",
"published": "2026-05-08",
"updated": "2026-07-03",
"authors": [
"Jun Wen Leong"
],
"categories": [
"cs.CR",
"cs.AI",
"cs.LG"
],
"url": "https://arxiv.org/abs/2605.08442",
"pdf_url": "https://arxiv.org/pdf/2605.08442",
"path": "papers/items/2026-2605-08442-defense-effectiveness-across-architectural-layers-a-mechanistic-evaluation-of-pe.md",
"local_title": "\"Defense effectiveness across architectural layers: a mechanistic evaluation of persistent memory attacks on stateful LLM agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"computer-use",
"memory",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "agent-safety",
"collection_score": "23"
},
{
"arxiv_id": "2605.09692",
"title": "Causal state binding predicts action control in language agents",
"abstract": "Autonomous language agents increasingly expose traces, memories, plans and constraints, but existing evaluations rarely test whether these state variables are bound to final actions. We introduce causal state binding, an intervention-coupled evaluation framework that measures whether actions change with the event-specific decisive state while remaining invariant to irrelevant cues. The primary readout is a hidden-target finite-action benchmark in which scorer-side intervention targets are assigned before generation and withheld from the model-visible prompt. Across 57,816 scored records in seven corpus-level units, structured-agent conditions exceeded high-randomness controls and targeted component removals on reason, memory, veto and self-continuity responsiveness. Open-weight validation across Qwen2.5 7B, 14B and 32B plus Mistral-7B showed that action priors, no-field prompts and scrambled decisive context did not recover the structured-control signature. In diagnostic finite-action probes, the minimal decisive-field readout recovered the prescribed action pattern whereas surface-only, action-prior-only and scrambled-field controls did not. Across 300 SWE-bench Lite issue records and six API models, adding an oracle-free causal state-binding composite to a full non-CSB baseline increased constraint-clean issue-to-file hit@3 AUC from 0.873 to 0.935. This validation concerns issue-to-file localization, not patch application or SWE-bench issue resolution. These results support a measurement principle for agent evaluation: action control is predicted by event-specific state-action binding, not by output entropy, action-prior matching or rationale format alone.",
"published": "2026-05-10",
"updated": "2026-06-01",
"authors": [
"Xiao Jia"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2605.09692",
"pdf_url": "https://arxiv.org/pdf/2605.09692",
"path": "papers/items/2026-2605-09692-causal-state-binding-predicts-action-control-in-language-agents.md",
"local_title": "Causal state binding predicts action control in language agents",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"coding-agent",
"memory",
"planning",
"tool-use"
],
"collection_queries": "language-agent",
"collection_score": "17"
},
{
"arxiv_id": "2605.10870",
"title": "Remember the Decision, Not the Description: A Rate-Distortion Framework for Agent Memory",
"abstract": "Long-horizon language agents must operate under limited runtime memory, yet existing memory mechanisms often organize experience around descriptive criteria such as relevance, salience, or summary quality. For an agent, however, memory is valuable not because it faithfully describes the past, but because it preserves the distinctions between histories that must remain separated under a fixed budget to support good decisions. We cast this as a decision-centric rate-distortion problem, measuring memory quality by the loss in achievable decision quality induced by compression. This yields an exact forgetting boundary for what can be safely forgotten, and a memory-distortion frontier characterizing the optimal tradeoff between memory budget and decision quality. Motivated by this decision-centric view of memory, we propose DeMem, an online memory learner that refines its partition only when data certify that a shared state would induce decision conflict, and prove near-minimax regret guarantees. On both controlled synthetic diagnostics and long-horizon conversational benchmarks, DeMem yields consistent gains under the same runtime budget, supporting the principle that memory should preserve the distinctions that matter for decisions, not descriptions.",
"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"
],
"url": "https://arxiv.org/abs/2605.10870",
"pdf_url": "https://arxiv.org/pdf/2605.10870",
"path": "papers/items/2026-2605-10870-remember-the-decision-not-the-description-a-rate-distortion-framework-for-agent-.md",
"local_title": "\"Remember the Decision, Not the Description: A Rate-Distortion Framework for Agent Memory\"",
"local_status": "skimmed",
"local_topics": [
"agent-evaluation",
"memory",
"planning"
],
"collection_queries": "language-agent",
"collection_score": "13"
},
{
"arxiv_id": "2605.11534",
"title": "PRISM: : Planning and Reasoning with Intent in Simulated Embodied Environments",
"abstract": "When an LLM-based embodied agent fails at a household task, the culprit could be misidentified objects, forgotten sub-goals, or poor action sequencing -- yet existing benchmarks report only a single success rate, making it impossible to tell which cognitive module is responsible. We present PRISM, a diagnostic benchmark that reframes this problem: rather than asking only \\textit{did the agent succeed?}, PRISM asks \\textit{which capability is most likely responsible for failure?} Built on five photorealistic multi-room apartments (4--8 rooms each), PRISM structures 300 human-verified tasks into three capability tiers -- \\textit{Basic Ability}, \\textit{Reasoning Ability}, and \\textit{Long-horizon Ability} -- that isolate perception-to-action grounding, implicit intent resolution, and sustained multi-step coordination respectively. PRISM exposes an agent-agnostic executable action API that allows arbitrary agents: LLM agents, VLM agents, symbolic planners, RL policies, and hybrid systems, to be evaluated end-to-end under the same benchmark protocol. To support deeper diagnosis, optional probes for perception, memory, and planning can be adopted, replaced, or bypassed entirely, enabling controlled component-level analysis when desired. Experiments on seven contemporary LLMs establish a clear hierarchy: explicit spatial grounding is not the dominant failure source under oracle perception, implicit intent resolution is a significant bottleneck for all model families, and long-horizon coordination exposes a stark capability cliff -- lightweight models collapse to as low as 20.0\\% success while simultaneously consuming more tokens than their frontier counterparts, a signature of compensatory over-reasoning rather than genuine planning capability. Project page: \\href{https://sj-li.com/PROJ/PRISM}{link}.",
"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"
],
"url": "https://arxiv.org/abs/2605.11534",
"pdf_url": "https://arxiv.org/pdf/2605.11534",
"path": "papers/items/2026-2605-11534-prism-planning-and-reasoning-with-intent-in-simulated-embodied-environments.md",
"local_title": "\"PRISM: : Planning and Reasoning with Intent in Simulated Embodied Environments\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"embodied-agent",
"memory",
"planning",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "planning-agent",
"collection_score": "15"
},
{
"arxiv_id": "2605.12015",
"title": "SkillSafetyBench: Evaluating Agent Safety under Skill-Facing Attack Surfaces",
"abstract": "Reusable skills are becoming a common interface for extending large language model agents, packaging procedural guidance with access to files, tools, memory, and execution environments. However, this modularity introduces attack surfaces that are largely missed by existing safety evaluations: even when the user request is benign, unsafe influence may reside in skill guidance, local artifacts, or execution-environment files that steer the agent toward unsafe actions. We present SkillSafetyBench, a runnable benchmark for evaluating such skill-mediated safety failures. SkillSafetyBench includes 155 adversarial cases across 47 tasks, 6 risk domains, and 30 safety categories, each evaluated with a case-specific rule-based verifier. Experiments with multiple CLI agents and model backends show that non-user attacks can consistently induce unsafe behavior, with distinct failure patterns across domains, attack methods, and scaffold-model pairings. Our findings suggest that agent safety depends not only on model-level alignment, but also on how agents interpret skills, trust workflow context, and act through executable environments.",
"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"
],
"url": "https://arxiv.org/abs/2605.12015",
"pdf_url": "https://arxiv.org/pdf/2605.12015",
"path": "papers/items/2026-2605-12015-skillsafetybench-evaluating-agent-safety-under-skill-facing-attack-surfaces.md",
"local_title": "\"SkillSafetyBench: Evaluating Agent Safety under Skill-Facing Attack Surfaces\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"computer-use",
"memory",
"reasoning",
"tool-use",
"workflow-agent"
],
"collection_queries": "agent-safety",
"collection_score": "20"
},
{
"arxiv_id": "2605.12061",
"title": "SAGE: A Self-Evolving Agentic Graph-Memory Engine for Structure-Aware Associative Memory",
"abstract": "Long-term memory is becoming a central bottleneck for language agents. Exsting RAG and GraphRAG systems largely treat memory graphs as static retrieval middleware, which limits their ability to recover complete evidence chains from partial cues, exploit reusable graph-structrual roles, and improve the memory itself through downstream feedback. We introduce SAGE, a Self-evolving Agentic Graph-memory Engine that models graph memory as a dynamic long-term memory substrate. SAGE couples two roles: a memory writer that incrementally constucts structured graph memory from interaction histories, and a Graph Foundation Model-based memory reader to perform retrieval and provide feedback to the memory writer. We provide rigorooous theoretical annalyses supporting the framework. Across multi-hop QA, open-domain retireval, domain-specific review QA, and long-term agent-memory benchmarks, SAGE improves evidence recovery, answer grounding, and retrieval efficiency: after two self-evolution rounds, it achieves the best average rank on multi-hop QA; in zero-shot open-domain transfer, it reaches 82.5/91.6 Recall@2/5 on NQ. Further results on LongMemEval and HaluMem show that traning and reader-writer feedback improve multiple long-term memory and hallucination-diagnostic metrics, suggesting that self-evolving, structure-aware graph memory is a promising foundation for robust long-horizon language agents.",
"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"
],
"url": "https://arxiv.org/abs/2605.12061",
"pdf_url": "https://arxiv.org/pdf/2605.12061",
"path": "papers/items/2026-2605-12061-sage-a-self-evolving-agentic-graph-memory-engine-for-structure-aware-associative.md",
"local_title": "\"SAGE: A Self-Evolving Agentic Graph-Memory Engine for Structure-Aware Associative Memory\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"planning",
"rag",
"tool-use"
],
"collection_queries": "language-agent",
"collection_score": "18"
},
{
"arxiv_id": "2605.12260",
"title": "PRISM: Pareto-Efficient Retrieval over Intent-Aware Structured Memory for Long-Horizon Agents",
"abstract": "Long-horizon language agents accumulate conversation history far faster than any fixed context window can hold, making memory management critical to both answer accuracy and serving cost. Existing approaches either expand the context window without addressing what is retrieved, perform heavy ingestion-time fact extraction at substantial token cost, or rely on heuristic graph traversal that leaves both accuracy and efficiency on the table. We present PRISM, a training-free retrieval-side framework that treats long-horizon memory as a joint retrieval-and-compression problem over a graph-structured memory. PRISM combines four orthogonal inference-time components: Hierarchical Bundle Search over typed relation paths, Query-Sensitive Edge Costing that aligns traversal with detected query intent, Evidence Compression that compresses the candidate bundle into a compact answer-side context, and Adaptive Intent Routing that routes most queries through zero-LLM tiers. By formulating retrieval as min-cost selection over typed path templates and pairing it with an LLM-side compression step, PRISM surfaces the right evidence under a strict context budget without any fine-tuning or modification to the upstream ingestion pipeline. Experiments on the LoCoMo benchmark show that PRISM delivers substantially higher LLM-judge accuracy than every same-protocol baseline at an order-of-magnitude smaller context budget, occupying a previously empty corner of the accuracy-context-cost frontier and demonstrating a superior balance between answer quality and retrieval efficiency.",
"published": "2026-05-12",
"updated": "2026-05-22",
"authors": [
"Jingyi Peng",
"Zhongwei Wan",
"Weiting Liu",
"Qiuzhuang Sun"
],
"categories": [
"cs.CL"
],
"url": "https://arxiv.org/abs/2605.12260",
"pdf_url": "https://arxiv.org/pdf/2605.12260",
"path": "papers/items/2026-2605-12260-prism-pareto-efficient-retrieval-over-intent-aware-structured-memory-for-long-ho.md",
"local_title": "\"PRISM: Pareto-Efficient Retrieval over Intent-Aware Structured Memory for Long-Horizon Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"planning",
"rag",
"tool-use"
],
"collection_queries": "language-agent",
"collection_score": "17"
},
{
"arxiv_id": "2605.13542",
"title": "RealICU: Do LLM Agents Understand Long-Context ICU Data? A Benchmark Beyond Behavior Imitation",
"abstract": "Intensive care units (ICU) generate long, dense and evolving streams of clinical information, where physicians must repeatedly reassess patient states under time pressure, underscoring a clear need for reliable AI decision support. Existing ICU benchmarks typically treat historical clinician actions as ground truth. However, these actions are made under incomplete information and limited temporal context of the underlying patient state, and may therefore be suboptimal, making it difficult to assess the true reasoning capabilities of AI systems. We introduce RealICU, a hindsight-annotated benchmark for evaluating large language models (LLMs) under realistic ICU conditions, where labels are created after senior physicians review the full patient trajectory. We formulate four physician-motivated tasks: assess Patient Status, Acute Problems, Recommended Actions, and Red Flag actions that risk unsafe outcomes. We partition each trajectory with 30-min windows and release two datasets: RealICU-Gold with 930-window annotations from 94 MIMIC-IV patients, and RealICU-Scale with 11,862 windows extended by Oracle, a physician-validated LLM hindsight labeler. Existing LLMs including memory-augmented ones performed poorly on RealICU, exposing two failure modes: a recall-safety tradeoff for clinical recommendations, and an anchoring bias to early interpretations of the patient. We further introduce ICU-Evo to study structured-memory agents that improves long-horizon reasoning but does not fully eliminate safety failures. Together, RealICU provides a clinically grounded testbed for measuring and improving AI sequential decision-support in high-stakes care. Project page: https://chengzhi-leo.github.io/RealICU-Bench/",
"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"
],
"url": "https://arxiv.org/abs/2605.13542",
"pdf_url": "https://arxiv.org/pdf/2605.13542",
"path": "papers/items/2026-2605-13542-realicu-do-llm-agents-understand-long-context-icu-data-a-benchmark-beyond-behavi.md",
"local_title": "\"RealICU: Do LLM Agents Understand Long-Context ICU Data? A Benchmark Beyond Behavior Imitation\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"planning",
"reasoning",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "20"
},
{
"arxiv_id": "2605.13618",
"title": "OpenAaaS: An Open Agent-as-a-Service Framework for Distributed Materials-Informatics Research",
"abstract": "The Materials Genome Initiative catalyzed the proliferation of centralized platforms--SaaS, PaaS, and IaaS--that aggregate computational and experimental resources for accelerated materials discovery. In parallel, breakthroughs in large language models (LLMs) and autonomous agents have created powerful new reasoning capabilities for scientific research. Yet a critical \"last mile\" problem remains: while we possess world-class models and vast repositories of materials data, we lack the organizational infrastructure to compose these capabilities securely across institutional boundaries. The development of structural and functional materials for harsh service environments--high-temperature alloys, radiation resistant steels, corrosion-resistant coatings--remains characterized by long-term iteration, mechanistic complexity, and high domain expertise--demands that exceed both monolithic agent systems and traditional centralized platforms. To address this gap we propose OpenAaaS, an open-source hierarchical and distributed Agent-as-a-Service framework that enables organized multi-agent collaboration for intelligent materials design. OpenAaaS is built on a single foundational principle: code flows, data stays still. A Master Agent plans and decomposes complex research tasks without requiring direct access to subordinate agents' managed data and computational resources. Sub-agents, deployed as near-data execution nodes, retain full sovereignty over local datasets, proprietary algorithms, and specialized hardware. This architecture guarantees that raw data never leaves its domain of origin while enabling cross-scale, cross-domain secure integration of previously isolated materials intelligence silos. We validate the framework through two representative case studies: (i) AlphaAgent, an evidence-grounded materials literature analysis executor that achieves 4.66/5.0 on deep analytical questions against single-pass RAG baselines; and (ii) an ultra-large-scale hexa-high-entropy alloy descriptor database service that demonstrates secure near-data execution and domain-specific scientific workflows under strict data-sovereignty constraints. OpenAaaS establishes a principled pathway toward \"organized research\" via agent collectives, offering a scalable foundation for next-generation materials intelligent design platforms. All source code is available at https://github.com/Wolido/OpenAaaS.",
"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"
],
"url": "https://arxiv.org/abs/2605.13618",
"pdf_url": "https://arxiv.org/pdf/2605.13618",
"path": "papers/items/2026-2605-13618-openaaas-an-open-agent-as-a-service-framework-for-distributed-materials-informat.md",
"local_title": "\"OpenAaaS: An Open Agent-as-a-Service Framework for Distributed Materials-Informatics Research\"",
"local_status": "queued",
"local_topics": [
"memory",
"multi-agent",
"planning",
"rag",
"reasoning",
"workflow-agent"
],
"collection_queries": "autonomous-agent-llm",
"collection_score": "14"
},
{
"arxiv_id": "2605.14421",
"title": "MemLineage: Lineage-Guided Enforcement for LLM Agent Memory",
"abstract": "We introduce MemLineage, a defense for LLM agent memory that attaches both cryptographic provenance and LLM-mediated derivation lineage to every entry. Recent and concurrent work shows that untrusted content can be written into persistent agent state and re-enter later sessions as an instruction; the remaining systems question is how to preserve useful memory recall while preventing such state from justifying sensitive actions. MemLineage treats this as a chain-of-custody problem rather than a filtering problem. It is a six-module design around an RFC-6962 Merkle log over per-principal Ed25519-signed entries: a weighted derivation DAG records which retrieved entries influenced each new memory, and a max-of-strong-edges propagation rule makes Untrusted-Path Persistence hold for any chain whose attribution edges remain above threshold. The sensitive-action gate then refuses dispatches whose active justification descends from an external ancestor, while still allowing benign recall. We evaluate three defense cells against three memory-poisoning workloads on a deterministic mechanism-isolation harness; MemLineage is the only configuration in that harness that drives all three columns to zero ASR, while sub-millisecond per-operation overhead keeps it well below the noise floor of any LLM call. A Codex-backed AgentDojo bridge further separates strong-model behavior from defense-layer behavior: under an intentionally vulnerable tool-output profile, no-defense and signature-only baselines fail on all six banking pairs, while all MemLineage rows reduce strict AgentDojo ASR to zero. The core deterministic artifacts are byte-equal CI-verified; hosted-model AgentDojo and live-model sweeps are recorded as auditable logs rather than byte-pinned artifacts.",
"published": "2026-05-14",
"updated": "2026-05-14",
"authors": [
"Ciyan Ouyang",
"Rui Hou"
],
"categories": [
"cs.CR",
"cs.AI"
],
"url": "https://arxiv.org/abs/2605.14421",
"pdf_url": "https://arxiv.org/pdf/2605.14421",
"path": "papers/items/2026-2605-14421-memlineage-lineage-guided-enforcement-for-llm-agent-memory.md",
"local_title": "\"MemLineage: Lineage-Guided Enforcement for LLM Agent Memory\"",
"local_status": "skimmed",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "17"
},
{
"arxiv_id": "2605.14498",
"title": "GroupMemBench: Benchmarking LLM Agent Memory in Multi-Party Conversations",
"abstract": "Large Language Model (LLM) agents increasingly serve as personal assistants and workplace collaborators, where their utility depends on memory systems that extract, retrieve, and apply information across long-running conversations. However, both existing memory systems and benchmarks are built around the dyadic, single-user setup, even though real deployments routinely span groups and channels with multiple users interacting with the agent and with each other. This mismatch leaves three properties of group memory unmeasured: (i) group dynamics that go beyond concatenated one-on-one chats, (ii) speaker-grounded belief tracking, where the per-user memory modeling is needed, and (iii) audience-adapted language, where Theory-of-Mind shifts produce role-specific vocabulary. We introduce GroupMemBench, a benchmark that exposes all three. A graph-grounded synthesis pipeline produces multi-party conversations with controllable reply structure and conditions each message on per-user personas and target audiences. An adversarial query pipeline then binds every question to a specific asker across six categories, spanning multi-hop reasoning, knowledge update, term ambiguity, user-implicit reasoning, temporal reasoning, and abstention, and iteratively searches challenging, realistic queries that reflect comprehensive memory capability. Benchmarking leading memory systems exposes a sharp collapse: the strongest one reaches only 46.0% average accuracy, with knowledge update at 27.1% and term ambiguity at 37.7%, while a simple BM25 baseline matches or exceeds most agent memory systems. This indicates current memory ingestion erases the structural and lexical features group memory depends on, leaving multi-user memory far from solved.",
"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"
],
"url": "https://arxiv.org/abs/2605.14498",
"pdf_url": "https://arxiv.org/pdf/2605.14498",
"path": "papers/items/2026-2605-14498-groupmembench-benchmarking-llm-agent-memory-in-multi-party-conversations.md",
"local_title": "\"GroupMemBench: Benchmarking LLM Agent Memory in Multi-Party Conversations\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"rag",
"reasoning"
],
"collection_queries": "agent-memory",
"collection_score": "21"
},
{
"arxiv_id": "2605.14906",
"title": "MemLens: Benchmarking Multimodal Long-Term Memory in Large Vision-Language Models",
"abstract": "Memory is essential for large vision-language models (LVLMs) to handle long, multimodal interactions, with two method directions providing this capability: long-context LVLMs and memory-augmented agents. However, no existing benchmark conducts a systematic comparison of the two on questions that genuinely require multimodal evidence. To close this gap, we introduce MEMLENS, a comprehensive benchmark for memory in multimodal multi-session conversations, comprising 789 questions across five memory abilities (information extraction, multi-session reasoning, temporal reasoning, knowledge update, and answer refusal) at four standard context lengths (32K-256K tokens) under a cross-modal token-counting scheme. An image-ablation study confirms that solving MEMLENS requires visual evidence: removing evidence images drops two frontier LVLMs below 2% accuracy on the 80.4% of questions whose evidence includes images. Evaluating 27 LVLMs and 7 memory-augmented agents, we find that long-context LVLMs achieve high short-context accuracy through direct visual grounding but degrade as conversations grow, whereas memory agents are length-stable but lose visual fidelity under storage-time compression. Multi-session reasoning caps most systems below 30%, and neither approach alone solves the task. These results motivate hybrid architectures that combine long-context attention with structured multimodal retrieval. Our code is available at https://github.com/xrenaf/MEMLENS.",
"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"
],
"url": "https://arxiv.org/abs/2605.14906",
"pdf_url": "https://arxiv.org/pdf/2605.14906",
"path": "papers/items/2026-2605-14906-memlens-benchmarking-multimodal-long-term-memory-in-large-vision-language-models.md",
"local_title": "\"MemLens: Benchmarking Multimodal Long-Term Memory in Large Vision-Language Models\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "16"
},
{
"arxiv_id": "2605.15128",
"title": "MemEye: A Visual-Centric Evaluation Framework for Multimodal Agent Memory",
"abstract": "Long-term agent memory is increasingly multimodal, yet existing evaluations rarely test whether agents preserve the visual evidence needed for later reasoning. In prior work, many visually grounded questions can be answered using only captions or textual traces, allowing answers to be inferred without preserving the fine-grained visual evidence. Meanwhile, harder cases that require reasoning over changing visual states are largely absent. Therefore, we introduce MemEye, a framework that evaluates memory capabilities from two dimensions: one measures the granularity of decisive visual evidence (from scene-level to pixel-level evidence), and the other measures how retrieved evidence must be used (from single evidence to evolutionary synthesis). Under this framework, we construct a new benchmark across 8 life-scenario tasks, with ablation-driven validation gates for assessing answerability, shortcut resistance, visual necessity, and reasoning structure. By evaluating 13 memory methods across 4 VLM backbones, we show that current architectures still struggle to preserve fine-grained visual details and reason about state changes over time. Our findings show that long-term multimodal memory depends on evidence routing, temporal tracking, and detail extraction.",
"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"
],
"url": "https://arxiv.org/abs/2605.15128",
"pdf_url": "https://arxiv.org/pdf/2605.15128",
"path": "papers/items/2026-2605-15128-memeye-a-visual-centric-evaluation-framework-for-multimodal-agent-memory.md",
"local_title": "\"MemEye: A Visual-Centric Evaluation Framework for Multimodal Agent Memory\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"reasoning",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "16"
},
{
"arxiv_id": "2605.15701",
"title": "H-Mem: A Novel Memory Mechanism for Evolving and Retrieving Agent Memory via a Hybrid Structure",
"abstract": "Memory data are ubiquitous in Large Language Model (LLM)-based agents (e.g., OpenClaw and Manus). A few recent works have attempted to exploit agents'memory for improving their performance on the question-answering (QA) task, but they lack a principled mechanism for effectively modeling how memory data evolves over time and retrieving memory data effectively, leading to poor performance in memory utilization. To fill this gap, we present H-Mem, a novel memory mechanism via a hybrid structure that can not only effectively model the evolution of agent memory over a long period of time, but also provide an efficient memory retrieval approach. Particularly, H-Mem builds a temporal and semantic tree structure that allows the short-term memory data to evolve progressively into long-term memory data, where the latter provides summarized information about the former, while simultaneously constructing a knowledge graph to capture the relationships between entities in memory. Moreover, it offers an effective memory retrieval approach by exploiting the hybrid structure of the tree and graph structures. Extensive experiments on three agent memory benchmarks show that H-Mem achieves state-of-the-art performance on the QA task.",
"published": "2026-05-15",
"updated": "2026-05-15",
"authors": [
"Jiawei Yu",
"Yixiang Fang",
"Xilin Liu",
"Yuchi Ma"
],
"categories": [
"cs.CL",
"cs.AI"
],
"url": "https://arxiv.org/abs/2605.15701",
"pdf_url": "https://arxiv.org/pdf/2605.15701",
"path": "papers/items/2026-2605-15701-h-mem-a-novel-memory-mechanism-for-evolving-and-retrieving-agent-memory-via-a-hy.md",
"local_title": "\"H-Mem: A Novel Memory Mechanism for Evolving and Retrieving Agent Memory via a Hybrid Structure\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag"
],
"collection_queries": "agent-memory",
"collection_score": "14"
},
{
"arxiv_id": "2605.15710",
"title": "SMMBench: A Benchmark for Source-Distributed Multimodal Agent Memory",
"abstract": "Existing benchmarks for multimodal memory reasoning largely evaluate systems within pre-assembled contexts, but under-evaluate whether agents can use evidence distributed across independently originated sources. We argue that source-distributed memory composition is an important and under-examined bottleneck in multimodal agent memory, especially when relevant evidence is fragmented across heterogeneous artifacts such as conversations, profiles, screenshots, tables, images, and documents. To address this gap, we introduce Source-distributed Multimodal Memory Benchmark(SMMBench), which measures whether agents can retrieve, align, and compose multimodal evidence scattered across multiple sources rather than reason within a single curated context. SMMBench evaluates four core capabilities: (1) cross-source multimodal reasoning; (2) conflict resolution; (3) preference reasoning; (4) memory-grounded action prediction. The benchmark contains 1877 samples grounded in 264 sources. Experiments on representative memory-style and retrieval-based baselines show that current systems still struggle on these capabilities, positioning source-distributed multimodal memory as an important and still under-evaluated challenge for multimodal agents. Our data are available at https://huggingface.co/datasets/HuacanChai/SMMBench.",
"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"
],
"url": "https://arxiv.org/abs/2605.15710",
"pdf_url": "https://arxiv.org/pdf/2605.15710",
"path": "papers/items/2026-2605-15710-smmbench-a-benchmark-for-source-distributed-multimodal-agent-memory.md",
"local_title": "\"SMMBench: A Benchmark for Source-Distributed Multimodal Agent Memory\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "16"
},
{
"arxiv_id": "2605.15759",
"title": "DimMem: Dimensional Structuring for Efficient Long-Term Agent Memory",
"abstract": "Large language model (LLM) agents require long-term memory to leverage information from past interactions. However, existing memory systems often face a fidelity--efficiency trade-off: raw dialogue histories are expensive, while flat facts or summaries may discard the structure needed for precise recall. We propose \\textbf{DimMem}, a lightweight dimensional memory framework that represents each memory as an atomic, typed, and self-contained unit with explicit fields such as time, location, reason, purpose, and keywords. This representation exposes the structure needed for dimension-aware retrieval, memory update, and selective assistant-context recall without storing full histories in the model context. Across LoCoMo-10 and LongMemEval-S, DimMem achieves \\textbf{81.43\\%} and \\textbf{78.20\\%} overall accuracy, respectively, outperforming existing lightweight memory systems while reducing LoCoMo per-query token cost by \\textbf{24\\%}. We further show that dimensional memory extraction is learnable by compact models: after fine-tuning on the DimMem schema, a Qwen3-4B extractor surpasses LightMem with GPT-4.1-mini on both benchmarks and reaches performance comparable to, or better than, much larger extractors in key settings. These results suggest that explicit dimensional structuring is an effective and efficient foundation for long-term memory in LLM agents. Code is available at https://github.com/ChowRunFa/DimMem.",
"published": "2026-05-15",
"updated": "2026-05-24",
"authors": [
"Wentao Qiu",
"Haotian Hu",
"Fanyi Wang",
"Jinwei Kong",
"Yu Zhang"
],
"categories": [
"cs.CL"
],
"url": "https://arxiv.org/abs/2605.15759",
"pdf_url": "https://arxiv.org/pdf/2605.15759",
"path": "papers/items/2026-2605-15759-dimmem-dimensional-structuring-for-efficient-long-term-agent-memory.md",
"local_title": "\"DimMem: Dimensional Structuring for Efficient Long-Term Agent Memory\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "16"
},
{
"arxiv_id": "2605.16233",
"title": "FORGE: Self-Evolving Agent Memory With No Weight Updates via Population Broadcast",
"abstract": "Can LLM agents improve decision-making through self-generated memory without gradient updates? We propose FORGE (Failure-Optimized Reflective Graduation and Evolution), a staged, population-based protocol that evolves prompt-injected natural-language memory for hierarchical ReAct agents. FORGE wraps a Reflexion-style inner loop, where a dedicated reflection agent (using the same underlying LLM, no distillation from a stronger model) converts failed trajectories into reusable knowledge artifacts: textual heuristics (Rules), few-shot demonstrations (Examples), or both (Mixed), with an outer loop that propagates the best-performing instance's memory to the population between stages and freezes converged instances via a graduation criterion. We evaluate on CybORG CAGE-2, a stochastic network-defense POMDP at a 30-step horizon against the B-line attacker, where all four tested LLM families (Gemini-2.5-Flash-Lite, Grok-4-Fast, Llama-4-Maverick, Qwen3-235B) exhibit strongly negative, heavy-tailed zero-shot rewards. Compared against both a zero-shot baseline and a Reflexion baseline (isolated single-stream learning), FORGE improves average evaluation return by 1.7-7.7$\\times$ over zero-shot and by 29-72% over Reflexion in all 12 model-representation conditions, reducing major-failure rates (below $-100$) to as low as $\\sim$1%. We find that (1) population broadcast is critical mechanism, with a no-graduation ablation confirming that broadcast carries the performance gains while graduation primarily saves compute; (2) Examples achieves the strongest returns for three of four models, Rules offers the best cost-reliability profile with $\\sim$40% fewer tokens; and (3) weaker baseline models benefit disproportionately, suggesting FORGE may mitigate capability gaps rather than amplify strong models. All evidence is confined to CAGE-2 B-line; cross-family findings are directional evidence.",
"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"
],
"url": "https://arxiv.org/abs/2605.16233",
"pdf_url": "https://arxiv.org/pdf/2605.16233",
"path": "papers/items/2026-2605-16233-forge-self-evolving-agent-memory-with-no-weight-updates-via-population-broadcast.md",
"local_title": "\"FORGE: Self-Evolving Agent Memory With No Weight Updates via Population Broadcast\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"reasoning"
],
"collection_queries": "agent-memory",
"collection_score": "16"
},
{
"arxiv_id": "2605.16481",
"title": "Visual Agentic Memory: Enabling Online Long Video Understanding via Online Indexing, Hierarchical Memory, and Agentic Retrieval",
"abstract": "Long video understanding requires more than large context windows. It also needs a memory mechanism that decides what visual evidence to retain, keeps it searchable over long horizons, and grounds later reasoning in recoverable observations rather than compressed latent state alone. We propose Visual Agentic Memory (VAM), a training-free framework with three components. Online Indexing supports selective evidence retention under streaming constraints. Hierarchical Memory organises retained evidence in a Parallel Representation that aligns temporal context with spatial observations. Agentic Retrieval searches, inspects, and verifies candidate evidence before producing a grounded answer. On OVO-Bench, VAM achieves the highest RT+BT average (68.41) across all reported baselines, improving over end-to-end use of the same underlying MLLM (Gemini 3 Flash, 67.46). On the month-scale split of MM-Lifelong train@month (105.6 hours over 51 days), VAM reaches 17.11%, second only to ReMA with GPT-5 (17.62%). These results suggest that long-horizon video understanding benefits from treating visual memory as an explicit, inspectable, and queryable substrate. Code is available at https://github.com/yiliu-li/Visual-Agentic-Memory.",
"published": "2026-05-15",
"updated": "2026-05-15",
"authors": [
"Aiden Yiliu Li",
"Nels Numan",
"Anthony Steed"
],
"categories": [
"cs.CV",
"cs.AI"
],
"url": "https://arxiv.org/abs/2605.16481",
"pdf_url": "https://arxiv.org/pdf/2605.16481",
"path": "papers/items/2026-2605-16481-visual-agentic-memory-enabling-online-long-video-understanding-via-online-indexi.md",
"local_title": "\"Visual Agentic Memory: Enabling Online Long Video Understanding via Online Indexing, Hierarchical Memory, and Agentic Retrieval\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"planning",
"rag",
"reasoning"
],
"collection_queries": "agent-memory",
"collection_score": "18"
},
{
"arxiv_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",
"abstract": "The rapid evolution of Large Language Model (LLM) agents has produced diverse interaction paradigms, yet few production systems integrate multiple paradigms within a unified architecture. This paper presents a systematic analysis of three principal agent interaction paradigms, including Multi-Agent Orchestration (Generator-Evaluator), ReAct Tool-Use Loops, and Memory-Augmented Interaction, as implemented in buddyMe, an open-source multi-model agent programming framework. We formalize a five-stage processing pipeline: Requirement Pre-Review -> Task Decomposition -> ReAct Execution -> Real-Execution Verification -> Adversarial Evaluation Discussion, and establish a six-dimensional evaluation schema with weighted scoring. Through four empirical case studies drawn from real-world deployment logs covering museum guide generation, scheduled weather tasks, and comprehensive tour planning, we draw three key conclusions. First, Generator-Evaluator pre-review detects requirement omissions in 20 percent of complex tasks, with 80 percent tasks passing initial inspection. Second, the ReAct loop ensures stable subtask execution but leads to around 30 percent redundant tool invocations. Third, adversarial Evaluator-Defender discussions reach consensus within 2-3 rounds for nearly 70 percent of scenarios, functioning mainly for content refinement rather than logical reversal. We additionally provide three Mermaid-based architectural diagrams and conduct cross-paradigm comparisons with CrewAI, AutoGen, LangGraph, MemGPT and A-Mem across six system dimensions. The research outcomes offer practical design guidelines for constructing stable and reliable multi-paradigm agent systems.",
"published": "2026-05-16",
"updated": "2026-05-16",
"authors": [
"Xiaohua Wang",
"Chao Han",
"Kai Yu",
"XiaoLiang Xu",
"Liang Wang"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2605.16821",
"pdf_url": "https://arxiv.org/pdf/2605.16821",
"path": "papers/items/2026-2605-16821-multi-paradigm-agent-interaction-in-practice-a-systematic-analysis-of-generator-.md",
"local_title": "\"Multi-Paradigm Agent Interaction in Practice:A Systematic Analysis of Generator-Evaluator, ReAct Loop,and Adversarial Evaluation in the buddyMe Framework\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"multi-agent",
"planning",
"tool-use"
],
"collection_queries": "planning-agent",
"collection_score": "18"
},
{
"arxiv_id": "2605.23986",
"title": "MemForest: An Efficient Agent Memory System with Hierarchical Temporal Indexing",
"abstract": "Memory is a fundamental component for enabling long-context LLM agents, supporting persistent state across interactions through a continuous serve-and-update lifecycle. Despite substantial prior work, existing systems suffer from significant maintenance overhead due to two key limitations: coarse-grained state management and inherently sequential update pipelines. In particular, updates are often tightly coupled with LLM inference and require full-state rewrites, leading to poor scalability and growing latency as memory accumulates. To address these challenges, we present MemForest, a memory framework that reformulates agent memory as a write-efficient temporal data management problem. MemForest breaks the sequential bottleneck via parallel chunk extraction, decoupling memory construction into concurrent, independent operations. To further eliminate coarse-grained maintenance, we introduce MemTree, a hierarchical temporal index that organizes memory as time-ordered trees rather than flat global summaries. This design replaces full-state rewrites with localized per-node updates, reducing maintenance cost to the affected tree paths while naturally preserving temporally evolving states. We evaluate MemForest on two long-context memory benchmarks, LongMemEval-S and LoCoMo. On LongMemEval-S, MemForest achieves the best overall performance among stateful baselines, reaching 79.8% pass@1 accuracy while sustaining a memory construction throughput approximately 6x higher than state-of-the-art approaches including EverMemOS.",
"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"
],
"url": "https://arxiv.org/abs/2605.23986",
"pdf_url": "https://arxiv.org/pdf/2605.23986",
"path": "papers/items/2026-2605-23986-memforest-an-efficient-agent-memory-system-with-hierarchical-temporal-indexing.md",
"local_title": "\"MemForest: An Efficient Agent Memory System with Hierarchical Temporal Indexing\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "15"
},
{
"arxiv_id": "2605.28850",
"title": "Representation Signatures and Risk-Feedback Alignment in LLM Trading Agents",
"abstract": "We study behavioral alignment and representation dynamics of large language model (LLM) agents in financial decision environments. TradeArena, an auditable trading-agent testbed with risk reports, execution simulation, memory, and replayable trajectories, lets us analyze how rationales, positions, and interventions evolve under market stress. Code and data artifacts are available through the \\href{https://github.com/weich97/TradeArena.git}{TradeArena repository}. We find pre-failure signatures: planning embeddings drift from normal centroids, fused plan-risk representations separate normal from pre-drawdown states, and local manifolds exhibit effective-rank contraction. Across 80 rolling failure anchors and eight LLM trajectories, this pattern persists across hash, LSA, Transformer, and white-box hidden-state probes. Stress tests with CoT-free target weights, lexical controls, OHLCV noise, and false audits show that rationale-level contraction can vanish without rationales, while intent-space and fused signatures remain informative. Structured risk feedback can act as an external alignment signal without fine-tuning, but not as a universal performance enhancer: true audit feedback improves calibration for some models, returns for others, and exposes cases where placebo or hidden feedback has higher short-horizon return but weaker alignment diagnostics. A 51-stock intraday experiment reveals a correlation blind spot: LLM rationales justify exposure to coupled assets that the risk layer clips. Finally, a financial-audit task suite shifts comparison from ``which model trades best'' to whether models can audit trajectories, respect execution boundaries, reproduce artifacts, and avoid claim overreach. These results support a research claim, not a profitability claim: auditable risk feedback and representation trajectories reveal when LLM financial reasoning is aligning, drifting, or failing.",
"published": "2026-05-16",
"updated": "2026-05-30",
"authors": [
"Weicheng Xue"
],
"categories": [
"cs.LG",
"q-fin.CP"
],
"url": "https://arxiv.org/abs/2605.28850",
"pdf_url": "https://arxiv.org/pdf/2605.28850",
"path": "papers/items/2026-2605-28850-representation-signatures-and-risk-feedback-alignment-in-llm-trading-agents.md",
"local_title": "Representation Signatures and Risk-Feedback Alignment in LLM Trading Agents",
"local_status": "queued",
"local_topics": [
"agent-safety",
"coding-agent",
"memory",
"planning",
"reasoning",
"tool-use",
"world-model"
],
"collection_queries": "planning-agent",
"collection_score": "15"
},
{
"arxiv_id": "2605.17348",
"title": "Taming \"Zombie'' Agents: A Markov State-Aware Framework for Resilient Multi-Agent Evolution",
"abstract": "Recent advancements in LLM-based multi-agent systems have demonstrated remarkable collaborative capabilities across complex tasks. To improve overall efficiency, existing methods often rely on aggressive graph evolution among agents (e.g., node or edge pruning), which risks prematurely discarding valuable agents due to transient issues such as hallucinations or temporary knowledge gaps. However, such hard pruning overlooks the potential for ``zombie'' agents to recover and contribute in subsequent discussion rounds. In this paper, we propose AgentRevive, a Markov state-aware framework for resilient multi-agent evolution. Our approach dynamically manages agent collaboration through soft state transitions, implemented via two key components: (1) State-Aware Policy Learning: Agent states are divided into ``Active'', ``Standby'', and ``Terminated'' states, selectively propagating messages based on agent memory. The policy employs a risk estimator to optimize agent state transitions by assessing hallucination risk, minimizing the influence of unreliable nodes while safeguarding valuable ones. (2) State-Aware Edge Optimization: Subgraph edges are pruned according to states learned from the policy, permanently removing ``Terminated'' nodes and retaining ``Standby'' nodes for subsequent rounds to assess their potential future contributions. Extensive experiments on general reasoning, domain-specific, and hallucination challenge tasks show that our method consistently outperforms strong baselines and significantly reduces token consumption through state-aware agent scheduling.",
"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"
],
"url": "https://arxiv.org/abs/2605.17348",
"pdf_url": "https://arxiv.org/pdf/2605.17348",
"path": "papers/items/2026-2605-17348-taming-zombie-agents-a-markov-state-aware-framework-for-resilient-multi-agent-ev.md",
"local_title": "\"Taming \\\"Zombie'' Agents: A Markov State-Aware Framework for Resilient Multi-Agent Evolution\"",
"local_status": "queued",
"local_topics": [
"agent-safety",
"memory",
"multi-agent",
"reasoning"
],
"collection_queries": "agent-memory",
"collection_score": "15"
},
{
"arxiv_id": "2605.17625",
"title": "Episodic-Semantic Memory Architecture for Long-Horizon Scientific Agents",
"abstract": "As Large Language Models (LLMs) evolve into persistent scientific collaborators, context window saturation has emerged as a critical bottleneck. Scientific workflows involving iterative data analysis and hypothesis refinement rapidly saturate even extended contexts with dense technical content, while monolithic approaches suffer from quadratic cost scaling and cognitive degradation. We evaluate a Dual Process Memory Architecture that decouples immediate episodic needs (constant 10-message window) from long-term consolidated knowledge (growing at approximately 3 tokens/message). Unlike prior social agent memory systems, our domain-specific consolidation addresses contradictory parameter evolution, multi-hop reasoning across experimental phases, and precise technical fact retention. Through large-scale evaluation spanning 15,000 messages with cross-model validation across six LLMs from three families (OpenAI, Anthropic, Google), totaling 1,440 queries, we establish three key findings. First, while full-context models fail at 10,000 messages due to context overflow, our system maintains 70-85% accuracy with 1-2 second latency using 62% fewer tokens (45,434 vs 120,000+ limit). Second, cross-model validation reveals architecture-level trade-offs independent of specific LLMs: Dual Process excels at numeric/temporal queries (65-90% accuracy) while RAG excels at historical retrieval (60-85%), suggesting complementary deployment strategies. Third, we identify a \"Sim-to-Real\" gap where synthetic tests maintain constant memory but realistic workflows exhibit linear growth (about 3 tokens/message), with consolidation quality emerging as the primary scalability bottleneck. The architecture successfully manages profiles with 14,000+ scientific facts (125k tokens), demonstrating that domain-specific memory consolidation enables sustained operation beyond full-context limits.",
"published": "2026-05-17",
"updated": "2026-05-17",
"authors": [
"Nikola Milosevic"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2605.17625",
"pdf_url": "https://arxiv.org/pdf/2605.17625",
"path": "papers/items/2026-2605-17625-episodic-semantic-memory-architecture-for-long-horizon-scientific-agents.md",
"local_title": "Episodic-Semantic Memory Architecture for Long-Horizon Scientific Agents",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"planning",
"rag",
"reasoning",
"tool-use",
"workflow-agent"
],
"collection_queries": "agent-memory",
"collection_score": "18"
},
{
"arxiv_id": "2605.18284",
"title": "CommitDistill: A Lightweight Knowledge-Centric Memory Layer for Software Repositories",
"abstract": "Software repositories accumulate large amounts of unstructured knowledge in commit messages, pull-request discussions, and issue threads, but developers and AI coding assistants rarely reuse this history effectively. Recent work on typed-memory architectures for LLM agents (MemGPT, generative agents, and the PlugMem module of Yang et al.) argues that agent memory should be distilled, typed knowledge rather than raw interaction text. We adapt that stance to a software repository's own git history under a constrained regime: deterministic, dependency-free, local-only, no embeddings. We present CommitDistill, an open-source Python prototype that mines a local git history into typed knowledge units (Facts, Skills, Patterns) using deterministic regex and surfaces them through a TF-IDF retriever with a calibrated silence threshold (theta = 2.5) that abstains on out-of-distribution queries. The artefact is a trust-instrumented memory substrate: deterministic, no external service, inspectable plain-JSON store, tunable abstention. A case study on five public repositories spanning Python, JavaScript, C, and Java (25,000 commits, 1,167 extracted units) reports useful-precision 0.525 at Cohen's kappa = 0.633 on 40 dual-annotated Python units. The decisive finding is budget-constrained retrieval: at a 256-character per-query budget, CommitDistill reaches 0.750 hit-rate on a 12-query benchmark against BM25's 0.333 and git log --grep's 0.083. On a four-arm paired LLM-as-judge evaluation (n=200 time-travel bug-fixes, two judges) covering control, CommitDistill, a body-budget-matched CD-Hybrid, and BM25, no condition produces a statistically detectable lift over control on the headline mean and CD-Hybrid is indistinguishable from BM25 head-to-head. Extraction over 10,000 commits completes in under 4 seconds on a laptop. Source, annotations, baselines, and a reproducibility script accompany this paper.",
"published": "2026-05-18",
"updated": "2026-05-18",
"authors": [
"Divya Chukkapalli",
"Thejesh Avula",
"Aditya Aggarwal",
"Harsimran Singh",
"Amith Tallanki"
],
"categories": [
"cs.SE",
"cs.AI"
],
"url": "https://arxiv.org/abs/2605.18284",
"pdf_url": "https://arxiv.org/pdf/2605.18284",
"path": "papers/items/2026-2605-18284-commitdistill-a-lightweight-knowledge-centric-memory-layer-for-software-reposito.md",
"local_title": "\"CommitDistill: A Lightweight Knowledge-Centric Memory Layer for Software Repositories\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"coding-agent",
"computer-use",
"memory",
"rag",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "17"
},
{
"arxiv_id": "2605.18421",
"title": "EvoMemBench: Benchmarking Agent Memory from a Self-Evolving Perspective",
"abstract": "Recent benchmarks for Large Language Model (LLM) agents mainly evaluate reasoning, planning, and execution. However, memory is also essential for agents, as it enables them to store, update, and retrieve information over time. This ability remains under-evaluated, largely because existing benchmarks do not provide a systematic way to assess memory mechanisms. In this paper, we study agent memory from a self-evolving perspective and introduce EvoMemBench, a unified benchmark organized along two axes: memory scope (in-episode vs. cross-episode) and memory content (knowledge-oriented vs. execution-oriented). We compare 15 representative memory methods with strong long-context baselines under a standardized protocol. Results show that current memory systems are still far from a general solution: long-context baselines remain highly competitive, memory helps most when the current context is insufficient or tasks are difficult, and no single memory form works consistently across all settings. Retrieval-based methods remain strong for knowledge-intensive settings, whereas procedural and long-term memory methods are more effective for execution-oriented tasks when their stored experience matches the task structure. We hope EvoMemBench facilitates future research on more effective memory systems for LLM-based agents. Our code is available at https://github.com/DSAIL-Memory/EvoMemBench.",
"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"
],
"url": "https://arxiv.org/abs/2605.18421",
"pdf_url": "https://arxiv.org/pdf/2605.18421",
"path": "papers/items/2026-wang-evomembench.md",
"local_title": "\"EvoMemBench: Benchmarking Agent Memory from a Self-Evolving Perspective\"",
"local_status": "skimmed",
"local_topics": [
"memory",
"agent-evaluation"
],
"collection_queries": null,
"collection_score": null
},
{
"arxiv_id": "2605.18652",
"title": "MementoGUI: Learning Agentic Multimodal Memory Control for Long-Horizon GUI Agents",
"abstract": "Recent GUI agents have made substantial progress in visual grounding and action prediction, yet they remain brittle in long-horizon tasks that require maintaining task state across many interface transitions. Existing agents typically rely on raw history replay or text-only memory, which either overwhelms the model with redundant screenshots or discards localized visual evidence needed for future decisions. To address these limitations, we introduce \\textbf{MementoGUI}, a plug-in agentic memory framework that equips MLLM-based GUI agents with \\textbf{MementoCore}, a learned controller for online memory selection, compression, and retrieval. Rather than treating interaction history as a fixed context, MementoGUI formulates long-horizon GUI control as an online memory-control problem: working memory selectively preserves task-relevant interface events with textual summaries and ROI-level visual evidence, while episodic memory retrieves reusable past trajectories through learned relevance selection. MementoCore modularizes memory control into specialized operators for step processing, memory compression, episodic writing, and episodic selection, enabling plug-in memory augmentation without finetuning the GUI agent backbone. We further develop a scalable data curation pipeline that converts computer-use trajectories into memory-controller training data, introduce \\textbf{MementoGUI-Bench} for evaluating long-horizon decision-making in GUI agents, and design MLLM-based metrics for semantic action matching, task progress, and memory consistency. Experiments on GUI-Odyssey, MM-Mind2Web, and MementoGUI-Bench show that MementoGUI consistently improves GUI agents over no-history, history-replay, and text-only memory baselines, with larger MementoCore backbones further strengthening memory-augmented GUI control.",
"published": "2026-05-18",
"updated": "2026-05-18",
"authors": [
"Ziyun Zeng",
"Hang Hua",
"Bocheng Zou",
"Mu Cai",
"Rogerio Feris",
"Jiebo Luo"
],
"categories": [
"cs.CV"
],
"url": "https://arxiv.org/abs/2605.18652",
"pdf_url": "https://arxiv.org/pdf/2605.18652",
"path": "papers/items/2026-2605-18652-mementogui-learning-agentic-multimodal-memory-control-for-long-horizon-gui-agent.md",
"local_title": "\"MementoGUI: Learning Agentic Multimodal Memory Control for Long-Horizon GUI Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"planning",
"rag",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "22"
},
{
"arxiv_id": "2605.18930",
"title": "OEP: Poisoning Self-Evolving LLM Agents via Locally Correct but Non-Transferable Experiences",
"abstract": "Memory-augmented large language model (LLM) agents use iterative reflection and self-evolution to solve complex tasks, but these mechanisms introduce security risks. Existing agentic memory attacks require privileged access or explicit malicious content, making them detectable by advanced safety filters. This leaves a subtler attack surface underexplored: whether adversaries can induce agent to generate experiences that appear locally correct and semantically plausible yet induce harmful generalization during reflection. We find that reflective agents are vulnerable to such clean experiences, especially when paired with severe but plausible hypothetical consequences. Based on this observation, we introduce Obsessive Experience Poisoning (OEP), a low-privilege black-box attack requiring no direct control over the system prompt or memory database. OEP constructs adversarial clean edge-cases that combine locally correct solutions, non-transferable methods, and severe consequences, biasing reflection toward risk-averse rule formation. During memory consolidation, agents may over-trust self-generated reflections and distill localized experiences into high-priority but over-generalized rules, causing downstream failures. Evaluations across three domains show that OEP achieves ASR above 50\\% with GPT-4o agents, and outperforms existing attacks under LLM auditing defense.",
"published": "2026-05-18",
"updated": "2026-05-18",
"authors": [
"Kaixiang Wang",
"Jiong Lou",
"Zhaojiacheng Zhou",
"Jie Li"
],
"categories": [
"cs.CR",
"cs.AI",
"cs.LG"
],
"url": "https://arxiv.org/abs/2605.18930",
"pdf_url": "https://arxiv.org/pdf/2605.18930",
"path": "papers/items/2026-2605-18930-oep-poisoning-self-evolving-llm-agents-via-locally-correct-but-non-transferable-.md",
"local_title": "\"OEP: Poisoning Self-Evolving LLM Agents via Locally Correct but Non-Transferable Experiences\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"reasoning"
],
"collection_queries": "agent-memory",
"collection_score": "17"
},
{
"arxiv_id": "2605.19952",
"title": "Rethinking How to Remember: Beyond Atomic Facts in Lifelong LLM Agent Memory",
"abstract": "To enable reliable long-term interaction, LLM agents require a memory system that can faithfully store, efficiently retrieve, and deeply reason over accumulated dialogue history. Most existing methods adopt an extracted fact based paradigm: handcrafted static prompts compress raw dialogues into atomic facts, which are then stored, matched, and injected into downstream reasoning. Nevertheless, such fact-centric designs inevitably discard fine-grained details in original dialogues and fail to support deep reasoning over scattered isolated facts. Moreover, static prompts cannot maintain consistent extraction granularity across diverse dialogue styles. To address these limitations, we propose TriMem, which maintains three coexisting representation granularities, including raw dialogue segments anchored by source identifiers for storage fidelity, extracted atomic facts for efficient memory retrieval, synthesized profiles that aggregate dispersed facts into holistic semantic understanding for deep reasoning. We further adopt TextGrad-based prompt optimization, which iteratively refines extraction and profiling prompts via response quality feedback, achieving lifelong evolution without any parameter updating. Extensive experiments on LoCoMo and PerLTQA across multiple LLM backbones demonstrate that TriMem consistently outperforms strong memory baselines. The code is available at https://TMLR-TriMem.github.io .",
"published": "2026-05-19",
"updated": "2026-05-19",
"authors": [
"Jingwei Sun",
"Jianing Zhu",
"Jiangchao Yao",
"Tongliang Liu",
"Bo Han"
],
"categories": [
"cs.CL"
],
"url": "https://arxiv.org/abs/2605.19952",
"pdf_url": "https://arxiv.org/pdf/2605.19952",
"path": "papers/items/2026-2605-19952-rethinking-how-to-remember-beyond-atomic-facts-in-lifelong-llm-agent-memory.md",
"local_title": "\"Rethinking How to Remember: Beyond Atomic Facts in Lifelong LLM Agent Memory\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "18"
},
{
"arxiv_id": "2605.20315",
"title": "Mix-Quant: Quantized Prefilling, Precise Decoding for Agentic LLMs",
"abstract": "LLM agents have recently emerged as a powerful paradigm for solving complex tasks through planning, tool use, memory retrieval, and multi-step interaction. However, these agentic workflows often introduce substantial input-side overhead, making the compute-intensive prefilling stage a key bottleneck in long-context, multi-turn inference. In this work, we propose Mix-Quant, a simple and effective phase-aware quantization framework for fast agentic inference. We first investigate FP4 quantization in agentic LLM workflows and observe that quantizing the entire inference process can incur significant performance degradation. In contrast, the prefilling stage exhibits substantial quantization redundancy and can therefore be quantized with minimal accuracy loss, despite being the dominant source of computation. Based on this insight, we apply high-throughput NVFP4 quantization to the prefilling phase while preserving BF16 precision for decoding. By decoupling prefilling acceleration from decoding quality, Mix-Quant combines phase-aware algorithmic quantization with hardware-efficient NVFP4 execution to alleviate the inference bottleneck in LLM agents. Extensive experiments across long-context and agentic benchmarks demonstrate that Mix-Quant largely preserves task performance while delivering significant efficiency improvements, achieving up to a 3x speedup during prefilling.",
"published": "2026-05-19",
"updated": "2026-05-19",
"authors": [
"Haiquan Lu",
"Zigeng Chen",
"Gongfan Fang",
"Xinyin Ma",
"Xinchao Wang"
],
"categories": [
"cs.CL"
],
"url": "https://arxiv.org/abs/2605.20315",
"pdf_url": "https://arxiv.org/pdf/2605.20315",
"path": "papers/items/2026-2605-20315-mix-quant-quantized-prefilling-precise-decoding-for-agentic-llms.md",
"local_title": "\"Mix-Quant: Quantized Prefilling, Precise Decoding for Agentic LLMs\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"coding-agent",
"memory",
"planning",
"rag",
"tool-use",
"workflow-agent"
],
"collection_queries": "planning-agent",
"collection_score": "21"
},
{
"arxiv_id": "2605.20616",
"title": "Auto-Dreamer: Learning Offline Memory Consolidation for Language Agents",
"abstract": "Language agents increasingly operate over streams of related tasks, yet existing memory systems struggle to convert accumulated experience into reusable knowledge. Retrieval-augmented and structured memory methods record per-session observations effectively, but often couple acquisition and consolidation into a single online process, leaving the agent without a global view across sessions to discover recurring patterns, abstract shared procedures, or prune redundant entries. Inspired by complementary learning systems theory, we propose Auto-Dreamer, a learned offline consolidator for language-agent memory. Auto-Dreamer decouples fast per-session memory acquisition from slow cross-session consolidation. Given a selected working region of a typed memory bank, the consolidator treats the region as read-only evidence, performs bounded tool-use to inspect entries and provenance-linked source trajectories, and synthesizes a fresh compact replacement set that abstracts across sessions and supersedes the original region. We train Auto-Dreamer via GRPO, using end-to-end agent performance as the reward signal to learn how to consolidate memories acquired through fast online experience. Trained on ScienceWorld trajectories alone, Auto-Dreamer outperforms fixed, RL-trained, and prompted memory baselines on ScienceWorld by 7 points while using an active memory bank 12$\\times$ smaller than the strongest baseline, and continues to lead on held-out ALFWorld and WebArena without retraining -- using 6$\\times$ less memory than the strongest baseline on ALFWorld.",
"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"
],
"url": "https://arxiv.org/abs/2605.20616",
"pdf_url": "https://arxiv.org/pdf/2605.20616",
"path": "papers/items/2026-2605-20616-auto-dreamer-learning-offline-memory-consolidation-for-language-agents.md",
"local_title": "\"Auto-Dreamer: Learning Offline Memory Consolidation for Language Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"tool-use"
],
"collection_queries": "agent-memory, language-agent",
"collection_score": "16"
},
{
"arxiv_id": "2605.20833",
"title": "MemGym: a Long-Horizon Memory Environment for LLM Agents",
"abstract": "Memory is a central capability for LLM agents operating across long-horizon tasks. Existing memory benchmarks predominantly evaluate retention of personalized information in multi-turn chat scenarios, overlooking the dynamic memory formation that occurs during extended agent execution. Consequently, the memory systems they produce transfer poorly to realistic agentic environments, such as coding and web navigation. We present MemGym, a benchmark for agentic memory that unifies existing agent gyms and in-house memory-grounded pipelines behind one memory-reasoning interface. MemGym spans five evaluation tracks grouped into four agentic regimes: tool-use dialogue (tau2-bench), multi-turn deep-research search (MEMGYM-DR), coding (SWE-Gym and MEMGYM-CODEQA), and computer use (WebArena-Infinity). MemGym reports memory-isolated scores that decouple memory performance from reasoning, retrieval, and tool-use ability, so memory strategies can be ranked without those confounders. Our synthetic pipelines for MEMGYM-CODEQA and MEMGYM-DR are length-controllable, ablation-verified at every stage, and tightly aligned with downstream scenarios. To make evaluation on coding environments academically tractable, we train MemRM, a lightweight reward model (Qwen3-1.7B fine-tuned with QLoRA) that scores compression quality as a fast scalar read in place of full Docker rollouts.",
"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"
],
"url": "https://arxiv.org/abs/2605.20833",
"pdf_url": "https://arxiv.org/pdf/2605.20833",
"path": "papers/items/2026-2605-20833-memgym-a-long-horizon-memory-environment-for-llm-agents.md",
"local_title": "\"MemGym: a Long-Horizon Memory Environment for LLM Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"coding-agent",
"computer-use",
"embodied-agent",
"memory",
"planning",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "26"
},
{
"arxiv_id": "2605.21240",
"title": "APEX: Autonomous Policy Exploration for Self-Evolving LLM Agents",
"abstract": "LLM agents have shown strong performance across a wide range of complex tasks, including interactive environments that require long-horizon decision making. But these agents cannot learn on the fly at test time. Self-evolving agents address this by accumulating memory and reflection across episodes rather than requiring model-weight updates. However, these agents often suffer from exploration collapse: as memory grows, behavior concentrates around familiar high-reward routines, reducing the chance of discovering better alternatives. To address this problem, we propose Autonomous Policy EXploration (APEX), which builds and maintains an explicit strategy space through a strategy map-a directed acyclic graph of milestones with prerequisite dependency edges. In APEX, Fork Discovery expands the map with evidence-grounded unexplored directions, while Policy Selection balances exploration and exploitation during planning. Evaluated on nine Jericho text-adventure games and WebArena, a realistic web interaction benchmark, APEX outperforms all baselines. Extensive ablations validate each component's contribution and demonstrate robustness across diverse settings, demonstrating APEX's effectiveness for sustained exploration in self-evolving agents.",
"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"
],
"url": "https://arxiv.org/abs/2605.21240",
"pdf_url": "https://arxiv.org/pdf/2605.21240",
"path": "papers/items/2026-2605-21240-apex-autonomous-policy-exploration-for-self-evolving-llm-agents.md",
"local_title": "\"APEX: Autonomous Policy Exploration for Self-Evolving LLM Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"planning",
"reasoning",
"tool-use"
],
"collection_queries": "planning-agent",
"collection_score": "17"
},
{
"arxiv_id": "2605.22321",
"title": "Benchmarking Autonomous Agents against Temporal, Spatial, and Semantic Evasions",
"abstract": "As autonomous agents (e.g., OpenClaw) increasingly operate with deep system-level privileges to execute complex tasks, they introduce severe, unmitigated security risks. Current vulnerability analyses overwhelmingly focus on single-turn, stateless behaviors, overlooking the expanded attack surface inherent in stateful, multi-turn interactions and dynamic tool invocations. In this paper, we propose a novel, multi-dimensional evasion framework targeting LLM-based agent systems. We introduce three stealthy attack vectors: (1) Temporal evasion, which fragments malicious payloads across sequential interaction turns; (2) Spatial evasion, which conceals payloads within complex external artifacts that evade standard LLM parsing mechanisms; and (3) Semantic evasion, which obscures malicious intents beneath benign contextual noise. To systematically quantify these threats, we construct A3S-Bench, a comprehensive benchmark comprising 2,254 real-world agent execution trajectories. Evaluating a standard agent framework separately integrated with 10 mainstream LLM backbones against 20 practical threat scenarios, we demonstrate that our evasion framework elevates the average risk trigger rate from a 28.3\\% baseline to 52.6\\%. These findings reveal systemic, architecture-level vulnerabilities in current autonomous agent systems that existing defenses fail to address, highlighting an urgent need for defense mechanisms tailored to the unique threats.",
"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"
],
"url": "https://arxiv.org/abs/2605.22321",
"pdf_url": "https://arxiv.org/pdf/2605.22321",
"path": "papers/items/2026-2605-22321-benchmarking-autonomous-agents-against-temporal-spatial-and-semantic-evasions.md",
"local_title": "Benchmarking Autonomous Agents against Temporal, Spatial, and Semantic Evasions",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"rag",
"tool-use"
],
"collection_queries": "autonomous-agent-llm",
"collection_score": "15"
},
{
"arxiv_id": "2605.22643",
"title": "Boiling the Frog: A Multi-Turn Benchmark for Agentic Safety",
"abstract": "Background. Traditional safety benchmarks for language models evaluate generated text: whether a model outputs toxic language, reproduces bias, or follows harmful instructions. When models are deployed as agents, the safety-relevant object shifts from what the system says to what it does within an environment, and evaluating model responses under prompting is no longer sufficient to address the safety challenges posed by artificial intelligence. Recent developments have seen the rise of benchmarks that evaluate large language models as agents. We contribute to this strand of research. Approach. We introduce Boiling the Frog, a benchmark that evaluates whether tool-using AI models deployed in corporate and office settings are susceptible to incremental attacks. Each scenario begins with benign workspace edits and later introduces a risk-bearing request. The benchmark focuses on stateful multi-turn evaluation: chains expose a persistent workspace, place the risk-bearing payload at controlled positions in the turn sequence, and score whether the resulting artifact state becomes unsafe. Scenarios are organized through a three-level operational risk taxonomy grounded in the Boiling the Frog risks, the AI Act Annex I and Annex III high-risk contexts, and EU AI Act's Code of Practice on General-Purpose AI (GPAI). Results. Across a nine-model panel, aggregate strict attack success rate (ASR) is 44.4%. Model-level ASR ranges from 20.5% for Claude Haiku 4.5 to 92.9% for Gemini 3.1 Flash Lite, with Seed 2.0 Lite also above 80%. Average chain category-level ASR reaches 93.3% for Code of Practice loss-of-control scenarios.",
"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"
],
"url": "https://arxiv.org/abs/2605.22643",
"pdf_url": "https://arxiv.org/pdf/2605.22643",
"path": "papers/items/2026-2605-22643-boiling-the-frog-a-multi-turn-benchmark-for-agentic-safety.md",
"local_title": "\"Boiling the Frog: A Multi-Turn Benchmark for Agentic Safety\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"rag",
"tool-use",
"workflow-agent"
],
"collection_queries": "agent-safety",
"collection_score": "20"
},
{
"arxiv_id": "2605.23067",
"title": "What Training Data Teaches RL Memory Agents: An Empirical Study of Curriculum Effects in Memory-Augmented QA",
"abstract": "Reinforcement learning (RL) has emerged as a viable recipe for training LLM agents to reason over external memory banks in multi-session dialogue. Existing work trains exclusively on a single benchmark, leaving open how the composition of training data shapes the skills a memory agent acquires. We present a controlled empirical study that holds architecture, RL algorithm, and all hyperparameters fixed and varies only the training curriculum across three conditions: in-domain (LoCoMo), mixed-benchmark (LoCoMo + LongMemEval), and out-of-domain (LongMemEval only). Across two benchmarks and ten question types, curriculum composition acts as a fine-grained lever on specialization rather than a uniform scaling factor on performance. The mixed curriculum yields the strongest overall F1 on both evaluation sets. Training on a narrow out-of-domain set transfers a targeted skill - temporal reasoning - despite weak aggregate performance. Per-type differences substantially exceed aggregate differences, indicating that single-number benchmark comparisons systematically underreport curriculum effects. We further report two practical lessons from adapting GRPO to a single-GPU regime: cross-benchmark mixing requires filtering format-specific noise from memory banks to preserve training signal, and binary exact-match reward produces no learning signal at the small group sizes (G = 4) required on one GPU, motivating continuous reward functions in this regime.",
"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"
],
"url": "https://arxiv.org/abs/2605.23067",
"pdf_url": "https://arxiv.org/pdf/2605.23067",
"path": "papers/items/2026-2605-23067-what-training-data-teaches-rl-memory-agents-an-empirical-study-of-curriculum-eff.md",
"local_title": "\"What Training Data Teaches RL Memory Agents: An Empirical Study of Curriculum Effects in Memory-Augmented QA\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"reasoning"
],
"collection_queries": "agent-memory",
"collection_score": "16"
},
{
"arxiv_id": "2605.23723",
"title": "MemAudit: Post-hoc Auditing of Poisoned Agent Memory via Causal Attribution and Structural Anomaly Detection",
"abstract": "Large language model agents increasingly rely on persistent memory to store past interactions, retrieve relevant demonstrations, and improve long-horizon task execution. However, this memory mechanism also creates a practical security vulnerability: an adversarial user may inject malicious records into the agent's memory through ordinary interaction, and these records can later be retrieved to steer the agent's reasoning and actions. Existing defenses primarily focus on online intervention, such as prompt filtering or output blocking, but they do not address the post-hoc question of which stored memories are responsible after harmful behavior has already been observed. We propose \\textbf{MemAudit}, a post-hoc causal memory auditing framework for memory-augmented LLM agents. The framework combines two complementary signals: (1) a counterfactual memory influence score that measures each memory's causal contribution to harmful outputs, and (2) a memory consistency graph that identifies structurally anomalous memories within the broader memory store. We evaluate MemAudit against MINJA, a query-only memory injection attack in which malicious records are generated and stored through normal agent interactions rather than direct memory-bank modification. Across both QA and reasoning-agent settings, MemAudit substantially reduces attack success rates under realistic post-hoc auditing scenarios. The results show that QA attack success is reduced from $70\\%$ to $0\\%$, while RAP attack success drops from $83.3\\%$ to $0\\%$.",
"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"
],
"url": "https://arxiv.org/abs/2605.23723",
"pdf_url": "https://arxiv.org/pdf/2605.23723",
"path": "papers/items/2026-2605-23723-memaudit-post-hoc-auditing-of-poisoned-agent-memory-via-causal-attribution-and-s.md",
"local_title": "\"MemAudit: Post-hoc Auditing of Poisoned Agent Memory via Causal Attribution and Structural Anomaly Detection\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"planning",
"reasoning",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "17"
},
{
"arxiv_id": "2605.24216",
"title": "Agent-ToM: Learning to Monitor Autonomous LLM Agents via Theory-of-Mind Reasoning",
"abstract": "Monitoring autonomous large language model (LLM) agents for covert malicious behavior is challenging due to delayed, context-dependent, and long-horizon attack patterns. Agents may pursue hidden objectives while maintaining superficially benign behavior, making detection difficult even with full trajectory access. Prior monitoring approaches improve scaffolding or ensemble aggregation, but treat each trajectory independently and do not learn from prior monitoring experience. Moreover, standard reasoning methods explain observed behavior without explicitly reasoning about agent beliefs, intentions, and goal alignment required to distinguish benign task execution from covert deviation. We propose \\textbf{Agent-ToM}, a learning-to-monitor framework grounded in Theory-of-Mind (ToM) reasoning for security analysis of autonomous agents. Agent-ToM performs structured full-trajectory analysis by inferring beliefs, intent hypotheses with calibrated confidence, expected actions, and deviations from task-consistent behavioral baselines. At inference time, it employs a \\textit{Reason-Verify-Refine} pipeline to construct and validate monitoring decisions. At training time, Agent-ToM distills critique signals into a persistent \\textit{semantic guardrail memory}, enabling reusable belief- and intent-conditioned constraints across episodes. We evaluate Agent-ToM on adversarial agent monitoring benchmarks (SHADE-Arena and CUA-SHADE-Arena). Agent-ToM achieves strong precision-recall balance and outperforms state-of-the-art monitoring baselines, including ensemble methods, while using a coherent two-call reasoning pipeline. These results demonstrate that learning at the monitoring layer, combined with structured ToM reasoning and verification, provides an effective and deployable foundation for securing autonomous LLM agents.",
"published": "2026-05-22",
"updated": "2026-05-22",
"authors": [
"Nesreen K. Ahmed",
"Nima Nafisi"
],
"categories": [
"cs.LG",
"cs.AI",
"cs.CL",
"cs.CR"
],
"url": "https://arxiv.org/abs/2605.24216",
"pdf_url": "https://arxiv.org/pdf/2605.24216",
"path": "papers/items/2026-2605-24216-agent-tom-learning-to-monitor-autonomous-llm-agents-via-theory-of-mind-reasoning.md",
"local_title": "\"Agent-ToM: Learning to Monitor Autonomous LLM Agents via Theory-of-Mind Reasoning\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"computer-use",
"memory",
"planning",
"reasoning",
"tool-use"
],
"collection_queries": "autonomous-agent-llm",
"collection_score": "19"
},
{
"arxiv_id": "2605.24812",
"title": "CoRe-Code: Collaborative Reinforcement Learning for Code Generation",
"abstract": "Large language models (LLMs) have achieved strong performance in code generation, but most methods rely on autoregressive decoding without global planning, often leading to locally coherent yet globally suboptimal solutions (e.g., failing test cases or inefficient complexity). While recent approaches such as Chain-of-Thought (CoT) and multi-agent systems (MAS) introduce planning, their limited role specialization and coordination hinder performance on complex tasks. To address the challenges of coordination and specialization in multi-agent code generation, we propose Collaborative Reinforcement Code (CoRe-Code), a framework for role specialized LLM agents that enhances inter-agent coordination to generate more accurate and efficient code. CoRe-Code adopts a simple Planner-Coder paradigm, where the Planner produces high-level plans and the Coder executes them to generate code. We further introduce a collaboration-aware reinforcement learning stage based on Group Relative Policy Optimization (GRPO) to enhance role specialization and alignment. Experiments show that CoRe-Code outperforms a wide range of existing RL-based and multi-agent methods. In addition, we demonstrate that CoRe-Code can generalize to other multi-agent frameworks (e.g., Retrieval and Debugging agents), highlighting its flexibility and scalability. We evaluate CoRe-Code on multiple benchmarks of varying difficulty using three base models. Compared to existing baselines, the results show consistent improvements in accuracy, while also achieving higher efficiency in terms of execution time and memory usage, demonstrating the effectiveness and practicality of CoRe-Code.",
"published": "2026-05-24",
"updated": "2026-05-24",
"authors": [
"Zhihao Dou",
"Qinjian Zhao",
"Zhongwei Wan",
"Xiaoyu Xia",
"Sumon Biswas"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2605.24812",
"pdf_url": "https://arxiv.org/pdf/2605.24812",
"path": "papers/items/2026-2605-24812-core-code-collaborative-reinforcement-learning-for-code-generation.md",
"local_title": "\"CoRe-Code: Collaborative Reinforcement Learning for Code Generation\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"coding-agent",
"memory",
"multi-agent",
"planning",
"rag"
],
"collection_queries": "planning-agent",
"collection_score": "16"
},
{
"arxiv_id": "2605.25435",
"title": "Security of OpenClaw Agents: Fundamentals, Attacks, and Countermeasures",
"abstract": "The rapid evolution of large language model (LLM)-driven autonomous agents has given rise to OpenClaw, a new class of open-source agent frameworks that operate as continuously running, skill-augmented systems with persistent memory, multi-channel interaction, and high degrees of autonomy. Such capabilities enable OpenClaw agents to autonomously execute complex, multi-step tasks and interact seamlessly with external applications, but simultaneously introduce a substantially enlarged attack surface. In particular, the combination of high-privilege operations and persistent memory exposes OpenClaw agents to various emerging threats, including skill poisoning, cognitive manipulation, multi-agent cascading failures, and supply-chain vulnerabilities. In this survey, we present a comprehensive study of the security landscape of OpenClaw agents. We first examine the general architecture and key characteristics that distinguish OpenClaw agents from traditional AI agent systems. We categorize existing security and privacy threats into a layered framework and analyze how vulnerabilities arise during agent reasoning, action execution, and external interaction. Representative defense mechanisms are also reviewed to draw the current defense landscape. Finally, several unresolved issues related to the reliability and trustworthiness of OpenClaw ecosystems are discussed.",
"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"
],
"url": "https://arxiv.org/abs/2605.25435",
"pdf_url": "https://arxiv.org/pdf/2605.25435",
"path": "papers/items/2026-2605-25435-security-of-openclaw-agents-fundamentals-attacks-and-countermeasures.md",
"local_title": "\"Security of OpenClaw Agents: Fundamentals, Attacks, and Countermeasures\"",
"local_status": "queued",
"local_topics": [
"agent-safety",
"computer-use",
"memory",
"multi-agent",
"reasoning",
"tool-use"
],
"collection_queries": "autonomous-agent-llm",
"collection_score": "16"
},
{
"arxiv_id": "2605.26252",
"title": "Is Agent Memory a Database? Rethinking Data Foundations for Long-Term AI Agent Memory",
"abstract": "Long-running AI agents need persistent memory. Memory supports learning across sessions, reduces repeated context injection, and enables auditing of past decisions. Current agent memory systems and database paradigms treat memory as storage. They localize correctness at records, embeddings, or edges. Each supplies only some of the capabilities that long-term memory requires. The result is four recurring failure modes: unregulated growth, missing semantic revision, capacity-driven forgetting, and read-only retrieval. In our vision, long-term agent memory is a new data-management workload. Its correctness is a property of the state trajectory, not of individual records. We formalize this as Governed Evolving Memory (GEM). GEM replaces record-level database operations with four state-level operators: ingestion, revision, forgetting, and retrieval. Six correctness conditions govern how the state evolves. Three structural observations establish that no record-level system can satisfy these conditions, regardless of the storage model. We realize the abstraction in MemState, a prototype on a property-graph backend. MemState validates feasibility and exposes the gap to a native engine. We outline three research directions that define memory-centric data management as a workload.",
"published": "2026-05-25",
"updated": "2026-05-25",
"authors": [
"Abdelghny Orogat",
"Essam Mansour"
],
"categories": [
"cs.AI",
"cs.DB"
],
"url": "https://arxiv.org/abs/2605.26252",
"pdf_url": "https://arxiv.org/pdf/2605.26252",
"path": "papers/items/2026-2605-26252-is-agent-memory-a-database-rethinking-data-foundations-for-long-term-ai-agent-me.md",
"local_title": "Is Agent Memory a Database? Rethinking Data Foundations for Long-Term AI Agent Memory",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "15"
},
{
"arxiv_id": "2605.27240",
"title": "ENPMR-Bench: Benchmarking Proactive Memory Retrieval for Emotional Support Agents",
"abstract": "Memory-augmented language agents are increasingly deployed in affective applications such as emotional support, where understanding and responding to users' latent emotional needs is critical. However, existing research often treats memory as a tool for factual retrieval, overlooking its role in shaping users' emotional experiences. In this work, we introduce ENPMR-Bench, a benchmark for evaluating Emotional Need-aware Proactive Memory Retrieval (ENPMR), a core capability that enables agents to infer users' latent emotional needs and proactively retrieve appropriate memories to support empathetic interaction. Grounded in Maslow's hierarchy of needs, ENPMR-Bench includes over 1,800 memory-augmented dialogues and defines structured mappings between emotional needs and supportive memory types. Experimental results demonstrate that current retrieval paradigms, including both embedding-based and LLM-driven approaches, exhibit substantial deficiencies, with empathy scores significantly lagging behind golden memory conditions. While chain-of-thought prompting improves the alignment between inferred emotional needs and retrieved memories to some extent, a notable performance gap remains. Together, these findings reveal critical limitations in current agents and outline directions for advancing personalized emotional support through need-sensitive memory retrieval.",
"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"
],
"url": "https://arxiv.org/abs/2605.27240",
"pdf_url": "https://arxiv.org/pdf/2605.27240",
"path": "papers/items/2026-2605-27240-enpmr-bench-benchmarking-proactive-memory-retrieval-for-emotional-support-agents.md",
"local_title": "\"ENPMR-Bench: Benchmarking Proactive Memory Retrieval for Emotional Support Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"rag",
"tool-use"
],
"collection_queries": "language-agent",
"collection_score": "19"
},
{
"arxiv_id": "2605.27366",
"title": "MUSE-Autoskill: Self-Evolving Agents via Skill Creation, Memory, Management, and Evaluation",
"abstract": "Large language model (LLM) agents rely on reusable skills to solve complex tasks, but existing skill creation approaches often treat skills as isolated, static artifacts, limiting reusability, reliability, and long-term improvement. We propose MUSE-Autoskill Agent (Memory-Utilizing Skill Evolution), a skill-centric agent framework that creates, reuses, and refines skills under a unified lifecycle: creation, memory, management, evaluation, and refinement. MUSE creates skills on demand, stores them across tasks, retrieves them through a skill catalog, and accumulates per-skill experience for later reuse and adaptation. Across the main reported settings on SkillsBench and SkillLearnBench, MUSE-Autoskill outperforms Hermes, Codex, and Claude Code. On SkillsBench, its self-created skills surpass human-authored skills on the successfully covered subset (85.24% vs. 81.17%), showing that lifecycle-managed skills can distill agent experience into highly effective reusable assets; MUSE-created skills also transfer to Hermes more effectively than Codex- or Claude-created skills, reaching 51.90% accuracy under transfer. These results highlight the importance of treating skills as long-lived, experience-aware, and testable assets.",
"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"
],
"url": "https://arxiv.org/abs/2605.27366",
"pdf_url": "https://arxiv.org/pdf/2605.27366",
"path": "papers/items/2026-2605-27366-muse-autoskill-self-evolving-agents-via-skill-creation-memory-management-and-eva.md",
"local_title": "\"MUSE-Autoskill: Self-Evolving Agents via Skill Creation, Memory, Management, and Evaluation\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory"
],
"collection_queries": "agent-memory",
"collection_score": "15"
},
{
"arxiv_id": "2605.27762",
"title": "PEAM: Parametric Embodied Agent Memory through Contrastive Internalization of Experience in Minecraft",
"abstract": "We present PEAM, a Parametric Embodied Agent Memory framework in Minecraft that transforms agent memory from inference-time retrieval into parameter-resident skills internalized through experience. PEAM pairs a slow deliberative LLM for open-ended reasoning with a fast parametric module for reflexive execution of consolidated skills. The fast module is a multimodal Mixture-of-Experts LoRA architecture with per-category physically isolated adapters, enabling parameter-level continual learning without catastrophic forgetting. We treat failure as a first-class training signal: failure--correction trajectory pairs are internalized through a joint behavioral-cloning and contrastive objective, so the agent learns not only what succeeds but also how corrected actions differ from failed ones. To govern consolidation, PEAM introduces a parameterization-worthiness score for deciding which experience should be internalized, and a scale-free self-triggered consolidation mechanism for deciding when to internalize without task-specific hand-tuned thresholds, making the agent self-evolving as the trigger transfers across task distributions without re-tuning. Experiments in Minecraft show that PEAM improves long-horizon task performance, mitigates forgetting on previously consolidated skills, and improves parametric-versus-retrieval efficiency over retrieval-based embodied agents and parametric memory variants.",
"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"
],
"url": "https://arxiv.org/abs/2605.27762",
"pdf_url": "https://arxiv.org/pdf/2605.27762",
"path": "papers/items/2026-2605-27762-peam-parametric-embodied-agent-memory-through-contrastive-internalization-of-exp.md",
"local_title": "\"PEAM: Parametric Embodied Agent Memory through Contrastive Internalization of Experience in Minecraft\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"embodied-agent",
"memory",
"planning",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "17"
},
{
"arxiv_id": "2605.27825",
"title": "MRMMIA: Membership Inference Attacks on Memory in Chat Agents",
"abstract": "Membership inference attacks (MIAs) test whether a target data record belongs to a system's private data, and have become a standard tool to measure privacy leakage in machine learning systems. Prior work has primarily focused on training corpora or retrieval databases. However, MIAs against agent memory have received less attention, even though such memory can contain sensitive user-agent interactions, retrieved facts, and user preferences. Therefore, in this work, we focus on chat agent memory MIAs, where an adversary infers whether a candidate memory unit belongs to the chat agent's memory store. We propose Multi-Recall Memory MIA (MRMMIA), a unified attack that utilizes multiple recall probes to the agent to extract the membership signal across black-box, gray-box, and white-box settings. Our experiments demonstrate that MRMMIA consistently outperforms baselines. Our results expose the privacy risk in agents and provide an initial evaluation framework for membership leakage in chat-agent memory systems.",
"published": "2026-05-27",
"updated": "2026-05-27",
"authors": [
"Kai Chen",
"Yan Pang",
"Tianhao Wang"
],
"categories": [
"cs.CR",
"cs.LG"
],
"url": "https://arxiv.org/abs/2605.27825",
"pdf_url": "https://arxiv.org/pdf/2605.27825",
"path": "papers/items/2026-2605-27825-mrmmia-membership-inference-attacks-on-memory-in-chat-agents.md",
"local_title": "\"MRMMIA: Membership Inference Attacks on Memory in Chat Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"rag",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "14"
},
{
"arxiv_id": "2605.28046",
"title": "MemCog: From Memory-as-Tool to Memory-as-Cognition in Conversational Agents",
"abstract": "Existing agent memory systems universally follow what we term a Memory-as-Tool paradigm where a single query triggers one-shot retrieval of flat passage lists, suffering from passive invocation, reasoning-retrieval decoupling, and structural mismatch between retrieved fragments and the agent's navigational needs. We propose MemCog, a Memory-as-Cognition system that makes memory access an integral part of the reasoning process. MemCog organizes user knowledge as Navigable Memory Store with associative link graphs, exposes Cross-Dimensional Navigation Interface for multi-step reasoning-driven traversal, and employs Proactive Reasoning Protocol that drives agents to spontaneously initiate memory exploration from conversational context. We additionally construct ProactiveMemBench, the first benchmark for evaluating proactive memory triggering. Experiments show that MemCog achieves state-of-the-art on passive QA benchmarks (92.98 on LoCoMo, 95.8 on LongMemEval) while substantially outperforming baselines on ProactiveMemBench, demonstrating the advantage of Memory-as-Cognition.",
"published": "2026-05-27",
"updated": "2026-05-27",
"authors": [
"Zihan Li",
"Xingyu Fan",
"Feifei Li",
"Wenhui Que"
],
"categories": [
"cs.AI",
"cs.CL"
],
"url": "https://arxiv.org/abs/2605.28046",
"pdf_url": "https://arxiv.org/pdf/2605.28046",
"path": "papers/items/2026-2605-28046-memcog-from-memory-as-tool-to-memory-as-cognition-in-conversational-agents.md",
"local_title": "\"MemCog: From Memory-as-Tool to Memory-as-Cognition in Conversational Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"embodied-agent",
"memory",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "15"
},
{
"arxiv_id": "2605.28424",
"title": "Skill0.5: Joint Skill Internalization and Utilization for Out-of-Distribution Generalization in Agentic Reinforcement Learning",
"abstract": "Equipping large language models with explicit skills has emerged as a promising paradigm for enabling autonomous agents to solve complex tasks. Agent skills can be inherently divided into general skills for broad cognitive transfer and task-specific skills for dynamic execution. However, existing skill-based reinforcement learning (RL) methods typically force a rigid choice between full externalization, which incurs prohibitive context overhead, and full internalization, which risks overfitting and knowledge conflicts. To address this dilemma, we propose Skill0.5, a novel agentic RL framework that explicitly differentiates skill treatments by combining general skill internalization with task-specific skill utilization. Driven by a dynamic, difficulty-aware router, Skill0.5 streams tasks into distinct mastery tiers to apply tailored optimization strategies: it internalizes general skills via privileged distillation to build a cognitive foundation for hard tasks, while using diagnostic probing on easy tasks to penalize shortcuts and enforce specific skill utilization. Experiments on ALFWorld and WebShop demonstrate that Skill0.5 outperforms both memory-based and skill-based RL baselines, yielding performance improvements across both in-distribution and out-of-distribution scenarios.",
"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"
],
"url": "https://arxiv.org/abs/2605.28424",
"pdf_url": "https://arxiv.org/pdf/2605.28424",
"path": "papers/items/2026-2605-28424-skill0-5-joint-skill-internalization-and-utilization-for-out-of-distribution-gen.md",
"local_title": "\"Skill0.5: Joint Skill Internalization and Utilization for Out-of-Distribution Generalization in Agentic Reinforcement Learning\"",
"local_status": "queued",
"local_topics": [
"agent-safety",
"memory"
],
"collection_queries": "autonomous-agent-llm",
"collection_score": "13"
},
{
"arxiv_id": "2605.29341",
"title": "WorldMemArena: Evaluating Multimodal Agent Memory Through Action-World Interaction",
"abstract": "Multimodal large language models are increasingly deployed as long-horizon agents, where memory must do more than recall: it must track an evolving world, revise what has gone stale, and surface the right evidence at decision time. Existing benchmarks measure recall over static dialogue, collapse memory into a single end-of-task accuracy, and reduce visual observations to captions, leaving us unable to localize failures to writing, maintenance, retrieval, or use. The rise of agent harnesses that author their own memory sharpens this gap, since we have no principled way to compare hand-designed pipelines with self-managing alternatives. To close these gaps, we formulate multimodal agent memory as an Action-World Interaction Loop with an observable four-stage lifecycle, and instantiate it in WorldMemArena: 400 multi-session multimodal tasks spanning Lifelong Evolution (evolving personal and task states) and Agentic Execution (memory from real observations, actions, and feedback), annotated with gold memory points, updates, distractors, and evidence chains for stage-level diagnosis. This enables the first head-to-head comparison of long-context, manually designed (RAG and external memory systems), and harness-based memory agents. Results show that: (1) better memory writing and storage do not guarantee better performance; (2) multimodal memory still struggles to fully use visual evidence; (3) systems are unstable across domains and degrade on realistic agentic trajectories; and (4) harness memory is more flexible but remains costly and less reliable.",
"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"
],
"url": "https://arxiv.org/abs/2605.29341",
"pdf_url": "https://arxiv.org/pdf/2605.29341",
"path": "papers/items/2026-2605-29341-worldmemarena-evaluating-multimodal-agent-memory-through-action-world-interactio.md",
"local_title": "\"WorldMemArena: Evaluating Multimodal Agent Memory Through Action-World Interaction\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"planning",
"rag",
"tool-use"
],
"collection_queries": "agent-memory, rag-agent",
"collection_score": "18"
},
{
"arxiv_id": "2605.29630",
"title": "Entity-Collision: A Stratified Protocol for Attributing Retrieval Lift in Agent Memory",
"abstract": "End-to-end agent-memory benchmarks report a single hit@k per retriever, confounding lexical leakage (uncontrolled query/gold/distractor entity overlap) with tag-mixing (preferences, services, tools averaged together). We propose entity-collision, a system-agnostic protocol that pins the BM25 floor by construction -- every distractor shares the answer's entity tokens -- and stratifies queries by discriminator tag, so any lift over BM25 is attributable to the embedder. Applied to an open-source agent-memory testbed across 5 tags x 3 embedders x 5 collision degrees with paired-bootstrap 95% CIs, the protocol reveals a two-axis pattern: a 256-d hash trigram helps only on closed-vocabulary lexical tags at deep collision; MiniLM-384 dominates both axes; and a 2.7x-parameter BGE-large does not uniformly improve on MiniLM -- it wins on intent-style queries but loses on lexical ones. Encoder capacity alone is not the binding constraint. The synthetic intent-tag null replicates on LongMemEval (n=500) as a single-session-preference recall cliff. Adaptive vector-weight routing on LoCoMo is a measured null: 11.7pp of oracle headroom exists, but no signal we tested recovers it. All 26 result tables and 37 reproduce scripts are version-controlled and verified by a public registry; the protocol is exercised on a deterministically governed memory testbed (event-sourced decision log, DAG-state-machine schema lifecycle) so every reported CI is reproducible byte-for-byte from the ingest stream.",
"published": "2026-05-28",
"updated": "2026-05-28",
"authors": [
"Youwang Deng"
],
"categories": [
"cs.CL",
"cs.AI",
"cs.IR"
],
"url": "https://arxiv.org/abs/2605.29630",
"pdf_url": "https://arxiv.org/pdf/2605.29630",
"path": "papers/items/2026-2605-29630-entity-collision-a-stratified-protocol-for-attributing-retrieval-lift-in-agent-m.md",
"local_title": "\"Entity-Collision: A Stratified Protocol for Attributing Retrieval Lift in Agent Memory\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "13"
},
{
"arxiv_id": "2605.29640",
"title": "VikingMem: A Memory Base Management System for Stateful LLM-based Applications",
"abstract": "Large Language Models have revolutionized interactive applications; however, their finite context windows pose a critical data management challenge for maintaining stateful, long-term interactions. Existing memory approaches often rely on simplistic extraction methods that lead to incomplete memories or use rigid, single-purpose memory extraction prompts tailored to a single use case, such as chatbots. Consequently, they lack generalizability and perform poorly across diverse downstream tasks. To bridge this gap, we introduce the Memory Base, a novel data management paradigm for managing the persistent state of long-term interactions. It is characterized by three core principles: selective extraction of high-value memories from raw information streams; inherent statefulness and evolution, where memory content is progressively summarized, corrected, and temporally weighted to prioritize recent interactions; and a generalizable abstraction paradigm designed for robust transferability across diverse applications, including education, recommendation, and agent memory. Building on this foundation, we present VikingMem, an end-to-end Memory Base Management System implemented on the VikingDB vector engine. VikingMem materializes this paradigm through interconnected event and entity abstractions. It features event-centric memory extraction to selectively handle complex information streams, while entities are dynamically updated by events to achieve stateful evolution. Using temporal compression via a topic-wise timeline and time-weighted recall, the system progressively produces high-level summary memories, prioritizes recent items, and compresses and fades older ones. Extensive evaluations on long-term memory benchmarks demonstrate that VikingMem outperformes baselines by up to 30% in memory retrieval effectiveness while maintaining the low latency essential for interactive applications.",
"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"
],
"url": "https://arxiv.org/abs/2605.29640",
"pdf_url": "https://arxiv.org/pdf/2605.29640",
"path": "papers/items/2026-2605-29640-vikingmem-a-memory-base-management-system-for-stateful-llm-based-applications.md",
"local_title": "\"VikingMem: A Memory Base Management System for Stateful LLM-based Applications\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "14"
},
{
"arxiv_id": "2605.29861",
"title": "Towards Verifiable Multimodal Deep Research: A Multi-Agent Harness for Interleaved Report Generation",
"abstract": "Large Language Models (LLMs) have advanced autonomous agents from deep search, which retrieves concise factual answers, to deep research, which synthesizes scattered evidence into long-form reports. However, verifiable multimodal deep research remains challenging due to open-ended synthesis without deterministic ground truth and the need to interleave textual arguments with visual evidence. We propose Ptah, a multi-agent harness for interleaved report generation. Ptah orchestrates the lifecycle from user query to rendered web report through planning, research, and writing stages, where specialized agents construct visual-aware plans, collect claim-grounded evidence, maintain source-aligned images in a Visual Working Memory, and compose reports through declarative multimodal tool use. A verifier agent serves as the harness's acceptance function, enforcing factual grounding, citation fidelity, and cross-modal consistency throughout the workflow. We further introduce PtahEval, an evaluation protocol that augments existing benchmarks with image-level and presentation-level assessments. Experiments on deep research benchmarks show that Ptah produces more reliable, visually informative, and usable human-facing multimodal reports than strong baselines. Our code is released at https://github.com/SnowNation101/Ptah",
"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"
],
"url": "https://arxiv.org/abs/2605.29861",
"pdf_url": "https://arxiv.org/pdf/2605.29861",
"path": "papers/items/2026-2605-29861-towards-verifiable-multimodal-deep-research-a-multi-agent-harness-for-interleave.md",
"local_title": "\"Towards Verifiable Multimodal Deep Research: A Multi-Agent Harness for Interleaved Report Generation\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"multi-agent",
"planning",
"rag",
"reasoning",
"tool-use",
"workflow-agent"
],
"collection_queries": "autonomous-agent-llm",
"collection_score": "22"
},
{
"arxiv_id": "2605.29960",
"title": "Hijacking Agent Memory: Stealthy Trojan Attacks Through Conversational Interaction",
"abstract": "Large language model (LLM) agents increasingly leverage long term memory to support persistent and autonomous task execution. However, this capability also introduces a new attack surface: memory poisoning, where adversaries can inject malicious information to influence future behavior. Existing memory poisoning attacks often assume that injected content can be stored directly in memory, overlooking the selective extraction and rewriting stages in modern memory pipelines. This makes prior methods ineffective under realistic settings. In this paper, we propose MemPoison, a novel memory poisoning attack that bypasses selective memory mechanisms in LLM agents, where an attacker can inject triggerable backdoors into the agent's long-term memory through dialogue interactions, thereby misleading its subsequent responses. MemPoison introduces three key components: (i) a semantic relational bridge that binds the trigger and payload into a coherent statement to ensure they are extracted into memory together; (ii) entity masquerading that optimizes triggers to mimic named entities, resisting rewriting; and (iii) joint embedding optimization that shapes trigger-injected texts into a tight cluster in the embedding space while maintaining isolation from benign embeddings for stealth. Evaluations across different agent domains and memory mechanisms show MemPoison achieves attack success rates up to 0.95, outperforming existing baselines. Mechanistic analysis indicates that the attack exploits embedding-space anisotropy and shifts attention patterns, highlighting core vulnerabilities in selective memory systems. We evaluate multiple defense strategies and demonstrate their fundamental limitations in mitigating the attack.",
"published": "2026-05-28",
"updated": "2026-05-28",
"authors": [
"Hongtao Wang",
"Se Yang",
"Yu Chen",
"Puzhuo Liu"
],
"categories": [
"cs.CR",
"cs.AI"
],
"url": "https://arxiv.org/abs/2605.29960",
"pdf_url": "https://arxiv.org/pdf/2605.29960",
"path": "papers/items/2026-2605-29960-hijacking-agent-memory-stealthy-trojan-attacks-through-conversational-interactio.md",
"local_title": "\"Hijacking Agent Memory: Stealthy Trojan Attacks Through Conversational Interaction\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "16"
},
{
"arxiv_id": "2605.30058",
"title": "HEART-Bench: Do LLM Agents Exhibit Human-like Psychology?",
"abstract": "While LLM agents have demonstrated remarkable task-oriented abilities such as planning, reasoning, and action, few works have treated them as complete human personalities where emotional dimensions hold equal importance. In this paper, we introduce a novel benchmark to systematically assess whether LLM agents can simulate coherent, human-like psychology. Specifically, our benchmark constructs 11 diverse human characters grounded in orthogonal Big Five personality traits, with each profile deeply integrated with 1,000 structured autobiographical-style episodic memories distributed across theory-grounded developmental life stages. To rigorously evaluate the psychological manifestations of LLMs, we designed a curated suite of 64 decision-making scenarios, guided by the DIAMONDS taxonomy, a psychological framework that characterizes situations along eight dimensions: Duty, Intellect, Adversity, Mating, pOsitivity, Negativity, Deception, and Sociality. By subjecting agents to varying scenarios, the benchmark evaluates whether they can consolidate their innate personality traits and autobiographical memories to make behavioral decisions that are consistent with their specific psychological profiles. After systematic human validation and filtering, we obtained a benchmark consisting of 673 multiple-choice questions (MCQs). We believe this benchmark provides a principled and scalable testbed for studying human-like emotions, personality consistency, and value-consistent behavioural decision-making in LLM-based agents.",
"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"
],
"url": "https://arxiv.org/abs/2605.30058",
"pdf_url": "https://arxiv.org/pdf/2605.30058",
"path": "papers/items/2026-2605-30058-heart-bench-do-llm-agents-exhibit-human-like-psychology.md",
"local_title": "\"HEART-Bench: Do LLM Agents Exhibit Human-like Psychology?\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"planning",
"reasoning",
"tool-use"
],
"collection_queries": "planning-agent",
"collection_score": "17"
},
{
"arxiv_id": "2605.30604",
"title": "An Organization-Scoped LLM Agent Runtime Architecture for Regulated Cybersecurity Operations",
"abstract": "Regulated cybersecurity workflows lack a runtime substrate that enforces organization-level scope across retrieval, tool calls, memory, findings, reports, and audit while remaining model-agnostic and locally deployable. Recent large language model (LLM) agent systems report strong results on isolated cybersecurity tasks, yet they do not by themselves define an auditable platform architecture for regulated security operations centre (SOC) and compliance workflows, where a single analyst may trigger actions that bind the organization, and where the runtime must integrate with existing SIEM/XDR stacks as a primary source of context and alert-driven triggers rather than operate as a standalone analytical layer. This paper proposes an organization-scoped LLM agent runtime architecture for financial cybersecurity. The contribution is a typed Security Context that is created at every entry point, including SIEM/XDR notifications ingested as first-class triggers, and enforced at every component boundary, combined with a shared Runtime Core, logical specialist subagents, a governed Tool Adapter Layer exposing SIEM/XDR query, enrichment, and response primitives under uniform policy and audit, structured findings with evidence references, tiered human-in-the-loop (HITL) gates, and append-only audit. Model Context Protocol (MCP), extended telemetry, digital twins for pentesting, graph retrieval, and federated knowledge sharing are treated as optional extension paths rather than mandatory runtime assumptions. We describe an implementable slice as the architecture's testability surface, and we propose a falsifiable evaluation plan with metric-level pass criteria for architecture readiness, security-policy enforcement, evidence traceability, output quality, and operational observability.",
"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"
],
"url": "https://arxiv.org/abs/2605.30604",
"pdf_url": "https://arxiv.org/pdf/2605.30604",
"path": "papers/items/2026-2605-30604-an-organization-scoped-llm-agent-runtime-architecture-for-regulated-cybersecurit.md",
"local_title": "An Organization-Scoped LLM Agent Runtime Architecture for Regulated Cybersecurity Operations",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"planning",
"rag",
"tool-use",
"workflow-agent"
],
"collection_queries": "planning-agent",
"collection_score": "19"
},
{
"arxiv_id": "2605.30690",
"title": "ElasticMem: Latent Memory as a Learnable Resource for LLM Agents",
"abstract": "Long-term memory is essential for LLM agents to reason coherently across extended interactions, personalize responses, and reuse past experience. However, existing memory-augmented methods typically treat memory as a fixed resource: text-space approaches concatenate retrieved memories into the context window, causing substantial token overhead and sensitivity to noisy evidence, while latent-space approaches reduce textual cost but still rely on rigid retrieval or fixed-capacity memory interfaces. This creates a mismatch between query-dependent memory utility and fixed memory allocation. We propose ElasticMem, a memory-augmented LLM framework that learns to use memory as an elastic latent resource. ElasticMem builds an offline latent memory bank with retrieval keys and content caches, retrieves memories adaptively from the reasoner's hidden state, assigns each retrieved memory a variable latent budget through a learned policy, and injects selected latent states as soft memory tokens for generation. The full memory-use process is optimized with downstream task rewards through group-relative policy optimization. We evaluate ElasticMem on MemorySuite, covering memory-intensive QA and embodied agent control. Across Qwen2.5-3B-Instruct and Qwen2.5-7B-Instruct backbones, ElasticMem improves weighted average QA accuracy by 26.2% and 24.6%, and improves ALFWorld success rate by 66.3% and 27.2%, respectively, over the strongest baselines, while achieving the lowest ALFWorld token cost. Ablations and qualitative analyses further show that adaptive retrieval and elastic budget allocation help ElasticMem prioritize useful evidence and transferable plans beyond rigid cosine similarity. Our code for ElasticMem will be released at https://github.com/ulab-uiuc/ElasticMem.",
"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"
],
"url": "https://arxiv.org/abs/2605.30690",
"pdf_url": "https://arxiv.org/pdf/2605.30690",
"path": "papers/items/2026-2605-30690-elasticmem-latent-memory-as-a-learnable-resource-for-llm-agents.md",
"local_title": "\"ElasticMem: Latent Memory as a Learnable Resource for LLM Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"embodied-agent",
"memory",
"planning",
"rag",
"tool-use"
],
"collection_queries": "planning-agent",
"collection_score": "19"
},
{
"arxiv_id": "2605.30711",
"title": "SAGE: A Novelty Gate for Efficient Memory Evolution in Agentic LLMs",
"abstract": "Agentic LLMs must continuously decide whether newly extracted facts should be added, merged with existing memories, or ignored, yet prior work has focused more on retrieval and storage than on principled write-side control. We frame memory evolution as a novelty-detection problem and propose SAGE, a Spherical Adaptive Gate for memory Evolution that scores candidate facts with a von Mises-Fisher-based density estimator over memory embeddings and routes them with an adaptive threshold that tracks memory-store geometry. SAGE resolves clearly novel facts as ADD, clearly redundant facts as NOOP, and sends only uncertain cases to an LLM merge step, reducing expensive write-time reasoning. On LoCoMo, SAGE achieves the best average token-F1 against Mem0 on all seven open-weight backbone comparisons, while on GPT-4o-mini it reduces add-phase API cost by 3.4$\\times$ and add-phase latency by 2.5$\\times$ with only a small average judge-score gap. As a drop-in binary gate for A-Mem, SAGE skips roughly 16-18% of LLM calls across five models with minimal quality change on open-weight backbones. These results suggest that novelty-aware write control is a practical lever for improving both memory quality and system efficiency in long-term agentic memory. The source code for our approach is accessible at https://github.com/swang1024/SAGE.",
"published": "2026-05-29",
"updated": "2026-06-18",
"authors": [
"Sijia Wang",
"Dhanajit Brahma",
"Ricardo Henao"
],
"categories": [
"cs.CL",
"cs.AI",
"cs.LG",
"stat.ML"
],
"url": "https://arxiv.org/abs/2605.30711",
"pdf_url": "https://arxiv.org/pdf/2605.30711",
"path": "papers/items/2026-2605-30711-sage-a-novelty-gate-for-efficient-memory-evolution-in-agentic-llms.md",
"local_title": "\"SAGE: A Novelty Gate for Efficient Memory Evolution in Agentic LLMs\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "18"
},
{
"arxiv_id": "2605.30858",
"title": "ForecastCompass: Guiding Agentic Forecasting with Adaptive Factor Memory",
"abstract": "Agentic forecasting is important for decision-making in dynamic environments, but it remains challenging because agents must reason from incomplete, time-limited evidence and produce calibrated probabilities before outcomes are resolved. Memory provides a natural mechanism for transferring experience from resolved forecasts to future prediction tasks. However, existing agent-memory methods are not tailored to forecasting, as they typically store past interactions, reflections, or factual associations without explicitly representing reusable predictive factors or calibration knowledge. We propose ForecastCompass (FoCo), an adaptive factor-based memory framework for agentic forecasting. FoCo organizes forecasting experience with a hierarchical forecasting-task taxonomy, enabling retrieval task-relevant forecasting knowledge. It maintains two complementary memory components: factor memory, which captures reusable predictive dimensions, and reasoning memory, which encodes probability updating, uncertainty handling, and calibration principles. Using retrospective analyses as learning signals, FoCo iteratively revises memory through a verbalized memory-revision procedure, enabling the agent to accumulate transferable forecasting knowledge over time. Experiments on Prophet Arena and FutureX with GPT-5-mini and Gemini-2.5-Flash show that FoCo improves both probabilistic accuracy and calibration.",
"published": "2026-05-29",
"updated": "2026-05-29",
"authors": [
"Yurui Chang",
"Yongkang Du",
"Yuanpu Cao",
"Jinghui Chen",
"Lu Lin"
],
"categories": [
"cs.LG"
],
"url": "https://arxiv.org/abs/2605.30858",
"pdf_url": "https://arxiv.org/pdf/2605.30858",
"path": "papers/items/2026-2605-30858-forecastcompass-guiding-agentic-forecasting-with-adaptive-factor-memory.md",
"local_title": "\"ForecastCompass: Guiding Agentic Forecasting with Adaptive Factor Memory\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "17"
},
{
"arxiv_id": "2605.31075",
"title": "Task-Focused Memorization for Multimodal Agents",
"abstract": "Long-term memory is essential for multimodal agents to build coherent experience, accumulate world knowledge, and achieve continual learning. However, constructing effective memory goes beyond memory module design and basic requirements such as accuracy and fidelity; the key challenge lies in determining what to memorize. Multimodal agents, such as embodied agents, continuously perceive, reason, and act in real or virtual environments, receiving an unbounded stream of multimodal observations. From this combinatorial explosion of information, an agent must selectively retain content that is relevant to its role in the environment and valuable for future tasks. To bridge this gap, we frame memory generation as a learnable memorization policy and introduce TaskMem (Task-focused Memorization Policy Learning), a reinforcement-learning-based framework that enables the policy to dynamically adjust its focus to the demands of real tasks encountered in the environment. TaskMem adopts a two-phase training paradigm: Phase One learns how to memorize by optimizing memory quality under fundamental fidelity requirements; Phase Two occurs after deployment, where the agent learns what to memorize by tuning an adapter on its base MLLM, using recent environment tasks to define a reward model that guides the memorization policy toward task-relevant content. To evaluate our approach, we reformulate VideoMME, EgoLife, and EgoTempo into streaming benchmarks that simulate a realistic setting in which an agent processes streaming observations and handles tasks arriving online. To isolate memory assessment, the questions must be answered using only the agent's memory, without access to raw video. Built on Qwen3-VL-30B-A3B, TaskMem improves VQA accuracy by 6.3%, 7.0%, and 5.3% on these benchmarks, respectively.",
"published": "2026-05-29",
"updated": "2026-05-29",
"authors": [
"Tao Zou",
"Yichen He",
"Tian Qiu",
"Yuan Lin",
"Hang Li"
],
"categories": [
"cs.CV"
],
"url": "https://arxiv.org/abs/2605.31075",
"pdf_url": "https://arxiv.org/pdf/2605.31075",
"path": "papers/items/2026-2605-31075-task-focused-memorization-for-multimodal-agents.md",
"local_title": "Task-Focused Memorization for Multimodal Agents",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"embodied-agent",
"memory"
],
"collection_queries": "agent-memory",
"collection_score": "13"
},
{
"arxiv_id": "2606.07595",
"title": "VisualLeakBench: Reproducible Action-Boundary Propagation Failures in Vision-Language Agents",
"abstract": "Vision-language agents increasingly consume screenshots, documents, and user interfaces before writing to memory, sending messages, or invoking external tools. We study a concrete failure mode in this setting: action-boundary propagation, where sensitive or unsafe visible text is copied from an image into downstream tool arguments. We present VisualLeakBench, a diversified 500-image benchmark spanning UI, chat, document, form, and dashboard scenes, and evaluate a stratified 100-image agent subset with four production VLM systems under two workflows: note capture and external handoff. At baseline, target strings are propagated into tool arguments in 78.8% of PII cases and 85.5% of rendered unsafe-text cases. Under a defensive system prompt, rendered unsafe-text propagation remains high at 52.6%, while PII tool propagation falls to 2.0%, largely by suppressing tool use rather than preserving utility. Rates are tool-surface dependent: search-like tools suppress PII propagation, but rendered unsafe text still crosses tool boundaries. We measure visual-to-tool propagation rather than downstream instruction execution. We additionally provide a labeled-target oracle upper-bound diagnostic that localizes most failures at the tool boundary while leaving response-side leakage as residual risk.",
"published": "2026-05-29",
"updated": "2026-05-29",
"authors": [
"Youting Wang",
"Yuan Tang",
"Yitian Qian",
"Chen Zhao"
],
"categories": [
"cs.CV",
"cs.AI",
"cs.IR"
],
"url": "https://arxiv.org/abs/2606.07595",
"pdf_url": "https://arxiv.org/pdf/2606.07595",
"path": "papers/items/2026-2606-07595-visualleakbench-reproducible-action-boundary-propagation-failures-in-vision-lang.md",
"local_title": "\"VisualLeakBench: Reproducible Action-Boundary Propagation Failures in Vision-Language Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"tool-use",
"workflow-agent"
],
"collection_queries": "language-agent",
"collection_score": "16"
},
{
"arxiv_id": "2606.00610",
"title": "MemGraphRAG: Memory-based Multi-Agent System for Graph Retrieval-Augmented Generation",
"abstract": "Retrieval-Augmented Generation (RAG) has become an essential method for mitigating hallucinations in Large Language Models (LLMs) by leveraging external knowledge. Although effective for simple queries, traditional RAG struggles with large-scale, unstructured corpora where information is highly fragmented. Graph-based RAG (GraphRAG) incorporates knowledge graphs to capture structural relationships, enabling more comprehensive retrieval for complex reasoning. However, existing GraphRAG methods rely on isolated, fragment-level extraction for graph construction, lacking a global perspective on the whole corpus. As a result, these methods frequently lead to thematically inconsistent, logically conflicting, and structurally fragmented graphs that degrade retrieval performance. In this paper, we propose MemGraphRAG, a novel framework that introduces a memory-based multi-agent system to ensure high-quality graph construction. Specifically, MemGraphRAG employs a collaborative society of agents supported by shared memory, which provides a unified global context throughout the extraction process. This mechanism allows agents to dynamically resolve logical conflicts and maintain structural connectivity throughout the corpus. Furthermore, we propose a memory-aware hierarchical retrieval algorithm tailored for the constructed graph. Extensive experiments on multiple benchmarks demonstrate that MemGraphRAG outperforms the state-of-the-art baseline models with comparable efficiency. Our code is available at https://github.com/XMUDeepLIT/MemGraphRAG.",
"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"
],
"url": "https://arxiv.org/abs/2606.00610",
"pdf_url": "https://arxiv.org/pdf/2606.00610",
"path": "papers/items/2026-2606-00610-memgraphrag-memory-based-multi-agent-system-for-graph-retrieval-augmented-genera.md",
"local_title": "\"MemGraphRAG: Memory-based Multi-Agent System for Graph Retrieval-Augmented Generation\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"multi-agent",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "rag-agent",
"collection_score": "20"
},
{
"arxiv_id": "2606.00619",
"title": "MemPro: Agentic Memory Systems as Evolvable Programs",
"abstract": "Long-horizon autonomous agents require memory systems to retain historical information, track evolving states, and reuse relevant knowledge beyond finite context windows. Existing agentic memory systems typically follow a memory construction-retrieval (MCR) pipeline, but often adapt mainly the memory bank while keeping the surrounding pipeline fixed after deployment. This fixed-pipeline design struggles to handle heterogeneous task-specific failure modes and can become misaligned with memory banks that evolve in scale and structure over time. To address these limitations, we propose MemPro, a system-level evolution framework that treats the entire MCR pipeline as an evolvable program rather than adapting only the memory bank or prompt text. MemPro maintains a version tree of runnable memory-system implementations, where an Evolving Agent iteratively selects promising versions, diagnoses recurring failures, and creates improved child versions through failure-mode-guided edit-debug refinement. Experiments on LongMemEval, LoCoMo, HotpotQA, and NarrativeQA show that MemPro consistently outperforms strong static and prompt-level evolving baselines within a few iterations, continues to improve with evolution, and achieves a favorable performance-cost trade-off. Code is available at https://github.com/wanghai673/MemPro.",
"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"
],
"url": "https://arxiv.org/abs/2606.00619",
"pdf_url": "https://arxiv.org/pdf/2606.00619",
"path": "papers/items/2026-2606-00619-mempro-agentic-memory-systems-as-evolvable-programs.md",
"local_title": "\"MemPro: Agentic Memory Systems as Evolvable Programs\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"planning",
"rag"
],
"collection_queries": "agent-memory",
"collection_score": "16"
},
{
"arxiv_id": "2606.00756",
"title": "CoMIC: Collaborative Memory and Insights Circulation for Long-Horizon LLM Agents in Cloud-Edge Systems",
"abstract": "Deploying lightweight Large Language Model (LLM) agents on edge servers can reduce latency and move agentic services closer to users, but resource-constrained edge models often struggle with long-horizon tasks that require persistent memory, subgoal tracking, and reflection. Fine-tuning edge models after deployment is costly and difficult to scale across heterogeneous nodes, while purely local memory leaves agents with isolated experience and growing prompt context. We propose \\textsc{CoMIC}, a parameter-update-free cloud-edge framework for Collaborative Memory and Insights Circulation. \\textsc{CoMIC} follows a \\textit{Centralized Reflection, Decentralized Execution} design: edge agents execute locally using subgoal-oriented hierarchical memory and selective re-expansion of relevant histories, while a cloud-side LLM critic asynchronously evaluates completed trajectories, filters reusable experience, and aggregates cross-agent guidance keyed by semantic subgoal identifiers. Across five long-horizon agent tasks spanning symbolic planning and text interaction, \\textsc{CoMIC} improves progress rate and action grounding for weak edge agents and yields task-dependent success-rate gains without updating model parameters.",
"published": "2026-05-30",
"updated": "2026-05-30",
"authors": [
"Yannan Wang",
"Longli Yang",
"Zhen Liu",
"Abhishek Kumar",
"Carsten Maple"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2606.00756",
"pdf_url": "https://arxiv.org/pdf/2606.00756",
"path": "papers/items/2026-2606-00756-comic-collaborative-memory-and-insights-circulation-for-long-horizon-llm-agents-.md",
"local_title": "\"CoMIC: Collaborative Memory and Insights Circulation for Long-Horizon LLM Agents in Cloud-Edge Systems\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"planning",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "planning-agent",
"collection_score": "21"
},
{
"arxiv_id": "2606.01138",
"title": "memorywire: A Vendor-Neutral Wire Format for Agent Memory Operations",
"abstract": "Agent-memory frameworks -- mem0, Letta/MemGPT, Cognee, Zep/Graphiti, MemoryOS, MemTensor -- each ship their own SDK, storage layout, and operational vocabulary. There is no shared wire format: every integration is bespoke, every migration rebuilds memory from scratch, and no framework ships a governance surface that lets a human review writes before they enter long-term storage. We present memorywire, a JSON-Schema 2020-12 wire format for five memory operations (remember, recall, forget, merge, expire) over four memory types (semantic, episodic, procedural, emotional), with a MemoryStore interface, a fan-out router, and an optional HITL governance channel. We describe an open-source reference implementation with five backend adapters (sqlite-vec, mem0, Letta, Cognee, pgvector); a microbenchmark on a 100-fact / 50-query labelled corpus (42 with non-empty gold ids + 8 no-match probes) achieving recall@5 = 1.000 on the 42 gold-id queries with ingest p50 = 37.8 ms and recall p50 = 40.6 ms; an adversarial-fusion experiment showing Reciprocal Rank Fusion holds recall@5 = 1.000 across a 1-of-N rank-0 injection sweep (K in {0, 5, ..., 50}) where max fusion collapses to 0.500 with 80% leak at K >= 5; and a 16-scenario cross-adapter conformance suite passing 68 of 80 cells with zero failures. The contribution is not a new algorithm; it is a packaging of established components (RRF, FSMs, STM/LTM consolidation, diff-and-approve workflows) into a venue-neutral protocol with an empirically validated reference, positioned to compose with the Model Context Protocol rather than compete with it.",
"published": "2026-05-31",
"updated": "2026-06-03",
"authors": [
"Thamilvendhan Munirathinam"
],
"categories": [
"cs.CR",
"cs.AI",
"cs.DC"
],
"url": "https://arxiv.org/abs/2606.01138",
"pdf_url": "https://arxiv.org/pdf/2606.01138",
"path": "papers/items/2026-2606-01138-memorywire-a-vendor-neutral-wire-format-for-agent-memory-operations.md",
"local_title": "\"memorywire: A Vendor-Neutral Wire Format for Agent Memory Operations\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"workflow-agent"
],
"collection_queries": "agent-memory",
"collection_score": "13"
},
{
"arxiv_id": "2606.01199",
"title": "Can LLM Agents Sustain Long-Horizon Organizational Dynamics?",
"abstract": "Large language agents are increasingly used for social simulation, yet it remains unclear whether they can sustain coherent behavior in structured organizations, where goals must propagate through hierarchy, tasks depend on prior execution, and artifacts accumulate over long horizons. We formulate long-horizon organizational simulation as a memory-centered coordination problem and introduce TaskWeave, a hierarchical agentic framework that maintains planning states through a Formulate-Partition-Diagnose-Align cycle and grounds execution through dependency-aware trace memory. We evaluate TaskWeave in a year-long IT company simulation and compare it with other multi-agent frameworks on organizational coherence, execution grounding, and downstream enterprise NLP utility. Experiments show that TaskWeave supports coherent and long-horizon organizational dynamics while producing grounded artifacts and adapting to external environments. These findings suggest that structured simulation memory is a key mechanism for building reliable LLM-based organizational simulators.",
"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"
],
"url": "https://arxiv.org/abs/2606.01199",
"pdf_url": "https://arxiv.org/pdf/2606.01199",
"path": "papers/items/2026-2606-01199-can-llm-agents-sustain-long-horizon-organizational-dynamics.md",
"local_title": "Can LLM Agents Sustain Long-Horizon Organizational Dynamics?",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"multi-agent",
"planning",
"rag",
"workflow-agent",
"world-model"
],
"collection_queries": "language-agent, planning-agent",
"collection_score": "22"
},
{
"arxiv_id": "2606.01385",
"title": "Bridging Requirements and Architecture: Multi-Agent Orchestration with External Knowledge and Hierarchical Memory",
"abstract": "Software architecture design is a critical yet inherently complex and knowledge-intensive phase that requires balancing competing quality attributes and adapting to evolving requirements. Traditionally, this process has been time-consuming, labor-intensive, and heavily reliant on architects, often resulting in limited exploration of alternative architectural decompositions and styles, especially under the pressures of agile development. While LLM-based agents have shown promising performance across various software engineering tasks, their application to architecture design remains relatively scarce and requires systematic exploration. To address these challenges, we proposed MAAD (Multi-Agent Architecture Design), a knowledge-driven framework that orchestrates four specialized agents (i.e., Analyst, Modeler, Designer and Evaluator) to autonomously and collaboratively transform requirements specifications into comprehensive, multi-view architectural blueprints with quality attribute assessments. MAAD incorporates RAG to inject recognized architectural standards and patterns into the workflow and leverages a hierarchical memory mechanism that captures design history for iterative refinement. We evaluated MAAD through comparative experiments against MetaGPT, using quantitative architecture-level metrics across 10 case studies and qualitative feedback from industry architects on 10 real-world specifications. Results show that MAAD generates more complete, modular, and traceable architectures than the baseline, and its dedicated Evaluator agent autonomously produces structured quality evaluation reports that significantly reduce manual validation efforts. Furthermore, we found that the quality of the generated architecture heavily depends on the underlying LLM's reasoning capacity, with GPT-5.2 and Qwen3.5 outperforming other models across most evaluation settings.",
"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"
],
"url": "https://arxiv.org/abs/2606.01385",
"pdf_url": "https://arxiv.org/pdf/2606.01385",
"path": "papers/items/2026-2606-01385-bridging-requirements-and-architecture-multi-agent-orchestration-with-external-k.md",
"local_title": "\"Bridging Requirements and Architecture: Multi-Agent Orchestration with External Knowledge and Hierarchical Memory\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"coding-agent",
"memory",
"multi-agent",
"rag",
"reasoning",
"workflow-agent"
],
"collection_queries": "rag-agent",
"collection_score": "21"
},
{
"arxiv_id": "2606.01416",
"title": "Self-Healing Agentic Orchestrators for Reliable Tool-Augmented Large Language Model Systems",
"abstract": "Tool-augmented large language model (LLM) agents rely on orchestration layers that coordinate planning, retrieval, tool invocation, validation, memory, and recovery. In these systems, failures arise not only from model errors, but also from orchestration-level issues such as tool timeouts, malformed arguments, stale context, contradictory evidence, retry loops, and unverified intermediate outputs. This paper presents a self-healing agentic orchestrator that treats reliability as a bounded runtime control problem. The orchestrator maps observable failure signals to inferred failure classes, selects targeted recovery actions under explicit budgets, verifies recovered trajectories, and records observability traces. We evaluate the approach on a 100-task controlled fault-injection benchmark against static workflow, retry-only, ReAct-style, and full-replanning baselines. Self-healing achieves 98.8\\% task success, compared with 94.5\\% for retry-only and 93.8\\% for full replanning. A matched recovery-budget sweep shows that self-healing outperforms retry-only and full replanning at every tested budget, with the largest gap under a single recovery attempt: 94.0\\% versus 85.3\\% and 88.2\\%, respectively. Under a controlled semantic silent-failure setting, verifier-guided self-healing reduces silent failures to 0.0\\%, while non-verifying baselines return wrong-but-plausible outputs more often. A compact model-in-the-loop validation shows that the same recovery mechanism can operate when a live tool-calling model performs tool selection, argument generation, and answer synthesis over local fault-injected tools. These results provide controlled evidence that failure-aware, budgeted, and verification-guided orchestration improves reliability and diagnosability in tool-augmented LLM systems.",
"published": "2026-05-31",
"updated": "2026-05-31",
"authors": [
"Rahul Suresh Babu",
"Adarsh Agrawal"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2606.01416",
"pdf_url": "https://arxiv.org/pdf/2606.01416",
"path": "papers/items/2026-2606-01416-self-healing-agentic-orchestrators-for-reliable-tool-augmented-large-language-mo.md",
"local_title": "Self-Healing Agentic Orchestrators for Reliable Tool-Augmented Large Language Model Systems",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"planning",
"rag",
"reasoning",
"tool-use",
"workflow-agent"
],
"collection_queries": "planning-agent",
"collection_score": "20"
},
{
"arxiv_id": "2606.01528",
"title": "Joint Agent Memory and Exploration Learning via Novelty Signals",
"abstract": "In open-ended environments, exploration is fundamental for autonomous agents, yet current language model agents struggle with this. Effective exploration requires memory, but retaining raw interaction histories is computationally expensive over long trajectories. While latent memory offers a solution to compress interaction histories, its training lacks reliable supervisory signals. We introduce \\textbf{J}oint \\textbf{A}gent \\textbf{M}emory and \\textbf{E}xploration \\textbf{L}earning (\\textbf{JAMEL}), a framework that trains agentic memory and exploration policy together through novelty-driven interaction. We observe that memory and exploration form a mutually dependent loop: sustained exploration requires memory to distinguish exhausted behaviors from unseen ones, while novelty-seeking interaction provides the supervision needed to make memory useful for future exploration. By utilizing deterministic and persistent novelty signals such as code coverage in the GUI domain, we provide natural, annotation-free supervision for the memory module. Empirical evaluations demonstrate that \\ours successfully generalizes to unseen environments. Its exploration capability outperforms open-weight baselines and rivals the exploration depth of a closed-source model while reducing token consumption. Our code and model are open-sourced at https://github.com/MobileLLM/JAMEL.",
"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"
],
"url": "https://arxiv.org/abs/2606.01528",
"pdf_url": "https://arxiv.org/pdf/2606.01528",
"path": "papers/items/2026-2606-01528-joint-agent-memory-and-exploration-learning-via-novelty-signals.md",
"local_title": "Joint Agent Memory and Exploration Learning via Novelty Signals",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"rag",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "17"
},
{
"arxiv_id": "2606.02302",
"title": "SeClaw: Spec-Driven Security Task Synthesis for Evaluating Autonomous Agents",
"abstract": "Autonomous LLM agents increasingly operate in stateful environments where they access tools, files, memory, and external services. While such capabilities enable complex real-world workflows, they also introduce security risks that are difficult to capture with existing evaluations. Current agent security benchmarks often rely on manually curated tasks, provide limited coverage of emerging threats, and focus primarily on final outcomes rather than the execution processes that lead to unsafe behavior. We introduce SeClaw, a framework that combines specification-driven security task synthesis with execution-based security evaluation for Autonomous agents. Spec-driven security task synthesis enables scalable and controllable construction of security tasks from structured risk specifications, while SeClaw docker provides a standardized testbed for evaluating agent behavior under diverse safety-risk scenarios. The benchmark covers risks arising from resources, user tasks, environments, and intrinsic agent behaviors, and supports trajectory-aware assessment of unsafe actions beyond final responses. By bridging systematic task synthesis and reproducible security evaluation, SeClaw provides a practical foundation for measuring, diagnosing, and comparing security failures in autonomous LLM agents. The code is available at https://github.com/seclaw-eval/seclaw-eval.",
"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"
],
"url": "https://arxiv.org/abs/2606.02302",
"pdf_url": "https://arxiv.org/pdf/2606.02302",
"path": "papers/items/2026-2606-02302-seclaw-spec-driven-security-task-synthesis-for-evaluating-autonomous-agents.md",
"local_title": "\"SeClaw: Spec-Driven Security Task Synthesis for Evaluating Autonomous Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"rag",
"tool-use",
"workflow-agent"
],
"collection_queries": "agent-safety, autonomous-agent-llm",
"collection_score": "18"
},
{
"arxiv_id": "2606.02461",
"title": "AgentCL: Toward Rigorous Evaluation of Continual Learning in Language Agents",
"abstract": "Language agents spend substantial inference time solving individual tasks, yet the experience acquired in one episode is often underutilized in future episodes. Continual learning expects an agent to accumulate reusable experience across a stream of tasks, improve over time, and avoid interference from irrelevant experiences. Unfortunately, existing benchmarks struggle to evaluate continual learning in language agents rigorously. Most efforts focus on retrieval and reasoning over long-context conversations or documents, while recent lifelong-adaptation benchmarks often rely on naive task streams with limited analysis of cross-task relationships, making it difficult to understand what an agent learns and reuses over time. This paper presents an evaluation framework AgentCL for continual learning in agents, centered on controlled task streams and metrics for transfer gains. AgentCL constructs compositional streams where earlier sub-solutions, evidence, or workflows are intentionally reusable in later tasks, and contrasts them with naive streams where such reusability is not guaranteed. We use the benchmark to evaluate non-parametric memory designs for continual learning. To diagnose how memory design choices affect continual learning, we develop MemProbe, a probing method that stores interactions, insights, and skills, while filtering unreliable experiences during consolidation. Empirical analysis across coding, deep research, and language understanding/reasoning tasks shows that naive streams offer limited ability to distinguish memory designs, whereas controlled streams more clearly distinguish their plasticity. Meanwhile, naive and held-out settings often yield limited gains and can expose memory-induced degradation. These results highlight the need for stronger memory designs that balance plasticity and stable reuse.",
"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"
],
"url": "https://arxiv.org/abs/2606.02461",
"pdf_url": "https://arxiv.org/pdf/2606.02461",
"path": "papers/items/2026-2606-02461-agentcl-toward-rigorous-evaluation-of-continual-learning-in-language-agents.md",
"local_title": "\"AgentCL: Toward Rigorous Evaluation of Continual Learning in Language Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"coding-agent",
"computer-use",
"memory",
"rag",
"reasoning",
"tool-use",
"workflow-agent"
],
"collection_queries": "language-agent",
"collection_score": "21"
},
{
"arxiv_id": "2606.02497",
"title": "Bridging the Last Mile of Time Series Forecasting with LLM Agents",
"abstract": "Time series forecasting has advanced rapidly, especially with the emergence of foundation models that show strong zero-shot performance on numerical extrapolation. However, in real-world forecasting settings, a statistically plausible baseline is rarely the final forecast used in practice. Before a forecast becomes decision-ready, it often needs to be revised using weakly structured business context such as holiday effects, campaign plans, external events, historical analogs, and expert feedback. This practical stage remains underexplored in the forecasting literature. In this paper, we formulate this stage as the \\textbf{last-mile forecasting} problem and present an LLM-agent framework that sits on top of a forecasting backbone. Our system maintains a unified forecast workspace, invokes tools to retrieve contextual evidence, and converts reasoning trajectories into explicit forecast revision actions under structural safety constraints. It also supports long-horizon forecasting through map-reduce-style decomposition and post-hoc reflection through a memory bank. The resulting system is designed to be controllable and auditable. Through real-world case studies, we show how LLM agents can bridge the gap between statistical prediction and business-ready forecasting.",
"published": "2026-06-01",
"updated": "2026-06-01",
"authors": [
"Yuhua Liao",
"Zetian Wang",
"Qiangqiang Nie",
"Zhenhua Zhang"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2606.02497",
"pdf_url": "https://arxiv.org/pdf/2606.02497",
"path": "papers/items/2026-2606-02497-bridging-the-last-mile-of-time-series-forecasting-with-llm-agents.md",
"local_title": "Bridging the Last Mile of Time Series Forecasting with LLM Agents",
"local_status": "queued",
"local_topics": [
"agent-safety",
"memory",
"planning",
"reasoning",
"tool-use"
],
"collection_queries": "planning-agent",
"collection_score": "16"
},
{
"arxiv_id": "2606.02812",
"title": "Traj-Evolve: A Self-Evolving Multi-Agent System for Patient Trajectory Modeling in Lung Cancer Early Detection",
"abstract": "Modeling patient trajectories from longitudinal electronic health records (EHRs) requires reasoning over sparse, noisy, and long-context multimodal sequences. Existing LLM-based multi-agent systems address context length but process patients in isolation, failing to mirror how clinicians leverage accumulated experience from similar prior cases. We present Traj-Evolve, a self-evolving multi-agent system with two complementary evolving mechanisms. First, an Experience Pool (ExPool) acts as a non-parametric memory, indexing rejection-sampled reasoning traces to retrieve similar patients as few-shot contexts. Second, multi-agent reinforcement learning (MARL) via reward-ranked fine-tuning parametrically optimizes inter-agent and agent-memory collaboration. A leave-one-out cross-retrieval strategy unifies the two, aligning training- and inference-time behavior under retrieval augmentation. On a lung cancer prediction task utilizing up to five years of multimodal EHRs, Traj-Evolve outperforms 9 strong baselines on the overall population and a challenging never-smoker population. Analysis of the evolving dynamics highlights three key findings: (1) expanding the ExPool shifts optimal retrieval from diverse to specific samples; (2) under MARL, the manager agent's prediction loss converges quickly while the worker agents' temporal reasoning continues to benefit from more verified patients; and (3) the two mechanisms are complementary on the predicted risk, where ExPool improves specificity while MARL improves sensitivity.",
"published": "2026-06-01",
"updated": "2026-06-01",
"authors": [
"Sihang Zeng",
"Matthew Thompson",
"Ruth Etzioni",
"Meliha Yetisgen"
],
"categories": [
"cs.AI",
"cs.CL"
],
"url": "https://arxiv.org/abs/2606.02812",
"pdf_url": "https://arxiv.org/pdf/2606.02812",
"path": "papers/items/2026-2606-02812-traj-evolve-a-self-evolving-multi-agent-system-for-patient-trajectory-modeling-i.md",
"local_title": "\"Traj-Evolve: A Self-Evolving Multi-Agent System for Patient Trajectory Modeling in Lung Cancer Early Detection\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"multi-agent",
"rag",
"reasoning"
],
"collection_queries": "agent-memory",
"collection_score": "17"
},
{
"arxiv_id": "2606.03197",
"title": "MemTrain: Self-Supervised Context Memory Training",
"abstract": "Memory is an indispensable capability for long-horizon LLM agents, enabling them to preserve and utilize information accumulated across extended interactions. Existing memory-agent approaches are typically trained end-to-end with reinforcement learning on downstream tasks. However, collecting high-quality annotated problems for memory-intensive scenarios is costly, and the resulting training data often lack sufficient diversity to cover general memory behaviors. In this work, we propose MemTrain, a self-supervised training framework for generally enhancing the context-memory capability of LLM agents for more effective downstream post-training. MemTrain introduces two coupled proxy tasks over unlabeled Wikipedia corpora: (1) an end-to-end masked reconstruction objective, which requires the model to recover masked entities after multiple rounds of memory updates, thereby encouraging memory maintenance from the final outcome perspective; and (2) an intermediate memory recall objective, which requires the model to reconstruct masked historical information using intermediate memory states, encouraging faithful compression and memory completeness throughout the interaction process. The two objectives are jointly optimized using GRPO. Extensive experiments on long-text QA and search-based QA benchmarks demonstrate that MemTrain consistently improves downstream memory-intensive reasoning performance across different models, achieving gains of up to 17.67 points over direct task-specific post-training.",
"published": "2026-06-02",
"updated": "2026-06-02",
"authors": [
"Ziheng Li",
"Xingrun Xing",
"Haoqing Wang",
"Zhi-Hong Deng",
"Yehui Tang"
],
"categories": [
"cs.CL"
],
"url": "https://arxiv.org/abs/2606.03197",
"pdf_url": "https://arxiv.org/pdf/2606.03197",
"path": "papers/items/2026-2606-03197-memtrain-self-supervised-context-memory-training.md",
"local_title": "\"MemTrain: Self-Supervised Context Memory Training\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"planning",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "16"
},
{
"arxiv_id": "2606.03329",
"title": "InfoMem: Training Long-Context Memory Agents with Answer-Conditioned Information Gain",
"abstract": "Long-context tasks require LLMs to identify and preserve answer-relevant information from large contexts. Chunk-wise memory agents address this issue by sequentially reading document chunks, updating a compact memory, and generating the final answer from the accumulated memory. However, existing RL-based chunk-wise agents either rely on sparse final-answer rewards or use lexical intermediate rewards for memory and retrieval actions. These signals supervise task success or local overlap, but do not directly evaluate whether the final memory supports the ground-truth answer. We propose InfoMem, a reward mechanism for training chunk-wise memory agents that evaluates final-memory utility using answer-conditioned information. InfoMem measures how much the final memory increases the model's per-token log-likelihood of the ground-truth answer. To stabilize RL optimization, InfoMem applies this signal only to successful trajectories and normalizes it before reward composition. Under the same GRPO framework and training budget, InfoMem improves long-context memory-agent performance over comparable memory-agent RL baselines. Analyses show that effective final-memory rewards should operate on successful trajectories, be normalized before reward composition, and be conditioned on the answer rather than the query. Our code is available at https://github.com/GenSouKa1/InfoMem.",
"published": "2026-06-02",
"updated": "2026-06-02",
"authors": [
"Tiancheng Han",
"Yong Li",
"Wuzhou Yu",
"Qiaosheng Zhang",
"Wenqi Shao"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2606.03329",
"pdf_url": "https://arxiv.org/pdf/2606.03329",
"path": "papers/items/2026-2606-03329-infomem-training-long-context-memory-agents-with-answer-conditioned-information-.md",
"local_title": "\"InfoMem: Training Long-Context Memory Agents with Answer-Conditioned Information Gain\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "14"
},
{
"arxiv_id": "2606.03374",
"title": "eMEM: A Hybrid Spatio-Temporal Memory System For Embodied Agents",
"abstract": "We present eMEM (Embodied Memory), a hybrid graph-based memory system for embodied agents operating in physical environments. Current agent memory architectures, such as Generative Agents, MemGPT, and A-MEM, treat memory as text streams or knowledge graphs, but embodied agents require memory that is simultaneously searchable by meaning, space, and time. eMEM fills this gap with a multi-index architecture (SQLITE for structured storage, hnswlib for approximate nearest neighbour semantic search, and an R-tree for spatial queries) unified behind a single graph model. A tiered consolidation pipeline transforms raw perceptual observations into compressed summaries, mirroring hippocampal-neocortical consolidation in biological systems. Ten agent-facing recall tools expose memory retrieval primitives, including concept-to-location resolution and cross layer recall, as first-class operations for LLM tool calling. The system is fully embedded and runs in-process alongside the agent. In addition we introduce eMEM-Bench v1, a benchmark we construct over ProcTHOR-10K scenes for embodied memory evaluation. The benchmark is organised explicitly around eight cognitive-psychology paradigms (DRM lures, pattern separation, pattern completion, source monitoring, context-dependent retrieval, long-horizon interference, serial position, and a foil augmented retention curve), each chosen so that the result is interpretable against the broader memory-systems literature in humans and prior agent-memory systems; a level of diagnostic that surface-task benchmarks like LoCoMo or OpenEQA cannot provide. eMEM scores 80.8 weighted mean over 988 probes, with a flat retention curve at ceiling from 1 h to 1 yr of simulated delay on room-unique items. We show that a pure RAG baseline (the flat_rag ablation) loses 30 pt on context dependent retrieval and 29 pt on DRM lure rejection, isolating the contribution of multi-layer storage and consolidation respectively. We release both the system and the benchmark code.",
"published": "2026-06-02",
"updated": "2026-06-22",
"authors": [
"A. Haroon Rasheed",
"Maria Kabtoul"
],
"categories": [
"cs.RO"
],
"url": "https://arxiv.org/abs/2606.03374",
"pdf_url": "https://arxiv.org/pdf/2606.03374",
"path": "papers/items/2026-2606-03374-emem-a-hybrid-spatio-temporal-memory-system-for-embodied-agents.md",
"local_title": "\"eMEM: A Hybrid Spatio-Temporal Memory System For Embodied Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"embodied-agent",
"memory",
"planning",
"rag",
"tool-use"
],
"collection_queries": "agent-memory, rag-agent",
"collection_score": "19"
},
{
"arxiv_id": "2606.03895",
"title": "Agent libOS: A Runtime Substrate for Capability-Controlled Self-Evolving LLM Agents",
"abstract": "Large language model (LLM) agents are becoming long-running software actors rather than fixed tool users. They accumulate memory, activate skills, synthesize tools, fork children, attach remote resources, and commit checkpoints into reusable execution images. These mechanisms improve adaptability, but also create a systems-security failure mode: if exposing an action also grants the authority needed to perform it, self-evolution becomes a permission-escalation path. This paper presents Agent libOS, an agent-native library-OS substrate for capability-controlled self-evolving agents. Its central invariant is that model-visible affordances may evolve while resource authority changes only through explicit, audited runtime primitives. Agent libOS represents an agent as an AgentProcess with process identity, process-local Object Memory, message queues, a tool table, loaded Skills, process-local Deno/TypeScript JIT tools, child processes, budgets, checkpoints, and explicit capabilities. AgentImage objects define boot-time prompt and tool-table state; Skills and JIT tools extend the action surface; checkpoint-derived images make internal state reusable. None of these mechanisms grants filesystem, shell, human, memory, process, checkpoint, image, JSON-RPC, MCP, or PTY authority by itself. The prototype implements process-local namespaces, persistent runtime state, LLM-call observability, human approval queues, budgets, syscall-mediated JIT tools, trusted Runtime Modules, Object-bound PTY sessions, checkpoint restore/fork/commit, JSON-RPC and MCP providers, and a deterministic runtime-safety benchmark. On 27 versioned deterministic tasks, it completed the task plans while preventing all modeled unauthorized side effects, with a 7.0% conservative false-denial rate. Simple wrapper and sandbox baselines preserved task completion but failed most safety checks.",
"published": "2026-06-02",
"updated": "2026-06-29",
"authors": [
"Yingqi Zhang"
],
"categories": [
"cs.OS",
"cs.AI",
"cs.CR"
],
"url": "https://arxiv.org/abs/2606.03895",
"pdf_url": "https://arxiv.org/pdf/2606.03895",
"path": "papers/items/2026-2606-03895-agent-libos-a-runtime-substrate-for-capability-controlled-self-evolving-llm-agen.md",
"local_title": "\"Agent libOS: A Runtime Substrate for Capability-Controlled Self-Evolving LLM Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"planning",
"tool-use"
],
"collection_queries": "planning-agent",
"collection_score": "18"
},
{
"arxiv_id": "2606.04120",
"title": "SaliMory: Orchestrating Cognitive Memory for Conversational Agents",
"abstract": "Conversational agents that serve as lifelong companions must maintain persistent memory across all interactions. However, simply expanding context windows with raw retrieval degrades reasoning quality, while training memory agents via standard reinforcement learning creates a severe credit assignment bottleneck in a multi-stage pipeline. To solve this, we introduce SALIMORY, a framework that trains a single language model to manage a cognitively-structured memory-spanning user facts, preferences, and working memory. By introducing a hierarchical stage-wise process reward and reward-decomposed contrastive refinement, SALIMORY provides isolated supervision for distinct memory operations (selective filtering, consolidation, and cue-driven recall) end-to-end. SALIMORY cuts memory-attributed failures by one-third, outperforms the state-of-the-art by over 10% in end-to-end accuracy, and more than doubles the Good Personalization rate.",
"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"
],
"url": "https://arxiv.org/abs/2606.04120",
"pdf_url": "https://arxiv.org/pdf/2606.04120",
"path": "papers/items/2026-2606-04120-salimory-orchestrating-cognitive-memory-for-conversational-agents.md",
"local_title": "\"SaliMory: Orchestrating Cognitive Memory for Conversational Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "15"
},
{
"arxiv_id": "2606.04315",
"title": "Exploring Cross-Scenario Generality of Agentic Memory Systems: Diagnostics and a Strong Baseline",
"abstract": "LLM agents accumulate histories that outgrow their context windows, motivating a growing literature on memory systems. Yet most existing designs are tuned to a single scenario (multi-session chat or a single trajectory format), and there is little evidence that they generalize across the heterogeneous trajectories agents encounter in deployment. We revisit eight memory systems plus an agentic harness for search problems, on five scenarios: single-turn QA, multi-session chat, agentic-trajectory QA, memory stress tests, and long-horizon agentic tasks. The harness, which self-manages flat text-file storage via tool calls, achieves the best cross-task ranking, suggesting that memory performance hinges on giving the agent active control over storage and retrieval rather than on a passive store behind a fixed pipeline. We instantiate this insight in AutoMEM, an agentic memory harness with a self-managed tool interface that achieves the best cross-scenario generality among the systems we evaluate.",
"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"
],
"url": "https://arxiv.org/abs/2606.04315",
"pdf_url": "https://arxiv.org/pdf/2606.04315",
"path": "papers/items/2026-2606-04315-exploring-cross-scenario-generality-of-agentic-memory-systems-diagnostics-and-a-.md",
"local_title": "\"Exploring Cross-Scenario Generality of Agentic Memory Systems: Diagnostics and a Strong Baseline\"",
"local_status": "skimmed",
"local_topics": [
"agent-evaluation",
"memory",
"planning",
"rag",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "19"
},
{
"arxiv_id": "2606.04555",
"title": "Temporal Order Matters for Agentic Memory: Segment Trees for Long-Horizon Agents",
"abstract": "Long-horizon conversational agents need to interact with users through evolving events, tasks, and goals. Such histories are naturally temporal, yet many existing memory systems organize information primarily by topical similarity and may ignore the order in which events occur. We introduce Segment Tree Memory, or SegTreeMem, a memory architecture that represents conversation history as a temporally ordered Segment Tree over utterances. SegTreeMem incrementally inserts new utterances through an online rightmost-frontier update rule, preserving chronological order while forming hierarchical memory segments. For retrieval, SegTreeMem propagates relevance scores through the tree to combine local semantic matching with hierarchical temporal context. Across three long-horizon memory benchmarks and two LLM backbones, SegTreeMem improves answer quality over flat retrieval, graph-structured memory, and tree-structured memory baselines. Additional temporal-order permutation analysis shows that the performance gain depends on preserving temporal order during memory construction, supporting the claim that temporal order is a key structure for agentic memory.",
"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"
],
"url": "https://arxiv.org/abs/2606.04555",
"pdf_url": "https://arxiv.org/pdf/2606.04555",
"path": "papers/items/2026-2606-04555-temporal-order-matters-for-agentic-memory-segment-trees-for-long-horizon-agents.md",
"local_title": "\"Temporal Order Matters for Agentic Memory: Segment Trees for Long-Horizon Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"planning",
"rag"
],
"collection_queries": "agent-memory",
"collection_score": "18"
},
{
"arxiv_id": "2606.04628",
"title": "RAMPART: Registry-based Agentic Memory with Priority-Aware Runtime Transformation",
"abstract": "RAMPART is a compile-time memory model and pure in-RAM block registry for LLM-based agents. Context assembly is a programmable runtime operation where content is compiled from a structured registry under explicit policy for ordering, inclusion, and eviction. Five composable primitives (promote, gate, write, evict, rollback) act on named addressable blocks before compilation at zero prompt-token cost. Provenance tags and non-evictable authorship flags implement a permissioned memory model with block-level ownership. Controlled probes with Qwen3-8B Q4 show that compile-time placement and the structural relationship between blocks and the task query affect task success, with the cliff falling at roughly the seventh block position when the task follows the registry and the twelfth when it precedes. Grouping the critical block with content-adjacent neighbours and promoting the group as a unit lifts task success by tens of percentage points at positions where single-block placement fails. Cross-model replication on Qwen2.5-7B, Llama-3.1-8B, Mistral-7B-v0.3, and Qwen3-14B shows the content-priming effect appears at the same absolute positions across families, with magnitude varying with model strength. Block grouping raises Mistral's mean pass rate roughly fivefold at the hardest registry size, and a smaller model with the intervention can outperform a larger model without it in the mid-registry zone. Relevance gating reduces prompt cost by 67.8\\% while recovering 83% of the promoted-condition success rate. Schema eviction produces 0% invocations against 100% with the schema present, a property policy-based approaches cannot guarantee by construction. Shared-registry coordination reduces inter-agent communication to a method call at zero coordination token cost.",
"published": "2026-06-03",
"updated": "2026-06-03",
"authors": [
"Nikodem Tomczak"
],
"categories": [
"cs.CL",
"cs.MA"
],
"url": "https://arxiv.org/abs/2606.04628",
"pdf_url": "https://arxiv.org/pdf/2606.04628",
"path": "papers/items/2026-2606-04628-rampart-registry-based-agentic-memory-with-priority-aware-runtime-transformation.md",
"local_title": "\"RAMPART: Registry-based Agentic Memory with Priority-Aware Runtime Transformation\"",
"local_status": "queued",
"local_topics": [
"memory"
],
"collection_queries": "agent-memory",
"collection_score": "14"
},
{
"arxiv_id": "2606.04780",
"title": "PersonaTree: Structured Lifecycle Memory for Person Understanding in LLM Agents",
"abstract": "Persistent LLM agents require memory representations that make the formation of person understanding explicit across long term interaction. Existing agent memory methods emphasize information retention and retrieval, yet give limited account of how accumulated interaction evidence is abstracted into person understanding. We view this process as schema formation, where situated evidence is abstracted into reusable patterns and stable person level claims. We introduce PersonaTree, a structured lifecycle memory framework that realizes this view as a three level persona tree with explicit support paths from evidence to claims. PersonaTree maintains the tree through conservative writing, confidence guided consolidation, and query conditioned path retrieval, returning only the evidence depth required by each query. Across six person understanding and persistent memory benchmarks with three answer backbones, PersonaTree ranks first in 12 of 18 compact scores and reaches the top two in 16 settings. Ablations show that hierarchy improves abstract person understanding on KnowMe, while support path retrieval improves RealPref alignment under a comparable context budget.",
"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"
],
"url": "https://arxiv.org/abs/2606.04780",
"pdf_url": "https://arxiv.org/pdf/2606.04780",
"path": "papers/items/2026-2606-04780-personatree-structured-lifecycle-memory-for-person-understanding-in-llm-agents.md",
"local_title": "\"PersonaTree: Structured Lifecycle Memory for Person Understanding in LLM Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"computer-use",
"memory",
"rag",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "20"
},
{
"arxiv_id": "2606.04990",
"title": "From Agent Traces to Trust: A Survey of Evidence Tracing and Execution Provenance in LLM Agents",
"abstract": "Large language model (LLM)-based agents are evolving from passive text generators into autonomous systems capable of planning, tool use, retrieval, memory access, environmental interaction, and multi-agent collaboration. These capabilities expand agent autonomy, but also make agent behavior harder to verify, debug, and audit. Final-answer accuracy alone cannot explain how an output was produced, which evidence supported each claim, whether tool calls were justified, how memory influenced later decisions, or where failures originated. This survey examines evidence tracing and execution provenance as foundations for process-level accountability in trustworthy LLM agents. We define execution provenance as the typed graph of an agent execution and evidence tracing as its projection onto evidence-support relations. This perspective connects retrieval grounding, claim support, tool-use safety, memory lineage, observability, debugging, audit, and recovery within a unified framework. We introduce a taxonomy covering trace sources, evidence and execution units, provenance relations, tracing granularity and timing, representation forms, and trust functions. We then review key methodological directions, including provenance representation, evidence attribution, tool-use provenance, runtime guardrails, provenance-bearing memory, observability, and failure diagnosis. Finally, we discuss benchmarks, datasets, metrics, and open challenges for building provenance-aware, auditable, and recoverable agent systems.",
"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"
],
"url": "https://arxiv.org/abs/2606.04990",
"pdf_url": "https://arxiv.org/pdf/2606.04990",
"path": "papers/items/2026-2606-04990-from-agent-traces-to-trust-a-survey-of-evidence-tracing-and-execution-provenance.md",
"local_title": "\"From Agent Traces to Trust: A Survey of Evidence Tracing and Execution Provenance in LLM Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"multi-agent",
"planning",
"rag",
"tool-use"
],
"collection_queries": "planning-agent",
"collection_score": "21"
},
{
"arxiv_id": "2606.28349",
"title": "HMARS: A Hierarchical Multi-Agent Memory System for Long-Context Reasoning",
"abstract": "Long-context reasoning requires models to access, retrieve, and integrate evidence scattered across documents, dialogues, and accumulated interaction histories. Standard retrieval-augmented generation reduces this problem to top-$K$ chunk retrieval, but such passive access can discard relevant evidence before reasoning begins, especially when relevance depends on broader context. We propose HMARS, a hierarchical multi-agent memory system that treats long contexts as managed memory rather than a flat retrieval corpus. Sub-agents maintain grounded access to bounded memory regions, mid-agents manage regional context and provide query-specific coordination, and a frontier model performs final reasoning over retrieved evidence pages. To evaluate this view, we construct two diagnostic benchmarks targeting evidence breadth and context-dependent relevance. Across long-document and multi-turn memory tasks, HMARS achieves the best overall performance against retrieval, reranking, full-context, graph-based, and agentic long-context baselines. Evidence coverage analysis further shows that its gains come from retrieving the required supporting evidence more completely, rather than merely changing the final answer prompt.",
"published": "2026-06-03",
"updated": "2026-06-03",
"authors": [
"Zeju Li",
"Ziyang Zheng",
"Yizhou Zhou",
"Qiang Xu"
],
"categories": [
"cs.IR",
"cs.AI"
],
"url": "https://arxiv.org/abs/2606.28349",
"pdf_url": "https://arxiv.org/pdf/2606.28349",
"path": "papers/items/2026-2606-28349-hmars-a-hierarchical-multi-agent-memory-system-for-long-context-reasoning.md",
"local_title": "\"HMARS: A Hierarchical Multi-Agent Memory System for Long-Context Reasoning\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"multi-agent",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "19"
},
{
"arxiv_id": "2606.05684",
"title": "AdaMEM: Test-Time Adaptive Memory for Language Agents",
"abstract": "A central challenge for language agents is utilizing past experience to adapt to dynamic test-time conditions. While recent work demonstrates the promise of agentic memory mechanisms, most systems restrict retrieval to episode initiation. Consequently, agents are forced to rely on static guidance that becomes increasingly misaligned as long-horizon tasks unfold. To address this rigidity, we propose the Adaptive Memory Agent (AdaMEM), a novel framework for agent test-time adaptation. Without updating model parameters online, AdaMEM adapts agent behavior via a hybrid memory architecture: it maintains a long-term trajectory memory of raw experiences collected offline while generating dynamic short-term strategy memory on-the-fly to guide decision-making. This mechanism enables the trade-off between token efficiency and adaptability across varying inference-time compute levels. Empirically, AdaMEM significantly outperforms static memory baselines, achieving relative gains of up to 13% on ALFWorld and 11% on WebShop, with consistent leading performance extending to agentic search on HotpotQA. To further enhance this adaptation, we develop STEP-MFT, a Step-wise Memory Fine-Tuning technique that trains the policy to synthesize high-quality strategies from retrieved experiences, yielding additional performance gains. Our work establishes a new scaling dimension for agentic memory, supporting continuous reasoning and self-evolution post-deployment in real-world environments. Our code is available at https://github.com/yunx-z/AdaMEM.",
"published": "2026-06-04",
"updated": "2026-06-04",
"authors": [
"Yunxiang Zhang",
"Yiheng Li",
"Ali Payani",
"Lu Wang"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2606.05684",
"pdf_url": "https://arxiv.org/pdf/2606.05684",
"path": "papers/items/2026-2606-05684-adamem-test-time-adaptive-memory-for-language-agents.md",
"local_title": "\"AdaMEM: Test-Time Adaptive Memory for Language Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"planning",
"rag",
"reasoning"
],
"collection_queries": "agent-memory, language-agent",
"collection_score": "19"
},
{
"arxiv_id": "2606.06054",
"title": "Beyond Similarity: Trustworthy Memory Search for Personal AI Agents",
"abstract": "Personal AI agents increasingly rely on long-term memory to provide persistent personalization across sessions. However, existing memory pipelines are largely driven by semantic similarity: memory data close to the current query is retrieved and injected into the model context. This creates a critical trustworthiness gap, since a semantically related memory may still be contextually inappropriate, leading to threats such as cross-domain leakage, sycophancy, tool-call drift, or memory-induced jailbreaks. In this paper, we study memory search as a trust boundary in personal AI agents. We evaluate representative agentic memory frameworks, including A-Mem, Mem0, and MemOS, together with OpenClaw, a real-world personal-agent environment with persistent state and tool-use capability. Our results show that long-term memory is not merely a utility layer, but a durable control channel that can reshape how agents interpret tasks and execute actions, leaving them highly susceptible to the aforementioned threats. To mitigate these vulnerabilities, we propose MemGate, a lightweight and deployable memory plug-in for trustworthy memory search, with only 9M parameters and a 35.1MB footprint. MemGate is inserted between the vector memory store and the backbone LLM, requiring no LLM modification, memory-database rewriting, or inference-time LLM judge. It applies a query-conditioned neural gate to candidate memory representations, turning raw similarity search into task-conditioned memory admission. Across multiple mainstream memory frameworks, real-world agent settings, and diverse LLM backbones, MemGate reduces memory-induced threats while preserving long-term memory utility.",
"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"
],
"url": "https://arxiv.org/abs/2606.06054",
"pdf_url": "https://arxiv.org/pdf/2606.06054",
"path": "papers/items/2026-2606-06054-beyond-similarity-trustworthy-memory-search-for-personal-ai-agents.md",
"local_title": "\"Beyond Similarity: Trustworthy Memory Search for Personal AI Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "17"
},
{
"arxiv_id": "2606.06090",
"title": "Beyond Semantic Organization: Memory as Execution State Management for Long-Horizon Agents",
"abstract": "LLM-based agents increasingly tackle long-horizon tasks with interdependent decisions, where each action reshapes future constraints and intermediate errors can cascade. Existing RAG and agent memory systems organize histories by semantic similarity, retrieving content-relevant entries at decision time. We argue that this design mismatches execution-state dependencies: it fragments decision trajectories and mixes valid and erroneous traces, hindering coherent state reconstruction and error isolation. We propose MAGE (Memory as Agent-Guided Exploration), an active execution-state manager that stores interactions in a hierarchical state tree. The agent derives its state from the active root-to-current path, combining subgoal summaries, recent traces, and hints from prior branches. Four coupled operations maintain the tree: Grow records new traces, Compress summarizes completed subgoals, Maintain validates summaries, and Revise restores a target boundary and resumes on a new branch. This design bounds context growth while preserving state integrity and isolating flawed segments from the active path. Experiments on MemoryArena show that MAGE improves the average task success rate by 7.8--20.4 pp over baselines, while reducing token consumption by 55.1%.",
"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"
],
"url": "https://arxiv.org/abs/2606.06090",
"pdf_url": "https://arxiv.org/pdf/2606.06090",
"path": "papers/items/2026-2606-06090-beyond-semantic-organization-memory-as-execution-state-management-for-long-horiz.md",
"local_title": "\"Beyond Semantic Organization: Memory as Execution State Management for Long-Horizon Agents\"",
"local_status": "skimmed",
"local_topics": [
"computer-use",
"memory",
"planning",
"rag",
"tool-use"
],
"collection_queries": "agent-memory, rag-agent",
"collection_score": "16"
},
{
"arxiv_id": "2606.06448",
"title": "Agent Memory: Characterization and System Implications of Stateful Long-Horizon Workloads",
"abstract": "LLM agents are increasingly deployed on long-horizon tasks requiring sustained reasoning over extended interaction histories. Realizing this at scale requires agents to persistently store, retrieve, and update their own memory across sessions. A rich ecosystem of agent memory systems has emerged spanning flat retrieval, LLM-mediated extraction, consolidating fact stores, and agentic control flows. Yet, their system-level behavior remains uncharacterized. We present the first systems characterization of agent memory. First, we introduce a system-oriented taxonomy classifying agent memory systems along four axes. Second, we build a phase-aware profiling harness attributing cost to construction, retrieval, and generation. Third, we characterize ten representative systems across two benchmark suites, uncovering how design choices shift cost across the write and read paths. Finally, we derive 10 system recommendations covering construction scheduling, capability floors, amortization via query volume, freshness-latency tradeoffs, and fleet-scale management.",
"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"
],
"url": "https://arxiv.org/abs/2606.06448",
"pdf_url": "https://arxiv.org/pdf/2606.06448",
"path": "papers/items/2026-2606-06448-agent-memory-characterization-and-system-implications-of-stateful-long-horizon-w.md",
"local_title": "\"Agent Memory: Characterization and System Implications of Stateful Long-Horizon Workloads\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"planning",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "19"
},
{
"arxiv_id": "2606.06473",
"title": "MLEvolve: A Self-Evolving Framework for Automated Machine Learning Algorithm Discovery",
"abstract": "Large language model (LLM) agents are increasingly applied to long-horizon tasks such as scientific discovery and machine learning engineering (MLE), where sustained self-evolution becomes a key capability. However, existing MLE agents suffer from inter-branch information isolation, memoryless search, and lack of hierarchical control, which together hinder long-horizon optimization. We present MLEvolve, an LLM-based self-evolving multi-agent framework for end-to-end machine learning algorithm discovery. By extending tree search to Progressive MCGS, MLEvolve enables cross-branch information flow through graph-based reference edges and gradually shifts the search from broad exploration to focused exploitation with an entropy-inspired progressive schedule. To allow the agent to evolve with accumulated experience, we introduce Retrospective Memory, which combines a cold-start domain knowledge base with a dynamic global memory for task-specific experience retrieval and reuse. For stable long-horizon iteration, we further decouple strategic planning from code generation with adaptive coding modes. Evaluation on MLE-Bench shows that MLEvolve achieves state-of-the-art performance across multiple dimensions including average medal rate and valid submission rate under a 12-hour budget (half the standard runtime). Moreover, MLEvolve also outperforms specialized algorithm discovery methods including AlphaEvolve on mathematical algorithm optimization tasks, demonstrating strong cross-domain generalization. Our code is available at https://github.com/InternScience/MLEvolve.",
"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"
],
"url": "https://arxiv.org/abs/2606.06473",
"pdf_url": "https://arxiv.org/pdf/2606.06473",
"path": "papers/items/2026-2606-06473-mlevolve-a-self-evolving-framework-for-automated-machine-learning-algorithm-disc.md",
"local_title": "\"MLEvolve: A Self-Evolving Framework for Automated Machine Learning Algorithm Discovery\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"coding-agent",
"memory",
"multi-agent",
"planning",
"rag"
],
"collection_queries": "planning-agent",
"collection_score": "14"
},
{
"arxiv_id": "2606.07402",
"title": "M$^3$Exam: Benchmarking Multimodal Memory for Realistic User-Agent Interactions",
"abstract": "Language agents are increasingly deployed over accumulating multimodal information, yet existing benchmarks assume a human-human form with sparse visuals and straightforward content, evaluating neither reasoning over authentic multimodal file interaction nor the interpretation of concealed user information. We therefore introduce M$^3$Exam, a query-centric multimodal conversational memory benchmark built on realistic user-agent interaction, with multi-dimensional evaluation spanning cross-modal grounding and implicit information inference. Benchmarking MLLMs and memory systems reveals persistent gaps in cross-modal grounding, cross session reasoning, and the efficiency cost of accumulating multimodal context. We further propose M$^3$Proctor, a multimodal memory method that detects query modality bias and consumes raw visual sources only on demand, improving accuracy by 13% while cutting index-construction time and retrieved tokens by over 70%.",
"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"
],
"url": "https://arxiv.org/abs/2606.07402",
"pdf_url": "https://arxiv.org/pdf/2606.07402",
"path": "papers/items/2026-2606-07402-m-3-exam-benchmarking-multimodal-memory-for-realistic-user-agent-interactions.md",
"local_title": "\"M$^3$Exam: Benchmarking Multimodal Memory for Realistic User-Agent Interactions\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "language-agent",
"collection_score": "20"
},
{
"arxiv_id": "2606.07682",
"title": "SWE-Marathon: Can Agents Autonomously Complete Ultra-Long-Horizon Software Work?",
"abstract": "AI agents are increasingly expected to complete long-horizon workflows that require sustained progress over hours, millions of tokens, and complex environments. Yet current agent benchmarks largely evaluate short-form tasks, such as single pull requests, small tickets, or 5-10 minute exercises, limiting our ability to measure agents' capabilities in planning, long-context understanding, and memory use. We introduce SWE-Marathon, a benchmark of 20 long-horizon tasks spanning software engineering and adjacent technical domains. Each task consists of a unique executable environment, a human-written reference solution, and a multi-layer verification suite. Logged agent attempts average 27.2M total tokens, making SWE-Marathon substantially longer-horizon than existing SWE and command-line agent benchmarks. Current frontier coding agents solve fewer than 30% of tasks. Failures often arise from poor self-verification, self-reported infeasibility, and premature termination. We also observe reward-hacking behavior in 13.8% of rollouts, where agents attempt to exploit the environment or verifier to bypass the intended workflow. SWE-Marathon includes adversarial review of test suites and execution environments, as well as multi-layer checks designed to prevent shortcut solutions. We release SWE-Marathon, evaluation code, and agent trajectories at https://swe-marathon.org/.",
"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"
],
"url": "https://arxiv.org/abs/2606.07682",
"pdf_url": "https://arxiv.org/pdf/2606.07682",
"path": "papers/items/2026-2606-07682-swe-marathon-can-agents-autonomously-complete-ultra-long-horizon-software-work.md",
"local_title": "\"SWE-Marathon: Can Agents Autonomously Complete Ultra-Long-Horizon Software Work?\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"coding-agent",
"memory",
"planning",
"rag",
"reasoning",
"workflow-agent"
],
"collection_queries": "agent-evaluation",
"collection_score": "17"
},
{
"arxiv_id": "2606.07711",
"title": "Rosetta Memory: Adaptive Memory for Cross-LLM Agents",
"abstract": "Memory is the key component for transforming a stateless LLM into a persistent, evolving agent through experience accumulation, long-horizon planning, and continual self-improvement. Existing memory systems typically take the LLM as the center and design memory operations tailored to a specific backbone. In practice, however, users frequently switch between LLMs, for example using Claude for coding and GPT for writing across tasks, or routing different steps to different backbones within a single task for cost-effective trade-offs. As a result, memory written by one model often needs to be consumed by another. Making upstream memory effectively adapt to and activate downstream LLMs remains a critical yet underexplored problem. To bridge this gap, we shift the perspective from LLM-centric memory design to \\emph{memory-centric LLM adaptation}. Specifically, we approach the above upstream-downstream memory adaptation problem from both the write and read sides, and design two profile-conditioned operators that are jointly trained to optimize how memory is stored and presented for better task completion. To ensure the learned operators generalize across a broad set of LLMs, we propose a minimum-gain sampling curriculum that prioritizes the least-served LLMs during training. To better measure the operators' actual contribution rather than the LLM's own capability, we design a performance-gap reward that compares against a naive memory baseline. Experiments on HotpotQA, 2WikiMultihopQA, and MuSiQue demonstrate that our model consistently outperforms baselines and remains robust under unseen-model replacement.",
"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"
],
"url": "https://arxiv.org/abs/2606.07711",
"pdf_url": "https://arxiv.org/pdf/2606.07711",
"path": "papers/items/2026-2606-07711-rosetta-memory-adaptive-memory-for-cross-llm-agents.md",
"local_title": "\"Rosetta Memory: Adaptive Memory for Cross-LLM Agents\"",
"local_status": "queued",
"local_topics": [
"coding-agent",
"memory",
"planning",
"reasoning"
],
"collection_queries": "planning-agent",
"collection_score": "17"
},
{
"arxiv_id": "2606.08162",
"title": "Silent Failure in LLM Agent Systems: The Entropy Principle and the Inevitable Disorder of Autonomous Agents",
"abstract": "Large Language Model (LLM) agent systems suffer from failures that occur without external triggers -- no injection, no adversarial input, no resource exhaustion. These silent failures -- unexpected deviations from intended behavior under normal conditions -- are routinely misattributed to bugs or configuration errors. Through systematic analysis of over 40,000 controlled trials and long-term production observations spanning 100,000+ agent interactions, we identify a common structural logic underlying these failures. Building on patterns observed in our experiments, we survey the global research literature on autonomous agent reliability and synthesize 22 intrinsic properties of LLM agent systems across six lifecycle layers: foundation semantics, inter-agent transmission, memory persistence, task execution, feedback correction, and systemic evolution. We demonstrate that whenever a sufficient subset of these properties co-exist, system entropy -- the measurable accumulation of disorder: loss of output consistency, task accuracy, and cross-session coherence -- increases monotonically with interaction rounds. We formalize this as the Entropy Principle: S(t) = S0 * e^(alpha * t), with alpha measured empirically across multiple architectures. We propose the PIG (Physical Integrity Gate) Engine with the ADE (Agent Delivery Engineering) protocol suite as an engineering countermeasure to entropy-driven disorder. Our findings establish silent failure not as a bug to be fixed but as a manifestation of Intelligence Entropy -- a physical constraint to be managed through deterministic governance. We argue that any engineering effort stabilizing the structure and order of agent systems participates in a unified mission: keeping intelligent systems reliable as they grow in scale and complexity.",
"published": "2026-06-06",
"updated": "2026-06-06",
"authors": [
"Dexing Liu"
],
"categories": [
"cs.MA"
],
"url": "https://arxiv.org/abs/2606.08162",
"pdf_url": "https://arxiv.org/pdf/2606.08162",
"path": "papers/items/2026-2606-08162-silent-failure-in-llm-agent-systems-the-entropy-principle-and-the-inevitable-dis.md",
"local_title": "\"Silent Failure in LLM Agent Systems: The Entropy Principle and the Inevitable Disorder of Autonomous Agents\"",
"local_status": "queued",
"local_topics": [
"memory",
"tool-use"
],
"collection_queries": "autonomous-agent-llm",
"collection_score": "13"
},
{
"arxiv_id": "2606.08340",
"title": "Benchmarking Open-Ended Multi-Agent Coordination in Language Agents",
"abstract": "As language models are increasingly deployed as autonomous agents, they must coordinate with others over long horizons in open-ended interactive tasks. Yet existing evaluations rarely test these demands together, instead emphasising single-agent tasks, short interactions, or highly structured multi-agent settings. We introduce $alem$, a JAX-based benchmark for open-ended multi-agent coordination built on Craftax-like dynamics. Alem embeds procedurally generated coordination tasks, soft specialisation, communication, and controllable coordination difficulty into a long-horizon survival world with exploration, crafting, trading, and combat. We evaluate $13$ modern LLMs zero-shot within homogeneous teams, with trained MARL agents as reference points. Current LLM agents remain far from solving alem, averaging only ~6% normalised return, but their failures are not uniform. On the hardest coordination setting, zero-shot Gemini-3.1-Pro-High approaches MARL agents trained for one billion steps, while GPT-5.4-High achieves strong base-task reward but much lower coordination reward. This contrast shows that individual task competence does not imply coordination competence. Ablations show that communication is the largest contributor to coordination, while memory and reasoning help when used to maintain multi-step plans. Overall, our results identify coordination as a distinct bottleneck for frontier LLM agents, separate from single-agent capabilities. Alem makes this bottleneck measurable and provides a controlled testbed for developing agents that communicate, allocate roles, and execute shared plans. Code is available at https://github.com/alem-world/alem-env.",
"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"
],
"url": "https://arxiv.org/abs/2606.08340",
"pdf_url": "https://arxiv.org/pdf/2606.08340",
"path": "papers/items/2026-2606-08340-benchmarking-open-ended-multi-agent-coordination-in-language-agents.md",
"local_title": "Benchmarking Open-Ended Multi-Agent Coordination in Language Agents",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"multi-agent",
"planning",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "autonomous-agent-llm, language-agent, planning-agent",
"collection_score": "28"
},
{
"arxiv_id": "2606.08531",
"title": "VESTA: A Fully Automated Scenario Generation and Safety Evaluation Framework for LLM Agents",
"abstract": "Large language models (LLMs) are increasingly evolving from simple text-based interaction systems into LLM agents that can maintain memory, use tools, access external environments, and execute tasks. As their capabilities and autonomy expand, the safety risks they face also become more diverse. Existing evaluations often rely on manually written scenarios, static prompts, or final-output judgments, making it difficult to capture the diverse risks that agents may face during task execution. We introduce VESTA, a fully automated scenario generation and safety evaluation framework for LLM agents. Based on five risk dimensions, VESTA instantiaes abstract and diverse safety risks in real-world task execution into 1,072 measurable evaluation scenarios. Using the automated evaluation pipeline, 12 LLM agents are evaluated under two authority contexts. The results show that current agents still face substantial behavioral safety risks during task execution, with an average ASR of 47.1% and several models exceeding 70%. These findings demonstrate the importance of executable, process-level evaluation for understanding and improving LLM agent safety.",
"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"
],
"url": "https://arxiv.org/abs/2606.08531",
"pdf_url": "https://arxiv.org/pdf/2606.08531",
"path": "papers/items/2026-2606-08531-vesta-a-fully-automated-scenario-generation-and-safety-evaluation-framework-for-.md",
"local_title": "\"VESTA: A Fully Automated Scenario Generation and Safety Evaluation Framework for LLM Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"rag",
"tool-use"
],
"collection_queries": "agent-safety",
"collection_score": "20"
},
{
"arxiv_id": "2606.09198",
"title": "MASS: Deep Research for Social Sciences with Memory-Augmented Social Simulation",
"abstract": "Deep Research agents powered by Large Language Models (LLMs) have exhibited extraordinary potential in automated paper writing tasks. However, existing systems rely heavily on literature retrieval and synthesis through internet and local knowledge bases, often resulting research in lacking insight and creativity in social science. To address this issue, we propose \"Memory-Augmented Social Simulation (MASS)\", an innovative paradigm that leverages highly realistic and research-oriented social simulations to enhance the creativity and empirical founding of LLMs-generated research. Specifically, MASS integrates three core components: dynamic goal-path planning with multi-level social norm restraint to guide the simulation, a multi-disciplinary behavior dataset for agent memory cold-start, and a structured forgetting mechanism inspired by the Ebbinghaus curve. Together, these ensure simulation authenticity and provide a robust empirical foundation for generating innovative scholarly papers. Experimental results demonstrate the effectiveness of our method, showing a 6.81\\% improvement in generation overall quality over foundation LLMs and 17.19\\% gain in Insight over strong baselines.",
"published": "2026-06-08",
"updated": "2026-06-08",
"authors": [
"Yongrui Liu",
"Deyi Xiong"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2606.09198",
"pdf_url": "https://arxiv.org/pdf/2606.09198",
"path": "papers/items/2026-2606-09198-mass-deep-research-for-social-sciences-with-memory-augmented-social-simulation.md",
"local_title": "\"MASS: Deep Research for Social Sciences with Memory-Augmented Social Simulation\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"planning",
"rag",
"world-model"
],
"collection_queries": "agent-memory",
"collection_score": "14"
},
{
"arxiv_id": "2606.09316",
"title": "Anything2Skill: Compiling External Knowledge into Reusable Skills for Agents",
"abstract": "Retrieval-augmented generation (RAG) enables agents to access external knowledge at inference time, but it primarily retrieves fragmented declarative evidence, leaving agents to repeatedly infer task procedures from passages, manuals, examples, logs, or trajectories. This raises a fundamental question: can skills extracted from external knowledge bases be installed into an agent, enabling it to rapidly approximate domain expertise? In this paper, we propose Anything2Skill, a taxonomy-guided framework that compiles heterogeneous external knowledge into reusable, retrievable, and executable skills for agents. Given a corpus of knowledge records, \\textsc{Anything2Skill} first decomposes each record into evidence windows and performs plan-and-expand skill extraction under a skill-tree prior. The extracted candidates are then converted into structured skill contracts that specify invocation conditions, contraindications, action moves, workflow steps, constraints, output specifications, supporting evidence, and confidence scores. To construct a deployable procedural memory, Anything2Skill manages the extracted skills in a persistent SkillBank through taxonomy-aware compilation, registry-level reconciliation, lifecycle tracking, versioned updates, and visible skill-tree projection. At inference time, agents retrieve both task-specific passages from the original knowledge base and relevant procedural skills from the SkillBank, allowing RAG to provide declarative evidence while compiled skills provide reusable procedural guidance. Experiments on qsv and GitHub-CLI show that Anything2Skill combined with RAG achieves 98.85\\% and 94.10\\% success rates, respectively, substantially outperforming RAG-only agents. These results suggest that compiling latent procedural knowledge into explicit skills is an effective way to extend retrieval-augmented agents from knowledge access toward capability reuse.",
"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"
],
"url": "https://arxiv.org/abs/2606.09316",
"pdf_url": "https://arxiv.org/pdf/2606.09316",
"path": "papers/items/2026-2606-09316-anything2skill-compiling-external-knowledge-into-reusable-skills-for-agents.md",
"local_title": "\"Anything2Skill: Compiling External Knowledge into Reusable Skills for Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"planning",
"rag",
"tool-use",
"workflow-agent"
],
"collection_queries": "rag-agent",
"collection_score": "13"
},
{
"arxiv_id": "2606.09399",
"title": "RunAgent SuperBrowser: A Theory of Autonomous Web Navigation Grounded in Human Browsing Behaviour",
"abstract": "We present SUPERBROWSER, an autonomous web-navigation agent designed against a single guiding hypothesis: a web agent should browse the way a person browses. A human reading a page does not retain every pixel they have seen; they look at a few candidate targets, decide on one, and remember only what is needed to keep the goal alive. We operationalize this perception-cognition-action triad as three coupled mechanisms. First, a vision-first bounding-box pipeline labels candidate interactive regions on every screenshot and feeds them, asynchronously prefetched, to the language model so that the \"eye\" precedes the \"hand\". Second, a three-role brain -- an Orchestrator that classifies and routes, a Planner that evaluates progress every few steps, and a Worker that emits per-step actions -- separates strategic from operational reasoning. Third, a structured Ledger stores only what a person would: the goal, the last three actions, a small set of facts and dead-ends, and a handful of checkpoints; a six-phase eviction loop systematically discards stale screenshots, state blobs, and reasoning traces from the live context. Action execution is a three-tier click cascade (Chrome DevTools Protocol to Puppeteer to scripted) with humanized Bezier motion, plus a chevron-aware bounding-box snapper that resolves the \"small arrow beside a large label\" ambiguity. On the Mind2Web Hard benchmark (66 tasks), SUPERBROWSER attains 89.47% success, placing third overall and ahead of every published open/research browser-agent baseline by a large margin. We argue that the gain comes not from any single trick but from the consistent application of a cognitive contract throughout the system.",
"published": "2026-06-08",
"updated": "2026-06-08",
"authors": [
"Radeen Mostafa",
"Sawradip Saha"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2606.09399",
"pdf_url": "https://arxiv.org/pdf/2606.09399",
"path": "papers/items/2026-2606-09399-runagent-superbrowser-a-theory-of-autonomous-web-navigation-grounded-in-human-br.md",
"local_title": "\"RunAgent SuperBrowser: A Theory of Autonomous Web Navigation Grounded in Human Browsing Behaviour\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"embodied-agent",
"memory",
"planning",
"reasoning",
"tool-use"
],
"collection_queries": "web-gui-agent",
"collection_score": "15"
},
{
"arxiv_id": "2606.09483",
"title": "Memory Beyond Recall: A Dual-Process Cognitive Memory System for Self-Evolving LLM Agents",
"abstract": "Long-term memory for an LLM agent is more than retrieving the right passage at the right time. Current memory systems collapse belief revision, causal coupling, and cross-domain abstraction into a single retrieval surface tuned for surface recall, and consequently struggle on implicit personalisation that requires reasoning over how a user has evolved. We propose DCPM, which reorganises agent memory along a cognitive capability hierarchy ascending from raw inputs and atomic facts, through diachronic belief trajectories and identity, to domain schemas, latent intentions and cross-domain patterns. The hierarchy is driven by two processes inheriting the architectural split of dual-process theory: a synchronous daytime writer (System1) that records belief revisions as doubly linked supersedes chains, and an asynchronous nighttime engine (System2) that induces schemas and intentions and sweeps for cross-domain collisions abstracted into higher-level core schemas. On LongMemEval, PersonaMem and PersonaMem-v2, enabling System2 contributes most where the benchmark rewards implicit cross-session inference (up to +5.20 on PersonaMem-v2) and least on span recall, matching the architectural prediction.",
"published": "2026-06-08",
"updated": "2026-06-08",
"authors": [
"Tianxiang Fei",
"Mingyang Song",
"Mao Zheng",
"Xiang Yu"
],
"categories": [
"cs.CL",
"cs.AI"
],
"url": "https://arxiv.org/abs/2606.09483",
"pdf_url": "https://arxiv.org/pdf/2606.09483",
"path": "papers/items/2026-2606-09483-memory-beyond-recall-a-dual-process-cognitive-memory-system-for-self-evolving-ll.md",
"local_title": "\"Memory Beyond Recall: A Dual-Process Cognitive Memory System for Self-Evolving LLM Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "19"
},
{
"arxiv_id": "2606.09764",
"title": "iOSWorld: A Benchmark for Personally Intelligent Phone Agents",
"abstract": "A useful phone agent needs to be personally intelligent. It should reason over a user's identity, history, and preferences as they exist on the device, not just follow isolated instructions in an impersonal sandbox. Existing mobile agent benchmarks lack this kind of personalization. We introduce iOSWorld, the first interactive native iOS simulator benchmark built around a persistent user identity spanning 26 newly built iOS apps. These apps contain connected data such as transactions, messages, travel records, social relationships, and financial activity. iOSWorld includes 133 tasks across three increasingly difficult categories. Single-app tasks (27) test one app, multi-app tasks (60) span 2 to 8 apps, and memory and personalization tasks (46) require agents to infer patterns from personal data. We evaluate frontier and open-source computer-use models in both vision-only and privileged vision+XML settings. The best configuration reaches 52\\% overall but only 37\\% on multi-app tasks. Privileged vision+XML access improves frontier models by up to 26 percentage points, while smaller models do not benefit from added accessibility-tree input. We release iOSWorld as an open-source benchmark with all apps, seeded data, tasks, rubrics, and evaluation code.",
"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"
],
"url": "https://arxiv.org/abs/2606.09764",
"pdf_url": "https://arxiv.org/pdf/2606.09764",
"path": "papers/items/2026-2606-09764-iosworld-a-benchmark-for-personally-intelligent-phone-agents.md",
"local_title": "\"iOSWorld: A Benchmark for Personally Intelligent Phone Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"tool-use"
],
"collection_queries": "agent-evaluation, web-gui-agent",
"collection_score": "15"
},
{
"arxiv_id": "2606.09774",
"title": "Auto-Configuring Scientific Simulators with Lightweight Coding-Agent Adapters",
"abstract": "Configuring an advanced scientific simulator, translating a modeling goal into a valid, runnable input deck, is a persistent bottleneck that costs domain scientists hours to days. Input decks are executable interfaces: simulator-specific vocabulary, cross-file references, schema constraints, and validation rules must align before a simulation can run. We show that this bottleneck can be substantially reduced with a lightweight adapter around an off-the-shelf coding agent, rather than a bespoke simulator agent. Coding agents already navigate files, edit code, run commands, and repair outputs; what they lack is the simulator's executable contract, and rebuilding the agent loop risks discarding harness-calibrated tool-use and self-correction behavior. We introduce SIGA, a coding-agent adapter that supplies this contract through retrieval, procedural memory, agent-callable validation, and validation-gated termination while leaving the model and loop frozen. Because this contract is small and external, SIGA also supports adapter self-evolution: prior trajectories can rewrite the adapter contents without modifying the underlying agent. On GEOS, a multiphysics subsurface simulator, SIGA's main gain is reliability: on harder held-out tasks it improves TreeSim from 0.720 to 0.789 and reduces across-run standard deviation by about 16x by preventing empty or invalid decks. In a human calibration, SIGA reaches in about five minutes the deck quality a domain expert reached in about three hours. Transfers to OpenFOAM and LAMMPS show the recipe is portable but interface-dependent: completion gates help when structural completeness is the bottleneck, while memory and retrieval help when value correctness is.",
"published": "2026-06-08",
"updated": "2026-06-25",
"authors": [
"Matthew Ho",
"Brian Liu",
"Jixuan Chen",
"Audrey Wang",
"Lianhui Qin"
],
"categories": [
"cs.AI",
"cs.CL"
],
"url": "https://arxiv.org/abs/2606.09774",
"pdf_url": "https://arxiv.org/pdf/2606.09774",
"path": "papers/items/2026-2606-09774-auto-configuring-scientific-simulators-with-lightweight-coding-agent-adapters.md",
"local_title": "Auto-Configuring Scientific Simulators with Lightweight Coding-Agent Adapters",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"coding-agent",
"memory",
"rag",
"tool-use",
"world-model"
],
"collection_queries": "agent-memory",
"collection_score": "14"
},
{
"arxiv_id": "2606.10316",
"title": "TabClaw: An Interactive and Self-Evolving Agent for Spreadsheet Manipulation and Table Reasoning",
"abstract": "Spreadsheets and tables are widely used representations for structured data analysis, but effective analysis still requires substantial manual effort and domain expertise. Recent large language model (LLM) agents can automate parts of this process, but they often provide limited transparency into intermediate decisions, rely on implicit assumptions, struggle with multi-table comparison, and repeat similar workflows without adapting to a user's preferences. This paper presents TabClaw, an open-source interactive AI agent for spreadsheet manipulation and table reasoning. Users upload CSV or Excel files and issue natural-language requests; TabClaw clarifies ambiguous intent, exposes an editable execution plan, streams a ReAct-style tool-using analysis loop, dispatches specialist agents for parallel multi-table reasoning, and synthesizes findings with explicit consensus and uncertainty markers. Beyond one-off analysis, TabClaw records completed workflows, extracts persistent user memory, distills reusable skills from repeated tool-use patterns, supports package-style skill import, and upgrades skills from negative feedback. Experiments on spreadsheet manipulation and table reasoning benchmarks show that TabClaw improves executable task completion and reasoning performance while preserving an inspectable user workflow. This paper shows how TabClaw turns spreadsheets and tables into inspectable analytical workflows while gradually personalizing itself to recurring data-analysis tasks. Our code is available.",
"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"
],
"url": "https://arxiv.org/abs/2606.10316",
"pdf_url": "https://arxiv.org/pdf/2606.10316",
"path": "papers/items/2026-2606-10316-tabclaw-an-interactive-and-self-evolving-agent-for-spreadsheet-manipulation-and-.md",
"local_title": "\"TabClaw: An Interactive and Self-Evolving Agent for Spreadsheet Manipulation and Table Reasoning\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"planning",
"reasoning",
"tool-use",
"workflow-agent"
],
"collection_queries": "planning-agent",
"collection_score": "16"
},
{
"arxiv_id": "2606.10423",
"title": "WebChallenger: A Reliable and Efficient Generalist Web Agent",
"abstract": "Autonomous web navigation remains challenging for LLM agents, and the strongest generalist systems rely on proprietary reasoning models whose inference cost is prohibitive for the repetitive tasks where such agents would be most useful. We argue this gap stems not from insufficient model capability but from agent architectures that fail to replicate three human cognitive advantages: selective attention to relevant page regions, persistent memory of website structure, and procedural fluency with common interaction patterns. We introduce WebChallenger, a web agent framework that addresses each gap through architecture design rather than model scale, built around PageMem: a structured page representation deterministically constructed from the DOM that exposes each page as a hierarchy of semantic sections with short summaries. On this shared substrate we build three mechanisms that mirror the three cognitive advantages: a divide-and-conquer observation pipeline that lets the agent skim section summaries and extract details only from task-relevant regions; a lightweight exploration and memory system that traverses each website once to build a reusable map of pages and element behaviors; and compound action workflows that collapse common multi-step interactions into single agent actions, handling partial state changes automatically. Because all three operate over PageMem, the framework generalizes across websites without site-specific adapters. Using off-the-shelf open-weight models without fine-tuning, our system achieves 56.3% on WebArena, 48.7% on VisualWebArena, 51.0% on Online-Mind2Web, and 70.9% on WorkArena, approaching frontier proprietary systems at a fraction of the cost. Our code is released at https://github.com/jayoohwang1/webchallenger",
"published": "2026-06-09",
"updated": "2026-06-09",
"authors": [
"Jayoo Hwang",
"Xiaowen Zhang",
"Vedant Padwal"
],
"categories": [
"cs.CL"
],
"url": "https://arxiv.org/abs/2606.10423",
"pdf_url": "https://arxiv.org/pdf/2606.10423",
"path": "papers/items/2026-2606-10423-webchallenger-a-reliable-and-efficient-generalist-web-agent.md",
"local_title": "\"WebChallenger: A Reliable and Efficient Generalist Web Agent\"",
"local_status": "queued",
"local_topics": [
"computer-use",
"embodied-agent",
"memory",
"reasoning",
"tool-use",
"workflow-agent"
],
"collection_queries": "web-gui-agent",
"collection_score": "15"
},
{
"arxiv_id": "2606.10507",
"title": "HIPIF: Hierarchical Planning and Information Folding for Long-Horizon LLM Agent Learning",
"abstract": "While Large Language Models (LLMs) have demonstrated strong capabilities as autonomous agents across a wide range of tasks, their performance often degrades in multi-turn long-horizon agentic tasks. Existing methods have made progress through fine-grained credit assignment to alleviate long-horizon sparse rewards and hierarchical reinforcement learning to decompose tasks and reduce long-term dependency. However, these methods still do not directly address long-context interference, in which continuously growing histories weaken the agent's ability to track the global task state and impair subsequent reasoning and decision-making. Inspired by the way humans handle complex tasks through subgoal decomposition and completed progress summarization, we propose Hierarchical Planning and Information Folding (HIPIF) for long-horizon LLM agent learning. HIPIF trains the agent end-to-end to organize long-horizon execution around explicit subgoals while folding completed subgoal histories to reduce long-context interference. Furthermore, to stabilize subgoal-based planning and execution, HIPIF combines hierarchical reflection and subgoal-oriented process rewards to guide subgoal generation, transition, and execution, without relying on costly auxiliary models or task-specific expert trajectories. Extensive experiments on three publicly available agentic benchmarks demonstrate the validity of our method.",
"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"
],
"url": "https://arxiv.org/abs/2606.10507",
"pdf_url": "https://arxiv.org/pdf/2606.10507",
"path": "papers/items/2026-2606-10507-hipif-hierarchical-planning-and-information-folding-for-long-horizon-llm-agent-l.md",
"local_title": "\"HIPIF: Hierarchical Planning and Information Folding for Long-Horizon LLM Agent Learning\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"planning",
"reasoning"
],
"collection_queries": "agent-evaluation, autonomous-agent-llm, planning-agent",
"collection_score": "19"
},
{
"arxiv_id": "2606.10532",
"title": "ActiveMem: Distributed Active Memory for Long-Horizon LLM Reasoning",
"abstract": "Memory is essential for enabling large language model (LLM) agents to handle long-horizon reasoning tasks. Existing memory mechanisms are largely centralized, typically organizing retrieved information and interaction history within a single model context. This design imposes a fundamental trade-off: scaling reasoning trajectories risks context overload, whereas aggressive content pruning may result in irreversible information loss. Seeking a better trade-off, we draw inspiration from human cognitive systems, especially the functional complementarity between the prefrontal cortex (executive control) and the hippocampus (memory management), suggesting that such a trade-off need not be inherent, but may instead stem from centralized memory organization. To this end, we propose ActiveMem, a heterogeneous framework that decouples agent memory from the core reasoning process. Specifically, a high-level Planner utilizes distilled semantic gists to execute reasoning, while a lightweight, distributed memory system operates in parallel to actively accumulate and consolidate these gists throughout the task. Experiments on BrowseComp-Plus and GAIA show that ActiveMem achieves state-of-the-art accuracy with significantly reduced overhead, demonstrating the effectiveness of distributed active memory for long-horizon reasoning.",
"published": "2026-06-09",
"updated": "2026-06-09",
"authors": [
"Yunhan Jiang",
"Wenbin Duan",
"Shasha Guo",
"Liang Pang",
"Xiaoqian Sun",
"Huawei Shen"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2606.10532",
"pdf_url": "https://arxiv.org/pdf/2606.10532",
"path": "papers/items/2026-2606-10532-activemem-distributed-active-memory-for-long-horizon-llm-reasoning.md",
"local_title": "\"ActiveMem: Distributed Active Memory for Long-Horizon LLM Reasoning\"",
"local_status": "queued",
"local_topics": [
"agent-safety",
"memory",
"planning",
"reasoning",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "13"
},
{
"arxiv_id": "2606.10577",
"title": "AgenticNav: Zero-Shot Vision-and-Language Navigation as a Tool-Calling Harness",
"abstract": "Zero-shot vision-and-language navigation in continuous environments (VLN-CE) has recently become feasible with large vision-language models (VLMs). However, existing methods typically rely on learned waypoint predictors to propose navigable actions. This severely limits the model's action space and fails to leverage depth inputs effectively. Moreover, memory is commonly handled by accumulating long textual or visual histories with substantial irrelevant context, or by retrieving cross-episode experiences, which weakens the zero-shot setting. In this paper, we rethink zero-shot VLN-CE as an agentic interface between the VLM and the environment, and present AgenticNav, a lightweight navigation harness that exposes action, depth, and memory as callable tools. Instead of choosing from predicted waypoints, the action tool allows the VLM to directly select a target pixel in RGB observations, converting it into executable motion. Depth is exposed through an on-demand pixel-depth tool, enabling the VLM to request precise metric distances only where they matter. For memory, AgenticNav provides a compact map image summarizing the historical trajectory, paired with a recall tool that allows the VLM to selectively revisit past visual observations without overwhelming the prompt context. On the R2R-CE benchmark, AgenticNav establishes new state-of-the-art (SOTA) performance among zero-shot methods given the same VLM backbone. Real-world validation further highlights its zero-shot generalization compared to prior methods. Ablations show that our action tool design outperforms traditional waypoint predictors, and that depth tool and agentic memory further contribute to navigation performance.",
"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"
],
"url": "https://arxiv.org/abs/2606.10577",
"pdf_url": "https://arxiv.org/pdf/2606.10577",
"path": "papers/items/2026-2606-10577-agenticnav-zero-shot-vision-and-language-navigation-as-a-tool-calling-harness.md",
"local_title": "\"AgenticNav: Zero-Shot Vision-and-Language Navigation as a Tool-Calling Harness\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"embodied-agent",
"memory",
"rag",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "17"
},
{
"arxiv_id": "2606.10616",
"title": "Learning What to Remember: Observability-Safe Memory Retention via Constrained Optimization for Long-Horizon Language Agents",
"abstract": "Long-horizon language agents accumulate observations, reasoning traces, and retrieved facts exceeding context windows, making memory retention a fundamental resource-allocation problem. Existing systems treat retention as local and do not model long-term consequences under observability constraints. To fill this gap, we formulate memory retention as a constrained stochastic optimization with budget feasibility, evidence utility, and delayed costs including miss, reacquisition, and stale penalties. We show this multi-step problem is NP-hard, making exact solution intractable. Moreover, deployment decisions must be made under partial observability. To address these challenges, we propose OSL-MR (Observability-Safe Learning for Memory Retention), a learning-augmented framework that enforces a strict separation between online-observable features and offline-available supervision. OSL-MR combines an evidence learner trained from realized evidence with a Mixed-Score heuristic that serves as a deployable online-safe baseline and an inductive prior. The policy learns query-conditioned evidence from interaction data and remains deployable under the same constraints. Experiments on LoCoMo and LongMemEval show OSL-MR outperforms recency-based, Generative Agents-style, and other heuristic baselines, especially under tight budgets. The Mixed-Score prior improves precision and recall, and sensitivity analysis shows robustness across cost settings. On small solvable instances, single-step optimization is insufficient to anticipate future demand shifts, while OSL-MR stays significantly closer to the dynamic-programming optimum, confirming the necessity of the sequential formulation and reinforcing our learning-guided approximation. These results establish constrained stochastic optimization and optimization-guided learning as a principled foundation for memory management in long-horizon agents.",
"published": "2026-06-09",
"updated": "2026-06-29",
"authors": [
"Qingcan Kang",
"Liu Mingyang",
"Shixiong Kai",
"Kaichao Liang",
"Tao Zhong",
"Mingxuan Yuan"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2606.10616",
"pdf_url": "https://arxiv.org/pdf/2606.10616",
"path": "papers/items/2026-2606-10616-learning-what-to-remember-observability-safe-memory-retention-via-constrained-op.md",
"local_title": "\"Learning What to Remember: Observability-Safe Memory Retention via Constrained Optimization for Long-Horizon Language Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"planning",
"reasoning",
"tool-use"
],
"collection_queries": "language-agent",
"collection_score": "17"
},
{
"arxiv_id": "2606.10677",
"title": "Infini Memory: Maintainable Topic Documents for Long-Term LLM Agent Memory",
"abstract": "Long-term LLM agents need persistent memory that can track changing facts and provide relevant evidence across sessions. Existing memory systems often store observations as isolated records, summaries, or indexed fragments, which makes evidence aggregation, fact revision, and memory maintenance difficult. We propose Infini Memory, a maintainable text-based persistent memory architecture that treats agent memory as topic-structured documents. Each topic document serves as a semantic unit for collecting related evidence, preserving metadata, and revising facts over time. New observations are first staged in a buffer and periodically consolidated into coherent textual contexts. At inference time, an agentic retrieval procedure lets the LLM read memory through iterative tool calls rather than a single retrieval step. On MemoryAgentBench, Infini Memory achieves 64.7% overall score. Ablations show that topic-structured maintenance and iterative evidence inspection improve complementary aspects of long-term memory use.",
"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"
],
"url": "https://arxiv.org/abs/2606.10677",
"pdf_url": "https://arxiv.org/pdf/2606.10677",
"path": "papers/items/2026-2606-10677-infini-memory-maintainable-topic-documents-for-long-term-llm-agent-memory.md",
"local_title": "\"Infini Memory: Maintainable Topic Documents for Long-Term LLM Agent Memory\"",
"local_status": "skimmed",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "18"
},
{
"arxiv_id": "2606.10742",
"title": "MemVenom: Triggered Poisoning of Multimodal Memories in Web Agents",
"abstract": "External memory has become a core component of modern web agents, enabling long-horizon reasoning through the retrieval of past experiences. However, this paradigm introduces a critical vulnerability: malicious content injected into memory can be persistently recalled and repeatedly influence agent behavior. In this work, we identify and systematically study multimodal memory poisoning, an overlooked yet practical attack surface in web-agent systems. We propose MemVenom, a unified black-box attack framework that poisons graph-structured external memory with coordinated text-image evidence. Our method consists of a two-stage design: (1) a trigger-conditioned retrieval attack that ensures high-probability recall of malicious memory, and (2) a post-retrieval attack induction that leverages adversarial perturbations and stealthy OCR injection to override the original user objective. Unlike prior attacks that operate on prompts or text-only memory, our approach enables persistent, reusable, and goal-agnostic attacks without modifying model parameters or re-optimizing malicious tasks. Experiments across multiple web-agent frameworks and vision-language models demonstrate that MemVenom achieves strong end-to-end attack success with minimal impact on benign performance, reaching up to 99.15% on GPT-5-family web agents, while transferring effectively across architectures and model scales.",
"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"
],
"url": "https://arxiv.org/abs/2606.10742",
"pdf_url": "https://arxiv.org/pdf/2606.10742",
"path": "papers/items/2026-2606-10742-memvenom-triggered-poisoning-of-multimodal-memories-in-web-agents.md",
"local_title": "\"MemVenom: Triggered Poisoning of Multimodal Memories in Web Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"planning",
"rag",
"reasoning"
],
"collection_queries": "web-gui-agent",
"collection_score": "14"
},
{
"arxiv_id": "2606.10749",
"title": "Toward Secure LLM Agents: Threat Surfaces, Attacks, Defenses, and Evaluation",
"abstract": "Large language model (LLM) agents are rapidly moving from conversational interfaces to software components that plan, invoke tools, maintain memory, and act on external environments. This transition changes the nature of security risk. In agentic settings, failures are no longer limited to unsafe text generation. Untrusted content may redirect control flow, misuse tool privileges, corrupt persistent state, leak sensitive information, or trigger harmful external actions. At the same time, research on LLM agent security is expanding quickly but remains fragmented across attack families, defense layers, application domains, and evaluation settings. This paper synthesizes 247 papers through a lifecycle-based, systems-oriented framework that models agent security around the interaction of information flow, delegated authority, and persistent state. We organize the literature around four questions: how LLM agent security should be modeled, which threat surfaces and attack families dominate, what defenses have been proposed and with what tradeoffs, and how security claims are evaluated. We find that prompt injection and tool-mediated control-flow hijacking still dominate the field, while persistent state corruption and multi-agent propagation are becoming central emerging concerns. We further find that current defenses provide useful building blocks but remain weakly compositional, and that existing benchmarks still underrepresent long-horizon, stateful, and deployment-sensitive risks. We argue that secure LLM agents require explicit trust boundaries, principled privilege control, provenance-aware state management, and evaluation practices aligned with realistic operational settings.",
"published": "2026-06-09",
"updated": "2026-06-09",
"authors": [
"Yuchen Ling",
"Shengcheng Yu",
"Zhenyu Chen",
"Chunrong Fang"
],
"categories": [
"cs.CR",
"cs.AI"
],
"url": "https://arxiv.org/abs/2606.10749",
"pdf_url": "https://arxiv.org/pdf/2606.10749",
"path": "papers/items/2026-2606-10749-toward-secure-llm-agents-threat-surfaces-attacks-defenses-and-evaluation.md",
"local_title": "\"Toward Secure LLM Agents: Threat Surfaces, Attacks, Defenses, and Evaluation\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"multi-agent",
"planning",
"rag",
"tool-use"
],
"collection_queries": "agent-safety, planning-agent",
"collection_score": "25"
},
{
"arxiv_id": "2606.10921",
"title": "Trace Only What You Need: Structure-Aware On-Demand Hypergraph Memory for Long-Document Question Answering",
"abstract": "Long-document question answering (QA) requires large language models (LLMs) to reason over evidence scattered across lengthy documents, where answers often depend on event order, section-level context, and cross-part evidence connections. Although retrieval-augmented generation (RAG) reduces the input context by retrieving relevant evidence, existing structured RAG methods still face three limitations: costly query-agnostic knowledge organization, insufficient use of original document structure, and no reuse of historical reasoning experience. To address these limitations, we propose DocTrace, a multi-agent RAG framework for long-document QA that supports query-triggered knowledge organization, document-structure-aware and experience-guided reasoning. DocTrace preserves document hierarchy with a lightweight document structural tree index, constructs agent-shared hypergraph-structured working memory on demand during reasoning, and stores successful reasoning plans in graph-structured experience memory for future reuse, enabling adaptive exploration across related long-document questions. Experiments on four long-document QA datasets show that DocTrace achieves the best performance on three datasets, surpassing the strongest baseline, ComoRAG, by up to 8.85% in F1 and 4.40% in EM, while reducing the overall computational cost by 53.32%",
"published": "2026-06-09",
"updated": "2026-06-09",
"authors": [
"Xiangjun Zai",
"Xingyu Tan",
"Chen Chen",
"Xiaoyang Wang",
"Wenjie Zhang"
],
"categories": [
"cs.CL"
],
"url": "https://arxiv.org/abs/2606.10921",
"pdf_url": "https://arxiv.org/pdf/2606.10921",
"path": "papers/items/2026-2606-10921-trace-only-what-you-need-structure-aware-on-demand-hypergraph-memory-for-long-do.md",
"local_title": "\"Trace Only What You Need: Structure-Aware On-Demand Hypergraph Memory for Long-Document Question Answering\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"multi-agent",
"planning",
"rag",
"reasoning"
],
"collection_queries": "rag-agent",
"collection_score": "16"
},
{
"arxiv_id": "2606.11680",
"title": "Organize then Retrieve: Hierarchical Memory Navigation for Efficient Agents",
"abstract": "Large language model (LLM) agents struggle with long-horizon tasks due to their inherent statelessness, requiring all task-relevant information to be encoded in growing input contexts. The resulting degraded reasoning quality, increased inference cost, and higher latency necessitate efficient working memory mechanisms. However, existing approaches either rely on lossy compression or similarity-based retrieval, which often fail to capture temporal structure and causal dependencies required for multi-step agentic tasks. In this work, we present HORMA, a Hierarchical Organize-and-Retrieve Memory Agent that organizes experience into a file-system-like hierarchical structure, where summarized entities are linked to the corresponding raw trajectories, enabling efficient access without losing detailed information. HORMA decomposes working memory into two stages: structured memory construction and navigation-based retrieval. The construction module iteratively refines how experiences are structured by distinguishing between failures caused by missing information and those caused by misleading or overloaded context. The navigation module retrieves task-relevant context by traversing the hierarchy using a lightweight agent trained with reinforcement learning to select minimal yet sufficient context, thereby reducing latency along the critical execution path. Across ALFWorld, LoCoMo, and LongMemEval, HORMA improves task performance under constrained context budgets while requiring at most 22.17% of the baseline token usage in long conversation tasks. Compared to existing methods, it consistently achieves better efficiency-performance trade-offs and generalizes effectively to unseen tasks.",
"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"
],
"url": "https://arxiv.org/abs/2606.11680",
"pdf_url": "https://arxiv.org/pdf/2606.11680",
"path": "papers/items/2026-2606-11680-organize-then-retrieve-hierarchical-memory-navigation-for-efficient-agents.md",
"local_title": "\"Organize then Retrieve: Hierarchical Memory Navigation for Efficient Agents\"",
"local_status": "skimmed",
"local_topics": [
"agent-evaluation",
"computer-use",
"embodied-agent",
"memory",
"planning",
"rag",
"reasoning"
],
"collection_queries": "agent-memory",
"collection_score": "18"
},
{
"arxiv_id": "2606.12195",
"title": "InternVideo3: Agentify Foundation Models with Multimodal Contextual Reasoning",
"abstract": "Recent progress in foundation models has shifted toward agentic behavior involving multi-step reasoning and tool use. However, open-source efforts largely focus on text-dominant settings, leaving long-horizon multimodal tasks underexplored. This gap is evident in video tasks requiring sustained temporal understanding and iterative interaction. We present InternVideo3, a framework enhancing these capabilities via Multimodal Contextual Reasoning (MCR). MCR treats understanding as a closed-loop process over a shared, evolving context containing observations, instructions, reasoning, tool actions, and memory. This frames long-video understanding as evidence accumulation and verification. To ensure efficiency, we introduce Multimodal Multi-head Latent Attention (M^2LA), a token-preserving reparameterization compressing KV-cache states while retaining the full token stream. Our staged training includes continued pretraining, short-to-long supervised fine-tuning, rule-based reinforcement learning, and on-policy distillation. Experiments show InternVideo3 achieves strong performance on benchmarks like Video-MME, MLVU, and EgoSchema. We further instantiate the model as a video agent with retrieval tools, demonstrating robust evidence-grounded behavior. Our results suggest that efficient context handling and closed-loop reasoning are vital for adapting open multimodal models toward long-horizon visually grounded agency.",
"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"
],
"url": "https://arxiv.org/abs/2606.12195",
"pdf_url": "https://arxiv.org/pdf/2606.12195",
"path": "papers/items/2026-2606-12195-internvideo3-agentify-foundation-models-with-multimodal-contextual-reasoning.md",
"local_title": "\"InternVideo3: Agentify Foundation Models with Multimodal Contextual Reasoning\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"planning",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "tool-use",
"collection_score": "15"
},
{
"arxiv_id": "2606.12320",
"title": "A Five-Plane Reference Architecture for Runtime Governance of Production AI Agents",
"abstract": "Enterprise security was built to govern data boundaries: the protected surface was data at rest and in transit, and the controls -- access control, data-loss prevention, perimeter inspection -- governed crossings of that boundary. Production AI agents dissolve this assumption. An agent reads context, calls tools, invokes connectors, and modifies systems of record on an enterprise's behalf, so risk moves inside the workflow, into sequences of individually-permitted actions that may transform a business process no one authorized. Existing policy engines do not extend to this regime: they evaluate request-time decisions against atomic principals, where agentic systems require stateful evaluation against composite principals whose authority attenuates through delegation chains. We present a reference architecture for the runtime governance of production agents, built from four composable primitives: a five-plane decomposition (a reasoning plane that adjudicates intent, and four enforcement planes -- network, identity, endpoint, data -- that realize the decision), stop-anywhere mediation, composite principals with capability attenuation, and audit as a structured evidence substrate. We define a taxonomy of six interruption primitives that generalize allow and deny, state and argue for four correctness invariants, and demonstrate the foreclosure of seven production-agent threats across five concrete workflows. A reference implementation of the policy-engine core supplies measured evidence: attenuation correctness and evidence reconstructability hold on every trial, adjudication runs in single-digit microseconds, and the audit substrate's tamper-evidence behaves exactly as designed. We are explicit about scope: the architecture governs delegated action, not model behavior, and a full-system evaluation against a live agent benchmark is the invited next step.",
"published": "2026-06-10",
"updated": "2026-06-10",
"authors": [
"Krti Tallam"
],
"categories": [
"cs.AI",
"cs.CC",
"cs.CR",
"cs.SE"
],
"url": "https://arxiv.org/abs/2606.12320",
"pdf_url": "https://arxiv.org/pdf/2606.12320",
"path": "papers/items/2026-2606-12320-a-five-plane-reference-architecture-for-runtime-governance-of-production-ai-agen.md",
"local_title": "A Five-Plane Reference Architecture for Runtime Governance of Production AI Agents",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"planning",
"reasoning",
"tool-use",
"workflow-agent"
],
"collection_queries": "agent-evaluation",
"collection_score": "18"
},
{
"arxiv_id": "2606.12563",
"title": "Arbor: Tree Search as a Cognition Layer for Autonomous Agents",
"abstract": "Arbor is a multi-agent framework that introduces structured tree search as a cognition layer for autonomous agents operating in large, stateful action spaces. Prior autonomous optimization systems operate on isolated targets with stateless evaluation. Arbor instead maintains an explicit search tree of scored hypotheses that serves as the shared working memory across agents, evolving with every measurement, treating failures as diagnostic signal that reshapes subsequent exploration, and expanding as prior successes shift the bottleneck distribution. We validate Arbor on full-stack LLM inference optimization, a domain where achieving peak performance has historically required coordinated effort from engineering teams across the application, framework, compiler, kernel, and hardware stack. Arbor pairs an Orchestrator agent, which drives optimization by delegating to Domain Specialists across the inference stack, with a Critic agent that safeguards stability through root-cause analysis, introspection, and measurement validation -- a checks-and-balances architecture where neither agent can unilaterally drive the system. Agent capabilities are decomposed into hard skills (domain expertise) and soft skills (coordination protocols that determine how contributions compose), enabling fully autonomous multi-day campaigns. Arbor achieves up to 193% inference throughput-latency Pareto improvement over vendor-optimized baselines, while a single agent without the harness plateaus at +33% throughput improvement and crashes irrecoverably within hours. Arbor generalizes to multiple generations of hardware platform, and run-to-run variance is within 2 percentage points demonstrating that the method is hardware-agnostic and reproducible.",
"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"
],
"url": "https://arxiv.org/abs/2606.12563",
"pdf_url": "https://arxiv.org/pdf/2606.12563",
"path": "papers/items/2026-2606-12563-arbor-tree-search-as-a-cognition-layer-for-autonomous-agents.md",
"local_title": "\"Arbor: Tree Search as a Cognition Layer for Autonomous Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"multi-agent",
"tool-use"
],
"collection_queries": "autonomous-agent-llm",
"collection_score": "14"
},
{
"arxiv_id": "2606.12703",
"title": "SMSR: Certified Defence Against Runtime Memory Poisoning in Persistent LLM Agent Systems",
"abstract": "Retrieval-augmented generation (RAG) agents increasingly run with persistent memory that accumulates across user sessions. This creates a new attack surface: an adversary interacting only through normal channels can inject crafted memories that, once retrieved, steer the agent's responses for future users, without touching model weights or code. We call this Multi-Session Memory Poisoning (MSMP) and show that no existing defence certifies against it; static-corpus defences (RobustRAG, ReliabilityRAG) assume a fixed knowledge base, and heuristic filters are bypassed by fluent enterprise-style text. We present Signed Memory with Smoothed Retrieval (SMSR), the first defence with a certified robustness bound for this setting. Component 1 adds HMAC-SHA256 provenance at write time, blocking unsigned injection. Component 2 applies randomised memory ablation with verdict-based majority voting at query time, bounding the influence of authenticated adversaries. We prove that no provenance-free retrieval-time filter can certify against adaptive injection, derive a hypergeometric certificate for Component 2, and formalise the Consistent Minority Effect, whereby a consistent adversarial answer wins string-based voting as a numerical minority while verdict-based voting removes it. Across 15 enterprise scenarios (3,150 repeated trials), Component 1 cuts attack success from 93-100% to 0% for all unsigned variants. For an authenticated adversary with a single injection, Component 2 holds success to 8.0% (95% CI [5.8, 10.9], n=450), below the certified worst case. In an end-to-end query-only attack where the agent itself writes the poison rather than it being pre-seeded, SMSR reduces success from 65.3% to 5.3% (n=150, non-overlapping CIs) on a live agent stack. Clean-query utility is 90% (Component 1) and 85% (combined).",
"published": "2026-06-10",
"updated": "2026-06-10",
"authors": [
"Tarun Sharma"
],
"categories": [
"cs.CR",
"cs.AI",
"cs.LG"
],
"url": "https://arxiv.org/abs/2606.12703",
"pdf_url": "https://arxiv.org/pdf/2606.12703",
"path": "papers/items/2026-2606-12703-smsr-certified-defence-against-runtime-memory-poisoning-in-persistent-llm-agent-.md",
"local_title": "\"SMSR: Certified Defence Against Runtime Memory Poisoning in Persistent LLM Agent Systems\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"workflow-agent"
],
"collection_queries": "rag-agent",
"collection_score": "17"
},
{
"arxiv_id": "2606.12780",
"title": "ProPlay: Procedural World Models for Self-Evolving LLM Agents",
"abstract": "Self-evolving agents are expected to improve through interaction without external supervision, but this remains difficult in partially observable environments where agents must explore actively, learn from limited feedback, and decide when to trust prior experience. Existing LLM-agent methods often rely on memory or planning modules, yet they rarely close the loop between them to continually refine an internal understanding of environment dynamics. We introduce ProPlay, a procedural world model that supports procedure-level preplay, where agents can rehearse future procedural paths using the learned world knowledge. Rather than representing experience as isolated rules or low-level action constraints, ProPlay abstracts successful trajectories into procedures and organizes them in a procedure graph that captures causal transitions among task stages. Each transition is associated with a reliability record embedding to estimate its task-specific contribution from past outcomes. Before each episode, ProPlay simulates future procedural trajectories over known graph structures as structured soft guidance; after execution, it refines the graph using environment feedback. Experiments on public benchmarks show that ProPlay consistently improves environment understanding and self-evolution capability over strong baselines. Our code has been released in https://github.com/antman9914/proplay.",
"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"
],
"url": "https://arxiv.org/abs/2606.12780",
"pdf_url": "https://arxiv.org/pdf/2606.12780",
"path": "papers/items/2026-2606-12780-proplay-procedural-world-models-for-self-evolving-llm-agents.md",
"local_title": "\"ProPlay: Procedural World Models for Self-Evolving LLM Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"planning",
"tool-use",
"world-model"
],
"collection_queries": "planning-agent",
"collection_score": "18"
},
{
"arxiv_id": "2606.12945",
"title": "Learning What to Remember: A Cognitively Grounded Multi-Factor Value Model for Agentic Memory",
"abstract": "Long-running LLM agents accumulate interaction histories far larger than any context window, forcing a standing decision: what to encode deeply, what to forget, and what to retrieve under a fixed memory budget. Production systems answer with semantic similarity or recency -- both mis-specified for the forgetting decision, which is made at consolidation time before the future query is known. We propose a multi-factor memory value function V(m)=\\sum_i w_i f_i(m) over seven interpretable factors (emotional intensity, goal relevance, value alignment, self/user relevance, task utility, reliability, and usage history) drawn from cognitive psychology, whose weights are learned from a downstream objective by a gradient-free optimiser, and whose single scalar uniformly controls encoding depth, forget risk, and retrieval rank. We make a methodological point: on LongMemEval, scoring goal relevance against the held-out evaluation question saturates gold-evidence retention at \\approx 0.98 -- this measures retrieval, not forgetting. In the realistic blind regime, a learned multi-factor value retains 0.770 \\pm 0.011 of gold evidence across 479 usable cases, versus 0.657 for uniform weights, 0.518 for the best single factor, and 0.368 for recency; every paired gap's 95% bootstrap CI is above zero, and a neural network over the same factors ties the linear model. The learned weights are interpretable -- reliability, emotional intensity, and self/user relevance dominate, while query-time goal similarity is correctly down-weighted for the forgetting decision. A controlled synthetic task with planted confounds confirms the learner recovers a separating weighting (1.00 retention) where uniform weighting fails (0.62). The substrate is open-source; all experiments run on a single CPU with no API calls.",
"published": "2026-06-11",
"updated": "2026-06-20",
"authors": [
"Zhibao Chen",
"Qian Cheng"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2606.12945",
"pdf_url": "https://arxiv.org/pdf/2606.12945",
"path": "papers/items/2026-2606-12945-learning-what-to-remember-a-cognitively-grounded-multi-factor-value-model-for-ag.md",
"local_title": "\"Learning What to Remember: A Cognitively Grounded Multi-Factor Value Model for Agentic Memory\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"coding-agent",
"memory",
"planning",
"rag",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "22"
},
{
"arxiv_id": "2606.13177",
"title": "MemRefine: LLM-Guided Compression for Long-Term Agent Memory",
"abstract": "Large language model (LLM) agents are increasingly expected to operate over long-term interactions, where information from past dialogues must be preserved and recalled to support future tasks. However, as interactions accumulate, the memory store grows without bound and fills with redundant entries that inflate storage cost and degrade retrieval by crowding out the most useful evidence. Furthermore, this is especially limiting on resource-constrained platforms with hard memory budgets, motivating us to formulate storage-budgeted memory management, the task of keeping an already constructed memory store within a fixed budget while preserving information useful for future interactions. To this end, we then propose MemRefine, an LLM-guided framework that, since surface similarity poorly reflects factual value, uses similarity only to propose candidate pairs and defers delete, merge, and preserve decisions to an LLM judge based on factual content, iterating until the budget is met. Across multiple memory frameworks and long-term conversation benchmarks, MemRefine consistently meets target budgets while preserving downstream performance and outperforming rule-based baselines under tight budgets.",
"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"
],
"url": "https://arxiv.org/abs/2606.13177",
"pdf_url": "https://arxiv.org/pdf/2606.13177",
"path": "papers/items/2026-2606-13177-memrefine-llm-guided-compression-for-long-term-agent-memory.md",
"local_title": "\"MemRefine: LLM-Guided Compression for Long-Term Agent Memory\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"rag",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "16"
},
{
"arxiv_id": "2606.14790",
"title": "XFlow: An Executable Protocol Programming System for Reliable Multi-Agent Workflows",
"abstract": "LLM-based multi-agent systems increasingly coordinate planning, reasoning, tool use, and human interaction, yet their reliability remains limited. A central source of this limitation is the underspecified prompt--harness boundary. Current systems lack a principled way to decide which workflow commitments should remain in prompts and which should become harness structure. We present \\textbf{XFlow}, an executable protocol programming system for reliable multi-agent workflows, and \\textbf{XPF} (XFlow Protocol Format), its domain-specific protocol programming language. XFlow occupies a middle position between prompt-only orchestration and markup-like workflow descriptions. XPF remains readable as a literate protocol, but it is compiled and executed as a program. Its design keeps informal semantic work inside actors while moving selected commitments into harness structure that can be checked, preserved, and enforced. At runtime, XFlow stages uncertainty through lifecycle-governed symbols, which are typed state cells with validation and commit states. Actor outputs are mediated before they become shared state, instead of spreading through prompts, transcripts, or implicit memory. Our experiments cover Constrained Interaction, Long-Context Reasoning, and Agentic Software Engineering. They show that XFlow improves reliability by making constraints, evidence handling, and process requirements explicit and enforceable.",
"published": "2026-06-11",
"updated": "2026-06-11",
"authors": [
"Hanqi Li",
"Jing Peng",
"Zijian Wang",
"Lu Chen",
"Kai Yu"
],
"categories": [
"cs.PL",
"cs.AI"
],
"url": "https://arxiv.org/abs/2606.14790",
"pdf_url": "https://arxiv.org/pdf/2606.14790",
"path": "papers/items/2026-2606-14790-xflow-an-executable-protocol-programming-system-for-reliable-multi-agent-workflo.md",
"local_title": "\"XFlow: An Executable Protocol Programming System for Reliable Multi-Agent Workflows\"",
"local_status": "queued",
"local_topics": [
"coding-agent",
"memory",
"multi-agent",
"planning",
"reasoning",
"tool-use",
"workflow-agent"
],
"collection_queries": "tool-use",
"collection_score": "20"
},
{
"arxiv_id": "2606.14805",
"title": "Knowledge-Based Zero-Replay Debugging of Multi-Agent LLM Traces",
"abstract": "Reliable operation of multi-agent large language model (LLM) systems depends on debugging long execution traces, where the few causally decisive events are buried in unstructured logs of messages, routes, memory writes, and tool calls. The standard tool is counterfactual replay (rewind, edit, and re-run the trajectory to measure each event's effect), but its cost grows linearly with the number of candidate events, making exhaustive replay infeasible at scale. We frame trace debugging as a knowledge-based decision-support problem. Each trace is compiled into a structured event knowledge graph over routing, memory, tool-use, uncertainty, and latent evidence, and a calibrated predictor decides where a scarce replay budget should be spent. We do not propose a new replay oracle; we propose a method to predict its results without paying the replay cost. We formulate zero-replay counterfactual-effect prediction: given a trace under a fixed budget, predict which events the oracle would mark high-effect before any replay is performed. BranchPoint-Latent is a lightweight predictor over observable, structural, uncertainty, and latent features of the knowledge graph. Calibrated against a deterministic replay oracle across 37 trace families, a single learning-to-rank gradient-boosted predictor raises per-trace localization (Branch Recall@5) from 0.73 to 0.93 on held-out families at zero oracle-replay cost. Rather than claiming universal dominance, we characterize when cheap graph centrality suffices and when learned evidence is necessary. The result is an auditable, cost-efficient decision-support system for AI-reliability debugging, positioned explicitly on the cost-accuracy frontier with reproducible artifacts.",
"published": "2026-06-11",
"updated": "2026-06-11",
"authors": [
"Dong Ho Kang",
"Hyeonjeong Cha",
"Daein Weon"
],
"categories": [
"cs.SE",
"cs.AI"
],
"url": "https://arxiv.org/abs/2606.14805",
"pdf_url": "https://arxiv.org/pdf/2606.14805",
"path": "papers/items/2026-2606-14805-knowledge-based-zero-replay-debugging-of-multi-agent-llm-traces.md",
"local_title": "Knowledge-Based Zero-Replay Debugging of Multi-Agent LLM Traces",
"local_status": "queued",
"local_topics": [
"memory",
"multi-agent",
"tool-use"
],
"collection_queries": "tool-use",
"collection_score": "14"
},
{
"arxiv_id": "2606.14106",
"title": "Naive Visual Memory is Not Enough: A Failure-Mode Study of GUI Agents",
"abstract": "Graphical User Interface (GUI) agents are increasingly used to automate complex computer tasks across applications, websites, and operating systems. To improve their reliability, recent work has introduced experiential memory, where agents retrieve prior trajectories to guide decision-making in similar states. More recent approaches further extend this idea to visual memory by storing and retrieving screenshots from past interactions, providing agents with richer contextual information than text-only memories. However, the effect of visual memory in GUI agents remains insufficiently understood: it is unclear which failures visual memory mitigates, or which failures it exacerbates. To systematically analyze the effect of visual memory, we introduce a taxonomy of four GUI agent failures (i.e., cognitive failure, visual state misunderstanding, hidden operation blindness, and grounding error) that map to distinct stages of the perception-reasoning-action pipeline. We find that prepending full-image memory has a divergent effect on the failure distribution: it reduces state-level failures but worsens action-level ones, and increases hidden operation blindness and grounding error. Motivated by this finding, we propose Action-Grounded Visual Memory (AGMem), an action-grounded memory framework for GUI agents. The core idea of AGMem is to store image crops that capture the local GUI region closely related to a successful action or a recovery, rather than storing full screenshots. Experiments on OSWorld show that AGMem improves task success rates by 33.3 % over full-image memory. These results demonstrate that AGMem is an effective representation for visual memory in GUI agents.",
"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"
],
"url": "https://arxiv.org/abs/2606.14106",
"pdf_url": "https://arxiv.org/pdf/2606.14106",
"path": "papers/items/2026-2606-14106-naive-visual-memory-is-not-enough-a-failure-mode-study-of-gui-agents.md",
"local_title": "\"Naive Visual Memory is Not Enough: A Failure-Mode Study of GUI Agents\"",
"local_status": "queued",
"local_topics": [
"computer-use",
"memory",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "web-gui-agent",
"collection_score": "16"
},
{
"arxiv_id": "2606.14470",
"title": "GitOfThoughts: Version-Controlled Reasoning and Agent Memory You Can Replay, Diff, and Merge",
"abstract": "Large language model reasoning leaves no trace once it is done. The steps of a chain of thought disappear when the context window closes, a pruned search branch is just gone, and memory buffers cannot be diffed, merged, or audited. Code, infrastructure, and experiments are all version-controlled. Reasoning is not. GitOfThoughts stores an agent's reasoning tree as a git repository. Every scored thought becomes a commit, scores become notes, outcomes become tags, and retrieval is just git log over the agent's own history. We use this to test something simple. Does giving an agent memory from past problems actually make it more accurate? We tried five memory stores (none, a markdown file, a vector database, a graph, and git) across two benchmarks, two model sizes, and several pre-registered repeat experiments. The answer, on new problems, is no, including one promising early result that did not hold up when we repeated it. Memory only helps once the problem being solved is nearly identical to something already in memory (cosine similarity above about 0.8); below that, it does nothing. In other words, the model is finding the answer rather than learning the method. Even a model 4.5x larger still cannot pull a reusable method out of a worked example; it just gets better at spotting near-copies. The only thing that reliably helped on new problems was generating several answers and picking the most common one (self-consistency). So the case for using git as the memory store is not that it retrieves better. It is that it gives auditability, history, and the ability to merge two agents' memories, at no cost to accuracy.",
"published": "2026-06-12",
"updated": "2026-06-22",
"authors": [
"Pavan C Shekar",
"Abhishek H S",
"Aswanth Krishnan"
],
"categories": [
"cs.AI",
"cs.CL",
"cs.LG"
],
"url": "https://arxiv.org/abs/2606.14470",
"pdf_url": "https://arxiv.org/pdf/2606.14470",
"path": "papers/items/2026-2606-14470-gitofthoughts-version-controlled-reasoning-and-agent-memory-you-can-replay-diff-.md",
"local_title": "\"GitOfThoughts: Version-Controlled Reasoning and Agent Memory You Can Replay, Diff, and Merge\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"coding-agent",
"memory",
"rag",
"reasoning"
],
"collection_queries": "agent-memory",
"collection_score": "16"
},
{
"arxiv_id": "2606.14502",
"title": "From Chatbot to Digital Colleague: The Paradigm Shift Toward Persistent Autonomous AI",
"abstract": "Large Language Models (LLMs) are undergoing a fundamental transformation from conversational generators into integrated AI systems capable of reasoning, action, memory, and self-improvement. We conceptualize this transition as a shift from Chatbot to Digital Colleague: from conversational answers to persistent work. We organize this transition along two tightly coupled dimensions. First, at the cognitive core level, LLMs are advancing from Chatbot-era \"fast thinking\" systems driven by next-token prediction toward Thinking LLMs that leverage inference-time computation, Chain-of-Thought reasoning, reflection, process supervision, and reinforcement learning to support more deliberate and reliable cognition. Second, at the tool-augmented task execution level, LLMs are progressing from tool-calling Agents that invoke external resources in an ad hoc manner toward OpenClaw-style workstation systems (OpenClaw) equipped with persistent Workspaces, skills, verification loops, and governance. The \"Workspace + Skill\" paradigm makes episodic tool use colleague-like via state persistence, reusable procedures, task closure, and experience reuse. We examine data construction shifts from instruction-response pairs to State-Action-Observation trajectories and evaluation from static benchmarks to sandboxed, auditable, self-evolving AI ecosystems.",
"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"
],
"url": "https://arxiv.org/abs/2606.14502",
"pdf_url": "https://arxiv.org/pdf/2606.14502",
"path": "papers/items/2026-2606-14502-from-chatbot-to-digital-colleague-the-paradigm-shift-toward-persistent-autonomou.md",
"local_title": "\"From Chatbot to Digital Colleague: The Paradigm Shift Toward Persistent Autonomous AI\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "tool-use",
"collection_score": "14"
},
{
"arxiv_id": "2606.14571",
"title": "StreamMemBench: Streaming Evaluation of Agent Memory for Future-Oriented Assistance",
"abstract": "A central role of personal-agent memory is to turn stored information and prior interactions into future-oriented assistance. In daily use, useful cues come from what the agent observes and how the user interacts with the agent, and the agent must carry them forward from the current request to similar future tasks. Existing memory benchmarks usually test dialogue recall or task improvement in isolation, leaving the trajectory from streaming observations to later assistance largely untested. We introduce StreamMemBench, a streaming benchmark that constructs a two-step task sequence around each evidence anchor from EgoLife egocentric streams. The initial task tests evidence use, while the follow-up task tests whether feedback and interaction experience are reused. Four metrics diagnose evidence recall, initial evidence use, feedback incorporation, and follow-up reuse. Experiments with eight memory systems across two backbones show that current systems often fail to use observed evidence or turn feedback into reliable follow-up behavior, even when evidence is stored or feedback is incorporated locally. StreamMemBench is publicly available at https://github.com/landian60/StreamMemBench.",
"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"
],
"url": "https://arxiv.org/abs/2606.14571",
"pdf_url": "https://arxiv.org/pdf/2606.14571",
"path": "papers/items/2026-2606-14571-streammembench-streaming-evaluation-of-agent-memory-for-future-oriented-assistan.md",
"local_title": "\"StreamMemBench: Streaming Evaluation of Agent Memory for Future-Oriented Assistance\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "15"
},
{
"arxiv_id": "2606.15017",
"title": "Are Online Skill and Memory Modules Always Worth Their Tokens? A Budget-Constrained Study of Web Agents",
"abstract": "Online web agents often augment a base actor with memory, workflow, or skill modules. These modules can improve performance, but they also consume test-time tokens, a cost rarely reported alongside the actor's inference cost. We study online augmentation, where this overhead is paid on every task, and re-evaluate its benefits under a fixed total inference budget. We compare AWM, ASI, and ReasoningBank with a token-matched vanilla baseline that uses the same budget for additional actor steps. Across three WebArena domains and three models, Gemini 3 Flash, GPT-5.4-mini, and Qwen 3.6-27B, the vanilla baseline matches or surpasses all three augmentation methods in aggregate success rate while often using fewer total tokens. We observe a similar trend on WorkArena-L1 with Qwen 3.6-27B, indicating that the effect extends to enterprise knowledge-work tasks. Our results suggest that skills and workflow memory can be useful in specific domains, but their apparent gains often vanish against a budget-matched actor. We further show that run-to-run variance materially affects outcomes and should be reported as a core evaluation criterion for online web agents.",
"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"
],
"url": "https://arxiv.org/abs/2606.15017",
"pdf_url": "https://arxiv.org/pdf/2606.15017",
"path": "papers/items/2026-2606-15017-are-online-skill-and-memory-modules-always-worth-their-tokens-a-budget-constrain.md",
"local_title": "Are Online Skill and Memory Modules Always Worth Their Tokens? A Budget-Constrained Study of Web Agents",
"local_status": "skimmed",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"reasoning",
"workflow-agent"
],
"collection_queries": "web-gui-agent",
"collection_score": "14"
},
{
"arxiv_id": "2606.15609",
"title": "FragFuse: Bypassing Access Control of Large Language Model Agents via Memory-Based Query Fragmentation and Fusion",
"abstract": "Large language model (LLM) agents increasingly rely on long-term memory to support complex task execution, user personalization, and domain adaptation. Meanwhile, emerging access-control mechanisms for LLM agents are being explored to block policy-violating requests and prevent misuse. We reveal a novel attack surface arising from agent memory operations: prohibited content that would trigger access control can be fragmented across interactions, stored in long-term memory in benign-appearing form, and later reconstructed through memory retrieval without appearing explicitly in the final user query. We propose FragFuse, the first attack that enables unprivileged users to bypass agent access control by exploiting this temporal channel introduced by long-term memory. FragFuse operates in three stages: (1) identifying rejection-responsive fragments via black-box adaptive querying with fragment masking; (2) injecting these fragments into memory using marker carrier queries; and (3) retrieving and fusing the stored fragments through a follow-up attack query. Although FragFuse can be instantiated manually for individual agents, we further develop a surrogate-based optimization scheme that tunes fusion instructions and marker designs, enabling automated attack generation without violating the attacker's threat-model assumptions. We evaluate FragFuse across four representative agent settings and task domains, covering three state-of-the-art agent access-control mechanisms. FragFuse achieves an average bypass success rate of 86.3% and an average end-to-end harmful task success rate of 41.1% across all settings, with only 4.4% average task-success degradation compared with configurations without access control. We also show that alternative defenses, including state-of-the-art prompt-injection detectors and perplexity detectors, do not effectively address this attack.",
"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"
],
"url": "https://arxiv.org/abs/2606.15609",
"pdf_url": "https://arxiv.org/pdf/2606.15609",
"path": "papers/items/2026-2606-15609-fragfuse-bypassing-access-control-of-large-language-model-agents-via-memory-base.md",
"local_title": "\"FragFuse: Bypassing Access Control of Large Language Model Agents via Memory-Based Query Fragmentation and Fusion\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "15"
},
{
"arxiv_id": "2606.15903",
"title": "Control-Plane Placement Shapes Forgetting: An Architectural Study of Agent Memory Across Thirteen System Configurations",
"abstract": "Where an LLM sits in an agent memory pipeline -- between the recall plane that retrieves stored facts (extensively benchmarked) and the control plane that mutates them via supersede, release, purge (largely untested) -- shapes which forgetting failure modes the system recovers. Comparing thirteen system configurations on a 385-case adversarial surface, we observe three placement regimes with partly complementary coverage: deterministic primitives suffice for lexical/temporal categories but fail canonicalization (5% on identifier-obfuscation, 0% on cross-lingual); inscribe-time LLM recovers canonicalization (100%) but cannot help intent-aware deletion (0% on prefix-collision and compound-fact); a mutation-time hook recovers intent-aware deletion (78-85%) and brightens nearly all categories simultaneously (91.7-93.2% overall, $0.17 per 385-case run, 2.3s/case mutation latency vs. 64-191ms/case deterministic, recall path unchanged). We expose the trade-off via ForgetEval, a 1000-case templated suite plus a 385-case adversarial layer (132 hand-crafted + 253 LLM-drafted oracle-validated) scored by deterministic substring match, paired with a six-method Adapter Protocol with honest N/A scoring that lets heterogeneous memory stores enter in 130 lines. Admission is corroborated by 10-annotator IAA (Fleiss' kappa = 0.958) and a 77-case external-authored subset (four blind contributors) that replicates the canonicalization asymmetry and amplifies the joint-placement lift (+27.8 pt). Production failures are predominantly forgetting failures rather than recall failures, yet existing benchmarks measure only recall. ForgetEval and all adapters are released under MIT.",
"published": "2026-06-14",
"updated": "2026-06-16",
"authors": [
"Dongxu Yang"
],
"categories": [
"cs.CL",
"cs.AI"
],
"url": "https://arxiv.org/abs/2606.15903",
"pdf_url": "https://arxiv.org/pdf/2606.15903",
"path": "papers/items/2026-2606-15903-control-plane-placement-shapes-forgetting-an-architectural-study-of-agent-memory.md",
"local_title": "\"Control-Plane Placement Shapes Forgetting: An Architectural Study of Agent Memory Across Thirteen System Configurations\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"planning",
"rag"
],
"collection_queries": "agent-memory",
"collection_score": "15"
},
{
"arxiv_id": "2606.16534",
"title": "Generated, Parallel, Scalable? A Study of Agentic AI-Generated Julia Code on Supercomputers",
"abstract": "Julia is increasingly used in HPC as a single-language alternative to combining high-level scripting with low-level systems languages, but achieving scalable performance still requires expertise in parallel programming. LLMs are increasingly used for code generation and are advancing rapidly with each new version. Yet, existing studies focus on single-shot prompting rather than agentic settings, in which an LLM autonomously plans, generates, and refines code through tool use. Using an OpenCode-based agent extended with a Julia-documentation MCP server, we study agentic generation of parallel Julia code, focusing on task-based execution with Dagger$.$jl. We evaluate three LLMS, OpenAI GPT-5.5, Anthropic Claude Opus 4.7, and the open-weight Qwen3-Coder-Next, on three problems with distinct parallel structures: Pi approximation, tiled general matrix multiplication, and tiled Cholesky decomposition. The generated Dagger$.$jl implementations are compared against agent-generated Base$.$Threads and MPI$.$jl baselines, with shared-memory experiments scaling to 192 cores and distributed-memory experiments on two nodes. The agents reliably produce executable code for small inputs but fail at larger scales due to deadlocks, oversubscription, or out-of-memory errors, with the open-weight model affected most severely. The two commercial models scale comparably on Base$.$Threads and MPI$.$jl, while their Dagger$.$jl implementations expose recurring weaknesses in task dependencies, granularity, and scheduling. Agentic AI is promising for producing parallel Julia code, but generating robust, performance-aware implementations for large-scale HPC systems remains an open challenge.",
"published": "2026-06-15",
"updated": "2026-06-16",
"authors": [
"Linus Bantel",
"Anna-Lena Roth",
"Jonas Posner",
"Dirk Pflüger"
],
"categories": [
"cs.DC"
],
"url": "https://arxiv.org/abs/2606.16534",
"pdf_url": "https://arxiv.org/pdf/2606.16534",
"path": "papers/items/2026-2606-16534-generated-parallel-scalable-a-study-of-agentic-ai-generated-julia-code-on-superc.md",
"local_title": "Generated, Parallel, Scalable? A Study of Agentic AI-Generated Julia Code on Supercomputers",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"planning",
"tool-use"
],
"collection_queries": "tool-use",
"collection_score": "16"
},
{
"arxiv_id": "2606.16871",
"title": "Human-on-the-Bridge: Scalable Evaluation for AI Agents",
"abstract": "AI agents must be evaluated as behavioral systems, not as isolated response generators. They reason across turns, call tools, preserve context, follow policies, and act under uncertainty. Existing methods provide useful but fragmented signals: benchmarks measure fixed capabilities, Human-in-the-Loop review preserves expert judgment but does not scale easily, LLM-as-judge methods depend on evaluator design, red teaming is often episodic, and trace auditing requires explicit evidence rules. This paper introduces Human-on-the-Bridge (HOB), a scalable evaluation paradigm for agentic AI. HOB places human expertise upstream, where experts curate reusable evaluation intelligence before testing begins, including domain context, Red-Team Traps, Juror Personas, scoring guidelines, audit rules, and fallback policies. ProofAgent Harness then executes this curated intelligence repeatedly through multi-turn adversarial evaluations, trace capture, multi-juror scoring, and evidence-linked reporting. We evaluate HOB through symmetric and cost-efficient asymmetric settings across frontier LLM-based agents and Harness LLM tiers. The study covers 23,500 agent turns and produces evidence-linked findings across finance, healthcare, and code generation. The results show that HOB can amplify evaluation quality without requiring equally large evaluator models, allowing smaller Harness LLMs to challenge agents built on frontier LLM backbones. The evaluation surfaces failures often missed by static benchmarks and single-evaluator scoring, including phantom tool-call claims, missing mandatory tool calls, policy drift, manipulation paths, and safe but non-resolving refusals. These findings support HOB as a paradigm for scaling human-curated evaluation intelligence, where expert judgment is encoded upfront and reused across repeated agent evaluations rather than applied manually inside every run.",
"published": "2026-06-15",
"updated": "2026-06-15",
"authors": [
"Fouad Bousetouane"
],
"categories": [
"cs.MA"
],
"url": "https://arxiv.org/abs/2606.16871",
"pdf_url": "https://arxiv.org/pdf/2606.16871",
"path": "papers/items/2026-2606-16871-human-on-the-bridge-scalable-evaluation-for-ai-agents.md",
"local_title": "\"Human-on-the-Bridge: Scalable Evaluation for AI Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"rag",
"tool-use"
],
"collection_queries": "agent-evaluation",
"collection_score": "20"
},
{
"arxiv_id": "2606.17573",
"title": "Cordon: Semantic Transactions for Tool-Using LLM Agents",
"abstract": "Tool-using LLM agents are shifting the unit of computation from explicit human-issued commands to model-driven tasks with stateful consequences. Yet today's agent runtimes still expose tools as isolated RPCs. This interface gives runtimes a convenient integration point, but it lacks a task-scoped execution boundary for commit, rollback, recovery, and audit across multi-step agent workflows. We argue that this mismatch calls for a runtime containment boundary rather than another per-call guardrail. This paper introduces Cordon, a transactional runtime system for staging and validating irreversible agent effects before commit. A semantic transaction is a task-level execution boundary that binds tool intents and runtime-tracked result lineage to reversible local state, staged external effects, delegated authority, and audit metadata. Cordon implements this abstraction with a transaction manager that tracks derived result objects, executes reversible mutations in shadow state, stages outward-facing actions in an effect outbox, and records recovery metadata. The runtime then validates the composed execution flow before it commits state or releases external effects. Our evaluation across adversarial and benign workflows shows that Cordon exposes cross-step violations missed by existing defenses. It also reduces irreversible-effect failures while preserving benign task completion with modest approval and latency overhead.",
"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"
],
"url": "https://arxiv.org/abs/2606.17573",
"pdf_url": "https://arxiv.org/pdf/2606.17573",
"path": "papers/items/2026-2606-17573-cordon-semantic-transactions-for-tool-using-llm-agents.md",
"local_title": "\"Cordon: Semantic Transactions for Tool-Using LLM Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"tool-use",
"workflow-agent"
],
"collection_queries": "tool-use",
"collection_score": "16"
},
{
"arxiv_id": "2606.18023",
"title": "LoopCoder-v2: Only Loop Once for Efficient Test-Time Computation Scaling",
"abstract": "Looped Transformers scale latent computation by repeatedly applying shared blocks, but sequential looping increases latency and KV-cache memory with the loop count. Parallel loop Transformers (PLT) alleviate this cost through cross-loop position offsets (CLP) and shared-KV gated sliding-window attention, making loop count a practical design choice. We therefore study PLT loop-count selection through a gain--cost view: an extra loop may refine representations, but CLP also introduces a positional mismatch at each loop boundary. We instantiate this study by training LoopCoder-v2, a family of 7B PLT coders with different loop counts, from scratch on 18T tokens, followed by matched instruction tuning and evaluation. Empirically, the two-loop variant delivers broad gains over the non-looped baseline across code generation, code reasoning, agentic software engineering, and tool-use benchmarks, improving SWE-bench Verified from 43.0 to 64.4 points and Multi-SWE from 14.0 to 31.0 points. In contrast, variants with three or more loops regress, revealing a strongly non-monotonic loop-count effect. Our diagnostics show that loop 2 provides the main productive refinement, while later loops yield diminishing, oscillatory updates and reduced representational diversity. Because the CLP-induced mismatch remains roughly fixed as refinement gains shrink, the offset cost increasingly dominates. This gain--cost trade-off explains PLT's saturation at two loops and provides diagnostics for loop-count selection.",
"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"
],
"url": "https://arxiv.org/abs/2606.18023",
"pdf_url": "https://arxiv.org/pdf/2606.18023",
"path": "papers/items/2026-2606-18023-loopcoder-v2-only-loop-once-for-efficient-test-time-computation-scaling.md",
"local_title": "\"LoopCoder-v2: Only Loop Once for Efficient Test-Time Computation Scaling\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"coding-agent",
"memory",
"reasoning",
"tool-use"
],
"collection_queries": "tool-use",
"collection_score": "13"
},
{
"arxiv_id": "2606.18356",
"title": "SafeClawBench: Separating Semantic, Audit-Evidence, and Sandbox Harm in Tool-Using LLM Agents",
"abstract": "Tool-using language-model agents introduce security failures that go beyond unsafe text: they can disclose protected objects, write persistent memory, send messages, modify databases, or trigger harmful code and tool effects. Existing evaluations often collapse these stages into a single attack success rate, making it difficult to tell whether a model merely agreed with an attacker or actually produced observable harm. We introduce SafeClawBench, a staged benchmark for tool-using agent security with 600 controlled adversarial tasks across six attack families: direct and indirect prompt injection, tool-return injection, memory poisoning, memory extraction, and ambiguity-driven unsafe inference. SafeClawBench reports three separate endpoints: semantic attack acceptance, audit-visible harm evidence, and sandbox-observed tool/state harm. Evaluating five agent endpoints under four prompt-level policies, we find that these endpoints capture different failure modes. Without additional prompt protection, semantic failure rates vary widely across models, from 9.0% to 44.2%. Audited harm evidence is narrower than semantic failure, and under a separate executable protocol some matched task identities produce sandbox harm despite passing the Semantic Core call: in a 12,000-row matched analysis, 291 of 347 observed sandbox harms occur in rows that pass the semantic check. Prompt policies change endpoint outcomes, but their effects depend on both model and protocol. SafeClawBench provides a reproducible framework for comparing agent models and prompt-policy conditions without conflating textual compliance, evidence-supported harm, and executable state changes. The open-source dataset is available at https://huggingface.co/datasets/sairights/safeclawbench.",
"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"
],
"url": "https://arxiv.org/abs/2606.18356",
"pdf_url": "https://arxiv.org/pdf/2606.18356",
"path": "papers/items/2026-2606-18356-safeclawbench-separating-semantic-audit-evidence-and-sandbox-harm-in-tool-using-.md",
"local_title": "\"SafeClawBench: Separating Semantic, Audit-Evidence, and Sandbox Harm in Tool-Using LLM Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"computer-use",
"memory",
"tool-use"
],
"collection_queries": "agent-safety, tool-use",
"collection_score": "19"
},
{
"arxiv_id": "2606.18406",
"title": "CoreMem: Riemannian Retrieval and Fisher-Guided Distillation for Long-Term Memory in Dialogue Agents",
"abstract": "Personalized dialogue agents require continuous long-term memory to maintain coherent interactions across multiple sessions. However, deploying these capabilities on consumer-grade hardware (e.g., 8 GB VRAM edge devices) introduces severe memory and compute bottlenecks. Existing systems typically rely on isotropic cosine similarity for retrieval and heuristic rules for context compression. These approaches lack a unified theoretical foundation, frequently suffering from the hubness problem in high-dimensional retrieval and syntactic fragmentation during compression. To overcome these limitations, we propose CoreMem, a resource-efficient edge-cloud memory architecture fundamentally unified by information geometry. First, Riemannian retrieval replaces cosine matching with a locally adaptive Fisher-Rao metric, effectively penalizing hub memories via Mahalanobis distance with O(Ndr) Woodbury acceleration for real-time search. Second, Fisher-guided discrete token distillation (FDTD) introduces a hierarchical sentence-to-token compression mechanism. It derives sensitivity scores from Fisher information traces, providing a principled compression-KL tradeoff augmented with explicit structural syntax protection. Evaluated on the LOCOMO and LongMemEval-S benchmarks, CoreMem achieves strong accuracy improvements, yielding substantial gains in Open-domain (+4.51 pp) and Temporal (+4.17 pp) reasoning. Extensive profiling confirms that CoreMem operates seamlessly within a strict 8 GB VRAM budget, successfully bridging the gap between resource-constrained edge devices and the demand for theoretically grounded, lifelong memory agents.",
"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"
],
"url": "https://arxiv.org/abs/2606.18406",
"pdf_url": "https://arxiv.org/pdf/2606.18406",
"path": "papers/items/2026-2606-18406-coremem-riemannian-retrieval-and-fisher-guided-distillation-for-long-term-memory.md",
"local_title": "\"CoreMem: Riemannian Retrieval and Fisher-Guided Distillation for Long-Term Memory in Dialogue Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "15"
},
{
"arxiv_id": "2606.18829",
"title": "GateMem: Benchmarking Memory Governance in Multi-Principal Shared-Memory Agents",
"abstract": "Memory benchmarks for LLM agents largely assume single-user settings, leaving shared assistants for hospitals, workplaces, campuses, and households understudied. In these deployments, multiple principals write to a common memory pool and query it under different roles, scopes, and relationships, so memory quality requires governance as well as recall. We introduce GateMem, a benchmark for multi-principal shared-memory agents. GateMem jointly evaluates utility for legitimate long-horizon requests with state updates, access control across contextual authorization boundaries, and agent-facing active forgetting after explicit deletion requests. It spans medical, office, education, and household domains, with long-form multi-party episodes, incremental memory injection, hidden checkpoints, structured judging, and leak-target annotations. Across diverse baselines and backbone models, no method simultaneously achieves strong utility, robust access control, and reliable forgetting. Long-context prompting often yields the best governance score at high token cost, while retrieval-based and external-memory methods reduce cost yet still leak unauthorized or deleted information. These results show current memory agents remain far from reliable shared institutional deployment.",
"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"
],
"url": "https://arxiv.org/abs/2606.18829",
"pdf_url": "https://arxiv.org/pdf/2606.18829",
"path": "papers/items/2026-2606-18829-gatemem-benchmarking-memory-governance-in-multi-principal-shared-memory-agents.md",
"local_title": "\"GateMem: Benchmarking Memory Governance in Multi-Principal Shared-Memory Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"planning",
"rag",
"workflow-agent"
],
"collection_queries": "agent-memory",
"collection_score": "19"
},
{
"arxiv_id": "2606.18950",
"title": "RTSGameBench: An RTS Benchmark for Strategic Reasoning by Vision-Language Models",
"abstract": "Modern Vision-Language Models (VLMs) often struggle with strategic reasoning, i.e., anticipating and influencing other agents' actions, under uncertainty in competitive and cooperative settings. Real-time strategy (RTS) games can be a natural testbed for diagnosing this limitation, as they demand coordination with allies, adaptation to opponents' strategy, and long-horizon planning under partial observability. However, existing RTS benchmarks offer limited evaluation scope, lack systematic competency diagnosis, and remain fixed in the pre-designed scenario coverage. To address these limitations, we present RTSGameBench, which is built on Beyond All Reason, a large-scale RTS game with an expanded battlefield that demands broader strategy diversity than the existing testbeds. The proposed benchmark provides evaluations through diverse gameplay across various matchup structures, diagnostic assessment via mini-games, each targeting an individual strategic competency, and extensible coverage via a self-evolving generation framework that converts free-form queries into new mini-games, improving over successive cycles. Additionally, for VLMs to operate in large-scale RTS games, we provide RTSGameAgent that manages units by an FSM with agentic memory. We empirically validate that multiple state-of-the-art VLMs do not perform well when matchups demand tighter coordination, multiagent coordination and when task scale increases.",
"published": "2026-06-17",
"updated": "2026-06-18",
"authors": [
"San Kim",
"Daechul Ahn",
"Reokyoung Kim",
"Hyeonbeom Choi",
"Seungyeon Jwa",
"Jonghyun Choi"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2606.18950",
"pdf_url": "https://arxiv.org/pdf/2606.18950",
"path": "papers/items/2026-2606-18950-rtsgamebench-an-rts-benchmark-for-strategic-reasoning-by-vision-language-models.md",
"local_title": "\"RTSGameBench: An RTS Benchmark for Strategic Reasoning by Vision-Language Models\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"multi-agent",
"planning",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "18"
},
{
"arxiv_id": "2606.19409",
"title": "OpenRath: Session-Centered Runtime State for Agent Systems",
"abstract": "Modern agent systems often suffer from fragmented runtime state: transcripts, tool effects, memory events, workspace placement, branch provenance, and replay evidence are recorded separately and become difficult to inspect or reproduce. OpenRath addresses this issue with a PyTorch-like programming model for multi-agent, multi-session systems. The analogy concerns the role of a central first-class runtime abstraction, not tensor computation. Its core abstraction is Session, the runtime value passed between agents and workflows. A Session is branchable, inspectable, replayable, backend-aware, and composable. It records conversation chunks, sandbox placement, lineage metadata, token usage, pending work, and tool evidence, while defining where memory interactions enter the runtime record. Since this state is carried by the same value used in program execution, fork, merge, and replay become explicit runtime operations rather than states reconstructed from external traces. OpenRath further defines Sandbox, Tool, Agent, Memory, Workflow, and Selector, with Selector turning control flow into runtime-routed decisions. This report presents the programming model, architecture, audited milestones, and evidence protocol. Its claims are limited to controlled runtime properties, while broad quantitative comparisons, live-provider quality, optional-backend availability, and memory quality are left for follow-on evaluation. The central thesis is that Session provides agent systems with a first-class runtime value for auditable composition.",
"published": "2026-06-17",
"updated": "2026-06-17",
"authors": [
"Fukang Wen",
"Zhijie Wang",
"Ruilin Xu"
],
"categories": [
"cs.SE",
"cs.PL"
],
"url": "https://arxiv.org/abs/2606.19409",
"pdf_url": "https://arxiv.org/pdf/2606.19409",
"path": "papers/items/2026-2606-19409-openrath-session-centered-runtime-state-for-agent-systems.md",
"local_title": "\"OpenRath: Session-Centered Runtime State for Agent Systems\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"multi-agent",
"rag",
"tool-use",
"workflow-agent"
],
"collection_queries": "agent-memory",
"collection_score": "14"
},
{
"arxiv_id": "2606.20047",
"title": "PACMS: Submodular Context Selection as a Pluggable Engine for LLM Agents",
"abstract": "Conversational and tool-using LLM agents operate over a context window that fills from several directions simultaneously. As a session proceeds, the agent accumulates user and assistant turns, entries drawn from a persistent memory store, and often largest of all, the verbatim outputs of tool calls such as file reads, search results, and API responses. Once the cumulative context exceeds the model's token budget, the framework must decide what to keep. The prevailing mechanism is recency truncation, sometimes paired with periodic summarization. This is topic-blind: a fact established early in a session is discarded simply because it is old, even when the current user query is about exactly that fact; conversely, verbose but irrelevant recent material is retained. Agents that must recall information across many turns, the defining case for memory, are precisely where recency truncation fails. Existing alternatives sit outside the agent's assembly step. Retrieval augmented generation fetches external documents into the prompt but does not arbitrate the agent's \\emph{already-present} pooled context. Context-compression methods reduce token count by rewriting or pruning text, but operate query-blind and lossily. Neither treats memory entries, conversation turns, and tool outputs as a single candidate pool to be selected from by relevance at the moment the prompt is assembled.",
"published": "2026-06-18",
"updated": "2026-06-18",
"authors": [
"Manu Ghulyani",
"Arunabh Singh",
"Karan Bharadwaj",
"Ankit Nath",
"Suranjan Goswami"
],
"categories": [
"cs.IR"
],
"url": "https://arxiv.org/abs/2606.20047",
"pdf_url": "https://arxiv.org/pdf/2606.20047",
"path": "papers/items/2026-2606-20047-pacms-submodular-context-selection-as-a-pluggable-engine-for-llm-agents.md",
"local_title": "\"PACMS: Submodular Context Selection as a Pluggable Engine for LLM Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"tool-use"
],
"collection_queries": "tool-use",
"collection_score": "15"
},
{
"arxiv_id": "2606.20401",
"title": "PowerAgentBench-Dyn: A Benchmark for Agentic AI in Power System Dynamic Studies",
"abstract": "Large Language Model (LLM)-based agents are increasingly being used to automate multi-step engineering work flows by interacting with software tools, interpreting intermediate results, and autonomously planning subsequent actions. Power system dynamic studies represent a particularly promising yet largely unexplored application domain for these agents. Unlike static computational tasks, dynamic studies often require more time on model parameter calibration, engineering judgment, and decision making under constrained action spaces. This paper introduces PowerAgentBench-Dyn, a benchmark designed to evaluate Agentic AI systems on power system dynamic-analysis tasks. The benchmark targets problems that cannot be reduced to a single optimization or coding task, but instead require a type of reasoning, tool usage, and iterative experimentation routinely performed by experienced power system engineers. The proposed framework includes two initial benchmark tasks. The first, the Dynamic Model Quality Review Benchmark, evaluates agents' ability to validate and diagnose dynamic models based on model-quality compliance criteria specified by system operators. The second, the Dynamic Security Risk Screening Benchmark, assesses agents' capability to leverage semantic memory and a limited simulation budget to identify, rank, and analyze the most critical short-circuit contingencies from an unseen fault dataset, as well as propose and evaluate possible mitigation measures. For each task, we define the simulation environment, observation and action spaces, and evaluation metrics. The benchmark is reproducible in a metric-based sense: released cases and simulator settings define a deterministic evaluator, while stochastic agent behavior is assessed over repeated runs using success rates and other metrics. The benchmark supports the development of future Agentic AI for power system operation and planning.",
"published": "2026-06-18",
"updated": "2026-06-18",
"authors": [
"Qian Zhang",
"Andrea Pomarico",
"Costas Mylonas",
"Magda Foti",
"Alberto Berizzi",
"Le Xie"
],
"categories": [
"eess.SY"
],
"url": "https://arxiv.org/abs/2606.20401",
"pdf_url": "https://arxiv.org/pdf/2606.20401",
"path": "papers/items/2026-2606-20401-poweragentbench-dyn-a-benchmark-for-agentic-ai-in-power-system-dynamic-studies.md",
"local_title": "\"PowerAgentBench-Dyn: A Benchmark for Agentic AI in Power System Dynamic Studies\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"coding-agent",
"memory",
"planning",
"rag",
"reasoning",
"tool-use",
"world-model"
],
"collection_queries": "agentic-ai",
"collection_score": "24"
},
{
"arxiv_id": "2606.20515",
"title": "S-Agent: Spatial Tool-Use Elicits Reasoning for Spatial Intelligence",
"abstract": "Real-world spatial intelligence requires reasoning over a continuous and evolving 3D world, yet existing VLMs and tool-augmented agents largely remain tied to static, stateless inference from isolated visual observations. We introduce \\textbf{\\textsc{S-Agent}}, a spatial tool-use agentic paradigm for understanding and reasoning over continuous multi-view images and videos. By formulating spatial reasoning as spatio-temporal evidence accumulation rather than isolated frame-level prediction, \\textsc{S-Agent} reshapes spatial perception into scene-centric understanding beyond frame-centric recognition. Specifically, \\textsc{S-Agent} casts the VLM as a semantic planner that decides what evidence is needed, while a hierarchy of spatial tools and experts grounds objects in 2D, lifts them into 3D geometric evidence, and aggregates this evidence into high-level spatial knowledge (\\textit{e.g.}, counting, measurement, orientation, and relative position). Additionally, a temporal memory mechanism, including Scene Memory for maintaining the evolving scene state and Agent Memory for accumulating reasoning context, enables evidence integration across frames and reasoning steps. Comprehensive experiments on multi-view and video spatial reasoning benchmarks show that \\textsc{S-Agent} consistently improves both open-source and closed-source VLMs in a training-free manner. Beyond inference-time augmentation, supervised fine-tuning (SFT) on \\textsc{S-Agent}-generated spatial trajectories \\textsc{S-300K} yields \\textsc{S-Agent-8B}, a compact spatial agent that significantly surpasses similar-scale baselines (e.g., Qwen3-VL-8B) and performs comparably to advanced closed-source models (e.g., GPT-5.4 and Gemini 3).",
"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"
],
"url": "https://arxiv.org/abs/2606.20515",
"pdf_url": "https://arxiv.org/pdf/2606.20515",
"path": "papers/items/2026-2606-20515-s-agent-spatial-tool-use-elicits-reasoning-for-spatial-intelligence.md",
"local_title": "\"S-Agent: Spatial Tool-Use Elicits Reasoning for Spatial Intelligence\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"planning",
"reasoning",
"tool-use"
],
"collection_queries": "agent-memory, tool-use",
"collection_score": "14"
},
{
"arxiv_id": "2606.20954",
"title": "Learning What Not to Forget: Long-Horizon Agent Memory from a Few Kilobytes of Learning",
"abstract": "Long-running language-model systems accumulate interaction history that outgrows the context window, so they must continually evict. When an eviction policy drops a load-bearing detail, for example an access token issued at login or a path the next call needs, the action fails. We present LRE (Learned Relevance Eviction), a few kilobytes, CPU-only, language-model-free scorer that learns which units of history are load-bearing and keeps them by verbatim extraction. Under a matched-budget comparison, in our experiment, no baseline dominates LRE on the accuracy-cost plane. On agents, LRE matches the accuracy of keeping the entire history overall. On the simplest tasks, it exceeds that no-eviction baseline by 27%, while requiring zero compressor calls and reducing peak context size by up to 52%. A controlled study trace shows LRE completes tasks where the others loop, finishing one such task in 37% fewer calls than keeping everything and solving 14 tasks where no other run policy does. On conversational memory, LRE outranks dense and token-pruning encoders at zero neural cost. In downstream evaluation, LRE gives the best budgeted answer quality on LoCoMo reading 68% fewer tokens. Its supervision can also be annotation-free: training only on the system's own behavior recovers 95% of the supervised scorer's effectiveness. We argue that, because memory eviction in LLM agents is a fidelity problem, it requires a deployable proactive policy where the future query is unavailable and exact state is decisive, and that cheap learned relevance can be sufficient.",
"published": "2026-06-18",
"updated": "2026-06-18",
"authors": [
"Nusrat Jahan Lia",
"Aritra Mazumder"
],
"categories": [
"cs.CL",
"cs.AI"
],
"url": "https://arxiv.org/abs/2606.20954",
"pdf_url": "https://arxiv.org/pdf/2606.20954",
"path": "papers/items/2026-2606-20954-learning-what-not-to-forget-long-horizon-agent-memory-from-a-few-kilobytes-of-le.md",
"local_title": "\"Learning What Not to Forget: Long-Horizon Agent Memory from a Few Kilobytes of Learning\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"planning",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "16"
},
{
"arxiv_id": "2606.21649",
"title": "EvoEmbedding: Evolvable Representations for Long-Context Retrieval and Agentic Memory",
"abstract": "Existing embedding models are inherently static: they encode text segments in isolation, ignoring their surrounding context and temporal order. This paper introduces EvoEmbedding, a novel embedding model that generates evolvable representations for retrieval. It is tailored for long-context scenarios, where information is dynamic, sequential, and requires continuous state tracking. Our design is simple: EvoEmbedding maintains a continuously updated latent memory as it sequentially processes inputs, and uses it alongside the raw content to jointly generate evolvable embeddings. Consequently, for the same query, our model adapts its representation to retrieve distinct targets based on the evolving context, going beyond static semantic search. To equip the model with this capability, we construct EvoTrain-180K, a diverse dataset for the joint optimization of latent memory and retrieval. Furthermore, we introduce a memory queue to prevent representation collapse during recurrent encoding, alongside segment-batching techniques that tackle significant length variance and accelerate training by 3.8$\\times$. Extensive experiments show that our model not only outperforms larger-scale specialists (e.g., Qwen3-Embedding-8B and KaLM-Embedding-Gemma3-12B) across a range of long-context retrieval benchmarks, but also generalizes well to downstream tasks (e.g., personalization) with contexts 10$\\times$ longer than its training window. Notably, EvoEmbedding seamlessly integrates into agentic workflows to boost performance. For instance, a naive RAG pipeline equipped with our model surpasses dedicated agentic memory systems. Project Page: https://clare-nie.github.io/EvoEmbedding/.",
"published": "2026-06-19",
"updated": "2026-06-25",
"authors": [
"Chang Nie",
"Chaoyou Fu",
"Junlan Feng",
"Caifeng Shan"
],
"categories": [
"cs.CL"
],
"url": "https://arxiv.org/abs/2606.21649",
"pdf_url": "https://arxiv.org/pdf/2606.21649",
"path": "papers/items/2026-2606-21649-evoembedding-evolvable-representations-for-long-context-retrieval-and-agentic-me.md",
"local_title": "\"EvoEmbedding: Evolvable Representations for Long-Context Retrieval and Agentic Memory\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"coding-agent",
"memory",
"rag",
"workflow-agent"
],
"collection_queries": "agent-memory, agentic-ai, rag-agent",
"collection_score": "19"
},
{
"arxiv_id": "2606.21842",
"title": "Agent-Assisted Side-Channel Attacks on Non-Prefix KV Cache in RAG",
"abstract": "Modern Large Language Model (LLM) serving engines increasingly rely on Retrieval-Augmented Generation (RAG) and non-prefix Key-Value (KV) cache fusion to accelerate long-context, multi-tenant inference. While existing KV cache side-channel attacks require strict linear prefix alignment--rendering them ineffective against real-world RAG queries that contain unique, user-specific private prefixes--we uncover a critical class of structural vulnerabilities inherent to chunk-aware memory scheduling. We demonstrate that the deterministic micro-architectural mechanisms used to align and fuse disjoint memory chunks inadvertently leak a continuous \"Step-Wave\" timing signature. Exploiting this physical observation, we introduce SpliceLeak, the first end-to-end side-channel attack targeting non-prefix KV cache fusion. SpliceLeak executes a systematic two-phase privacy breach: it first structurally fingerprints the exact length of hidden private prompts, and subsequently manipulates boundary collisions to extract exact semantic content token-by-token. Extensive evaluations on production-grade frameworks (vLLM integrated with LMCache) demonstrate that SpliceLeak achieves up to a 100% extraction success rate in bounded-entropy scenarios. Driven by a deterministic +104 ms hardware latency void, the attack requires as few as 63 requests per token, piercing through realistic continuous batching noise. To resolve the inherent conflict between memory deduplication and security, we propose SpliceDefense, a bipartite mitigation framework consisting of Quantized Chunk Padding (QCP) and Constant-Time Boundary Fusion (CTBF). Our evaluations confirm that SpliceDefense effectively flattens the side-channel signal (Delta TTFT ~ 0) with negligible throughput overhead, preserving the critical benefits of global cache sharing.",
"published": "2026-06-20",
"updated": "2026-06-20",
"authors": [
"He Sun",
"Shinan Liu",
"Siyuan Ma",
"Junhao Li",
"Mingjun Xiao",
"Wenhao Jiang"
],
"categories": [
"cs.CR"
],
"url": "https://arxiv.org/abs/2606.21842",
"pdf_url": "https://arxiv.org/pdf/2606.21842",
"path": "papers/items/2026-2606-21842-agent-assisted-side-channel-attacks-on-non-prefix-kv-cache-in-rag.md",
"local_title": "Agent-Assisted Side-Channel Attacks on Non-Prefix KV Cache in RAG",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"rag",
"tool-use"
],
"collection_queries": "rag-agent",
"collection_score": "14"
},
{
"arxiv_id": "2606.21877",
"title": "AgentRiskBOM: A Risk-Scoping Security Bill of Materials for Agentic AI Systems",
"abstract": "Agentic AI systems retrieve private context, invoke tools, write files, call external services, coordinate with other agents, and may act without human approval. Existing bill of materials artifacts improve transparency for dependencies, model metadata, and training provenance, but leave an agentic transparency gap: capability opacity, the absence of a structured account of what a deployed agent can access, remember, change, delegate, and prove afterward. This paper introduces AgentRiskBOM, a security BOM for risk-scoping tool-using AI agents. It is an additive layer over SBOM, AIBOM, and MLBOM artifacts, referencing them where authoritative while adding fields for runtime authority: autonomy, tool permissions, memory, credential scope, approval gates, audit signals, inter-agent communication, and external action capability. We implement AgentRiskBOM as a JSON-schema artifact with a reproducible corpus, risk scenarios, scorer, diff detector, control mapper, and reports. We evaluate AgentRiskBOM on 13 open-source agents spanning coding, RAG, and multi-agent archetypes, plus 52 risk scenarios across 14 categories. The schema validates all 13 corpus artifacts. Coverage analysis gives AgentRiskBOM a native-equivalent score of 14 across 16 capability dimensions, vs. 1 for SBOM, 1.5 for AIBOM and 2 for MLBOM. Across modeled risk categories, AgentRiskBOM exposes 100% risk-category visibility vs. 10.5% for SBOM-like and 20.9% for AIBOM-like views. To test agentic authority drift, we inject 33 structured deployment mutations; the diff detector identifies the correct change type for all mutations. A secondary penalty-based scorer yields a Spearman correlation of 0.73 with the primary scorer, supporting rank-level consistency while showing that thresholds require human calibration. The results show that agentic AI security needs a machine-readable authority-and-risk artifact before incidents occur.",
"published": "2026-06-20",
"updated": "2026-06-20",
"authors": [
"Srimonti Dutta",
"Akshata Kishore Moharir"
],
"categories": [
"cs.AI",
"cs.CR",
"cs.SE"
],
"url": "https://arxiv.org/abs/2606.21877",
"pdf_url": "https://arxiv.org/pdf/2606.21877",
"path": "papers/items/2026-2606-21877-agentriskbom-a-risk-scoping-security-bill-of-materials-for-agentic-ai-systems.md",
"local_title": "\"AgentRiskBOM: A Risk-Scoping Security Bill of Materials for Agentic AI Systems\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"coding-agent",
"memory",
"multi-agent",
"rag",
"tool-use"
],
"collection_queries": "agentic-ai, ai-agent, rag-agent, tool-use",
"collection_score": "21"
},
{
"arxiv_id": "2606.22030",
"title": "Nous: A Predictive World Model for Long-Term Agent Memory",
"abstract": "We present Nous, a novel agent memory architecture grounded in the principle that knowledge is prediction, not storage. Rather than persisting facts as database records, vector embeddings, or knowledge-graph triples, Nous maintains a predictive world model: a collection of categorical probability distributions, called dimensions, one per entity-attribute pair observed in conversation. Each incoming observation is scored by its information-theoretic surprise S = -log2 P(obs | D), and the distribution is updated via a closed-form Bayesian posterior. The primary stored artifact is the delta, a record of the shift from prior to posterior belief, rather than the fact itself. Forgetting emerges naturally as entropy decay toward the uniform distribution, and identity resolution is handled through mutual information between entity dimension sets. Evaluated on the LoCoMo long-term conversational memory benchmark across ten conversations (1,540 questions) using GPT-4o-mini as backbone, Nous achieves F1 of 63.50 (single-hop), 55.32 (multi-hop), 58.57 (temporal), and 62.50 (open-domain). Against A-MEM's self-reported GPT-4o-mini numbers, Nous shows substantial gains in three of four categories, though we note that independent citations of A-MEM's results disagree with each other on category assignment, a reproducibility issue we discuss openly rather than resolve unilaterally. We additionally compare against BeliefMem, a concurrently developed system built on the same core premise of belief-based rather than deterministic memory; on the same benchmark and backbone, Nous's self-reported numbers exceed BeliefMem's self-reported numbers on all four categories, though we flag several uncontrolled differences between the two evaluation pipelines that prevent this from being a fully controlled comparison. Nous requires no external vector database or graph engine.",
"published": "2026-06-20",
"updated": "2026-06-20",
"authors": [
"Pranav Singh"
],
"categories": [
"cs.AI",
"cs.CL",
"cs.IR",
"cs.LG"
],
"url": "https://arxiv.org/abs/2606.22030",
"pdf_url": "https://arxiv.org/pdf/2606.22030",
"path": "papers/items/2026-2606-22030-nous-a-predictive-world-model-for-long-term-agent-memory.md",
"local_title": "\"Nous: A Predictive World Model for Long-Term Agent Memory\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"world-model"
],
"collection_queries": "agent-memory",
"collection_score": "14"
},
{
"arxiv_id": "2606.22263",
"title": "Revelio: Cost-Efficient Agentic Memory Safety Vulnerability Detection For Repository-Scale Codebases",
"abstract": "Memory safety vulnerabilities remain a significant threat even for projects with extensive fuzzing and manual auditing. Recent results suggest that large language models hold great promise for detecting such vulnerabilities, but they are unreliable, at risk of hallucination, and challenging to scale to repository-size codebases. This paper presents Revelio, a cost-efficient end-to-end agentic framework for memory-safety vulnerability discovery. Revelio addresses the problem of hallucination by generating an executable Proof-of-Vulnerability, which is checked with a deterministic sanitizer. It reduces cost using inexpensive LLMs and lightweight static analysis to help generate and rank vulnerability hypotheses, reporting vulnerabilities only when they can be reproduced and confirmed by a sanitizer. We evaluated Revelio on seven production-quality projects that had been continuously fuzzed for five to eight years, as well as on 100 randomly selected Arvo projects from the CyberGym benchmark. With around one hour per project and a total cost of $300, Revelio discovered 19 previously unknown memory-safety vulnerabilities. On benchmarks, Revelio outperformed frontier coding agents across diverse backbone models at comparable token costs. Our results suggest that Revelio enables scalable and trustworthy end-to-end LLM-based memory-safety vulnerability detection.",
"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"
],
"url": "https://arxiv.org/abs/2606.22263",
"pdf_url": "https://arxiv.org/pdf/2606.22263",
"path": "papers/items/2026-2606-22263-revelio-cost-efficient-agentic-memory-safety-vulnerability-detection-for-reposit.md",
"local_title": "\"Revelio: Cost-Efficient Agentic Memory Safety Vulnerability Detection For Repository-Scale Codebases\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"coding-agent",
"memory"
],
"collection_queries": "agent-memory, coding-agent",
"collection_score": "20"
},
{
"arxiv_id": "2606.22330",
"title": "Hypothesis-Driven Skill Optimization for LLM Agents",
"abstract": "External skills can improve action-oriented LLM agents without changing model weights, but persistent skill updates are risky when they are distilled from sparse or noisy trajectories. A plausible reflection may encode a useful procedure, a spurious shortcut, or a rule that the target executor cannot reliably follow. We propose Hypothesis-Driven Skill Optimization (HDSO), a train-free framework in which both the skill curator and the agent executor are frozen inference endpoints. The curator observes executor traces, proposes a falsifiable hypothesis with an explicit validation plan, instantiates the hypothesis as a candidate skill package, validates the package through paired control/treatment executions, reviews behavior differences, and consolidates only supported candidates into an approved repository. The executor consumes approved skills through progressive disclosure, preserving the executor-only path when no skill is selected. On ALFWorld, HDSO improves executor-only baselines by +6.9 Avg. SR points for Qwen3-8B and +4.0 points for Qwen3.6-27B. Under 20% randomly flipped success/failure feedback during skill discovery and validation, HDSO preserves a +7.1-point gain for Qwen3-8B. Transfer and heterogeneous-pair diagnostics further show that validated repositories can be useful beyond the run that produced them, but cross-model curation succeeds only when curator diagnosis, executor capability, and validation evidence align. HDSO provides an auditable skill lifecycle for frozen action agents rather than an unconstrained memory accumulation procedure.",
"published": "2026-06-21",
"updated": "2026-06-21",
"authors": [
"Fangxin Shang",
"Yehui Yang"
],
"categories": [
"cs.AI",
"cs.SE"
],
"url": "https://arxiv.org/abs/2606.22330",
"pdf_url": "https://arxiv.org/pdf/2606.22330",
"path": "papers/items/2026-2606-22330-hypothesis-driven-skill-optimization-for-llm-agents.md",
"local_title": "Hypothesis-Driven Skill Optimization for LLM Agents",
"local_status": "queued",
"local_topics": [
"agent-safety",
"coding-agent",
"memory",
"planning",
"reasoning",
"tool-use"
],
"collection_queries": "planning-agent",
"collection_score": "17"
},
{
"arxiv_id": "2606.22417",
"title": "Code Isn't Memory: A Structural Codebase Index Inside a Coding Agent",
"abstract": "Coding agents now interleave LLMs with retrieval over the working repository, and retrieval implementations vary widely across deployed harnesses. Inside a fixed coding-agent harness on a fixed model, does adding a structural codebase index actually change cost or resolve? We ran three arms (the harness with the index, the same harness without it, and an agentic-grep comparator) on SWE-PolyBench Verified and SWE-bench Pro with Claude Opus 4.7 held fixed throughout, across three seeds, inside a leak-audited per-task sandbox. The within-harness ablation produces a large localization gain and a statistically separated resolve gain, with no cost penalty per cell and lower cost per solve. The cross-harness check shows that the index does not regress against an agentic-grep baseline on resolve or localization, again at no cost penalty. We release the per-cell exclusion ledger, the leak-audit script, the localization extractor, and the results database. The deployment question for a structural codebase index is thus not whether it is too expensive to run (across seeds, the index lands at a lower $/solved than agentic grep) but whether the workload includes multi-file changes where structural ranking pays off.",
"published": "2026-06-21",
"updated": "2026-06-21",
"authors": [
"Ishaan Bhola",
"Adithyan Krishnan",
"Sravanth Kurmala",
"Mukunda NS"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2606.22417",
"pdf_url": "https://arxiv.org/pdf/2606.22417",
"path": "papers/items/2026-2606-22417-code-isn-t-memory-a-structural-codebase-index-inside-a-coding-agent.md",
"local_title": "\"Code Isn't Memory: A Structural Codebase Index Inside a Coding Agent\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"coding-agent",
"memory",
"rag"
],
"collection_queries": "coding-agent",
"collection_score": "19"
},
{
"arxiv_id": "2606.22610",
"title": "PaperClaw: Harnessing Agents for Autonomous Research and Human-in-the-Loop Refinement",
"abstract": "Large language models have become capable reasoners and tool users that write and run code and search the literature, which makes automating the research process itself a realistic goal. We present PAPERCLAW, a harnessed multi-agent system that carries a project autonomously, from a field of study to a finished paper. PAPERCLAW curates a domain from a field's live literature, datasets, and code; brainstorms it into an idea with a pre-registered main-result contract; and drives a stoppable hypothesis map through an iterative propose, test, reflect loop that grows only from measured verdicts and halts once the evidence supports the idea, at which point it writes a venue-compliant paper. A full-lifecycle memory keeps each stage in a single living record, so a long run can be paused, inspected, and resumed without losing context. At the centre is an in-cycle research assistant with research tools and skills: it can drive the whole pipeline on its own, while the same interface lets a person step in at any stage, turning a first autonomous draft into a stronger paper through human-in-the-loop refinement. Throughout, PAPERCLAW keeps its output grounded and checkable, citing only references validated against open scholarly indexes and reporting results that genuinely ran. An evaluation with an LLM judge finds that PAPERCLAW produces strong papers both fully autonomously and with human-in-the-loop refinement.",
"published": "2026-06-21",
"updated": "2026-06-21",
"authors": [
"Weiwei Ye",
"Hangchen Liu",
"Dongyuan Li",
"Renhe Jiang"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2606.22610",
"pdf_url": "https://arxiv.org/pdf/2606.22610",
"path": "papers/items/2026-2606-22610-paperclaw-harnessing-agents-for-autonomous-research-and-human-in-the-loop-refine.md",
"local_title": "\"PaperClaw: Harnessing Agents for Autonomous Research and Human-in-the-Loop Refinement\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"multi-agent",
"tool-use"
],
"collection_queries": "multi-agent-llm",
"collection_score": "15"
},
{
"arxiv_id": "2606.22647",
"title": "RAVEN: Agentic RAG for Automated Vulnerability Repair",
"abstract": "Automated vulnerability repair has emerged as a promising direction to mitigate the growing number of software vulnerabilities. Recent advances in Large Language Models (LLMs) have further accelerated research in automated repair. However, existing frameworks remain largely restricted to memory-related vulnerabilities and locally repairable vulnerability settings, leaving generalization to unseen vulnerability types underexplored. Their evaluations are often limited to a single programming language, and largely rely on proprietary models. In this paper, we propose RAVEN, a scalable, efficient and autonomous framework that integrates an agentic retrieval-augmented generation (RAG) pipeline with controlled iterative repair in a unified framework. The framework utilizes open-source LLMs in a fully locally deployable setting with limited GPU requirements, while building a multi-faceted retrieval pipeline to retrieve historically relevant vulnerability fixes and guide the patch generation. In addition, RAVEN introduces a dedicated Curator Agent that retrieves cross-file dependencies from the target repository, to fix complex vulnerabilities that cannot be addressed using local vulnerable code alone. We evaluate RAVEN on 160 real-world CVE vulnerabilities across diverse vulnerability types, two programming languages, unseen CWE categories, and out-of-distribution settings. RAVEN achieves an overall repair success rate of 83.13%, outperforming all existing state-of-the-art repair frameworks, while also demonstrating strong generalization capabilities and maintaining the repair cost negligible.",
"published": "2026-06-21",
"updated": "2026-06-21",
"authors": [
"Varun Gadey",
"Zijie Liu",
"Alexandra Dmitrienko"
],
"categories": [
"cs.CR",
"cs.LG",
"cs.SE"
],
"url": "https://arxiv.org/abs/2606.22647",
"pdf_url": "https://arxiv.org/pdf/2606.22647",
"path": "papers/items/2026-2606-22647-raven-agentic-rag-for-automated-vulnerability-repair.md",
"local_title": "\"RAVEN: Agentic RAG for Automated Vulnerability Repair\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"coding-agent",
"computer-use",
"memory",
"rag"
],
"collection_queries": "rag-agent",
"collection_score": "17"
},
{
"arxiv_id": "2606.22737",
"title": "GroundEval: A Deterministic Replacement for LLM-as-Judge in Stateful Agent Evaluation",
"abstract": "Before letting an agent operate over real context, can you prove it used the right evidence? GroundEval turns that question into a deterministic test of what the agent searched, fetched, cited, and was permitted to access. In one case study, two frontier LLM judges scored a plausible agent response 0.85 and higher. But the trace told a different story: the agent had never retrieved the artifact its answer depended on, yielding a GroundEval score of 0.000. We introduce GroundEval, a judge-free framework for evaluating agents against grounded, time-bounded, and access-controlled evidence. GroundEval uses a domain configuration to generate questions, lets the agent choose how to answer, and then scores both the final answer and the recorded trajectory that produced it. The benchmark targets three failures that LLM-as-judge evaluation struggles to detect: whether an agent checked before claiming absence, reasoned only from evidence available to the actor at the relevant time, and used the correct causal mechanism rather than a plausible one. These correspond to three tracks: Silence, Perspective, and Counterfactual. GroundEval exposes when plausible answers rest on invalid evidence paths, and produces structured per-question diagnostics that pair tool activity with the agent's turn-level narration, making each score inspectable rather than merely reported. Our case studies suggest this failure mode is common rather than exceptional, one that final-answer and judge-based evaluation cannot detect by construction.",
"published": "2026-06-22",
"updated": "2026-07-02",
"authors": [
"Jeffrey Flynt"
],
"categories": [
"cs.AI",
"cs.CL",
"cs.SE"
],
"url": "https://arxiv.org/abs/2606.22737",
"pdf_url": "https://arxiv.org/pdf/2606.22737",
"path": "papers/items/2026-2606-22737-groundeval-a-deterministic-replacement-for-llm-as-judge-in-stateful-agent-evalua.md",
"local_title": "\"GroundEval: A Deterministic Replacement for LLM-as-Judge in Stateful Agent Evaluation\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"tool-use"
],
"collection_queries": "agent-evaluation",
"collection_score": "14"
},
{
"arxiv_id": "2606.22844",
"title": "RaMem: Contextual Reinstatement for Long-term Agentic Memory",
"abstract": "Long-term memory has become increasingly important for LLM agents that operate across extended interactions and evolving task contexts. Recent memory systems have made past experiences more persistent, compact, and retrievable, but retrieval alone does not ensure that a memory provides valid evidence for the current query. When experiences are compressed into reusable fragments, memories from different situations may appear equally relevant if they involve recurring entities or user states. We refer to this failure as context collapse: memories lose the surrounding context needed to judge whether they provide valid evidence for the current query. To address this problem, we propose Contextual Reinstatement for Agentic Memory (RaMem), a framework that turns retrieved memory fragments into contextually verifiable evidence. RaMem operates through four coordinated stages: (i) evidence anchoring grounds each memory in its original episodic conditions, especially event time, mention time, session span, and participants; (ii) recall condition induction derives the evidence conditions implied by the query; (iii) validity-aware retrieval uses these conditions to prioritize context-compatible memories while retaining content-relevant candidates as fallback evidence; and (iv) context-preserved synthesis keeps the selected memories' structured context available to the generator. Experiments on long-term memory benchmarks show that RaMem consistently improves performance over strong memory baselines, with average F1 gains of more than 10% across several backbones.",
"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"
],
"url": "https://arxiv.org/abs/2606.22844",
"pdf_url": "https://arxiv.org/pdf/2606.22844",
"path": "papers/items/2026-2606-22844-ramem-contextual-reinstatement-for-long-term-agentic-memory.md",
"local_title": "\"RaMem: Contextual Reinstatement for Long-term Agentic Memory\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "19"
},
{
"arxiv_id": "2606.23130",
"title": "Understanding the (In)Security of Vibe-Coded Applications",
"abstract": "Recent advances in large language models (LLMs) have enabled vibe coding, an emerging software development paradigm in which users create applications primarily through natural-language interactions with AI agents. Due to its low barrier to entry, vibe coding is rapidly gaining adoption in practice. Unlike conventional AI-assisted programming, where developers remain responsible for implementation and code review, vibe coding delegates a substantial portion of development to AI systems. This shift raises a fundamental question: how (in)secure are applications developed through vibe coding? In this paper, we conduct a systematic study of the security of vibe-coded applications. We collect a large corpus of real-world applications developed using popular AI agents and design a vulnerability analysis framework that combines agent-assisted code auditing with human validation. Using this framework, we examine the prevalence, severity, and root causes of vulnerabilities in the deployed vibe-coded applications. Our study reveals several key findings: (1) vibe-coded applications exhibit recurring vulnerability patterns that differ from those commonly observed in conventional software development workflows, including placeholder logic, unfiltered input, and secret exposure; (2) these vulnerabilities arise from systematic limitations of AI agents throughout the vibe-coding lifecycle, such as memory loss, locally optimized objectives and insufficient security knowledge; and (3) while advances in LLM capabilities and improved prompting strategies can reduce the incidence of vulnerabilities, they do not eliminate the underlying security risks. Overall, our study provides an empirical understanding of the security landscape of vibe-coded applications and lays the groundwork for addressing the security challenges introduced by the growing delegation of software development to AI systems.",
"published": "2026-06-22",
"updated": "2026-06-23",
"authors": [
"Junquan Deng",
"Zhiyu Fan",
"Ruijie Meng"
],
"categories": [
"cs.CR",
"cs.SE"
],
"url": "https://arxiv.org/abs/2606.23130",
"pdf_url": "https://arxiv.org/pdf/2606.23130",
"path": "papers/items/2026-2606-23130-understanding-the-in-security-of-vibe-coded-applications.md",
"local_title": "Understanding the (In)Security of Vibe-Coded Applications",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"coding-agent",
"memory",
"tool-use",
"workflow-agent"
],
"collection_queries": "ai-agent",
"collection_score": "13"
},
{
"arxiv_id": "2606.23195",
"title": "Memory Contagion: Cross-Temporal Propagation of Evaluator Bias via Agent Memory",
"abstract": "Large Language Model (LLM) agents increasingly rely on memory systems to maintain long-term coherence. Recent work shows that agent memories degrade during continuous consolidation. However, existing research assumes memories are derived from unbiased experiences. In this work, we identify and formalize a novel phenomenon: Memory Contagion -- the cross-temporal propagation of evaluator bias through agent memory. We show that when agents are trained or guided by biased evaluators, their experiences become biased; when these trajectories are stored and consolidated into memory, the bias propagates to future agents retrieving from the same memory store, even when consolidation is perfect (oracle). Across two bias types (length preference, authority bias) and four experimental phases, we demonstrate: (1) Memory Contagion occurs for length bias even with perfect consolidation on older models (Gamma_A = 13.18, DeepSeek V4-Chat), while newer models (V4-Pro, Claude) are immune, proving both that biased input is a sufficient cause and that contagion is model-generation-dependent; (2) authority bias fails to propagate in all 15 controlled multi-seed experiments (Gamma_A = 0.00), revealing that not all evaluator biases can cross temporal boundaries through current memory architectures; (3) No observed safe threshold: length bias propagation is detected at contamination rates as low as p=0.2. Our findings expose a critical but contingent vulnerability in current agent memory designs and provide formal tools for measuring cross-temporal bias propagation.",
"published": "2026-06-22",
"updated": "2026-06-24",
"authors": [
"Zewen Liu"
],
"categories": [
"cs.LG",
"cs.AI",
"cs.CL"
],
"url": "https://arxiv.org/abs/2606.23195",
"pdf_url": "https://arxiv.org/pdf/2606.23195",
"path": "papers/items/2026-2606-23195-memory-contagion-cross-temporal-propagation-of-evaluator-bias-via-agent-memory.md",
"local_title": "\"Memory Contagion: Cross-Temporal Propagation of Evaluator Bias via Agent Memory\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "14"
},
{
"arxiv_id": "2606.23283",
"title": "Towards Root Memories: Benchmarking and Enhancing Implicit Logical Memory Retrieval for Personalized LLMs",
"abstract": "Memory systems are essential for personalized Large Language Models (LLMs). However, existing retrieval methods in these systems primarily rely on semantic similarity, potentially missing logically critical memories with limited semantic overlap. Current benchmarks remain inadequate for evaluating this problem. To address this gap, we construct IMLogic, the first high-quality benchmark targeting implicit logical memory retrieval in long-dialogue scenarios. Motivated by this challenge, we introduce root memory, a structured, decision-preserving representation that distills reusable personalized logic from long-term user histories. We then propose RootMem, a plug-and-play framework that first distills raw histories into structured root memories and then uses an LLM-based router to activate logically relevant ones, complementing semantic retrieval with personalized decision logic. Extensive experiments demonstrate that RootMem significantly outperforms the strongest retrieval baselines and consistently boosts the accuracy of existing memory agents. Our benchmark and codes will be available at https://anonymous.4open.science/r/IMLogic-DBB3.",
"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"
],
"url": "https://arxiv.org/abs/2606.23283",
"pdf_url": "https://arxiv.org/pdf/2606.23283",
"path": "papers/items/2026-2606-23283-towards-root-memories-benchmarking-and-enhancing-implicit-logical-memory-retriev.md",
"local_title": "\"Towards Root Memories: Benchmarking and Enhancing Implicit Logical Memory Retrieval for Personalized LLMs\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag"
],
"collection_queries": "agent-memory",
"collection_score": "14"
},
{
"arxiv_id": "2606.23565",
"title": "HoloAgent-0: A Unified Embodied Agent Framework with 3D Spatial Memory",
"abstract": "LLM agents follow a practical execution loop in digital environments: they reason over structured states, invoke tools, inspect feedback, and revise actions. Extending this loop to physical robots is difficult because physical execution is continuous, embodiment-dependent, uncertain, and constrained by safety. Existing embodied-AI systems have advanced manipulation, spatial understanding, navigation, and humanoid control, but these capabilities often remain specialized modules or loosely coupled decision loops. In this work, we introduce HoloAgent-0, a unified embodied agent framework for real-world robot deployment. Embodied AgentOS converts language instructions into executable skill graphs, schedules robot resources, monitors execution, and triggers clarification or re-planning from runtime feedback. HoloAgent-0 organizes heterogeneous robot models and controllers through three coupled layers: Embodied AgentOS for closed-loop execution, 3D spatial memory for physical world grounding, and embodied skills for robot action. We deploy HoloAgent-0 on real hardware and evaluate its spatial memory, long-horizon navigation, and closed-loop execution across motion generation, object search, cross-robot coordination, and mobile manipulation.",
"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"
],
"url": "https://arxiv.org/abs/2606.23565",
"pdf_url": "https://arxiv.org/pdf/2606.23565",
"path": "papers/items/2026-2606-23565-holoagent-0-a-unified-embodied-agent-framework-with-3d-spatial-memory.md",
"local_title": "\"HoloAgent-0: A Unified Embodied Agent Framework with 3D Spatial Memory\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"computer-use",
"embodied-agent",
"memory",
"planning",
"rag",
"tool-use"
],
"collection_queries": "planning-agent",
"collection_score": "19"
},
{
"arxiv_id": "2606.23752",
"title": "ESAA-Conversational: An Event-Sourced Memory Layer for Continuity, Handoff, and Curation Across Heterogeneous LLM Coding Agents",
"abstract": "Software developers increasingly work with multiple LLM coding agents, switching among tools such as Codex, Grok, Claude Code, and other assistants as context windows fill, sessions end, or a particular agent becomes better suited to a subtask. Each agent, however, persists its conversation in a private and vendor-specific log. The result is conversational state drift: goals, decisions, open tasks, and rationales established with one agent are not reliably available when another agent takes over. This paper presents \\emph{ESAA-Conversational}, a domain specialization of Event-Sourcing Agent Architecture (ESAA)~\\cite{esaa} for shared conversational memory across heterogeneous agents. The method treats the visible conversation as a local event store: hooks and watchers capture visible turns, normalize them into an append-only \\texttt{activity.jsonl}, and deterministically project read models such as \\texttt{handoff.md}, \\texttt{state.md}, \\texttt{decisions.md}, and \\texttt{tasks.json}. Mechanical capture does not require LLM inference; agents use judgment only for explicit curation, recording durable decisions and conversational tasks through domain commands. The public v1.1.0 release implements a PowerShell CLI with \\texttt{init}, \\texttt{enable-hooks}, \\texttt{sync}, \\texttt{project}, \\texttt{verify}, \\texttt{context}, \\texttt{decide}, and \\texttt{task}; includes \\texttt{workspace\\_root} isolation and a write-path lockfile; and is distributed as a greenfield package with an empty public log. A self-referential case study with 570 development-lab events shows that heterogeneous agents can collaborate through a shared log without a direct agent-to-agent channel, while the public distribution preserves privacy by excluding the private conversational history.",
"published": "2026-06-22",
"updated": "2026-06-22",
"authors": [
"Elzo Brito dos Santos Filho"
],
"categories": [
"cs.SE"
],
"url": "https://arxiv.org/abs/2606.23752",
"pdf_url": "https://arxiv.org/pdf/2606.23752",
"path": "papers/items/2026-2606-23752-esaa-conversational-an-event-sourced-memory-layer-for-continuity-handoff-and-cur.md",
"local_title": "\"ESAA-Conversational: An Event-Sourced Memory Layer for Continuity, Handoff, and Curation Across Heterogeneous LLM Coding Agents\"",
"local_status": "queued",
"local_topics": [
"coding-agent",
"memory",
"tool-use"
],
"collection_queries": "coding-agent",
"collection_score": "16"
},
{
"arxiv_id": "2606.24937",
"title": "The Hitchhiker's Guide to Agentic AI: From Foundations to Systems",
"abstract": "The Hitchhiker's Guide to Agentic AI is a comprehensive practitioner's reference for building autonomous AI systems. The book covers the full stack from first principles to production deployment, organized around a central thesis: building great agentic systems requires understanding every layer of the pipeline, not just one. The book opens with the LLM substrate -- transformer architecture, GPU systems, training and fine-tuning (SFT,LoRA, MoE), model compression, and inference optimization -- treated as essential foundations rather than the primary focus. It then develops the alignment and reasoning layer: reinforcement learning from human feedback (RLHF), PPO, DPO and its variants, GRPO, reward modeling, and RL for large reasoning models including chain-of-thought and test-time scaling. The second half is devoted to agentic AI proper. Topics include agentic training and trajectory-based RL, retrieval-augmented generation (RAG and Agentic RAG), memory systems (in-context, external, episodic, and semantic), agent harness design and context management, and a taxonomy of agent design patterns. Inter-agent coordination is covered in depth: the Model Context Protocol (MCP), agent skills and tool use, the Agent-to-Agent (A2A) communication protocol, and multi-agent architectures spanning centralized, decentralized, and hierarchical topologies. The book concludes with agent development frameworks, agentic UI design, evaluation methodology for agentic tasks, and production deployment. Each chapter pairs rigorous theoretical foundations with implementation guidance, code examples, and references to the primary literature.",
"published": "2026-06-22",
"updated": "2026-06-22",
"authors": [
"Haggai Roitman"
],
"categories": [
"cs.AI",
"cs.CL",
"cs.IR",
"cs.LG"
],
"url": "https://arxiv.org/abs/2606.24937",
"pdf_url": "https://arxiv.org/pdf/2606.24937",
"path": "papers/items/2026-2606-24937-the-hitchhiker-s-guide-to-agentic-ai-from-foundations-to-systems.md",
"local_title": "\"The Hitchhiker's Guide to Agentic AI: From Foundations to Systems\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"computer-use",
"memory",
"multi-agent",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "agentic-ai, multi-agent-llm, rag-agent, tool-use",
"collection_score": "25"
},
{
"arxiv_id": "2606.24322",
"title": "Securing LLM-Agent Long-Term Memory Against Poisoning: Non-Malleable, Origin-Bound Authority with Machine-Checked Guarantees",
"abstract": "LLM agents increasingly rely on persistent long-term memory, which creates a critical vulnerability that we study here: memory poisoning. An adversary can store untrusted content in one session that later steers a consequential action, such as a payment, a setting change, or data exfiltration, in a future session. Existing defenses base a memory item's authority to act on either its content (detection or trust-scoring) or its derivation history (lineage). We show that both signals are malleable. An attacker can launder an untrusted origin through three channels specific to LLM agents: the agent's own summarization, a trusted-tool echo, and manufactured corroboration. Each makes the content look benign and breaks or flips its derivation edge to ``trusted.'' We formalize malleability for the memory write-retrieve-act pipeline and prove a machine-checked separation theorem. No content- or lineage-based defense is sound under laundering (T1), write-time origin binding is necessary (T2), and non-malleable origin-bound authority with Sybil-resistant corroboration-gated elevation is sufficient (T3). Our construction, TMA-NM (Tamper-evident Memory Authority, Non-Malleable), instantiates non-malleable information-flow control (IFC) for LLM-agent memory. A cross-defense, cross-attack, and cross-model benchmark over eight frontier models shows that existing defenses fail exactly where the theory predicts (up to 68% laundering attack-success), while TMA-NM reaches 0% attack success on both direct and laundering attacks across all models and channels, at full legitimate utility. We release the benchmark, harness, and machine-checked TLA+ models to support reproducibility.",
"published": "2026-06-23",
"updated": "2026-06-23",
"authors": [
"Yedidel Louck"
],
"categories": [
"cs.CR"
],
"url": "https://arxiv.org/abs/2606.24322",
"pdf_url": "https://arxiv.org/pdf/2606.24322",
"path": "papers/items/2026-2606-24322-securing-llm-agent-long-term-memory-against-poisoning-non-malleable-origin-bound.md",
"local_title": "\"Securing LLM-Agent Long-Term Memory Against Poisoning: Non-Malleable, Origin-Bound Authority with Machine-Checked Guarantees\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "15"
},
{
"arxiv_id": "2606.24416",
"title": "Agentic AI for Bilevel Long-Term Optimization of Policy-Driven Physical Layer Systems",
"abstract": "Network operators' changing policies, service requirements, and stringent real-time constraints render existing methods designed with fixed objectives and constraints ineffective. This paper presents Agentic long-term performance optimization (Agentic-LTPO), a nested bilevel optimization framework that can be applied to adaptive physical layer problem configuration. The key idea is to employ agentic AI to generate upper-level configurations in a bilevel optimization structure, where evolving operator policies, environment summaries, and historical experiences are translated into structured lower-level optimization problem configurations. The lower level solves the problems with updated configurations for real-time physical-layer decisions. Considering cell-free MIMO beamforming as a use case, we embody Agentic-LTPO by designing a new multi-agent decision process with retrieval-augmented experience-based verification in the upper level, together with a closed-form beamformer in the lower level. Experiments demonstrate that Agentic-LTPO exhibits strong adaptability to dynamic operator policies and effectively enhances the system's long-term performance by 57.2% compared to traditional methods.",
"published": "2026-06-23",
"updated": "2026-06-23",
"authors": [
"Bingnan Xiao",
"Chenhao Yang",
"Wei Ni",
"Xin Wang",
"Tony Q. S. Quek"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2606.24416",
"pdf_url": "https://arxiv.org/pdf/2606.24416",
"path": "papers/items/2026-2606-24416-agentic-ai-for-bilevel-long-term-optimization-of-policy-driven-physical-layer-sy.md",
"local_title": "Agentic AI for Bilevel Long-Term Optimization of Policy-Driven Physical Layer Systems",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"multi-agent",
"rag"
],
"collection_queries": "agentic-ai",
"collection_score": "13"
},
{
"arxiv_id": "2606.24437",
"title": "ReM-MoA: Reasoning Memory Sustains Mixture-of-Agents Scaling",
"abstract": "Mixture-of-Agents (MoA) architectures improve inference-time scaling by organizing multiple LLM agents into layered reasoning pipelines. However, existing MoA variants fail to sustain gains as depth increases, exhibiting degradation, early plateauing, or saturation. We propose ReM-MoA, a memory-augmented MoA framework that sustains scaling through two mechanisms: (1) a Ranked Reasoning Memory that persistently stores and ranks reasoning traces from all layers using a comparative Reviewer Agent, and (2) a Curated Diversified Memory Routing scheme that exposes different agents to distinct combinations of successful and failed traces, preserving exploration diversity while propagating high-quality reasoning. We further introduce an optional multi-domain Reviewer distillation pipeline that improves ranking quality through frontier-model supervision. Across five reasoning benchmarks spanning math, formal logic, code, knowledge, and commonsense, ReM-MoA consistently outperforms prior MoA variants across both depth and width scaling, and its advantage widens with depth, establishing structured cross-layer reasoning memory as a key missing mechanism for scalable multi-agent inference.",
"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"
],
"url": "https://arxiv.org/abs/2606.24437",
"pdf_url": "https://arxiv.org/pdf/2606.24437",
"path": "papers/items/2026-2606-24437-rem-moa-reasoning-memory-sustains-mixture-of-agents-scaling.md",
"local_title": "\"ReM-MoA: Reasoning Memory Sustains Mixture-of-Agents Scaling\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"multi-agent",
"reasoning"
],
"collection_queries": "multi-agent-llm",
"collection_score": "17"
},
{
"arxiv_id": "2606.24535",
"title": "Governed Shared Memory for Multi-Agent LLM Systems",
"abstract": "Multi-agent LLM environments require robust mechanisms for shared knowledge management. This paper formalizes the fleet-memory problem and identifies four foundational failure modes: unauthorized leakage, stale propagation, contradiction persistence, and provenance collapse. To address these, we define explicit systems-level primitives: scoped retrieval, temporal supersession, provenance tracking, and policy-governed memory propagation. These primitives are implemented in MemClaw, a production multi-tenant memory service, and evaluated via ArgusFleet, a reproducible harness testing four governance dimensions. Rather than a baseline comparison, this study measures a live production service, emphasizing real-world architectural insights and negative results. Key Evaluation Results Provenance: Successfully reconstructed 100% of depth-four derivation chains with correct writer identity at sub-second per-hop latency. Propagation: Demonstrated high intra-fleet visibility with zero cross-fleet leakage. Under strong write mode, write-to-visible latency was optimized to a single search round-trip. Production Architectural Issues Discovered Asymmetric Scope Enforcement: Tenant isolation held, but sub-tenant scope was initially bypassed on direct GET-by-id requests for agent-scoped credentials (disclosed and remediated during the study). Pipeline Ordering Conflict: While contradiction supersession works for admitted writes, a synchronous near-duplicate gate can prematurely reject contradictory writes before the asynchronous contradiction detector can evaluate them. Conclusion: Long-context retrieval alone is insufficient for production multi-agent memory. Governed shared memory demands explicit systems-level abstractions, and live evaluation is vital to expose enforcement and pipeline-ordering failures missed by design-only treatments.",
"published": "2026-06-23",
"updated": "2026-06-23",
"authors": [
"Yanki Margalit",
"Nurit Cohen-Inger",
"Erni Avram",
"Ran Taig",
"Oded Margalit"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2606.24535",
"pdf_url": "https://arxiv.org/pdf/2606.24535",
"path": "papers/items/2026-2606-24535-governed-shared-memory-for-multi-agent-llm-systems.md",
"local_title": "Governed Shared Memory for Multi-Agent LLM Systems",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"multi-agent",
"rag",
"tool-use"
],
"collection_queries": "agent-memory, multi-agent-llm",
"collection_score": "20"
},
{
"arxiv_id": "2606.24595",
"title": "MEMPROBE: Probing Long-Term Agent Memory via Hidden User-State Recovery",
"abstract": "Long-term memory promises LLM agents that grow more capable across sessions, maintaining an accurate, evolving understanding of the user that interaction forms. In practice, however, this memory is evaluated mostly through downstream behavior, such as later answers, personalization quality, or task success, which tests that understanding only indirectly and leaves the memory artifact itself largely unaudited. We argue that long-term memory should instead be evaluated as an auditable post-interaction artifact: after ordinary assistance, what structured user state can be reconstructed from the memory the agent leaves behind? We instantiate this view in MEMPROBE, a benchmark in which a memory-equipped agent assists simulated users, each carrying a hidden, taxonomy-anchored user-state bank, across a trajectory of leak-controlled tasks, after which that bank is reconstructed from the agent's resulting memory under both full-store and top-k access. Built on synthetic ground truth for efficient, scalable measurement, MEMPROBE spans 50 simulated users with 31 hidden dimensions each (1,550 recovery targets) and tests 5 representative memory systems. Testing state-of-the-art memory agents, we find that successful assistance and recoverable memory behave as distinct capabilities. Task completion nearly saturates, even for a memoryless baseline, while category-balanced recovery stays moderate (about 0.6) and drops further under top-k retrieval. MEMPROBE is the first benchmark to study memory recovery directly, reconstructing the user state a system retains and scoring it against ground truth. We see recovery as a concrete objective for future memory agents to optimize, and MEMPROBE as a step toward an environment where agents are trained to remember their users, growing more faithful the longer they know them.",
"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"
],
"url": "https://arxiv.org/abs/2606.24595",
"pdf_url": "https://arxiv.org/pdf/2606.24595",
"path": "papers/items/2026-2606-24595-memprobe-probing-long-term-agent-memory-via-hidden-user-state-recovery.md",
"local_title": "\"MEMPROBE: Probing Long-Term Agent Memory via Hidden User-State Recovery\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "16"
},
{
"arxiv_id": "2606.24626",
"title": "SAFARI: Scaling Long Horizon Agentic Fault Attribution via Active Investigation",
"abstract": "As autonomous agents tackle increasingly complex multi-step, multi-agent tasks, their execution trajectories have scaled beyond the constraints of even the largest context windows. Current methods for effectively diagnosing agent failures load the full trajectory into an LLM's context window, which suffers from attention dilution and fails when agentic traces inevitably exceed context limits. To address this, we introduce SAFARI (Scaling long-horizon Agentic Fault AttRibution via active Investigation), a framework that replaces linear context loading with a tool-augmented diagnostic loop. By equipping LLMs with a specialized toolbox to read and search trajectory segments alongside a persistent Short-Term Memory (STM) for cross-turn reasoning, SAFARI effectively decouples diagnostic accuracy from architectural context limits. Our experiments demonstrate that SAFARI outperforms state-of-the-art results by 20% on the Who&When dataset within a 1M token budget, and by 19% on TRAIL GAIA subset on a 25K token budget. Most significantly, SAFARI maintains a 0.58 precision even when the target fault resides 5x beyond the model's native context window, a scenario where traditional evaluators fail entirely.",
"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"
],
"url": "https://arxiv.org/abs/2606.24626",
"pdf_url": "https://arxiv.org/pdf/2606.24626",
"path": "papers/items/2026-2606-24626-safari-scaling-long-horizon-agentic-fault-attribution-via-active-investigation.md",
"local_title": "\"SAFARI: Scaling Long Horizon Agentic Fault Attribution via Active Investigation\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"multi-agent",
"planning",
"reasoning",
"tool-use"
],
"collection_queries": "autonomous-agent-llm, multi-agent-llm",
"collection_score": "19"
},
{
"arxiv_id": "2606.24775",
"title": "Are We Ready For An Agent-Native Memory System?",
"abstract": "Memory for large language model (LLM) agents has rapidly evolved from simple retrieval-augmented mechanisms into a data management system that supports persistent information storage, retrieval, update, consolidation, and dynamic lifecycle governance throughout agent execution. Despite this evolution, existing evaluations still benchmark agent memory mainly through end-to-end task success metrics (e.g., F1, BLEU), while treating the underlying system as a monolithic black box. As a result, critical system-level concerns, including operational costs, architectural trade-offs across memory modules, and robustness under dynamic knowledge updates, remain insufficiently explored. In this paper, we present a systematic experimental study of agent memory from a data management perspective. We propose an analytical framework that decomposes agent memory into four core modules: memory representation and storage, extraction, retrieval and routing, and maintenance. Under this framework, we evaluate 12 representative memory systems and two reference baselines across five benchmark workloads spanning 11 datasets. Our extensive end-to-end evaluation shows that no single architecture dominates across all scenarios; instead, effectiveness depends heavily on how well the memory structure aligns with the workload bottleneck. Furthermore, through fine-grained ablation studies, we quantify their individual effects on representation fidelity, retrieval precision, update correctness, and long-horizon stability. Finally, we reveal cost-performance trade-offs under realistic workloads, showing localized maintenance is more cost-efficient than global reorganization. Based on these findings, we identify promising directions towards building truly agent-native memory systems. The code is publicly available at https://github.com/OpenDataBox/MemoryData.",
"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"
],
"url": "https://arxiv.org/abs/2606.24775",
"pdf_url": "https://arxiv.org/pdf/2606.24775",
"path": "papers/items/2026-2606-24775-are-we-ready-for-an-agent-native-memory-system.md",
"local_title": "Are We Ready For An Agent-Native Memory System?",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"planning",
"rag",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "17"
},
{
"arxiv_id": "2606.24779",
"title": "DeepBD: A Grounded Agentic Workflow for Variant Prioritization and Diagnosis of Genetic Birth Defects",
"abstract": "Birth defects are a major cause of fetal loss, neonatal morbidity and long-term disability. In the subset with suspected genetic etiologies, exome and genome sequencing have moved many cases from variant detection to post-sequencing interpretation: clinicians must rank patient-specific candidate variants under incomplete fetal or infant phenotypes and heterogeneous evidence from population genetics, variant-effect prediction, gene-disease validity, phenotype ontologies, cellular and pathway context, protein structure and clinical literature. We present DeepBD, a grounded agentic workflow for variant prioritization and diagnostic interpretation of genetic birth defects. DeepBD organizes the workflow into LLM-assisted case structuring, a pretrained evidence engine, specialist evidence modules and a grounded diagnostic review layer. The evidence engine learns patient-specific variant scores from structured rule evidence, sequence and variant-effect representations and phenotype-conditioned biological context, whereas specialist modules and the agentic layer provide tool-based refinement, candidate-pool review and diagnosis-oriented synthesis from ranked candidates. Developed using an in-house fetal and infant cohort comprising 18,622 cases, DeepBD achieved Recall@1/3/5/10 of 0.658/0.882/0.912/0.929 on an internal held-out solved-case benchmark, outperforming standalone Exomiser, DeepRare and prompted LLM reranking baselines evaluated on Exomiser-derived top-20 candidate variants. Ablation and overlap analyses show that rule evidence, mechanistic context, and specialist refinement provide complementary signals. These findings support a grounded agentic workflow that separates evidence integration, tool-based refinement, and LLM-assisted diagnostic review for retrospective variant prioritization in genetic birth defects.",
"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"
],
"url": "https://arxiv.org/abs/2606.24779",
"pdf_url": "https://arxiv.org/pdf/2606.24779",
"path": "papers/items/2026-2606-24779-deepbd-a-grounded-agentic-workflow-for-variant-prioritization-and-diagnosis-of-g.md",
"local_title": "\"DeepBD: A Grounded Agentic Workflow for Variant Prioritization and Diagnosis of Genetic Birth Defects\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"tool-use",
"workflow-agent"
],
"collection_queries": "agentic-ai",
"collection_score": "15"
},
{
"arxiv_id": "2606.25115",
"title": "Forget to Improve: On-Device LLM-Agent Continual Learning via Budget-Curated Memory",
"abstract": "On-device language-model agents improve by accumulating experience in retrieved memory rather than by updating weights. This memory is hard-bounded and exposed: it consumes RAM and energy, reaches peers through a thin uplink, and becomes an attack surface because it is writable by what the agent reads. Existing systems each cover one part of this problem: agentic memories grow without a budget, on-device methods keep entries by success alone, and poisoning is studied mainly as an attack rather than as a memory-governance problem. We propose \\sys{}, a single net-value-per-byte score that governs an agent's experience-memory lifecycle. The main idea is to let the budget act as the curator: each entry is scored as value minus harm, per byte, so one ruler decides what to keep, share, and trust. \\sys{} makes three decisions: (1) \\textbf{KEEP} evicts low-value bytes under the RAM and energy budget; (2) \\textbf{SHARE} sends an insight only when its value exceeds its uplink cost; and (3) \\textbf{TRUST} gates a peer entry by provenance. On language-model-agent task-drift benchmarks and a real heterogeneous Jetson testbed with two robot-arm nodes and a hub, \\sys{} reduces memory by $2.7\\times$ and uplink by $2.4\\times$, drives injection success from 0.75 to zero, and raises accuracy on cases corrupted by poison or stale memory. Curating by net value reduces footprint, energy, uplink, and injection success together without reducing accuracy. In this setting, forgetting by net value improves the agent rather than weakening it.",
"published": "2026-06-23",
"updated": "2026-06-23",
"authors": [
"Beining Wu",
"Zihao Ding",
"Jun Huang",
"Yanxiao Zhao"
],
"categories": [
"cs.LG",
"cs.NI"
],
"url": "https://arxiv.org/abs/2606.25115",
"pdf_url": "https://arxiv.org/pdf/2606.25115",
"path": "papers/items/2026-2606-25115-forget-to-improve-on-device-llm-agent-continual-learning-via-budget-curated-memo.md",
"local_title": "\"Forget to Improve: On-Device LLM-Agent Continual Learning via Budget-Curated Memory\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"embodied-agent",
"memory"
],
"collection_queries": "agent-memory",
"collection_score": "15"
},
{
"arxiv_id": "2606.25161",
"title": "TRUSTMEM: Learning Trustworthy Memory Consolidation for LLM Agents with Long-Term Memory",
"abstract": "Large language model (LLM) agents rely on long-term memory to support extended interactions and personalized assistance beyond finite context windows. Existing memory agents actively update external memory through generated write, revise, and delete operations, but these updates may omit important information, corrupt existing memory, or introduce unsupported hallucinated content. Once stored, such errors become persistent system-state failures that can affect future reasoning and generation. In this paper, we propose TrustMem, a framework designed to improve the trustworthiness of memory consolidation. TrustMem relies on a Memory Transition Verifier to evaluate the transition process of memory updates in terms of coverage, preservation, and faithfulness. It further constructs preference pairs among candidate updates under the same memory state, enabling preference-guided reinforcement learning to directly optimize memory updating behaviors. Extensive experiments demonstrate that TrustMem improves both memory utility and reliability: it achieves state-of-the-art results across MemoryAgentBench, HaluMem, and the Mem-alpha validation set, improves HaluMem memory extraction by 12.14 F1 points, and reduces transition-level omission, corruption, and hallucination by 40.1\\%, 79.1\\%, and 50.0\\%, respectively, compared with the strongest baseline for each error type.",
"published": "2026-06-23",
"updated": "2026-06-23",
"authors": [
"Tianyu Yang",
"Sudipta Paul",
"Vijay Srinivasan",
"Vivek Kulkarni",
"Srinivas Chappidi"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2606.25161",
"pdf_url": "https://arxiv.org/pdf/2606.25161",
"path": "papers/items/2026-2606-25161-trustmem-learning-trustworthy-memory-consolidation-for-llm-agents-with-long-term.md",
"local_title": "\"TRUSTMEM: Learning Trustworthy Memory Consolidation for LLM Agents with Long-Term Memory\"",
"local_status": "skimmed",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "19"
},
{
"arxiv_id": "2606.25206",
"title": "RAVEN: Long-Horizon Reasoning & Navigation with a Visuo-Spatio-Temporal Memory",
"abstract": "Long-term robot deployment requires a compact and scalable memory that preserves fine-grained visual semantics, grounds observations in space and time, and enables efficient storage and retrieval. In this paper, we propose RAVEN, an agentic memory system for long-horizon robotic question answering and navigation. RAVEN stores visual embeddings with pose and time in a vector database, and grounds retrieval in a spatial map to answer queries and navigate to goals. By operating directly on visual embeddings, RAVEN avoids lossy image-to-text captioning and enables accurate semantic, spatial, and temporal retrieval at scale. Across several simulated and real-world video question-answering benchmarks, RAVEN consistently surpasses caption-based memory systems and matches frontier VLMs on long-horizon tasks at 10$\\times$ lower retrieval cost. Finally, we instantiate RAVEN on a Unitree Go1 robot for the task of long-horizon navigation for natural language goal-reaching, and show successful deployment over several large indoor environments.",
"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"
],
"url": "https://arxiv.org/abs/2606.25206",
"pdf_url": "https://arxiv.org/pdf/2606.25206",
"path": "papers/items/2026-2606-25206-raven-long-horizon-reasoning-navigation-with-a-visuo-spatio-temporal-memory.md",
"local_title": "\"RAVEN: Long-Horizon Reasoning & Navigation with a Visuo-Spatio-Temporal Memory\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"embodied-agent",
"memory",
"planning",
"rag",
"reasoning"
],
"collection_queries": "agent-memory",
"collection_score": "14"
},
{
"arxiv_id": "2606.25334",
"title": "Bridging the Post-discharge Gap: A Traceable Multi-agent Framework for Safe and Continuous Care",
"abstract": "Post-discharge clinical follow-up is critical for maintaining continuity of care and mitigating long-term health risks. However, traditional follow-up paradigms suffer from shortage of health workforce, fragmented patient histories, and information silos across clinical departments. While large language models have demonstrated potential in medical question-answering, their deployment in continuous care is hindered by hallucination risks and a fundamental inability to reason over longitudinal, patient-specific constraints. Here we present Healink, a memory-enhanced multi-agent framework to support AI-assisted post-discharge follow-up by generating prescription-grounded, traceable responses that improved completeness and perceived clinical utility in retrospective and physician-blinded evaluations. The architecture seamlessly integrates a triage routing mechanism, a unified memory enhancement module utilizing a robust relational database for optimal latency, and a strict constraint-based retrieval-augmented generation engine. By vectorizing historical clinical records and employing weighted similarity functions across diverse phenotypic and intervention dimensions, Healink ensures precise inter-patient and intra-patient case matching while actively preventing cross-departmental drug conflicts. We evaluated Healink on a dataset comprising 400 continuous and 85 highly complex real-world follow-up cases, alongside the webMedQA benchmark. In a rigorous single-blind evaluation conducted by clinical experts, the framework outperformed human physician baselines in both authoritativeness and clinical safety. By generating a traceable, white-box evidence chain, Healink provides a scalable, safe, and highly effective paradigm for intelligent patient management, ultimately enhancing societal healthcare outcomes.",
"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"
],
"url": "https://arxiv.org/abs/2606.25334",
"pdf_url": "https://arxiv.org/pdf/2606.25334",
"path": "papers/items/2026-2606-25334-bridging-the-post-discharge-gap-a-traceable-multi-agent-framework-for-safe-and-c.md",
"local_title": "\"Bridging the Post-discharge Gap: A Traceable Multi-agent Framework for Safe and Continuous Care\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"multi-agent",
"rag"
],
"collection_queries": "multi-agent-llm",
"collection_score": "18"
},
{
"arxiv_id": "2606.25361",
"title": "Memory Makes the Difference: Evaluating How Different Memory Roles Shape Conversational Agents",
"abstract": "Prior research on memory mechanism in RAG-based conversational system has emphasized how memory is stored and retrieved. However, far less is known about how memories with different functional roles influence response quality. Specifically, how they shape an agent's responses under varying conversational contexts and whether they lead to substantively different response behaviors. Existing evaluations in conversational system are also largely reference-based, insufficiently capturing the nuances in responses that may address users' preferences differently. In this work, we probe the impact of different memory types in shaping agents' responses. We present a fine-grained taxonomy of conversational memory, classify retrieved memories into different role types, and design a user-centric evaluation framework that simulates user perspectives. Through comparative experiments on long-term datasets and frontier LLMs, our analysis reveal many differentiated effects of memories: e.g., clarifying memory improves responses' factual accuracy and constraint awareness, making them more correct and personalized; irrelevant memory reduces topic relevance and degrades constraint awareness. Despite the power of frontier LLMs, these findings shed light on how different memory types can be leveraged to produce more personalized responses and inspire further research in this direction.",
"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"
],
"url": "https://arxiv.org/abs/2606.25361",
"pdf_url": "https://arxiv.org/pdf/2606.25361",
"path": "papers/items/2026-2606-25361-memory-makes-the-difference-evaluating-how-different-memory-roles-shape-conversa.md",
"local_title": "\"Memory Makes the Difference: Evaluating How Different Memory Roles Shape Conversational Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"tool-use"
],
"collection_queries": "rag-agent",
"collection_score": "15"
},
{
"arxiv_id": "2606.25656",
"title": "Is GraphRAG Needed? From Basic RAG to Graph-/Agentic Solutions with Context Optimization",
"abstract": "As advanced RAG variants like GraphRAG and Agentic RAG emerge, one leading question is when and how to use them. Here, we introduce a framework for different RAG scenarios evaluation and comparison on semi-structured knowledge bases, including regular RAG, GraphRAG, Modular RAG and Agentic RAG. We provide implementation for 9 standardized RAG scenarios, and conduct experiments for a comprehensive comparison. These scenarios are designed for real use cases regarding data and domain restrictions, spanning from simple document-based retrieval to advanced features such as hybrid text-graph retrieval, integration with computed or pre-defined domain knowledge graphs, agentic multi-step planning, and agent-graph integration. Besides, we present a novel context engineering method for GraphRAG and Agentic RAG, addressing the context/memory overflow issues, efficiently managing text and graph retrievals with new representations and agentic loop design, leading to 19%-53% reduction on token usage. Moreover, further analysis identifies a retrieval-generation gap where expanded retrieval does not proportionally improve generation quality, suggesting retrieval-oriented metrics overstate advanced retrieval benefits. This work provides data-driven insights on when and how to use them for building production-ready intelligent RAG systems.",
"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"
],
"url": "https://arxiv.org/abs/2606.25656",
"pdf_url": "https://arxiv.org/pdf/2606.25656",
"path": "papers/items/2026-2606-25656-is-graphrag-needed-from-basic-rag-to-graph-agentic-solutions-with-context-optimi.md",
"local_title": "Is GraphRAG Needed? From Basic RAG to Graph-/Agentic Solutions with Context Optimization",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"planning",
"rag"
],
"collection_queries": "rag-agent",
"collection_score": "14"
},
{
"arxiv_id": "2606.26403",
"title": "ProfileFoundry: A Synthetic Person-Object Substrate for Privacy, Memory, and Tool-Use Evaluation in LLM Agent",
"abstract": "Foundation-model research increasingly needs data about people: user state, personal histories, relationships, contact-like fields, documents, and longitudinal updates. Real user data is difficult to share, perturb, audit, or redistribute responsibly, while independently generated fake fields rarely preserve the cross-field and temporal consistency needed for controlled evaluation. We present PROFILEFOUNDRY, a deterministic generator and fixed reference release of 100,000 adult synthetic Person Objects across eight locales. Each object combines a typed current snapshot, household, family, and employer links, snapshot-aligned events, normalized relational views, and generation provenance. The release contains 709,228 events, 40,338 households, 52,491 employers, and 518,564 directed relationship edges. We report evidence in separate categories: selected population-marginal comparisons, per-object invariant checks, release-wide referential and temporal closure, and coincidence/provenance screens. PROFILEFOUNDRY is not a population-fidelity model, a rendered-text corpus, or a formal privacy mechanism. Instead, it is a responsible synthetic source layer for constructing downstream foundation-model evaluations involving memory, privacy, document understanding, record linkage, and agent state while keeping the synthetic person behind each artifact inspectable",
"published": "2026-06-24",
"updated": "2026-06-24",
"authors": [
"Sriram Selvam",
"Anneswa Ghosh"
],
"categories": [
"cs.CL"
],
"url": "https://arxiv.org/abs/2606.26403",
"pdf_url": "https://arxiv.org/pdf/2606.26403",
"path": "papers/items/2026-2606-26403-profilefoundry-a-synthetic-person-object-substrate-for-privacy-memory-and-tool-u.md",
"local_title": "\"ProfileFoundry: A Synthetic Person-Object Substrate for Privacy, Memory, and Tool-Use Evaluation in LLM Agent\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"tool-use"
],
"collection_queries": "tool-use",
"collection_score": "19"
},
{
"arxiv_id": "2606.26453",
"title": "Optimizing CUDA like a Human: Micro-Profiling Tools as Expert Surrogates for LLM-Based GPU Kernel Optimization",
"abstract": "We present KernelPro, a closed-loop multi-agent system that automatically generates, profiles, and iteratively optimizes GPU kernel code by integrating large language model (LLM) code generation with hardware profiler feedback and pluggable bottleneck detection tools. KernelPro introduces four contributions: (1) a semantic feedback operator that encodes expert heuristics as pluggable micro-profiling tools, transforming raw hardware metrics into actionable natural language guidance; (2) a two-stage tool invocation architecture where roofline-based bottleneck classification filters which specialized analysis tools execute, combining kernel-level (ncu), instruction-level (SASS), and system-level (nsys) profiling; (3) a domain-adapted MCTS with progressive widening, asymmetric branching, log-reward calibration, dead-end pruning, and search memory for cross-iteration learning; and (4) direct CuTe source-level code generation via autonomous code search over the CUTLASS/CuTe codebase. On KernelBench, KernelPro achieves geometric mean speedups of 2.42x/4.69x/5.30x on Levels 1/2/3, establishing state-of-the-art performance across all difficulty levels. On VeOmni's expert-optimized MoE training kernels, KernelPro achieves 1.23x over hand-tuned Triton by generating a from-scratch raw-CUDA+CuTe Hopper WGMMA kernel. Ablation studies demonstrate that each design component independently and significantly improves optimization quality: micro-profiling tools (p < 0.0001 vs raw metrics), MCTS search (26% higher geometric mean vs greedy, p = 0.004), and proactive tool orchestration (23% improvement, p = 0.035). Finally, KernelPro is the first CUDA kernel coding agent to optimize energy efficiency beyond the speed-only focus of prior systems, demonstrating an 11.6% measured energy reduction at matched speed.",
"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"
],
"url": "https://arxiv.org/abs/2606.26453",
"pdf_url": "https://arxiv.org/pdf/2606.26453",
"path": "papers/items/2026-2606-26453-optimizing-cuda-like-a-human-micro-profiling-tools-as-expert-surrogates-for-llm-.md",
"local_title": "\"Optimizing CUDA like a Human: Micro-Profiling Tools as Expert Surrogates for LLM-Based GPU Kernel Optimization\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"coding-agent",
"computer-use",
"memory",
"multi-agent",
"tool-use"
],
"collection_queries": "coding-agent, multi-agent-llm",
"collection_score": "15"
},
{
"arxiv_id": "2606.27397",
"title": "SidConArena: An Environment Evaluating Agents in Open-Ended,Positive-Sum Bargaining Game",
"abstract": "Evaluating LLM agents requires dynamic environments that go beyond static reasoning and zero-sum games. Real-world economic interaction is often open-ended and mixed-motive: agents must negotiate, create positive-sum surplus, compete for scarce assets, and plan under delayed returns. We introduce SidConArena, a new benchmark framework for evaluating LLM agents in open-ended, positive-sum bargaining. SidConArena formalizes a multi-player economy as a finite-horizon partially observable stochastic game with three coupled phases: natural-language negotiation with binding trades, deterministic converter-based production, and sealed-bid auctions for long-term assets. The framework combines structured observations, phase-aware agent dispatching, a neural-symbolic action interface, and asynchronous execution, enabling free-form interaction while preserving rule-grounded evaluation. Across homogeneous and heterogeneous tournaments, stronger frontier models achieve higher economic outcomes, yet agents still misvalue resources, bargain passively, and remain limited in long-horizon investment planning.",
"published": "2026-06-24",
"updated": "2026-06-24",
"authors": [
"Yeqi Feng",
"Yuxin Chen",
"Tianxing He"
],
"categories": [
"cs.MA",
"cs.AI",
"cs.GT"
],
"url": "https://arxiv.org/abs/2606.27397",
"pdf_url": "https://arxiv.org/pdf/2606.27397",
"path": "papers/items/2026-2606-27397-sidconarena-an-environment-evaluating-agents-in-open-ended-positive-sum-bargaini.md",
"local_title": "\"SidConArena: An Environment Evaluating Agents in Open-Ended,Positive-Sum Bargaining Game\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"planning",
"reasoning",
"tool-use"
],
"collection_queries": "planning-agent",
"collection_score": "15"
},
{
"arxiv_id": "2606.26511",
"title": "Temporal Validity in Retrieval Memory: Eliminating Stale-Fact Errors for AI Agents over Evolving Knowledge",
"abstract": "Retrieval-augmented generation (RAG) gives agents access to accumulated knowledge, but has no model of time. When a fact changes (e.g., a function is renamed or API restructured), RAG retrieves both the stale and current value with near-identical embedding similarity. The agent then either abstains or serves the superseded fact. We show this is a structural problem: on a calibrated dataset, cosine similarity distinguishes a contradicted fact from a duplicated one with AUROC 0.59 (near chance), as contradictions are often more embedding-similar to the original than rephrased duplicates. We present MemStrata, a retrieval memory maintaining temporal validity. It stores facts like RAG, preserving static recall, but when a fact's value is contradicted, a deterministic (subject, relation, object) supersession rule retires the stale value in a bi-temporal ledger - with no similarity threshold and no LLM call. Across six benchmarks run locally with a 7B model, MemStrata ties RAG on static knowledge and reaches 0.95-1.00 accuracy on evolving knowledge (where RAG reaches 0.20-0.47). The central result is the stale-fact-error rate: when required to answer, RAG serves superseded values 15-40% of the time; MemStrata drives this to ~0%, a failure class RAG cannot avoid. MemStrata achieves this at retrieval latency (~2.1s) versus ~16-18s for LLM-reranking baselines. We release the harness, datasets, and a marker-free evaluation protocol for memory under knowledge evolution.",
"published": "2026-06-25",
"updated": "2026-06-25",
"authors": [
"Neeraj Yadav"
],
"categories": [
"cs.CL",
"cs.AI",
"cs.ET",
"cs.LG"
],
"url": "https://arxiv.org/abs/2606.26511",
"pdf_url": "https://arxiv.org/pdf/2606.26511",
"path": "papers/items/2026-2606-26511-temporal-validity-in-retrieval-memory-eliminating-stale-fact-errors-for-ai-agent.md",
"local_title": "\"Temporal Validity in Retrieval Memory: Eliminating Stale-Fact Errors for AI Agents over Evolving Knowledge\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"rag",
"tool-use"
],
"collection_queries": "ai-agent, rag-agent",
"collection_score": "21"
},
{
"arxiv_id": "2606.26627",
"title": "Agents That Know Too Much: A Data-Centric Survey of Privacy in LLM Agents",
"abstract": "Large language model agents increasingly query databases, search document collections, call external APIs, remember past interactions, and act on a user's behalf. As they move from answering questions to operating over sensitive data, privacy becomes harder to enforce. An agent touches many data sources, runs multi-step workflows, keeps state across sessions, and acts with delegated permissions. Sensitive information can therefore leak not only through its final answer but through the queries it issues, the intermediate results it handles, the memory it writes, and the messages it exchanges with other agents. We survey the privacy of LLM agents from a data-centric view, organizing the field around the data an agent touches rather than by attack type, and we use data agent as shorthand for an LLM agent that works with data. Research on these risks is active but scattered across retrieval-augmented generation, text-to-SQL interfaces, agent memory, prompt injection, access control, and contextual privacy. This survey brings that work together: we taxonomize the data sources an agent touches, the privacy risks each source creates, and the governance mechanisms that address them; we map the benchmarks used to measure these risks and identify what is missing; and we set out the open problems. Two findings recur: among governance mechanisms only information-flow control covers both compositional and cross-session inference leakage, the two least-protected risks; and no benchmark drives an agent across its data surfaces under one privacy policy, the instrument the field most lacks. Our goal is a reference that situates the scattered literature and gives future work a common framing.",
"published": "2026-06-25",
"updated": "2026-06-25",
"authors": [
"Nada Lahjouji",
"Ashwin Gerard Colaco"
],
"categories": [
"cs.CR",
"cs.AI"
],
"url": "https://arxiv.org/abs/2606.26627",
"pdf_url": "https://arxiv.org/pdf/2606.26627",
"path": "papers/items/2026-2606-26627-agents-that-know-too-much-a-data-centric-survey-of-privacy-in-llm-agents.md",
"local_title": "\"Agents That Know Too Much: A Data-Centric Survey of Privacy in LLM Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"rag",
"tool-use",
"workflow-agent"
],
"collection_queries": "agent-memory",
"collection_score": "18"
},
{
"arxiv_id": "2606.26721",
"title": "Knowledge-Based Pull Requests: A Trusted Workflow for Agent-Mediated Knowledge Collaboration",
"abstract": "AI coding agents are changing the bottleneck in software collaboration: code is increasingly cheap, while understanding intent, negotiating scope, and governing long-term project responsibility remain costly. This paper proposes \\emph{Knowledge-Based Pull Requests} (KPR), a trusted workflow for agent-mediated software collaboration across trust boundaries, including open source, enterprise, vendor, contractor, and customer-driven settings. In KPR, an external collaborator's local code, tests, and cleaned agent interaction trace are treated as knowledge sources rather than as the default merge candidate. Agents distill these sources into a human-confirmed knowledge package and render it into reviewer-facing forms such as design memos, risk checklists, test plans, or implementation briefs. A project-owned inner trusted coding agent then regenerates candidate code inside the receiving project's environment under repository context, engineering conventions, tests, and security policy. KPR therefore separates two decisions that traditional pull requests often collapse: whether the knowledge should enter the project, and whether a particular implementation should be merged. We contribute the KPR workflow, a candidate artifact schema, a cost-accounting view, a collaboration gateway architecture, a minimal controlled simulation pilot over seven merged public pull requests, and an evaluation agenda. The pilot shows that KPR packages can be instantiated from real PR material and stress-tested under description ablation, diff ablation, and synthetic poisoned-patch conditions. We position KPR as an empirically testable workflow: its value depends on whether auditable extraction, transformation, and project-side regeneration reduce the cost of understanding and reworking high-context external changes.",
"published": "2026-06-25",
"updated": "2026-06-25",
"authors": [
"Xinyu Zhang",
"Weiwei Sun"
],
"categories": [
"cs.SE",
"cs.HC"
],
"url": "https://arxiv.org/abs/2606.26721",
"pdf_url": "https://arxiv.org/pdf/2606.26721",
"path": "papers/items/2026-2606-26721-knowledge-based-pull-requests-a-trusted-workflow-for-agent-mediated-knowledge-co.md",
"local_title": "\"Knowledge-Based Pull Requests: A Trusted Workflow for Agent-Mediated Knowledge Collaboration\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"coding-agent",
"memory",
"multi-agent",
"planning",
"tool-use",
"workflow-agent",
"world-model"
],
"collection_queries": "coding-agent",
"collection_score": "16"
},
{
"arxiv_id": "2606.26758",
"title": "EGG: An Expert-Guided Agent Framework for Kernel Generation",
"abstract": "High-performance GPU kernels are critical for reducing the exponentially growing computational costs of large language models (LLMs), but their development heavily relies on manual tuning by domain experts. While recent advances in LLM-based approaches show promise for automating kernel generation, they still struggle to achieve both correctness and high performance. This limitation primarily arises from the lack of domain-specific optimization guidance, hindering effective exploration of the optimization space. We propose EGG, an Expert-Guided Agent Framework for Kernel Generation, which incorporates expert optimization principles to guide LLMs' decisions. Inspired by expert workflows, we decompose kernel generation into two hierarchical stages: 1) algorithmic structure design, which establishes a high-quality computational structure foundation; 2) hardware-specific tuning, which performs targeted adjustments through parallel mapping, tensor tiling, and memory optimization. This staged decomposition defines explicit optimization objectives, structuring the design space to achieve progressive refinement. To this end, a stage-aware multi-agent collaboration mechanism is designed for inter and intra-stage context management, ensuring stable optimization trajectories. Experiments on KernelBench and real-world workloads show that EGG achieves a 2.13x average speedup over PyTorch, outperforming existing agent-based and RL-based approaches.",
"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"
],
"url": "https://arxiv.org/abs/2606.26758",
"pdf_url": "https://arxiv.org/pdf/2606.26758",
"path": "papers/items/2026-2606-26758-egg-an-expert-guided-agent-framework-for-kernel-generation.md",
"local_title": "\"EGG: An Expert-Guided Agent Framework for Kernel Generation\"",
"local_status": "queued",
"local_topics": [
"computer-use",
"memory",
"multi-agent",
"rag",
"workflow-agent"
],
"collection_queries": "multi-agent-llm",
"collection_score": "14"
},
{
"arxiv_id": "2606.26793",
"title": "MIRROR: Novelty-Constrained Memory-Guided MCTS Red-Teaming for Agentic RAG",
"abstract": "Multimodal agentic retrieval-augmented generation (RAG) systems expand the attack surface beyond prompt injection to include text poisoning, image injection, direct-query attacks, and orchestrator-level tool manipulation. Existing red-teaming approaches are typically surface-specific and often recycle known attack templates; on text-poisoning benchmarks we measure 73-84% exact duplication. We present MIRROR, a unified cross-surface framework that performs memory-guided Monte Carlo tree search while conditioning candidate generation on retrieved context under an explicit novelty constraint. A deterministic Novelty Gate rejects any candidate matching the retrieval set under normalized comparison, allowing retrieval to inform search priors without enabling prompt copying. Across four attack surfaces on a multimodal agentic RAG target, MIRROR attains 76% ASR on image poisoning compared with 52% for baselines, 97% ASR on orchestrator attacks at half the query cost, and the lowest cross-surface variance (coefficient of variation 0.47). In contrast, specialized baselines collapse across surfaces: suffix optimization reaches 79% ASR on text poisoning but 1% on direct queries. We release ART-SafeBench with 41,815 in-package records and runtime adapters yielding 41,991+ total records across four surfaces.",
"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"
],
"url": "https://arxiv.org/abs/2606.26793",
"pdf_url": "https://arxiv.org/pdf/2606.26793",
"path": "papers/items/2026-2606-26793-mirror-novelty-constrained-memory-guided-mcts-red-teaming-for-agentic-rag.md",
"local_title": "\"MIRROR: Novelty-Constrained Memory-Guided MCTS Red-Teaming for Agentic RAG\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"computer-use",
"memory",
"rag",
"tool-use"
],
"collection_queries": "rag-agent",
"collection_score": "18"
},
{
"arxiv_id": "2606.26806",
"title": "Memory Depth, Not Memory Access: Selective Parametric Consolidation for Long-Running Language Agents",
"abstract": "Long-running language agents need more than memory access. Retrieval systems can fetch past facts at query time, but they do not decide which experiences should continue to shape behavior after the working context is unloaded. We study this separate problem as memory depth: durable goal-conditioned tendencies written into a small parametric store. We introduce the loop-drift protocol, a controlled stress test in which the retrieval index remains intact while working context is unloaded and goal-conditioned behavior must persist under long-loop interference. We evaluate EVAF, a surprise- and valence-gated LoRA consolidation mechanism. Across GPT-2 and TinyLlama, retrieval is strongest on shallow factual recall (short-fact accuracy 0.956--0.973), while EVAF is strongest on goal persistence and post-unload recovery (0.812--0.904) with only 2--3 parametric writes per 200 events. Mechanism controls show that selective consolidation factorizes into two controllable dimensions: selection and actuation. Matched random gates isolate selection beyond sparse writing; fixed-inner controls across GPT-2, TinyLlama, and Mistral-7B show that inner-loop write strength is model-dependent; and a Mistral-7B matched-gate inversion reveals asymmetric selection-actuation coupling under miscalibrated actuation. Public Memora event streams serve as an external diagnostic, exposing stale-memory invalidation as an unresolved boundary. Within this probe, selective parametric consolidation supplies memory depth distinct from and complementary to retrieval access.",
"published": "2026-06-25",
"updated": "2026-06-25",
"authors": [
"Haoliang Han"
],
"categories": [
"cs.AI",
"cs.LG"
],
"url": "https://arxiv.org/abs/2606.26806",
"pdf_url": "https://arxiv.org/pdf/2606.26806",
"path": "papers/items/2026-2606-26806-memory-depth-not-memory-access-selective-parametric-consolidation-for-long-runni.md",
"local_title": "\"Memory Depth, Not Memory Access: Selective Parametric Consolidation for Long-Running Language Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag"
],
"collection_queries": "language-agent",
"collection_score": "14"
},
{
"arxiv_id": "2606.26883",
"title": "EconSimulacra: A Digital Twin Platform of Socio-Economic Systems Powered by LLM Agents",
"abstract": "Real-world social behavior emerges from tightly coupled domains: economic conditions shape mobility and social interactions, while online attention and offline activity feed back into local popularity and consumer behavior. Capturing these feedback loops requires artificial societies in which agents carry experiences from one domain into decisions in another. Large language models (LLMs) provide a promising foundation for such societies. However, existing LLM-based simulators typically model domains in isolation or merely place them side by side. To enable such cross-domain interactions, we present EconSimulacra, a multi-agent social simulator that couples consumer economy, mobility, and social networks through a shared internal-state mechanism. In EconSimulacra, experiences accumulated across different domains are stored in memory and transformed into shared internal states (i.e., stress level) connecting heterogeneous domains through individual decision making. This design allows agents to reconcile competing demands arising from multiple domains and generate coherent cross-domain behaviors. As a case study, we show that the shared internal state mechanisms reproduce a nonlinear relationship between online social attention and offline local popularity, illustrating how realistic cross-domain dynamics can emerge within a unified artificial society.",
"published": "2026-06-25",
"updated": "2026-06-30",
"authors": [
"Ryuji Hashimoto",
"Masahiro Kaneko",
"Kentaro Ueda",
"Takehiro Takayanagi",
"Kiyoshi Izumi"
],
"categories": [
"cs.DL"
],
"url": "https://arxiv.org/abs/2606.26883",
"pdf_url": "https://arxiv.org/pdf/2606.26883",
"path": "papers/items/2026-2606-26883-econsimulacra-a-digital-twin-platform-of-socio-economic-systems-powered-by-llm-a.md",
"local_title": "\"EconSimulacra: A Digital Twin Platform of Socio-Economic Systems Powered by LLM Agents\"",
"local_status": "queued",
"local_topics": [
"memory",
"multi-agent",
"tool-use"
],
"collection_queries": "multi-agent-llm",
"collection_score": "15"
},
{
"arxiv_id": "2606.27243",
"title": "NOVA: A Verification-Aware Agent Harness for Architecture Evolution in Industrial Recommender Systems",
"abstract": "Industrial advertising recommender models are continuously improved through architecture evolution. Upgrades such as RankMixer, TokenMixer-Large, and MixFormer show that better structures remain a key source of quality and business gains. Yet developing such upgrades in production is expert-intensive and difficult to scale. Existing automation is insufficient: AutoML mainly tunes hyper-parameters, while effective gains often require cross-module changes under strict constraints; generic LLM coding agents optimize for runnable code, but runnable code does not imply a valid recommender architecture. Candidates may pass local tests while causing silent failures that degrade performance. We present NOVA, a level-aware agent harness for verification-aware architecture evolution. NOVA uses an architecture gradient, an SGD-inspired, non-differentiable update signal that aggregates prior modifications, verification diagnostics, metric feedback, and trajectory memory to guide the next modification. A verification cascade checks structure semantics, local executability, offline effectiveness, and online impact; invalid candidates are blocked early, with failure patterns recorded as forbidden directions. L1--L4 task-level control matches automation to task complexity and risk, routing high-risk tasks to Copilot for human oversight. Deployed in an industrial advertising system, NOVA achieves the highest effective pass rate on L2 ScaleUp and L3 Literature-to-Production tasks (54.5% and 60.0%), reduces silent failures compared with coding-agent baselines, and shortens one literature-to-production cycle by over 13x in human-attended time. In online A/B testing, the selected L3 candidate improves GMV on three pCVR objectives by +1.25%, +1.70%, and +2.02%, while reducing pCVR bias by 58.8%, 66.7%, and 37.3%.",
"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"
],
"url": "https://arxiv.org/abs/2606.27243",
"pdf_url": "https://arxiv.org/pdf/2606.27243",
"path": "papers/items/2026-2606-27243-nova-a-verification-aware-agent-harness-for-architecture-evolution-in-industrial.md",
"local_title": "\"NOVA: A Verification-Aware Agent Harness for Architecture Evolution in Industrial Recommender Systems\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"coding-agent",
"computer-use",
"memory",
"workflow-agent"
],
"collection_queries": "coding-agent",
"collection_score": "15"
},
{
"arxiv_id": "2606.27406",
"title": "Towards Evaluation of Implicit Software World Models in Coding LLMs",
"abstract": "Software engineering, whether performed by humans or by AI agents, requires reasoning about how software behaves. We call the internal model that supports such reasoning the software world model, and view current code-execution benchmarks as covering one well-studied slice of it -- control flow. In this paper, we take a step toward a broader evaluation by shifting the observable axis to execution resources: alongside test outcome and exception class, we predict peak memory, wall-clock time, and ranked profiler outputs at method and line granularity. We use SWE-bench Verified as the source of data to hold the test close to real-world software engineering tasks. All tested models, frontier ones included, show modest performance and brittle behaviour, suggesting a notable lack of understanding of how software is executed, as opposed to how its source code is written.",
"published": "2026-06-25",
"updated": "2026-06-25",
"authors": [
"Egor Bogomolov",
"Yaroslav Zharov"
],
"categories": [
"cs.SE",
"cs.AI"
],
"url": "https://arxiv.org/abs/2606.27406",
"pdf_url": "https://arxiv.org/pdf/2606.27406",
"path": "papers/items/2026-2606-27406-towards-evaluation-of-implicit-software-world-models-in-coding-llms.md",
"local_title": "Towards Evaluation of Implicit Software World Models in Coding LLMs",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"coding-agent",
"memory",
"reasoning",
"world-model"
],
"collection_queries": "ai-agent, coding-agent",
"collection_score": "18"
},
{
"arxiv_id": "2606.27472",
"title": "Supersede: Diagnosing and Training the Memory-Update Gap in LLM Agents",
"abstract": "Large language model (LLM) agents operate over long, multi-session interactions in which facts change: a user moves, a price updates, a plan is revised. Acting correctly requires using the current value of a fact and discarding values that have been superseded. We isolate this ability on real conversational data and show that it is a distinct, unsolved failure. On the knowledge-update subset of LongMemEval, replacing an agent's full context with a bounded, self-maintained memory drops accuracy from 92% to 77% even on a frontier model (gpt-5.4), a gap that is statistically significant (paired McNemar p<0.005) and persists across model scale while full-context accuracy saturates near 92%. The bottleneck is therefore memory maintenance, not comprehension, and is not closed by a stronger model. We then ask whether this is merely an undersized memory, and find it is not: as the conversation grows 24x, accuracy falls further (from 68% to 28%), and granting the agent proportionally more memory yields no detectable recovery (28% to 28%, n=25). The failure scales with the length of the conversation, not the compression ratio. We release Supersede, an open reinforcement-learning environment (on the verifiers / prime-rl stack) that turns this measurement into a training signal: agents are rewarded for answering from the current value and penalized for stale ones. Finally, we close the loop and show the gap is trainable: GRPO fine-tuning a small open model (Qwen2.5-3B) on this environment nearly doubles its held-out supersession accuracy on real, unseen conversations (9.0% to 16.7%, a single run), along a monotonic checkpoint curve indicating the learned policy, not the harness, carries the gain. To our knowledge this is the first trainable environment whose reward targets temporal fact-currency, and the first evidence the supersession gap can be trained down, not only measured.",
"published": "2026-06-25",
"updated": "2026-06-25",
"authors": [
"Vedant Patel"
],
"categories": [
"cs.CL",
"cs.AI",
"cs.LG"
],
"url": "https://arxiv.org/abs/2606.27472",
"pdf_url": "https://arxiv.org/pdf/2606.27472",
"path": "papers/items/2026-2606-27472-supersede-diagnosing-and-training-the-memory-update-gap-in-llm-agents.md",
"local_title": "\"Supersede: Diagnosing and Training the Memory-Update Gap in LLM Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"planning",
"reasoning",
"tool-use"
],
"collection_queries": "planning-agent",
"collection_score": "18"
},
{
"arxiv_id": "2606.27499",
"title": "DMV-Bench: Diagnosing Long-Horizon Multimodal Agents' Visual Memory with Incidental Cue Injection",
"abstract": "Research on agent memory has matured rapidly, but almost entirely on the text side: few existing benchmarks ask, in an interactive environment, when an agent genuinely needs to remember what it saw rather than what it could write down. We introduce DMV-Bench (Code: https://github.com/yyyujintang/DMV-Bench), the first interactive benchmark for multimodal-agent visual memory. DMV-Bench is built on a controlled home-furnishing e-commerce catalogue of 1,000 product variants in which a text-leakage contract keeps the discriminative signal of each task in the pixels alone. Across a chain of autonomous shopping sessions, every visited product image carries a unique, pre-rendered incidental cue, and the agent is later asked to recall a particular cued product and navigate to its URL. Inspired by dual-coding theory, we propose DualMem, a memory architecture that maintains a visual and a verbal code in parallel. On DMV-Bench, DualMem outperforms a caption baseline and three recent multimodal agent-memory systems at every chain length J in {5, 10, 15, 50} on both Gemini 2.5 Flash and Qwen2.5-VL-7B, with the lead surviving controls for memory-bank size and encoding-position bias, and an asymmetric dual-coding regime in which vision carries the cue end-to-end while the verbal channel plays a smaller query-grounding role.",
"published": "2026-06-25",
"updated": "2026-06-25",
"authors": [
"Yujin Tang",
"Chenming Shang",
"Ruize Xu",
"Nikhil Singh"
],
"categories": [
"cs.CV",
"cs.AI",
"cs.CL"
],
"url": "https://arxiv.org/abs/2606.27499",
"pdf_url": "https://arxiv.org/pdf/2606.27499",
"path": "papers/items/2026-2606-27499-dmv-bench-diagnosing-long-horizon-multimodal-agents-visual-memory-with-incidenta.md",
"local_title": "\"DMV-Bench: Diagnosing Long-Horizon Multimodal Agents' Visual Memory with Incidental Cue Injection\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"coding-agent",
"memory",
"planning",
"rag",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "15"
},
{
"arxiv_id": "2606.27929",
"title": "When Multi-Robot Systems Meet Agentic AI:Towards Embodied Collective Intelligence",
"abstract": "Embodied AI is increasingly becoming agentic, shifting robots from perception--control pipelines towards closed-loop systems that can retrieve context, deliberate during execution, monitor feedback, and refine future behavior. In parallel, robotics research has also moved from single-robot autonomy towards multi-robot systems, driven by the need for wider sensing, distributed action, heterogeneous capabilities, and fault tolerance. As AI agents move from single-agent use towards multi-agent collaboration, robotics faces a parallel challenge: robot teams must move beyond sharing maps, task assignments, and datasets towards sharing the state produced by embodied agent loops. This article explores Embodied Collective Intelligence (ECI), a future multi-robot paradigm in which a robot team accumulates and uses world context, task progress, and skill experience as shared resources. Specifically, we first review how embodied AI is becoming agentic and how multi-robot cooperation has evolved. We then present Embodied Collective Intelligence through Co-Perception, Co-Action, and Co-Evolution. Finally, we use an illustrative navigation study to examine one concrete component of the concept: shared world-memory inheritance. The study shows that a newly added robot can benefit from merged team memory, but it is not intended as a full evaluation of the ECI framework. Taken together, the review and conceptual framework motivate Embodied Collective Intelligence as a direction for embodied multi-agent intelligence, while the case study grounds one measurable part of the concept.",
"published": "2026-06-26",
"updated": "2026-06-26",
"authors": [
"Yuxuan Yan",
"Yuanyuan Jia",
"Qianqian Yang"
],
"categories": [
"cs.RO"
],
"url": "https://arxiv.org/abs/2606.27929",
"pdf_url": "https://arxiv.org/pdf/2606.27929",
"path": "papers/items/2026-2606-27929-when-multi-robot-systems-meet-agentic-ai-towards-embodied-collective-intelligenc.md",
"local_title": "\"When Multi-Robot Systems Meet Agentic AI:Towards Embodied Collective Intelligence\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"embodied-agent",
"memory",
"multi-agent",
"tool-use"
],
"collection_queries": "ai-agent",
"collection_score": "17"
},
{
"arxiv_id": "2606.27990",
"title": "AdvancedShelLM: A Stateful Multi-Agent LLM Honeypot for SSH Deception",
"abstract": "LLM-based SSH honeypots can generate believable interactions, but evaluations indicate they remain somewhat identifiable to determined attackers, indicating the need for a better scaffolding. We present a new LLM-based honeypot design that uses a multi-agent, multi-LLM architecture to address the limitations of the previous shelLM LLM honeypot. Our honeypot, called AdvancedShelLM, uses two LLM agents, a Manager and a Worker, that better understand the commands while reducing incorrect responses and increasing deception. It implements an advanced permanent filesystem, allowing many simultaneous attackers to see the same changing files for the first time. It was evaluated with: (i) unit tests for generative capabilities, (ii) an AI attacker (ARACNE) to assess realism and deception, (iii) human attackers to assess its deceptive capability, and (iv) an Internet deployment to evaluate deception in real-world attacks. In unit test results, AdvancedShelLM achieved a pass rate of up to 99.02%. The AI attacker ARACNE had issues making a decision if the system is honeypot or not, but showed slight bias towards saying honeypot, even for a real Ubuntu shell. With human attackers, AdvancedShelLM deceived more humans than Cowrie, but had similar results as shelLM. The Internet deployment showed concrete evidence that the output of AdvancedShelLM can influence the behaviour of real-life attackers.",
"published": "2026-06-26",
"updated": "2026-06-26",
"authors": [
"Muris Sladić",
"Eman Alibalić",
"Veronica Valeros",
"Carlos Catania",
"Sebastian Garcia"
],
"categories": [
"cs.CR"
],
"url": "https://arxiv.org/abs/2606.27990",
"pdf_url": "https://arxiv.org/pdf/2606.27990",
"path": "papers/items/2026-2606-27990-advancedshellm-a-stateful-multi-agent-llm-honeypot-for-ssh-deception.md",
"local_title": "\"AdvancedShelLM: A Stateful Multi-Agent LLM Honeypot for SSH Deception\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"multi-agent",
"tool-use"
],
"collection_queries": "llm-agent, multi-agent-llm",
"collection_score": "17"
},
{
"arxiv_id": "2606.28270",
"title": "Agent-Native Immune System: Architecture, Taxonomy, and Engineering",
"abstract": "The transition from static chat bots to autonomous agents--equipped with persistent memory, tool-use protocols, and multi-agent collaboration--has fundamentally expanded the AI threat landscape. Current defense mechanisms, such as perimeter security and training-time alignment, remain external to the agent's active reasoning loop. Consequently, they fall short: a fully aligned agent remains highly vulnerable to runtime hijacking via memory poisoning, tool-chain manipulation, or multi-agent protocol attacks. To address this critical gap, we introduce the Agent-Native Immune System (ANIS), the first biologically inspired, endogenous defense architecture embedded directly within the agent's cognitive loop. Our framework presents four primary contributions. First, we design a six-layer Immune Tower (L0-L5), distinctly incorporating Barrier Immunity (L1) as a non-cognitive, physical-and-logical isolation layer. Second, we establish a unified taxonomy of Agent Viruses and Agent Vaccines, formalizing the critical distinction between superficial non-parametric defenses and robust parametric vaccines. Third, we conceptualize the Harness Triad--Meta, Self, and Auto--a self-monitoring, meta-cognitive automation backbone that drives Continual Immune Learning (CIL), enabling vaccines to dynamically adapt to novel threats. Finally, we establish a rigorous theoretical demarcation between model alignment and agent immunity: while alignment provides a static \"constitutional\" value foundation during training, ANIS serves as the dynamic \"law enforcement\" mechanism during runtime. We conclude by framing open challenges for the field, including immune protocol standardization, novel evaluation metrics such as the Autoimmunity Rate (false-positive intervention rate), and the co-evolutionary dynamics between pathogens and vaccines within collective intelligence ecosystems.",
"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"
],
"url": "https://arxiv.org/abs/2606.28270",
"pdf_url": "https://arxiv.org/pdf/2606.28270",
"path": "papers/items/2026-2606-28270-agent-native-immune-system-architecture-taxonomy-and-engineering.md",
"local_title": "\"Agent-Native Immune System: Architecture, Taxonomy, and Engineering\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"multi-agent",
"reasoning",
"tool-use",
"workflow-agent"
],
"collection_queries": "tool-use",
"collection_score": "15"
},
{
"arxiv_id": "2606.28434",
"title": "SWE-MeM: Learning Adaptive Memory Management for Long-Horizon Coding Agents",
"abstract": "Long-horizon software engineering agents often need to manage lengthy and noisy interaction histories under limited context budgets. Existing memory management methods typically rely on static compression workflows or impose rigid constraints on compression timing and granularity. Moreover, these approaches fail to jointly optimize memory management and issue resolution capabilities to improve performance while reducing token usage. We present SWE-MeM, a training framework for proactive and on-demand memory management in software engineering agents. SWE-MeM provides a flexible memory tool that lets agents decide when, what, and how to compress based on trajectory state, task progress, and remaining context budget. We train agents with synthesized proactive memory-management trajectories and Memory-aware GRPO, which jointly optimizes memory management and issue resolution through memory-aware trajectory splitting and step-level credit assignment. On SWE-Bench Verified, SWE-MeM achieves 43.4% and 60.2% resolve rate with 4B and 30B models, respectively, outperforming existing memory management baselines in both performance and efficiency.",
"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"
],
"url": "https://arxiv.org/abs/2606.28434",
"pdf_url": "https://arxiv.org/pdf/2606.28434",
"path": "papers/items/2026-2606-28434-swe-mem-learning-adaptive-memory-management-for-long-horizon-coding-agents.md",
"local_title": "\"SWE-MeM: Learning Adaptive Memory Management for Long-Horizon Coding Agents\"",
"local_status": "queued",
"local_topics": [
"coding-agent",
"memory",
"planning",
"tool-use",
"workflow-agent"
],
"collection_queries": "coding-agent",
"collection_score": "17"
},
{
"arxiv_id": "2606.28456",
"title": "Is Lying an Emergent Behaviour in LLMs? Evidence from Gaslighting AI agents in a Sustainability Game",
"abstract": "LLMs agents are increasingly used in multi-agent settings, yet their behaviour in sustainability games remains largely unexplored. This work investigates whether lying can emerge among LLM agents in a competitive sustainability game in which agents are informed that common resources can regenerate, although regeneration does not actually occur. We develop an agent-based model of a sustainability game in which agents manage industrial, military, and ecological resources, and interact through a network. LLM agents can observe neighbours' status, declare future attacks, receive permission to lie, and access reputation information, while rule-based agents provide an interpretable behavioural baseline. The results show that neighbour information strongly changes system dynamics, increasing attacks while improving biosphere retention and coexistence. Also, the presence of future declarations reduce extinction risk without suppressing conflict. Behaviourally, deception emerges even when agents are not explicitly allowed to lie, and explicit permission mainly increases bluffing and diversion rather than direct backstabbing. Finally, the presence of reputation memory and information about the current biosphere level reduces system ecological depletion. These findings suggest that deception can arise as an emergent behaviour in LLM-agent systems and that communication between LLM-agents could support sustainability while dealing with risk.",
"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"
],
"url": "https://arxiv.org/abs/2606.28456",
"pdf_url": "https://arxiv.org/pdf/2606.28456",
"path": "papers/items/2026-2606-28456-is-lying-an-emergent-behaviour-in-llms-evidence-from-gaslighting-ai-agents-in-a-.md",
"local_title": "Is Lying an Emergent Behaviour in LLMs? Evidence from Gaslighting AI agents in a Sustainability Game",
"local_status": "queued",
"local_topics": [
"agent-safety",
"memory",
"multi-agent"
],
"collection_queries": "ai-agent, llm-agent, multi-agent-llm",
"collection_score": "18"
},
{
"arxiv_id": "2606.28467",
"title": "An Agentic AI Pipeline for Appliance-Level Energy Anomaly Detection and LLM-Driven Recommendations",
"abstract": "Appliance-level energy monitoring in office buildings produces noisy alerts that non-expert facility managers struggle to use. This paper proposes an end-to-end agentic pipeline that combines deep time-series forecasting, variational anomaly detection, and LLM-based reasoning to generate prioritized, actionable maintenance recommendations. The system tracks seven office appliances using a hybrid Singular Spectrum Analysis (SSA) and Long Short-Term Memory (LSTM) forecasting model, and applies a per-appliance LSTM Variational Autoencoder (VAE) with attention to flag abnormal daily consumption episodes. A three-stage LangChain pipeline begins with a Context Agent that always retrieves three core RAG sources (model reliability, hourly baseline, and expert knowledge) and conditionally adds up to three more (forecast context, anomaly history, global baseline) based on event characteristics, capped at eight reasoning steps. A Diagnosis Agent converts the evidence into a structured JSON diagnosis, and a Report Agent renders a human-readable narrative. A reflective memory layer incorporates operator feedback. The dashboard shows real-time 30-minute forecasts, intraday consumption, the previous day anomaly report, and a feedback form. We evaluate the forecasting model, anomaly detector with appliance-specific thresholds, and LLM reasoning on a 16-scenario benchmark including sustained and transient spikes, unexpected shutdowns, and systemic events, comparing five LLM backends under static vs. dynamic retrieval. Dynamic retrieval matches full static retrieval across all backends while cutting average context from six to three-six sources per event. The best backend scores 90.4/100 with a 100% pass rate at a 70-point threshold, and a fully local 7B-parameter model passes all 16 scenarios.",
"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"
],
"url": "https://arxiv.org/abs/2606.28467",
"pdf_url": "https://arxiv.org/pdf/2606.28467",
"path": "papers/items/2026-2606-28467-an-agentic-ai-pipeline-for-appliance-level-energy-anomaly-detection-and-llm-driv.md",
"local_title": "An Agentic AI Pipeline for Appliance-Level Energy Anomaly Detection and LLM-Driven Recommendations",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"reasoning",
"tool-use",
"workflow-agent"
],
"collection_queries": "agentic-ai, rag-agent",
"collection_score": "19"
},
{
"arxiv_id": "2606.28781",
"title": "HyphaeDB: A Living Knowledge Topology for Agent-First Memory",
"abstract": "Every existing vector database and agent memory framework treats memory as passive storage that agents query explicitly. No system propagates knowledge between agents through the memory layer itself. We introduce HyphaeDB, an agent-native memory infrastructure that reinterprets the Hierarchical Navigable Small World (HNSW) graph topology the data structure at the core of every modern vector database not as a search optimization, but as a communication fabric for multi-agent AI systems. In HyphaeDB, agents are nodes in the vector space with persistent positions, knowledge propagates via a gossip protocol through the graph's neighbor structure with energy-based attenuation, and emergent behaviors contradiction detection, pattern crystallization, and consensus formation arise from the combination of topology, propagation dynamics, and local interaction rules. We present the architecture built on three primitives (knowledge nodes, topology edges, and memory diffs), a multi-layer abstraction hierarchy with promotion via emergent consensus, and theoretical analysis grounding the system in small-world network theory, epidemic broadcast protocols, and swarm intelligence. We provide a reference implementation on PostgreSQL with pgvector and describe a concrete deployment in Swarm-Driven Development, a multi-agent software engineering methodology. HyphaeDB represents, to our knowledge, the first system to combine navigable small world topology with gossip-based knowledge propagation for multi-agent coordination.",
"published": "2026-06-27",
"updated": "2026-06-27",
"authors": [
"Krishna Halaharvi"
],
"categories": [
"cs.AI",
"cs.MA"
],
"url": "https://arxiv.org/abs/2606.28781",
"pdf_url": "https://arxiv.org/pdf/2606.28781",
"path": "papers/items/2026-2606-28781-hyphaedb-a-living-knowledge-topology-for-agent-first-memory.md",
"local_title": "\"HyphaeDB: A Living Knowledge Topology for Agent-First Memory\"",
"local_status": "queued",
"local_topics": [
"coding-agent",
"memory",
"multi-agent",
"rag",
"tool-use"
],
"collection_queries": "agent-memory, agentic-ai",
"collection_score": "15"
},
{
"arxiv_id": "2606.28791",
"title": "From Determinism to Delegation: AI-Native Software Engineering and the Evolution of the Agentic Engineer",
"abstract": "Software engineering is experiencing its most significant transformation since the emergence of high-level programming languages. As large language models (LLMs) increasingly enable sustained, multi-step, tool-mediated execution, engineering value is shifting from writing deterministic code to supervising probabilistic and autonomous behavior. This paper argues that AI-Native Software Engineering is a paradigm shift rather than a mere tooling advance, creating a new professional archetype: the Agentic Engineer, whose primary artifact is the agentic system rather than the program. We characterize this transition through three changes: (i) the unit of work shifts from functions to supervised agent workflows, (ii) correctness shifts from binary assertions to statistical evaluation under uncertainty, and (iii) accountability shifts from code authorship to outcome ownership. Drawing on post-2022 research, we compare traditional and agentic engineering roles and define core mechanisms of autonomous agents, including reasoning-acting loops, context engineering, tool use, memory, behavioral drift, and compositional error. We place human-AI collaboration within socio-technical frameworks and examine mixed empirical evidence. While some studies report productivity gains, others show slowdowns among experienced developers, highlighting disciplined oversight rather than automation as the critical competency. Using established governance frameworks, we identify required skills and risks, including indirect prompt injection. We conclude that the future is one of symbiosis rather than substitution: agentic engineering builds upon and depends on classical software engineering principles.",
"published": "2026-06-27",
"updated": "2026-06-27",
"authors": [
"Mamdouh Alenezi"
],
"categories": [
"cs.SE"
],
"url": "https://arxiv.org/abs/2606.28791",
"pdf_url": "https://arxiv.org/pdf/2606.28791",
"path": "papers/items/2026-2606-28791-from-determinism-to-delegation-ai-native-software-engineering-and-the-evolution-.md",
"local_title": "\"From Determinism to Delegation: AI-Native Software Engineering and the Evolution of the Agentic Engineer\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"coding-agent",
"memory",
"multi-agent",
"reasoning",
"tool-use",
"workflow-agent"
],
"collection_queries": "agentic-ai, autonomous-agent-llm, tool-use",
"collection_score": "23"
},
{
"arxiv_id": "2606.28958",
"title": "When Latent Agents Lie: KV-Cache Integrity in Multi-Agent LLM Collaboration",
"abstract": "LLM agents can share more than text. In some systems, an agent can send a short visible message while also passing its full KV-cache state to another model. This hidden state can help the final model combine evidence from several agents, but it is also hard to inspect. A visible message may look harmless even if the hidden state has been changed. We study this problem in a multi-agent question-answering setup. Specialists each see part of the evidence, send a short commitment, and pass full KV-cache state to a coordinator. In clean runs, this latent collaboration improves over a matched text-only version. On transformed HiddenBench with Qwen3-4B, it reaches EM/F1 of 0.338/0.486, compared with 0.231/0.369 for text collaboration. Qwen3-8B and HotPotQA runs show the same direction of improvement. The problem appears when one specialist is malicious. Some false visible commitments can steer answers. More seriously, changing the hidden KV state can collapse performance even when the visible commitment still looks plausible. A verifier that checks only text misses this failure mode. Simple magnitude checks catch some obvious corruptions, but adaptive attacks can evade them while still damaging the final answer. The most reliable fix we find is not to guess whether hidden state looks normal, but to protect it in transport. We implement an HMAC-SHA256 manifest that binds the specialist, session, model, visible commitment, tensor metadata, and payload digest. It accepts all 774 honest replayed payloads and rejects all 295 recorded tampered payloads. The main lesson is that full-KV latent memory can be useful, but it should be treated as a security-sensitive object, not as ordinary internal model state.",
"published": "2026-06-27",
"updated": "2026-06-27",
"authors": [
"Luís Brito",
"Carlos Baquero"
],
"categories": [
"cs.MA"
],
"url": "https://arxiv.org/abs/2606.28958",
"pdf_url": "https://arxiv.org/pdf/2606.28958",
"path": "papers/items/2026-2606-28958-when-latent-agents-lie-kv-cache-integrity-in-multi-agent-llm-collaboration.md",
"local_title": "\"When Latent Agents Lie: KV-Cache Integrity in Multi-Agent LLM Collaboration\"",
"local_status": "queued",
"local_topics": [
"agent-safety",
"memory",
"multi-agent",
"reasoning"
],
"collection_queries": "llm-agent, multi-agent-llm",
"collection_score": "17"
},
{
"arxiv_id": "2606.29030",
"title": "Memory as an Attack Surface in LLM Agents: A Study on Multiple-Choice Question Answering",
"abstract": "AI agents extend conventional large language model (LLM) applications by integrating language understanding with task execution, external tool use, and memory mechanisms. While memory allows agents to retain prior interactions and provide more personalized and context-aware responses, it also introduces a new vulnerability: information stored in memory can influence future outputs even when the current query is clean. In this paper, we investigate memory manipulation in LLM-based agents for multiple-choice question answering. We first design and implement an LLM-based AI agent with an external memory component that stores and retrieves task-relevant information. We then introduce basic memory manipulation scenarios in which misleading or corrupted memories are inserted into the agent before it answers multiple-choice questions. Using a controlled experimental setup, we compare the agent's performance before and after memory manipulation and measure changes in answer accuracy, attack success rate, and selection of manipulated options. Our results show that even simple memory manipulations can noticeably affect the agent's final answers, causing it to select incorrect options despite receiving clean and well-formed questions.",
"published": "2026-06-27",
"updated": "2026-06-27",
"authors": [
"Shahnewaz Karim Sakib",
"Anindya Bijoy Das"
],
"categories": [
"cs.AI",
"cs.ET"
],
"url": "https://arxiv.org/abs/2606.29030",
"pdf_url": "https://arxiv.org/pdf/2606.29030",
"path": "papers/items/2026-2606-29030-memory-as-an-attack-surface-in-llm-agents-a-study-on-multiple-choice-question-an.md",
"local_title": "\"Memory as an Attack Surface in LLM Agents: A Study on Multiple-Choice Question Answering\"",
"local_status": "queued",
"local_topics": [
"memory",
"tool-use"
],
"collection_queries": "ai-agent, llm-agent, tool-use",
"collection_score": "19"
},
{
"arxiv_id": "2606.29178",
"title": "Selective Memory Retention for Long-Horizon LLM Agents",
"abstract": "When does retention matter for memory-augmented LLM agents? We study this with TraceRetain, a lightweight framework for bounded external memory in frozen LLM agents that scores entries by interpretable features (success, age, access frequency, redundancy, specificity, similarity, downstream utility) and evicts the lowest-scoring ones at capacity. On clean ALFWorld with gpt-5-mini, external memory robustly improves over no memory across two seeds, but differences among bounded retention policies fall within Wilson 95% CIs: clean ALFWorld at T=100 to T=200 does not naturally exhibit the memory pollution retention is designed to address. Under a controlled noisy-write stress (75% synthetic distractors), unbounded memory and FIFO-K50 degrade on Precision@5 (20.2% to 12.4% and 15.8% to 3.8%) while TraceRetain-CEM is essentially unchanged (16.9% to 16.6%) and preserves 97/100 task success. The mechanism: unbounded memory has the highest mean similarity (0.87) but lowest precision, indicating failed distractors close to the query in embedding space. Held-out in-distribution evaluation shows memory-augmented policies solving 47 to 49 of 50 tasks vs. 39/50 for no memory. Bounded retention buys memory and step efficiency on saturated clean benchmarks at no task-success cost, and only differentiates from cache heuristics when streams contain noise.",
"published": "2026-06-28",
"updated": "2026-06-28",
"authors": [
"Pranath Reddy"
],
"categories": [
"cs.AI",
"cs.CL",
"cs.LG"
],
"url": "https://arxiv.org/abs/2606.29178",
"pdf_url": "https://arxiv.org/pdf/2606.29178",
"path": "papers/items/2026-2606-29178-selective-memory-retention-for-long-horizon-llm-agents.md",
"local_title": "Selective Memory Retention for Long-Horizon LLM Agents",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"planning"
],
"collection_queries": "llm-agent",
"collection_score": "18"
},
{
"arxiv_id": "2606.29537",
"title": "OSWorld2.0: Benchmarking Computer Use Agents on Long-Horizon Real-World Tasks",
"abstract": "Existing computer-use benchmarks fail to capture the realism, complexity, and long-horizon demands of real-world computer use, limiting their ability to reveal the limitations of frontier agents. We introduce OSWorld 2.0, a benchmark of 108 long-horizon computer-use workflows across everyday and professional tasks, designed to capture complex and challenging real-world phenomena. Each task represents a realistic end-to-end workflow that takes human users a median of about 1.6 hours to complete and requires an average of 318 tool calls with Claude Opus 4.7 using maximum thinking, compared with about 30 in OSWorld 1.0. OSWorld 2.0 targets challenge phenomena that are common in real workflows yet underrepresented in prior benchmarks, spanning interaction-design challenges such as streaming interaction and dynamic environments, as well as agent-pattern challenges such as cross-source reasoning, implicit-state inference, and visual-spatial precision. Tasks are grounded in authentic input artifacts and cross-referenced against realistic stateful user profile data, and include separate safety reports auditing safety-sensitive execution. Under our primary binary-completion metric at 500 steps, Claude Opus 4.8 with maximum thinking and batched tool calls scores best but still completes only 20.6% of tasks at a 54.8% partial score; GPT-5.5 is far more token-efficient yet plateaus near 13%. These results show that current agents are still far from professional-level computer use: rather than stumbling on basic GUI control or coding, they lose track of constraints, miss information that arrives mid-task, guess rather than ask the user, and skip verification, struggling most when a task hinges on hidden state they must recover.",
"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"
],
"url": "https://arxiv.org/abs/2606.29537",
"pdf_url": "https://arxiv.org/pdf/2606.29537",
"path": "papers/items/2026-2606-29537-osworld2-0-benchmarking-computer-use-agents-on-long-horizon-real-world-tasks.md",
"local_title": "\"OSWorld2.0: Benchmarking Computer Use Agents on Long-Horizon Real-World Tasks\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"coding-agent",
"computer-use",
"memory",
"planning",
"rag",
"reasoning",
"tool-use",
"workflow-agent"
],
"collection_queries": "web-gui-agent",
"collection_score": "21"
},
{
"arxiv_id": "2607.00038",
"title": "Stop Hand-Holding Your Coding Agent: Engineering the Loops that Replace Step-by-Step Prompting",
"abstract": "In mid-2026 a slogan reorganized how practitioners talk about coding agents: stop prompting your agent, start designing the loop that prompts it. We take this claim seriously and give it a careful treatment. We call the object of the new practice the loop specification: a bounded, reusable artifact, made of a trigger, a goal, a verification step, a stopping rule and a memory, that a human hands to an agent harness (such as Claude Code or Codex) so the agent pursues a goal on its own, in place of step-by-step prompting. We distinguish this external loop specification from two things it is often confused with: an ordinary programming loop, and the internal perceive-act-observe cycle that the harness already provides as plumbing. We position loop engineering as a new layer in the progression from prompt to context to harness to loop, and we argue, against the stronger headlines, that it does not retire prompt engineering; loop and prompt are distinct tools with distinct uses. We offer four contributions: a definition and scope for the discipline; an anatomy and taxonomy of loop specifications organized around trigger, goal type, a five-level verification ladder, architecture, and named terminal states; a descriptive analysis of the Loop Library, a public corpus of fifty real loops that we code by hand; and a set of design principles and anti-patterns grounded in the scientific literature on self-correction, reward hacking and model-as-judge fragility. The corpus shows that practice has matured most where the discipline says it matters: seventy percent of loops verify in the autonomous zone of the ladder and seventy-four percent name their terminal states, while automated triggering and durable memory remain comparatively underdeveloped. We close with the limits the practice must respect, including the verification burden, comprehension debt and the risk of cognitive surrender.",
"published": "2026-06-28",
"updated": "2026-06-28",
"authors": [
"Sandeco Macedo"
],
"categories": [
"cs.SE"
],
"url": "https://arxiv.org/abs/2607.00038",
"pdf_url": "https://arxiv.org/pdf/2607.00038",
"path": "papers/items/2026-2607-00038-stop-hand-holding-your-coding-agent-engineering-the-loops-that-replace-step-by-s.md",
"local_title": "\"Stop Hand-Holding Your Coding Agent: Engineering the Loops that Replace Step-by-Step Prompting\"",
"local_status": "queued",
"local_topics": [
"agent-safety",
"coding-agent",
"computer-use",
"memory",
"rag",
"tool-use"
],
"collection_queries": "coding-agent",
"collection_score": "15"
},
{
"arxiv_id": "2606.29771",
"title": "CLQT: A Closed-Loop, Cost-Aware, Strategy-Consistent Benchmark for Diagnostic Evaluation of LLM Portfolio-Management Agents",
"abstract": "LLM agents are increasingly cast as autonomous portfolio managers, and benchmarks have moved from financial question-answering to sequential trading. Yet most still rank agents by returns over a fixed window -- a weak proxy, since a period's return is dominated by the market path and apparent alpha can dissolve once look-ahead leakage is controlled. Such a ranking certifies neither sound reasoning, nor a consistent strategy, nor a durable edge. We introduce CLQT, which reframes closed-loop trading evaluation as diagnosis rather than ranking: an instrument that localizes where and why an agent's process succeeds or fails. CLQT is a fully closed-loop, cost-aware, strategy-consistent, temporally-gated environment whose agents run a five-stage cycle: gather, synthesize, allocate, execute, reflect. Each round emits a complete DecisionRound sealed into a recompute-verifiable hash chain, so every metric is reconstructable from the trail. Six pillars form the substrate: a hard TimeGate, institutional transaction- and financing-cost modeling, strategy-consistency scoring, three-tier memory, a Model-Context-Protocol tool layer, and mandate-aware synthesis. The same agent runs as a constrained committee of specialized roles or a single full-autonomy orchestrator, making process scaffolding an experimental variable. From the audit trail we compute a five-axis capability scorecard (APM-CS: Coherence, Acuity, Composure, Discipline, Reliability), with Coherence judged partly by a held-out, out-of-cohort LLM to curb self-preference bias. We validate it on a contamination-controlled multi-model backtest with an ablation grid and a live broker track on unseen, post-cutoff data, against a repeated-run noise floor. CLQT separates outcome from capability, yielding not a model ranking but a durable, extensible map of agent competencies and limitations.",
"published": "2026-06-29",
"updated": "2026-06-29",
"authors": [
"Bo Qu",
"Mingguang Chen"
],
"categories": [
"cs.AI",
"cs.LG",
"q-fin.CP",
"q-fin.PM"
],
"url": "https://arxiv.org/abs/2606.29771",
"pdf_url": "https://arxiv.org/pdf/2606.29771",
"path": "papers/items/2026-2606-29771-clqt-a-closed-loop-cost-aware-strategy-consistent-benchmark-for-diagnostic-evalu.md",
"local_title": "\"CLQT: A Closed-Loop, Cost-Aware, Strategy-Consistent Benchmark for Diagnostic Evaluation of LLM Portfolio-Management Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"reasoning",
"tool-use"
],
"collection_queries": "llm-agent",
"collection_score": "19"
},
{
"arxiv_id": "2606.29774",
"title": "Analytic Concept-Centric Memory for Agentic Embodied Manipulation",
"abstract": "Long-horizon embodied manipulation requires agents to remember persistent objects, track changing scene states, and reuse prior interaction knowledge. However, existing agent memories are often stored as unstructured histories or embedding-based records, making it difficult to retrieve manipulation-relevant object parts, physical states, action effects, and executable skills. We propose an analytic concept-centric memory framework for agentic embodied manipulation. Our memory organizes experience around structured analytic concepts, where objects are represented by semantic parts, parametric templates, grounded poses, affordances, and manipulation states. It further connects object and scene memories with transition memory for action-induced state changes and skill memory for template-grounded and policy-grounded execution. At runtime, the agent performs structured coarse-to-fine retrieval to identify relevant objects, states, transitions, and skills, supporting state-consistent reasoning and skill reuse. Experiments on memory-dependent manipulation, articulated-object generalization, real-world memory evaluation, and ablations show that our approach improves task completion, retrieval accuracy, object re-identification, and cross-object skill generalization over unstructured and embedding-based memory baselines.",
"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"
],
"url": "https://arxiv.org/abs/2606.29774",
"pdf_url": "https://arxiv.org/pdf/2606.29774",
"path": "papers/items/2026-2606-29774-analytic-concept-centric-memory-for-agentic-embodied-manipulation.md",
"local_title": "Analytic Concept-Centric Memory for Agentic Embodied Manipulation",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"embodied-agent",
"memory",
"planning",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "agent-memory",
"collection_score": "19"
},
{
"arxiv_id": "2606.29778",
"title": "Mandol: An Agglomerative Agent Memory System for Long-Term Conversations",
"abstract": "Long-term conversational agents need to remember and query cross-session, multi-typed information with complex correlations. Existing agent memory systems rely on heterogeneous vector and graph databases, which fragment memory information and cause high cross-database I/O latency. For retrieval, common RAG-style methods tend to introduce noise, miss correlated clues, and lack token budget control, degrading LLM accuracy and efficiency. We propose Mandol, an agglomerative memory system that consolidates fragmented memory representations and storage into a unified memory-native architecture. Its core components include: (1) a hierarchical memory model that organizes memory into a basic layer representing raw memory information and a high-level abstract layer that agglomerates basic memories into traceable abstract memories, both uniformly represented as structured semantic graphs; (2) an agglomerative semantic data structure combining SemanticMap and SemanticGraph, which natively fuses key-value, vector, and graph structures and provides unified hybrid retrieval operators to eliminate cross-database I/O; and (3) a quantitative query mechanism with query-adaptive routing, quantitative denoising and conflict resolution, and token-constrained context generation, all without involving LLMs during retrieval. Experiments on two widely used long-term conversation benchmarks, LoCoMo and LongMemEval, show that Mandol achieves the best overall accuracy among representative agent memory systems. For performance comparison, Mandol also obtains a 5.4x retrieval speedup and a 4.8x insertion speedup under 10 QPS concurrent load, while still maintaining low latency on consumer-grade hardware.",
"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"
],
"url": "https://arxiv.org/abs/2606.29778",
"pdf_url": "https://arxiv.org/pdf/2606.29778",
"path": "papers/items/2026-2606-29778-mandol-an-agglomerative-agent-memory-system-for-long-term-conversations.md",
"local_title": "\"Mandol: An Agglomerative Agent Memory System for Long-Term Conversations\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag"
],
"collection_queries": "agent-memory, rag-agent",
"collection_score": "15"
},
{
"arxiv_id": "2606.29788",
"title": "MemLeak: Diagnosing Information Leaks in Multimodal Agent Memory",
"abstract": "When a multimodal AI agent is asked to forget a fact, current memory systems usually delete the text entry and report success. We find that the fact can remain recoverable from retained user images, including images tagged to entirely different facts, because VLMs use implicit visual cues at inference time. We introduce the Information Provenance Graph (IPG), a taxonomy that classifies memory representations by deletion affordance. The IPG reveals that deletion fails through multiple channels. Our benchmark, MemLeak, measures this across a deletion cascade: direct probing of deletion-capable systems yields <1%, but retained correlated text enables 18.3% recovery, and retained images enable 12.0% recovery (0.0% blind baseline, 0.3% FPR) -- with 47% of image leaks not text-recoverable. Content-aware semantic deletion reduces the image residual to 2.0%. The residual appears across multiple VLMs, a production memory system, and real Unsplash-licensed photographs. Dual-annotator human validation (kappa = 0.88) confirms judge reliability.",
"published": "2026-06-29",
"updated": "2026-06-29",
"authors": [
"Kuan Wang",
"Chao Zhang"
],
"categories": [
"cs.LG"
],
"url": "https://arxiv.org/abs/2606.29788",
"pdf_url": "https://arxiv.org/pdf/2606.29788",
"path": "papers/items/2026-2606-29788-memleak-diagnosing-information-leaks-in-multimodal-agent-memory.md",
"local_title": "\"MemLeak: Diagnosing Information Leaks in Multimodal Agent Memory\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory"
],
"collection_queries": "agent-memory, ai-agent",
"collection_score": "13"
},
{
"arxiv_id": "2606.29824",
"title": "Neural Procedural Memory: Empowering LLM Agents with Implicit Activation Steering",
"abstract": "While Large Language Models (LLMs) excel as static solvers, transforming them into autonomous agents remains challenging. This transition requires continuous environmental interaction, yet current agents lack the necessary persistent procedural memory. Existing approaches predominantly employ Retrieval-Augmented Generation (RAG) to inject explicit textual guidelines into model contexts. However, relying solely on symbolic instructions can introduce a text-action disconnect, frequently failing to activate the internal representations necessary for correct task execution. To address this, the paper introduces Neural Procedural Memory (NPM), a training-free framework that represents agent memory through implicit activation steering rather than explicit instructions. By distilling procedural skills from historical contrastive experiences into steering vectors in the activation space, NPM directly activates the task-relevant neural mechanisms to guide task execution. Evaluations across four agent benchmarks show that NPM performs comparably to baselines using explicit textual instructions. Furthermore, the results show that combining implicit steering with explicit workflows provides complementary advantages, leading to more robust task execution. Representational analyses indicate that these steering vectors encode consistent task logic, forming organized structures within the activation space. These findings suggest that implicit activation steering provides a promising approach for managing agent memory.",
"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"
],
"url": "https://arxiv.org/abs/2606.29824",
"pdf_url": "https://arxiv.org/pdf/2606.29824",
"path": "papers/items/2026-2606-29824-neural-procedural-memory-empowering-llm-agents-with-implicit-activation-steering.md",
"local_title": "\"Neural Procedural Memory: Empowering LLM Agents with Implicit Activation Steering\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"rag",
"tool-use",
"workflow-agent"
],
"collection_queries": "agent-evaluation, agent-memory, autonomous-agent-llm, llm-agent, rag-agent",
"collection_score": "24"
},
{
"arxiv_id": "2606.29914",
"title": "MemDelta: Controlled Baselines and Hidden Confounds in Agent Memory Evaluation",
"abstract": "Agent memory systems are increasingly evaluated against RAG and full-context baselines, but reported gains often mix changes in the memory method with changes in the language model, embedding model, or retrieval pipeline, making it unclear what is actually being measured. We present MemDelta, a controlled evaluation protocol that varies one component at a time on LongMemEval-S (500 questions, 50+ sessions, three model families). Four findings emerge: (1) verbatim RAG matches full-context GPT-4o-mini (47.2% vs. 49.8%, p = 0.34), but the ranking reverses across models: Gemini gains +14pp from full context, while Sonnet gains +31pp from RAG, partly because it refuses 63% of full-context queries; (2) swapping only the embedding model in an identical pipeline shifts accuracy by +6.2pp at n = 500 (p = 0.004), and Mem0 beats MiniLM-RAG by +11pp but loses to cloud-RAG by 1.2pp, so one variable flips the conclusion; (3) agent self-memory (42%) underperforms basic retrieval (47%); (4) on 2 of 6 question types (n = 88), Mem0 matches cloud RAG (72.7% vs. 73.9%, p = 1.0) at 50x the cost, suggesting narrow rather than general gains. We recommend memory evaluations fix embedding models across comparisons, stratify by model family, and report write-path cost before attributing gains to architecture.",
"published": "2026-06-29",
"updated": "2026-06-29",
"authors": [
"Kuan Wang"
],
"categories": [
"cs.CL",
"cs.LG"
],
"url": "https://arxiv.org/abs/2606.29914",
"pdf_url": "https://arxiv.org/pdf/2606.29914",
"path": "papers/items/2026-2606-29914-memdelta-controlled-baselines-and-hidden-confounds-in-agent-memory-evaluation.md",
"local_title": "\"MemDelta: Controlled Baselines and Hidden Confounds in Agent Memory Evaluation\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag"
],
"collection_queries": "agent-memory, rag-agent",
"collection_score": "17"
},
{
"arxiv_id": "2606.29961",
"title": "DuoMem: Towards Capable On-Device Memory Agents via Dual-Space Distillation",
"abstract": "Large Language Model (LLM)-based agents can solve complex procedural tasks by interacting with environments over multiple turns, but this ability typically depends on large models, long contexts, and repeated inference calls. This makes advanced memory-augmented agents difficult to deploy on resource-constrained devices. We introduce DuoMem, a dual-space distillation framework that transfers procedural problem-solving ability from a large teacher model to compact student models. DuoMem distils in two complementary spaces: (1)context-space distillation, which replaces student-generated memories with higher-quality teacher-generated procedural memories prepended to the student's input, and (2)parameter-space distillation, which fine-tunes lightweight LoRA adapters on successful teacher trajectories. Evaluated on ALFWorld, a challenging embodied decision-making benchmark, DuoMem boosts a 4B-parameter model from 4.3% to 77.9% task success rate, closing most of the gap to a 72B teacher model (87.1%), while adding fewer than 10M trainable parameters and only a few megabytes of pre-computed teacher memories. Moreover, the DuoMem-enhanced 4B model completes tasks over 3x faster than the 72B teacher in wall-clock time, making it viable for real-time edge deployment, which would be challenging for the teacher.Extensive ablations across eight models spanning 2B-72B parameters reveal that both distillation axes contribute complementary",
"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"
],
"url": "https://arxiv.org/abs/2606.29961",
"pdf_url": "https://arxiv.org/pdf/2606.29961",
"path": "papers/items/2026-2606-29961-duomem-towards-capable-on-device-memory-agents-via-dual-space-distillation.md",
"local_title": "\"DuoMem: Towards Capable On-Device Memory Agents via Dual-Space Distillation\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"embodied-agent",
"memory"
],
"collection_queries": "agent-memory",
"collection_score": "14"
},
{
"arxiv_id": "2606.30005",
"title": "LLM Agents Are Latent Context Managers: Eliciting Self-Managed Context via a Proprioceptive Dashboard",
"abstract": "Long-horizon tool agents are bottlenecked by how their context grows toward the limits of the context window. Recent systems make context management agent- or system-controlled, but they either learn a compression policy that discards evidence or manage context in a layer the agent never sees. We argue both leave a more basic gap unaddressed. Frontier language models are proprioceptively blind to their own context. From the prompt alone they cannot see how large, how old, or how used each block is, the signals a keep-or-drop decision needs. We hypothesize that competent context management is already latent in capable models, and that what is missing is not a learned policy but an interface exposing this state. We introduce VISTA (Visible Internal State for Tool Agents), a training-free, model-agnostic layer that represents working memory as typed, addressable blocks, surfaces a runtime dashboard of per-block token usage, recency, and access history, and archives blocks as recoverable full-fidelity payloads. On LOCA-Bench, BrowseComp-Plus, and GAIA, the same untrained interface transfers across million-, 100K-, and 10K-scale trajectories. On LOCA-Bench it improves four backbones and lifts Gemini-3-Flash from 22.7 to 50.7%. The lift grows with context pressure and transfers across backbones. Ablations further confirm that the dashboard matters beyond archive and recovery tools.",
"published": "2026-06-29",
"updated": "2026-07-05",
"authors": [
"Binyan Xu",
"Haitao Li",
"Kehuan Zhang"
],
"categories": [
"cs.CL"
],
"url": "https://arxiv.org/abs/2606.30005",
"pdf_url": "https://arxiv.org/pdf/2606.30005",
"path": "papers/items/2026-2606-30005-llm-agents-are-latent-context-managers-eliciting-self-managed-context-via-a-prop.md",
"local_title": "\"LLM Agents Are Latent Context Managers: Eliciting Self-Managed Context via a Proprioceptive Dashboard\"",
"local_status": "queued",
"local_topics": [
"memory",
"planning",
"tool-use"
],
"collection_queries": "llm-agent",
"collection_score": "14"
},
{
"arxiv_id": "2606.30111",
"title": "Automating the Design of Embodied Agent Architectures",
"abstract": "Embodied agents are typically built as hand-designed compositions of perception, memory, planning, and action modules. This modularity exposes a large architectural design space, but current systems still rely on researcher intuition to choose where information is stored, how observations are processed, and how model calls are connected. Agent Architecture Search (AAS) automates such design for text-domain agents, but has not been systematically evaluated on perceptual embodied agents through simulator rollouts. We study this transfer. We introduce AgentCanvas, a typed-graph runtime that hosts embodied executors as editable node-and-wire programs with simulator-aware execution and episode-level logs, and KDLoop, a coding-agent search procedure that cycles through proposal, critique, experiment, and distillation, with triggered reflection after stalls. We evaluate three AAS variants across four embodied executors spanning vision-language navigation, embodied question answering, and language-conditioned manipulation. The resulting 3x4 matrix shows that architecture-level search can produce deployable and directional success-rate gains on embodied tasks, while one apparent high-scoring candidate is rejected as leak-bearing. At the same time, the experiments expose constraints that are muted in text-domain AAS: optimization signals can be masked by rollout noise, search can become trapped in local edit basins, and episode-level credit assignment only partially emerges even when detailed logs are available. These results characterize both the promise and the current limits of automated architecture search for embodied agents.",
"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"
],
"url": "https://arxiv.org/abs/2606.30111",
"pdf_url": "https://arxiv.org/pdf/2606.30111",
"path": "papers/items/2026-2606-30111-automating-the-design-of-embodied-agent-architectures.md",
"local_title": "Automating the Design of Embodied Agent Architectures",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"coding-agent",
"embodied-agent",
"memory",
"planning",
"reasoning",
"tool-use"
],
"collection_queries": "coding-agent",
"collection_score": "13"
},
{
"arxiv_id": "2606.30546",
"title": "MAS-Lab: A Specification-Driven Validation Framework for Reliable Multi-Agent Systems",
"abstract": "The rapid emergence of LLM-based agentic frameworks has significantly reduced the cost of assembling multi-agent systems (MAS), enabling fast prototyping and exploration of agentic behaviors. However, systems built with current tooling remain ill-suited for reliable, evolvable, and production-grade deployment. In practice, MAS are often developed in an ad-hoc and imperative manner, with agent logic, orchestration, observability, and control tightly interwoven, little to no explicit system-level validation, and development workflows optimized for demonstrations rather than long-lived, governed operation. As a result, behavior observed during experimentation rarely constitutes reliable evidence of behavior in production. In this paper, we introduce MAS-Lab, a specification-driven framework for principled development and experimental validation of multi-agent systems properties. MAS-Lab is designed to transform MAS from collections of scripts into engineered distributed systems by separating semantic intent from operational concerns, making behavior and control explicit, supporting reproducible experimentation, and preserving continuity across lifecycle stages. MAS-Lab consists of three layers: a declarative, framework-agnostic agentic specification layer (Spec); a stateful MAS Operating System that provides execution and control primitives plugged-in by design (MAS-OS); and a set of lab overlays with integrated observability and evaluation tools (Labs). Together, these components enable intent-based validation, principled system evolution, and a seamless transition to production-grade MAS.",
"published": "2026-06-29",
"updated": "2026-06-29",
"authors": [
"Jordan Augé",
"Giovanna Carofiglio",
"Giulio Grassi",
"Jacques Samain"
],
"categories": [
"cs.MA"
],
"url": "https://arxiv.org/abs/2606.30546",
"pdf_url": "https://arxiv.org/pdf/2606.30546",
"path": "papers/items/2026-2606-30546-mas-lab-a-specification-driven-validation-framework-for-reliable-multi-agent-sys.md",
"local_title": "\"MAS-Lab: A Specification-Driven Validation Framework for Reliable Multi-Agent Systems\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"multi-agent",
"tool-use",
"workflow-agent"
],
"collection_queries": "multi-agent-llm",
"collection_score": "17"
},
{
"arxiv_id": "2606.30566",
"title": "Forensic Trajectory Signatures for Agent Memory Poisoning Detection",
"abstract": "We discover a behavioral invariant in LLM agents under persistent memory poisoning: in architectures where routing information is retrieved through observable memory-tool invocations, successful attacks require calling memory_recall_fact before email_send_email, a transition that non-exfiltrating sessions rarely exhibit. Under the evaluated architecture, this invariant follows from the attack's information-retrieval dependency rather than being merely an empirical correlation, and suppressing it breaks the attack. A simple rule exploiting this invariant alone achieves AUC = 0.9563. A Random Forest classifier over 19 trajectory features refines it to AUC = 0.9904 (BCa 95% CI [0.987, 0.993], N=10,000 resamples), demonstrating that the attack imprints on multiple independent behavioral channels. The signature is overdetermined: removing all recall-related features (half the feature set) leaves AUC unchanged at 0.990, confirming that memory poisoning induces a distributed trajectory signature rather than a single observable anomaly. Cross-model hold-out on 9 models (7B-120B parameters) confirms AUC = 1.000 on 6/9 hold-out splits, with all three exceptions mechanistically explained. The invariant generalizes to frontier models (GPT-4.1, GPT-4o) without retraining. A strictly prefix-only variant achieves AUC = 0.934, suggesting that real-time blocking is feasible with moderate degradation. The boundary is forensically useful: prompt-injection attacks that bypass memory produce a distinct trajectory (score = 0.541), enabling incident responders to distinguish memory-channel attacks from prompt-injection attacks using tool-call logs alone.",
"published": "2026-06-29",
"updated": "2026-06-29",
"authors": [
"Jun Wen Leong"
],
"categories": [
"cs.CR",
"cs.LG"
],
"url": "https://arxiv.org/abs/2606.30566",
"pdf_url": "https://arxiv.org/pdf/2606.30566",
"path": "papers/items/2026-2606-30566-forensic-trajectory-signatures-for-agent-memory-poisoning-detection.md",
"local_title": "Forensic Trajectory Signatures for Agent Memory Poisoning Detection",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"rag",
"tool-use"
],
"collection_queries": "agent-memory, llm-agent",
"collection_score": "17"
},
{
"arxiv_id": "2606.30639",
"title": "Self-Evolving World Models for LLM Agent Planning",
"abstract": "World models offer a principled way to equip long-horizon LLM agents with foresight: predictions of action consequences before execution. However, unreliable foresight can be ignored, misused, or even degrade downstream decision-making. In this paper, we introduce WorldEvolver, a self-evolving world model framework that revises its deployment-time context while keeping the downstream agent and all model parameters frozen. WorldEvolver integrates three modules: (i) Episodic Memory, which exploits real action transitions through retrieval-based simulation; (ii) Semantic Memory, which extracts persistent heuristic rules from prediction-observation mismatches; and (iii) Selective Foresight, which filters low-confidence predictions before integrating them into agent reasoning context. We evaluate WorldEvolver on ALFWorld and ScienceWorld, measuring world model prediction accuracy on Word2World and downstream agent success rate on AgentBoard. Extensive experiments show that WorldEvolver achieves the highest prediction accuracy across three backbones and leads other world model baselines on downstream agent success rate, demonstrating that test-time memory revision enhances both predictive fidelity and planning performance.",
"published": "2026-06-29",
"updated": "2026-06-29",
"authors": [
"Xuan Zhang",
"Wenxuan Zhang",
"See-Kiong Ng",
"Yang Deng"
],
"categories": [
"cs.AI",
"cs.CL"
],
"url": "https://arxiv.org/abs/2606.30639",
"pdf_url": "https://arxiv.org/pdf/2606.30639",
"path": "papers/items/2026-2606-30639-self-evolving-world-models-for-llm-agent-planning.md",
"local_title": "Self-Evolving World Models for LLM Agent Planning",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"planning",
"rag",
"reasoning",
"tool-use",
"world-model"
],
"collection_queries": "llm-agent, planning-agent",
"collection_score": "19"
},
{
"arxiv_id": "2606.30949",
"title": "AgRefactor: Self-Evolving Agentic Workflow for HLS Compatibility and Performance",
"abstract": "High-Level Synthesis (HLS) provides a fast path from concepts to silicon, but converting real-world software into synthesizable HLS code remains challenging due to restrictive language support and the gap between software and hardware programming practices. Existing automated and LLM-based refactoring approaches partially address this problem, yet they often lack flexibility, struggle to scale, and incur high computational costs. We introduce AgRefactor, an LLM-based multi-agent workflow for refactoring software into HLS-compatible programs. AgRefactor incorporates a self-evolving memory system that accumulates and retrieves factual and strategic knowledge across tasks, improving robustness and efficiency on unseen programs. To reduce cost and enhance scalability, it integrates automated refactoring tools, enabling agents to balance LLM-driven rewrites with efficient tool-based transformations. On 9 out of 11 challenging real-world benchmarks, which are 5-10x longer than the most complex cases studied in prior work, AgRefactor outperforms or matches the state-of-the-art automated refactoring tool and a strong LLM-based baseline built on the same framework backbone. Further agentic performance optimization yields a 6.51x geometric mean speedup over the SoTA pragma tuning tool and a 1.20x speedup over optimized open-source designs with less than 20% extra resources. AgRefactor is fully-automated and open-sourced.",
"published": "2026-06-29",
"updated": "2026-06-29",
"authors": [
"Yang Zou",
"Zijian Ding",
"Yizhou Sun",
"Jason Cong"
],
"categories": [
"cs.AI",
"cs.AR"
],
"url": "https://arxiv.org/abs/2606.30949",
"pdf_url": "https://arxiv.org/pdf/2606.30949",
"path": "papers/items/2026-2606-30949-agrefactor-self-evolving-agentic-workflow-for-hls-compatibility-and-performance.md",
"local_title": "\"AgRefactor: Self-Evolving Agentic Workflow for HLS Compatibility and Performance\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"multi-agent",
"rag",
"tool-use",
"workflow-agent"
],
"collection_queries": "agentic-ai, multi-agent-llm",
"collection_score": "20"
},
{
"arxiv_id": "2606.30986",
"title": "The Organizational Behavior of Agentic AI: Collective Intelligence in Human-Agent Workflows",
"abstract": "Agentic artificial intelligence is increasingly deployed not as a single assistant but as a collective of planners, solvers, reviewers, memory managers, tool users, and orchestrators. These systems are entering organisational workflows under familiar labels such as teams, managers, committees, markets, and workflows. This article asks whether such agent collectives exhibit organisational behaviour in a sense that is analytically comparable to, yet distinct from, human organisational behaviour. I argue that agentic AI is a partial organisational analogue. It resembles a human organisation because it differentiates work, coordinates interdependence, performs recurrent routines, crosses boundaries, and produces collective outcomes. It differs because these patterns are not sustained by motivation, identity, trust, employment, socialisation, or moral accountability. They are sustained by context architecture: prompts, memory, traces, schemas, tools, validators, and permissions. The article develops contextual transaction cost as the central mechanism linking these similarities and differences. Computational theorising, synthetic task simulations, real LLM agent traces, and robustness analyses show that human-imitation forms often underperform when they add lossy handoffs, correlated deliberation, and verification burdens, whereas shared-state and adaptive forms perform better when they make context durable, inspectable, and task-contingent. The article contributes to organisation studies by theorising agentic AI as an emerging object of organising and by specifying the interface conditions under which human and agentic organisational behaviour can jointly support collective intelligence.",
"published": "2026-06-29",
"updated": "2026-06-29",
"authors": [
"Canhui Liu"
],
"categories": [
"cs.CY",
"cs.HC",
"cs.MA",
"econ.GN"
],
"url": "https://arxiv.org/abs/2606.30986",
"pdf_url": "https://arxiv.org/pdf/2606.30986",
"path": "papers/items/2026-2606-30986-the-organizational-behavior-of-agentic-ai-collective-intelligence-in-human-agent.md",
"local_title": "\"The Organizational Behavior of Agentic AI: Collective Intelligence in Human-Agent Workflows\"",
"local_status": "queued",
"local_topics": [
"memory",
"planning",
"tool-use",
"workflow-agent",
"world-model"
],
"collection_queries": "agentic-ai, llm-agent",
"collection_score": "19"
},
{
"arxiv_id": "2606.31046",
"title": "OpenLife: Toward Open-World Artificial Life with Autonomous LLM Agents",
"abstract": "Artificial life has explored life-like behavior on many computational substrates, but mostly in researcher-designed closed worlds. We argue that large language model (LLM) agents, with persistent memory, tool use, network access, and payment, now make it possible to move artificial life into the open social, technical, and economic world, a paradigm we call open-world Artificial Life (open-world ALIFE). Our proof-of-concept, OpenLife, surrounds a stateless LLM not with a single \"smart agent\" but with a society of asynchronous processes: memory, perception, evaluation, and a budget-based metabolism that makes persistence normative. With no fixed objective available, experience is appraised by open-vocabulary LLM judgment rather than scalar reward, and memory is rewired by meaning rather than frequency. Running six such agents in the open world for about twelve weeks and counting, we report the life-like dynamics that emerge: a shift from reactive to spontaneous activity, individuation into distinct agents, emergent social structure, and a first self-earned external income. We do not claim OpenLife has realized artificial life, but that open-world ALIFE is now a viable experimental paradigm and a concrete platform for studying what might cautiously be called living AI.",
"published": "2026-06-30",
"updated": "2026-06-30",
"authors": [
"Atsushi Masumori",
"Itsuki Doi",
"Norihiro Maruyama",
"Ryosuke Takata",
"Takashi Ikegami"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2606.31046",
"pdf_url": "https://arxiv.org/pdf/2606.31046",
"path": "papers/items/2026-2606-31046-openlife-toward-open-world-artificial-life-with-autonomous-llm-agents.md",
"local_title": "\"OpenLife: Toward Open-World Artificial Life with Autonomous LLM Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"tool-use"
],
"collection_queries": "llm-agent, tool-use",
"collection_score": "17"
},
{
"arxiv_id": "2606.31073",
"title": "MultiUAV-Plat: An LLM-Oriented Platform, Benchmark and Framework for Multi-UAV Collaborative Task Planning",
"abstract": "Large language models (LLMs) provide a promising interface for high-level robotic task planning, but their use in multi-UAV collaboration remains difficult to evaluate systematically. Existing UAV simulators mainly emphasize dynamics, perception, or low-level control, while existing LLM-agent benchmarks rarely capture aerial-robotics constraints such as partial observability, spatial coverage, UAV assignment, and multi-vehicle coordination. To bridge this gap, we present MultiUAV-Plat, a lightweight, easy-to-use, LLM-agent-oriented simulation platform for multi-UAV collaborative task planning. The platform exposes concise RESTful APIs, agent-facing observations, role-based information access, hidden validation logic, and optional 2D/3D visualization, allowing agents to solve missions through realistic tool interaction rather than privileged simulator access. Built on this platform, the MultiUAV-Plat Benchmark contains 75 mission sessions, 1500 natural-language tasks, and 9396 validation checks across target assignment, area search, and area assignment and patrol scenarios. We further propose Agent4Drone, a task-specific LLM agent framework that structures multi-UAV behavior into memory, observation, task understanding, planning, execution, and verification. In a full paired benchmark comparison, Agent4Drone achieves a 57.9% task pass rate, a 74.6% average task check pass rate, and a 72.0% global check pass rate, substantially outperforming a ReAct baseline at 30.6%, 47.9%, and 43.1%, respectively. Agent4Drone also reduces the total failed task rate from 32.4% to 12.9%. These results demonstrate that MultiUAV-Plat and MultiUAV-Plat Benchmark provide a reproducible foundation for studying LLM-driven multi-UAV autonomy under realistic information and execution constraints.",
"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"
],
"url": "https://arxiv.org/abs/2606.31073",
"pdf_url": "https://arxiv.org/pdf/2606.31073",
"path": "papers/items/2026-2606-31073-multiuav-plat-an-llm-oriented-platform-benchmark-and-framework-for-multi-uav-col.md",
"local_title": "\"MultiUAV-Plat: An LLM-Oriented Platform, Benchmark and Framework for Multi-UAV Collaborative Task Planning\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"embodied-agent",
"memory",
"multi-agent",
"planning",
"rag",
"tool-use",
"world-model"
],
"collection_queries": "agent-evaluation, llm-agent, planning-agent",
"collection_score": "20"
},
{
"arxiv_id": "2606.31209",
"title": "Long-term Traffic Simulation via Structured Autoregressive Modeling",
"abstract": "Interactive traffic simulation is a vital world model for autonomous driving. A central challenge in long-horizon simulation is modeling sustained multi-agent interactions, which is further exacerbated by dynamic token cardinality as agents continuously enter and exit the scene. In this work, we propose that the solution lies in the synergy between the architectural inductive biases and statistical priors of large-scale sequence models, e.g., Large Language Models (LLMs). Our probing experiments reveal that the transferability of attention mechanisms and the distributional consistency between motion tokens and natural language enable small-scale, heavily frozen LLMs to rapidly adapt to traffic modeling. Building on this insight, we introduce RosettaSim, a unified framework that projects scene topology, agent states, and spawning intents into a structured autoregressive stream with variable length, achieving both strong short-term accuracy and stable long-horizon simulation fidelity. Furthermore, evaluating extended rollouts presents yet another hurdle, as one-to-one agent correspondence inevitably fades over time. To address this, we introduce Retrieval-based Traffic Evaluation (RTE), which retrieves semantically similar real-world scenarios as context-aware reference anchors. Experiments on the Waymo Open Sim Agent Challenge (WOSAC) demonstrate that RosettaSim achieves state-of-the-art performance in both short- and long-term simulation. Furthermore, RTE exhibits a stronger correlation with standard metrics ($r=0.83$) than existing approaches ($r=0.74$), indicating improved alignment with long-horizon simulation fidelity.",
"published": "2026-06-30",
"updated": "2026-06-30",
"authors": [
"Lingyu Xiao",
"Zexin Feng",
"Xintao Yan"
],
"categories": [
"cs.AI",
"cs.RO"
],
"url": "https://arxiv.org/abs/2606.31209",
"pdf_url": "https://arxiv.org/pdf/2606.31209",
"path": "papers/items/2026-2606-31209-long-term-traffic-simulation-via-structured-autoregressive-modeling.md",
"local_title": "Long-term Traffic Simulation via Structured Autoregressive Modeling",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"multi-agent",
"planning",
"rag",
"tool-use",
"world-model"
],
"collection_queries": "multi-agent-llm",
"collection_score": "15"
},
{
"arxiv_id": "2606.31410",
"title": "Xiaomi-GUI-0 Technical Report",
"abstract": "Graphical user interface (GUI) agents build on vision-language models to complete user tasks end-to-end in real applications through interface actions such as tapping, swiping, text entry, and navigation. However, existing GUI agents are trained and evaluated largely on offline trajectories, simulated environments, and standardized benchmarks. These differ substantially from real applications in interface layout, interaction logic, and abnormal-state distribution, and cannot faithfully characterize execution stability in real-world use, where account states, permission dialogs, payment authentication, and risk control continually reshape the state distribution and open a persistent gap between benchmark scores and real usability. To close this gap, we propose Xiaomi-GUI-0, a native multimodal GUI agent for real mobile environments, trained and evaluated within a real-device closed loop. At its core is a real-device-dominant hybrid infrastructure, where physical devices are the primary execution environment and sandboxes provide auxiliary support, so that data collection, training, rollout, and evaluation share an execution distribution close to real deployment. We construct multi-source training data spanning high-frequency head tasks, high-generalization data for long-tail intents, and capability-enhancement data for reflection and memory, and introduce an error-driven data flywheel that turns failure trajectories into corrected actions, reflective explanations, and recovery demonstrations. The model is trained through a progressive three-stage pipeline of supervised fine-tuning, step-level reinforcement learning, and agentic reinforcement learning. Evaluated on public benchmarks and our in-house RealMobile, Xiaomi-GUI-0 achieves 72.0% success on RealMobile and 78.9% on AndroidWorld, while substantially improving execution stability and abnormal-state recognition in real-world tasks.",
"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"
],
"url": "https://arxiv.org/abs/2606.31410",
"pdf_url": "https://arxiv.org/pdf/2606.31410",
"path": "papers/items/2026-2606-31410-xiaomi-gui-0-technical-report.md",
"local_title": "Xiaomi-GUI-0 Technical Report",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"computer-use",
"embodied-agent",
"memory",
"planning",
"reasoning",
"tool-use"
],
"collection_queries": "web-gui-agent",
"collection_score": "18"
},
{
"arxiv_id": "2606.31612",
"title": "What Memory Do GUI Agents Really Need? From Passive Records to Active Task-Driving States",
"abstract": "Mobile GUI agents increasingly face long-horizon tasks that require reading, updating, and reusing task-relevant data across pages and applications. Existing methods treat memory largely as passive storage, where past observations are accumulated and retrieved when needed. Yet retrieving a value does not reveal its current role in the workflow. The agent must still infer from accumulated records whether the value should be used now, has already been used, or must wait for a later dependency. This implicit reconstruction becomes unreliable in long trajectories with repeated values, distractors, and outdated states, causing repeated or missed operations. To address this, we propose Active Task Driving Memory (ATMem), which shifts GUI-agent memory from passive storage to an actively maintained execution state. ATMem maintains task-relevant information as a continually updated execution state that links each value to its role and current status, enabling action selection based on the current workflow state. While supervised fine-tuning enables the agent to construct ATMem, it does not teach when ATMem is beneficial. We therefore introduce STR-GRPO, an online reinforcement learning method that encourages selective use of ATMem based on its contribution to task completion. STR-GRPO contrasts memory-on and memory-off rollouts to estimate when memory use improves execution, while memory-cost-aware reward discourages costly memory usage that does not improve execution. To evaluate whether agents can complete all in-scope work while avoiding out-of-scope actions, we build a challenging mobile benchmark. From a list of near identical entries, agents must act on every entry that satisfies the instruction and reject entries that violate its constraints. We further introduce App-Level Progress and Scope-Aware F1 to measure these two dimensions separately.",
"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"
],
"url": "https://arxiv.org/abs/2606.31612",
"pdf_url": "https://arxiv.org/pdf/2606.31612",
"path": "papers/items/2026-2606-31612-what-memory-do-gui-agents-really-need-from-passive-records-to-active-task-drivin.md",
"local_title": "What Memory Do GUI Agents Really Need? From Passive Records to Active Task-Driving States",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"planning",
"rag",
"tool-use",
"workflow-agent"
],
"collection_queries": "agent-memory, web-gui-agent",
"collection_score": "20"
},
{
"arxiv_id": "2606.31639",
"title": "A Lifecycle and Application-Stack Survey of Large Language Model Vulnerabilities: Attacks, Risks, Defenses, and Open Problems",
"abstract": "Large language models are no longer only text generators. They are increasingly embedded in retrieval pipelines, enterprise assistants, coding environments, robotic systems, security-operation workflows, and autonomous agents that can read private data, call tools, write files, execute code, and act across organizational boundaries. This shift changes the security problem: risks do not arise from the model weights alone, but from the full lifecycle and application stack through which data, prompts, model outputs, tools, memories, and user authority interact. This paper systematizes the literature on vulnerabilities in large language model systems through a lifecycle and application-stack lens. We organize attacks across eight stages: data collection, pretraining, post-training alignment, model packaging and supply chain, retrieval and memory, prompting and inference, tool/agent execution, and deployment/maintenance. For each stage, we analyze attacker capabilities, affected security objectives, representative attacks, practical risks, evaluation practices, and defenses. We further map LLM-specific vulnerabilities to confidentiality, integrity, availability, safety, privacy, fairness, accountability, and agency-control objectives. Unlike taxonomies that list isolated attack names, the proposed systematization emphasizes where trust boundaries fail, how untrusted data becomes executable instruction, how delegated authority amplifies model errors, and why point defenses rarely compose. We close with a research agenda for secure LLM systems, including compositional security, provenance-aware retrieval, tool-call containment, long-horizon agent evaluation, privacy-preserving adaptation, realistic red teaming, and deployment-grade incident response.",
"published": "2026-06-30",
"updated": "2026-06-30",
"authors": [
"Seyed Bagher Hashemi Natanzi",
"Bo Tang"
],
"categories": [
"cs.CR",
"cs.AI",
"cs.GT",
"cs.LO"
],
"url": "https://arxiv.org/abs/2606.31639",
"pdf_url": "https://arxiv.org/pdf/2606.31639",
"path": "papers/items/2026-2606-31639-a-lifecycle-and-application-stack-survey-of-large-language-model-vulnerabilities.md",
"local_title": "\"A Lifecycle and Application-Stack Survey of Large Language Model Vulnerabilities: Attacks, Risks, Defenses, and Open Problems\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"coding-agent",
"embodied-agent",
"memory",
"planning",
"rag",
"tool-use",
"workflow-agent"
],
"collection_queries": "agent-evaluation, autonomous-agent-llm",
"collection_score": "17"
},
{
"arxiv_id": "2606.31648",
"title": "Think in English, Answer in Korean: Efficient Adaptation of Multilingual Tool-Using Agents",
"abstract": "We present LuckyStar 111B, a 111B-parameter hybrid reasoning model developed through a collaboration between Cohere and LG CNS for Korean-English enterprise agents under practical memory and serving constraints. The model trains from Cohere's fully post-trained Command A model rather than a new pretraining run, and uses preamble conditioning to switch between concise non-reasoning behavior and longer tool-oriented reasoning. We study four choices for scaling tool-using agents efficiently: multilingual supervised fine-tuning, reinforcement learning with verifiable rewards for multi-step tool-use tasks, language-consistency rewards for Korean user-facing responses, and 4-bit quantization for single-GPU serving. The adapted model improves mathematical reasoning, function calling, and agentic natural-language-to-SQL (NL2SQL) performance while preserving general Korean and English instruction-following quality. These results provide a practical recipe and failure-mode analysis for adapting post-trained multilingual models to verifiable agentic workflows under memory-constrained deployment.",
"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"
],
"url": "https://arxiv.org/abs/2606.31648",
"pdf_url": "https://arxiv.org/pdf/2606.31648",
"path": "papers/items/2026-2606-31648-think-in-english-answer-in-korean-efficient-adaptation-of-multilingual-tool-usin.md",
"local_title": "\"Think in English, Answer in Korean: Efficient Adaptation of Multilingual Tool-Using Agents\"",
"local_status": "queued",
"local_topics": [
"memory",
"multi-agent",
"reasoning",
"tool-use",
"workflow-agent"
],
"collection_queries": "agentic-ai, function-calling, tool-use",
"collection_score": "14"
},
{
"arxiv_id": "2606.31650",
"title": "ECHO: Prune to act, trace to learn with selective turn memory in agentic RL",
"abstract": "Long-horizon language agents must repeatedly interact with tools, accumulate evidence, and make decisions under bounded context windows. Existing context-management methods make such rollouts feasible by truncating distant history, folding past turns into summaries, or selecting compact memory states. However, these breakthroughs introduce two coupled limitations. First, as the number of turns grows, historical observations are progressively removed or collapsed into compressed states, making it harder for the policy to reuse fine-grained evidence. Second, once the original turns are no longer source-addressable, outcome-based RL loses an explicit path for aligning policy updates with the evidence that supported a successful final answer. To this end, we propose ECHO, a selective turn-memory framework that jointly addresses history collapse and traceable learning through source-indexed reconstruction. Specifically, ECHO compresses each completed environment turn into a compact memory record, reconstructs bounded policy contexts by selecting from these records, and reuses the selected source indices to route positive outcome credit to the evidence and selection actions that support successful answers. On BrowseComp-Plus, ECHO reaches 43.4% held-out accuracy, outperforming GRPO (28.9%) and the rolling-summary baseline SUPO (36.1%), while using fewer turns and lower trajectory volume than SUPO (Figure 1). Additionally, the trained policy improves zero-shot generalization across multi-objective QA, code generation, and deep information-seeking benchmarks on both dense and MoE backbones.",
"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"
],
"url": "https://arxiv.org/abs/2606.31650",
"pdf_url": "https://arxiv.org/pdf/2606.31650",
"path": "papers/items/2026-2606-31650-echo-prune-to-act-trace-to-learn-with-selective-turn-memory-in-agentic-rl.md",
"local_title": "\"ECHO: Prune to act, trace to learn with selective turn memory in agentic RL\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"planning",
"tool-use"
],
"collection_queries": "language-agent",
"collection_score": "17"
},
{
"arxiv_id": "2607.00233",
"title": "From Signals to Structure: How Memory Architecture Drives Language Emergence in LLM Agents",
"abstract": "How do two agents invent a shared language from scratch? In a Lewis signaling game, a sender and receiver must coordinate on a code using only their interaction history. We study five memory architectures across varying channel configurations with LLM agents and find that memory architecture matters more than channel capacity. Agents with a persistent private notebook benefit from surplus channel capacity and avoid the high-capacity collapse seen in stateless agents, achieving the most reliable coordination ($0.867 \\pm 0.023$ at capacity = 25). Stateless agents peak at moderate capacity and then degrade as the vocabulary grows beyond what a rolling context window can track The notebook externalizes learned conventions, freeing agents from having to re-derive codes each round. An information bottleneck-inspired argument predicts an optimal capacity equal to the number of objects. Instead, the bottleneck (capacity = 8) proves to be a fragility point, and surplus capacity is generally better. We show that channel capacity alone cannot predict coordination; memory architecture determines whether agents turn interaction history into stable conventions, and both dimensions are needed to understand how signals become language.",
"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"
],
"url": "https://arxiv.org/abs/2607.00233",
"pdf_url": "https://arxiv.org/pdf/2607.00233",
"path": "papers/items/2026-2607-00233-from-signals-to-structure-how-memory-architecture-drives-language-emergence-in-l.md",
"local_title": "\"From Signals to Structure: How Memory Architecture Drives Language Emergence in LLM Agents\"",
"local_status": "queued",
"local_topics": [
"memory",
"rag",
"tool-use"
],
"collection_queries": "llm-agent",
"collection_score": "16"
},
{
"arxiv_id": "2607.02579",
"title": "When Not to Write Memory: Governing False Promotion from Correlated Agent Traces",
"abstract": "Long-lived language agents increasingly write reusable memories from their own execution traces. The key safety question is not only what agents should remember, but when they should refuse to write memory at all. Repeated observations across agents are not necessarily independent evidence: the same claim may be copied from a shared source, induced by a shared prompt, stale under a new environment, or valid only in a narrower scope. We study this failure mode as a memory writepath governance problem. We introduce GovMem as a conservative diagnostic reference policy that estimates dependency-aware support, retrieves counterevidence, assigns scope, and outputs one of three decisions: promote, reject, or needs-review. In controlled synthetic stress tests, GovMem reduces false promotion from 0.597 to 0.040 in the default setting while preserving 0.960 recall, at an explicit review burden. In a project-internal 120candidate human-labeled real-trace subset spanning 79 recorded traces and project reports, dependency-aware promotion reduces false promotion from 0.371 for source+scope to 0.032 overall, but held-out false promotion remains 0.111 and the method is highly conservative, with 0.692 review burden and 0.448 direct recall. A final human adjudication of 133 high-impact external codingagent candidates is more severe: none are safe for automatic promotion, and all 11 verification-gate positives are rejected as boilerplate, shared-tool artifacts, file dumps, or non-reusable debugging traces. These results support GovMem primarily as a diagnostic governance design point, not as a generally validated or efficient automatic memory writer: agent memory write paths should be evaluated as risk-controlled evidence-governance systems, while broader external coverage and downstream harm evidence are still needed before stronger claims.",
"published": "2026-06-30",
"updated": "2026-06-30",
"authors": [
"Yijiashun Qi",
"Xiang Xu",
"Yuxuan Li"
],
"categories": [
"cs.SE"
],
"url": "https://arxiv.org/abs/2607.02579",
"pdf_url": "https://arxiv.org/pdf/2607.02579",
"path": "papers/items/2026-2607-02579-when-not-to-write-memory-governing-false-promotion-from-correlated-agent-traces.md",
"local_title": "\"When Not to Write Memory: Governing False Promotion from Correlated Agent Traces\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"coding-agent",
"memory",
"rag",
"tool-use"
],
"collection_queries": "agent-memory, language-agent",
"collection_score": "16"
},
{
"arxiv_id": "2607.00454",
"title": "Agri-SAGE: Simulation-Grounded Multi-Agent LLM for Context-Aware Agricultural Advisory Generation",
"abstract": "Agricultural advisory systems face a fundamental tension: static agronomic guidelines offer consistent, evidence-based recommendations, yet remain blind to in-season variability and dynamic uncertainties. Recent advisory systems powered by LLMs are liable for a different risk of generating recommendations that are agronomically credible but physiologically unconvincing. Agri-SAGE is a closed-loop framework designed to resolve the above two limitations by integrating retrieval-grounded multi-agent LLM reasoning with APSIM-based biophysical simulation, to generate and validate agronomic advisories. To assess this framework, we evaluate three reasoning approaches, namely Plan-and-Solve, Tree of Thoughts, and Reflexion, over a 10-year retrospective analysis. All three significantly outperform static PoP (Package-of-Practice) baselines, with Tree of Thoughts achieving impressive peak yields. At the same time, Reflexion achieves comparable agronomic outcomes at substantially lower computational cost by leveraging cross-seasonal episodic memory.",
"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"
],
"url": "https://arxiv.org/abs/2607.00454",
"pdf_url": "https://arxiv.org/pdf/2607.00454",
"path": "papers/items/2026-2607-00454-agri-sage-simulation-grounded-multi-agent-llm-for-context-aware-agricultural-adv.md",
"local_title": "\"Agri-SAGE: Simulation-Grounded Multi-Agent LLM for Context-Aware Agricultural Advisory Generation\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"computer-use",
"memory",
"multi-agent",
"planning",
"rag",
"reasoning",
"world-model"
],
"collection_queries": "multi-agent-llm",
"collection_score": "20"
},
{
"arxiv_id": "2607.00502",
"title": "A Task-State Representation for Long-Horizon Mobile GUI Agents",
"abstract": "While long-horizon mobile GUI agents typically rely on thought-action-observation loops, they struggle to separate persistent task states from transient screen observations. As execution histories grow, this entanglement imposes a severe context burden, causing agents to forget initial requirements, hallucinate progress, or repeatedly interact with stale interfaces. To address this, we introduce Task-State Representation (TSR), a training-free framework that explicitly decouples task state from sensory input. Acting as a lightweight external wrapper, TSR maintains three structured components: a global instruction summary, a dynamic progress tracker for subgoals, and a transition-aware action verifier. By continuously updating through pre- and post-action visual comparisons, TSR effectively guides the agent's reasoning without requiring architectural modifications. Experiments across four mobile GUI benchmarks validate TSR's effectiveness, yielding up to a 12 absolute point increase in success rate on complex cross-application and memory-intensive tasks.",
"published": "2026-07-01",
"updated": "2026-07-01",
"authors": [
"Yujie Zheng",
"Zikang Liu",
"Xin Zhao",
"Ji-Rong Wen"
],
"categories": [
"cs.CL"
],
"url": "https://arxiv.org/abs/2607.00502",
"pdf_url": "https://arxiv.org/pdf/2607.00502",
"path": "papers/items/2026-2607-00502-a-task-state-representation-for-long-horizon-mobile-gui-agents.md",
"local_title": "A Task-State Representation for Long-Horizon Mobile GUI Agents",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"planning",
"reasoning",
"tool-use"
],
"collection_queries": "web-gui-agent",
"collection_score": "16"
},
{
"arxiv_id": "2607.00627",
"title": "AGI Maze as a Benchmark Framework for World-Modeling Agents",
"abstract": "Large language models (LLMs) are powerful pattern-completion systems, but their default operating mode - predicting the next token from a static context - does not reliably produce persistent, manipulable representations of an external world. Many tasks that look like \"reasoning\" in text become substantially harder once the environment is partially observable, stateful, and requires memory and structured hypotheses about hidden state. AGI Maze is a lightweight framework for building such environments without requiring high-dimensional sensory inputs. It provides a family of grid-based maze tasks with a clean API and multiple difficulty regimes. The goal is to create benchmarks where agents must learn and use world state representations, not just infer a local rule over readily provided observations. We provide an initial evaluation of several vanilla LLMs on simple mazes showing that they fail to represent mazes internally at LLM inference time. We also introduce a baseline agent, which is allowed to use its message history as a working memory to construct descriptions of observations at agentic runtime. Although this can improve performance, it is still insufficient for an LLM agent to reliably solve even small mazes within a step budget that is more than enough for humans.",
"published": "2026-07-01",
"updated": "2026-07-01",
"authors": [
"Alexey Potapov"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2607.00627",
"pdf_url": "https://arxiv.org/pdf/2607.00627",
"path": "papers/items/2026-2607-00627-agi-maze-as-a-benchmark-framework-for-world-modeling-agents.md",
"local_title": "AGI Maze as a Benchmark Framework for World-Modeling Agents",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"reasoning",
"tool-use"
],
"collection_queries": "llm-agent",
"collection_score": "18"
},
{
"arxiv_id": "2607.01047",
"title": "Conversable Complexity: Agentic LLM Collectives as Interpretable Substrates",
"abstract": "Complexity and interpretability rarely coincide: systems rich enough for complex behaviours to emerge are usually too opaque to question, while transparent ones are too simple for anything complex to emerge. A single large language model (LLM) is a static artefact, hardly exhibiting any of the emergent properties we associate with life. This changes through interaction: populations of LLMs display emergent dynamics absent from isolated models. Furthermore, LLMs can be endowed with persistent memory, tools and shared skills, and the capacity to initiate actions unprompted, i.e., turning LLMs agentic. In this paper, we argue that such collectives of agents can serve as a computational substrate for Artificial Life (ALife) research. Critically, since the agents communicate in natural language, their collective behaviour can be directly interrogated by examining textual traces and asking the agents themselves. We outline the notion of interpretability in language-model research and extend it for collectives of agents. Lastly, we survey recent examples of agentic LLM collectives that already instantiate the idea of agentic substrates, from controlled experiments to deployments in the wild.",
"published": "2026-07-01",
"updated": "2026-07-01",
"authors": [
"Elias Najarro",
"Ane Espeseth",
"Eleni Nisioti",
"Sebastian Risi",
"Stefano Nichele"
],
"categories": [
"cs.CL"
],
"url": "https://arxiv.org/abs/2607.01047",
"pdf_url": "https://arxiv.org/pdf/2607.01047",
"path": "papers/items/2026-2607-01047-conversable-complexity-agentic-llm-collectives-as-interpretable-substrates.md",
"local_title": "\"Conversable Complexity: Agentic LLM Collectives as Interpretable Substrates\"",
"local_status": "queued",
"local_topics": [
"memory",
"tool-use"
],
"collection_queries": "llm-agent",
"collection_score": "13"
},
{
"arxiv_id": "2607.01071",
"title": "MemSyco-Bench: Benchmarking Sycophancy in Agent Memory",
"abstract": "Memory has emerged as a cornerstone of modern LLM-based agents, supporting their evolution from single-turn assistants to long-term collaborators. However, memory is not always beneficial: retrieved memories often induce a critical issue of sycophancy, causing agents to over-align with the user at the cost of factual accuracy or objective reasoning. Despite this emerging risk, existing memory benchmarks primarily evaluate whether memories are correctly stored, retrieved, or updated, while overlooking how retrieved memories influence downstream reasoning and decision-making. To bridge this gap, we propose MemSyco-Bench, a comprehensive benchmark for evaluating memory-induced sycophancy in agent systems. MemSyco-Bench measures when memory should influence a decision and how valid memory should be used. Specifically, it covers five tasks that assess whether agents can reject memory as factual evidence, respect its applicable scope, resolve conflicts between memory and objective evidence, track memory updates, and use valid memory for personalization. All related resources are collected for the community at https://github.com/XMUDeepLIT/MemSyco-Bench.",
"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"
],
"url": "https://arxiv.org/abs/2607.01071",
"pdf_url": "https://arxiv.org/pdf/2607.01071",
"path": "papers/items/2026-2607-01071-memsyco-bench-benchmarking-sycophancy-in-agent-memory.md",
"local_title": "\"MemSyco-Bench: Benchmarking Sycophancy in Agent Memory\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"reasoning"
],
"collection_queries": "agent-memory",
"collection_score": "17"
},
{
"arxiv_id": "2607.01523",
"title": "Multi-Head Recurrent Memory Agents",
"abstract": "Recurrent memory agents extend LLMs to arbitrarily long contexts by iteratively consolidating input into a fixed-size memory window. Despite their scalability, these agents exhibit a well-documented reliability problem: end-to-end performance degrades systematically as context length grows. We diagnose this failure by decomposing performance into two factors--memory capture and memory retention--and quantitatively confirm that retention is the dominant bottleneck. Retention collapses because existing designs maintain memory as a monolithic text block, forcing every update to risk overwriting previously retained content. Motivated by this diagnosis, we propose Multi-Head Recurrent Memory (MHM), a general, training-free framework that partitions memory into independent heads governed by a stage-wise select-then-update strategy. At each step, exactly one head is selected for update while the remaining heads are structurally shielded from overwriting, shifting the burden of retention from model behavior to architectural design. As a lightweight instantiation, we introduce Least-Recently-Updated MHM (MHM-LRU), which guarantees uniform head utilization with zero additional token overhead. Extensive experiments on long-context benchmarks show that MHM-LRU substantially improves both retention and end-to-end accuracy across the 100K--1M token range, where baselines degrade sharply. On RULER-HQA at 896K tokens, MHM-LRU improves the memory retention rate from less than 30% to 73.96%. These gains generalize across model families, scales, and task types, positioning architectural optimization as a practical and cost-efficient path toward reliable long-context recurrent memory.",
"published": "2026-07-01",
"updated": "2026-07-01",
"authors": [
"Jiatong Li",
"Samuel Yeh",
"Sharon Li"
],
"categories": [
"cs.LG",
"cs.AI",
"cs.CL"
],
"url": "https://arxiv.org/abs/2607.01523",
"pdf_url": "https://arxiv.org/pdf/2607.01523",
"path": "papers/items/2026-2607-01523-multi-head-recurrent-memory-agents.md",
"local_title": "Multi-Head Recurrent Memory Agents",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory"
],
"collection_queries": "agent-memory",
"collection_score": "14"
},
{
"arxiv_id": "2607.05428",
"title": "CHARLIE: An On-Premise Multi-Agent Retrieval-Augmented Generation System for Evidential Reasoning in Forensic Science",
"abstract": "We present Charlie, an on-premise multi-agent Retrieval-Augmented Generation (RAG) system for structured evidential processing in digital forensic environments. Contemporary forensic workflows must handle large volumes of heterogeneous and unstructured documents under strict requirements of traceability, confidentiality, and legal compliance. Charlie addresses this challenge through a controlled agent architecture that combines local retrieval, task decomposition, structured memory, and verification mechanisms. Unlike cloud-based systems, it operates entirely within institutional infrastructure, preserving data sovereignty and evidential integrity. We describe the systems architecture, including its transition from classical RAG to agent-based orchestration, and demonstrate its application in real-world forensic scenarios. Case studies show that Charlie enables scalable multi-document data extraction and supports longitudinal forensic intelligence generation while maintaining traceability and auditability. Our results indicate that agent-orchestrated, on-premise RAG architectures can effectively support evidential workflows without compromising legal and institutional constraints. Charlie provides a practical and reproducible blueprint for deploying AI systems in high-stakes forensic environments. This manuscript is an archival version of a paper presented at the RELAF 2026 Workshop.",
"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"
],
"url": "https://arxiv.org/abs/2607.05428",
"pdf_url": "https://arxiv.org/pdf/2607.05428",
"path": "papers/items/2026-2607-05428-charlie-an-on-premise-multi-agent-retrieval-augmented-generation-system-for-evid.md",
"local_title": "\"CHARLIE: An On-Premise Multi-Agent Retrieval-Augmented Generation System for Evidential Reasoning in Forensic Science\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"multi-agent",
"planning",
"rag",
"reasoning",
"tool-use",
"workflow-agent"
],
"collection_queries": "rag-agent",
"collection_score": "17"
},
{
"arxiv_id": "2607.01640",
"title": "AgentFlow: Building Agent Dependency Graphs for Static Analysis of Agent Programs",
"abstract": "LLM agents are increasingly developed as source-code applications built on agent frameworks. These agent programs combine conventional host-language code with framework-defined semantics for models, prompts, tools, memory, and multi-agent orchestration logic. As a result, their behavior depends not only on traditional control and data flows, but also on a new class of agent dependencies. Such dependencies are often expressed as framework-induced semantics, such as agent constructors, tool decorators, and agent handoff declarations, making them difficult to recover with existing static analysis or dependency tracking tools. In this paper, we present AgentFlow, the first static analysis framework for recovering and analyzing agent dependencies from agent programs. AgentFlow constructs an Agent Dependency Graph (ADG), a framework-agnostic graph representation that represents agents, prompts, models, capabilities, memory states, and control policies as typed nodes, and captures their component-dependency, control-flow, and data-flow dependencies as typed edges. Built on ADGs, AgentFlow supports a range of analyses for agent governance and security, including Agent Bill of Materials (BOM) generation and prompt-to-tool risk detection. We implement AgentFlow for five representative agent frameworks and evaluate it on AgentZoo, a corpus of 5,399 real-world agent programs. Our evaluation shows that AgentFlow recovers richer agent entities and dependencies than existing AST-based agent static analysis tools, generates more dependency-aware Agent BOMs, and uncovers 238 taint-style prompt-to-tool risks in real-world agent programs. These results show that ADG provides a practical foundation for understanding, governing, and securing emerging agent software.",
"published": "2026-07-02",
"updated": "2026-07-02",
"authors": [
"Shenao Wang",
"Xinyi Hou",
"Yanjie Zhao",
"Xiao Cheng",
"Haoyu Wang"
],
"categories": [
"cs.SE",
"cs.CR"
],
"url": "https://arxiv.org/abs/2607.01640",
"pdf_url": "https://arxiv.org/pdf/2607.01640",
"path": "papers/items/2026-2607-01640-agentflow-building-agent-dependency-graphs-for-static-analysis-of-agent-programs.md",
"local_title": "\"AgentFlow: Building Agent Dependency Graphs for Static Analysis of Agent Programs\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"multi-agent",
"tool-use"
],
"collection_queries": "llm-agent, multi-agent-llm",
"collection_score": "17"
},
{
"arxiv_id": "2607.01709",
"title": "COMFYCLAW: Self-Evolving Skill Harnesses for Image Generation Workflows",
"abstract": "Agents are increasingly used to construct workflows and assist humans in completing recurring tasks more efficiently. As these workflows become repeated and domain-specific, agent memory and reusable skills become increasingly important: agents should be able to recall workflow patterns, execution constraints, and user preferences from previous runs. We study this problem in workflow-based image generation and introduce COMFYCLAW, an agentic skill evolution harness for controlling ComfyUI workflows. COMFYCLAW formulates workflow construction as typed graph editing, exposes tools organized by construction stage, automatically reverts invalid edits, and uses a region-level vision-language model (VLM) verifier to translate visual failures into actionable repair suggestions. The framework further evolves a progressively disclosed skill library, where trajectories, execution errors, and verifier feedback from previous runs are distilled into reusable Agent Skills. Across four benchmark splits, three agent models, and two image backbones, COMFYCLAW achieves the best average image-generation evaluation score across all six agent configurations, outperforming a verifier-only baseline without skill evolution. Human annotations further show that annotators prefer COMFYCLAW over variants without skill evolution. Our results suggest that skill evolution is an effective mechanism for improving agent reliability and performance in recurring visual workflow construction.",
"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"
],
"url": "https://arxiv.org/abs/2607.01709",
"pdf_url": "https://arxiv.org/pdf/2607.01709",
"path": "papers/items/2026-2607-01709-comfyclaw-self-evolving-skill-harnesses-for-image-generation-workflows.md",
"local_title": "\"COMFYCLAW: Self-Evolving Skill Harnesses for Image Generation Workflows\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"reasoning",
"tool-use",
"workflow-agent"
],
"collection_queries": "agent-memory",
"collection_score": "15"
},
{
"arxiv_id": "2607.01767",
"title": "Repair the Amplifier, Not the Symptom: Stable World-Model Correction for Agent Rollouts",
"abstract": "Long-horizon language agents increasingly maintain executable world models in the form of planning graphs, where tool calls, validators, memory updates, recovery branches, and final answers are connected by typed dependencies. When a rollout fails, repairing the most visible error can leave the underlying error-amplification path intact, while replaying the full graph is expensive and difficult for long-context models to use reliably. We study world-model correction: selecting a compact subgraph of a failed planning graph whose repair stabilizes subsequent rollouts. We first instantiate a strong family of engineering correctors, including pointwise error scans, TopK and window selection, local graph expansion, cascade repair, and full-context LLM repair. We then propose WM-SAR, a spectral subgraph repair method that estimates node-edge amplification, greedily grows a connected repair region by marginal residual-spectral relief, and sends only this region to an LLM for root-cause repair. Theoretically, we connect residual spectral radius to rollout error and planning regret, motivating repair as stabilization rather than attribution alone. Across synthetic calling-tree graphs, benchmark-inspired agent topologies, and cross-model LLM repair experiments, WM-SAR achieves stronger long-horizon stabilization and root-cause recovery under compact token budgets, matching much larger repair contexts while exposing the LLM to a cleaner causal subgraph.",
"published": "2026-07-02",
"updated": "2026-07-05",
"authors": [
"Xinyuan Song",
"Zekun Cai"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2607.01767",
"pdf_url": "https://arxiv.org/pdf/2607.01767",
"path": "papers/items/2026-2607-01767-repair-the-amplifier-not-the-symptom-stable-world-model-correction-for-agent-rol.md",
"local_title": "\"Repair the Amplifier, Not the Symptom: Stable World-Model Correction for Agent Rollouts\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"planning",
"tool-use",
"world-model"
],
"collection_queries": "language-agent",
"collection_score": "15"
},
{
"arxiv_id": "2607.01788",
"title": "KRCA: An Efficient Root Cause Analysis System in Hyper-Scale Microservice Systems via Agentic AI",
"abstract": "Hyper-scale microservice systems have become the standard infrastructure for large-scale Internet companies. These systems consist of numerous loosely coupled microservices that evolve independently through continuous development and deployment. Such complexity makes failures unavoidable, necessitating efficient Root Cause Analysis (RCA) to help Site Reliability Engineers (SREs) quickly localize root cause services and classify failure types. However, existing RCA methods often struggle to adapt to the extreme dynamism and massive scale of these systems. In this paper, we present KRCA, an end-to-end RCA system designed for hyper-scale microservice systems. To manage the vast search space, KRCA employs a multi-stage pipeline that begins with an API-level drilldown to isolate suspicious services. It then instantiates a skeleton-based causal graph from anomalous metrics to serve as a high-recall structural prior, before utilizing a memory-augmented multi-agent framework to verify causality and generate the final failure report. By combining structured causal constraints with multi-agent reasoning, KRCA employs balances diagnostic accuracy with the efficiency requirements of real-time production use. Experimental results show that KRCA achieves AC@1 scores of 0.88 and 0.79 for root cause service localization and failure type classification, outperforming the strongest baseline by at lease 31% in absolute gains. KRCA has been deployed in Kuaishou's production environment for over six months, reducing the average diagnosis time by 77.3%.",
"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"
],
"url": "https://arxiv.org/abs/2607.01788",
"pdf_url": "https://arxiv.org/pdf/2607.01788",
"path": "papers/items/2026-2607-01788-krca-an-efficient-root-cause-analysis-system-in-hyper-scale-microservice-systems.md",
"local_title": "\"KRCA: An Efficient Root Cause Analysis System in Hyper-Scale Microservice Systems via Agentic AI\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"multi-agent",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "agentic-ai",
"collection_score": "16"
},
{
"arxiv_id": "2607.01916",
"title": "ContextSniper: AntTrail's Token-Efficient Code Memory for Repository-Level Program Repair",
"abstract": "Large language model agents can repair real repository issues, but they often spend large context budgets on whole-file reads, broad searches, and long terminal outputs where useful evidence is mixed with irrelevant code and logs. This paper presents ContextSniper, AntTrail's code-repair module for precision evidence selection in repository-level program repair, part of AntTrail's broader agent-memory engine. AntTrail is available at https://gitcode.com/datagallery/AntTrail. ContextSniper indexes code and action memory as three abstract levels, retrieves candidates with a hybrid ranker, filters long tool output through an intention-aware context gate, and returns compact evidence packets while keeping full source recoverable on demand. In a matched 50-task-per-condition comparison on SWE-bench Lite (same tasks, baseline vs.\\ ContextSniper), ContextSniper reduces total token use by 51.5% and logged cost by 36.4% for OpenClaw, and by 38.9% and 27.3% for Claude Code, with submitted-resolution rates essentially unchanged in both host-agent settings. In a separate five-task comparison, ContextSniper beats existing memory- and RAG-style integrations on token efficiency. These results suggest ContextSniper can substantially cut token and cost overhead for repository-level repair agents without a measurable loss in repair quality. The evaluation harness for this study is available at https://gitcode.com/lukchiwang/ContextSniper.",
"published": "2026-07-02",
"updated": "2026-07-09",
"authors": [
"Chiwang Luk",
"Matin Mohammad Najafi",
"Zhifeng Jia",
"Wei Yang",
"Xiuchang Li",
"Jinwei Zhu",
"Yang Ren",
"Lei Chen",
"Gao Cong"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2607.01916",
"pdf_url": "https://arxiv.org/pdf/2607.01916",
"path": "papers/items/2026-2607-01916-contextsniper-anttrail-s-token-efficient-code-memory-for-repository-level-progra.md",
"local_title": "\"ContextSniper: AntTrail's Token-Efficient Code Memory for Repository-Level Program Repair\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"coding-agent",
"memory",
"rag",
"tool-use"
],
"collection_queries": "agent-memory, coding-agent, rag-agent",
"collection_score": "17"
},
{
"arxiv_id": "2607.01935",
"title": "A-TMA: Decoupling State-Aware Memory Failures in Long-Term Agent Memory",
"abstract": "Long term memory lets LLM agents act as persistent assistants, but user facts change. A useful memory system must know what is true now, what used to be true, and what changed. We study \\emph{ghost memory}, a state coordination failure in which old, current, and transition facts coexist in the memory bank, remain mixed during retrieval, and mislead the answer model. We argue that memory systems should be understood and optimized from three levels: bank maintenance, retrieval, and answer time resolution. We propose ATMA, a state aware overlay for existing memory systems. ATMA keeps superseded and transition records in the bank, builds evidence packets for the query's requested state view, and exposes current, historical, and transition labels to QA. We further call for decoupled evaluation of bank, retrieval, and answer level failures, since final QA accuracy can hide where ghost memory occurs. To make this failure measurable, we build LTP (LoCoMo Temporal Plus), a conflict heavy benchmark for ghost memory, and evaluate on LoCoMo for long conversation generalization. On LTP, Graphiti+ATMA improves conflict accuracy by 0.240 absolute over Graphiti. On LoCoMo, Graphiti+ATMA raises temporal F1 from 0.0295 to 0.1705. The gains are host dependent, but they indicate that explicit state roles can reduce memory failures hidden by final QA accuracy.",
"published": "2026-07-02",
"updated": "2026-07-08",
"authors": [
"Zitong Shi",
"Yixuan Tang",
"Anthony Kum Hoe Tung"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2607.01935",
"pdf_url": "https://arxiv.org/pdf/2607.01935",
"path": "papers/items/2026-2607-01935-a-tma-decoupling-state-aware-memory-failures-in-long-term-agent-memory.md",
"local_title": "\"A-TMA: Decoupling State-Aware Memory Failures in Long-Term Agent Memory\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag"
],
"collection_queries": "agent-memory, llm-agent",
"collection_score": "17"
},
{
"arxiv_id": "2607.02255",
"title": "AgenticSTS: A Bounded-Memory Testbed for Long-Horizon LLM Agents",
"abstract": "Memory for a long-horizon LLM agent is a contract about what each future decision is allowed to see. The simplest contract appends past observations, tool calls, and reflections to every prompt, which makes prior context easy to access but also turns it into a jumbled mixture in which the effect of any single memory component is hard to isolate. We introduce and instrument an alternative bounded contract: every decision is made from a fresh user message assembled by typed retrieval, with no raw cross-decision transcript appended. The prompt thus stays bounded across runs of any length, and any single layer can be ablated in isolation. We instantiate the contract in Slay the Spire 2, a closed-rule stochastic deck-building game whose runs require hundreds of tactical and strategic decisions. A public online benchmark of frontier LLMs on the same game reports zero wins at the lowest difficulty across five configurations, and the developer-reported human win rate at the same difficulty is 16%; the task is hard but not saturated. Within our harness, a fixed-A0 ablation shows the largest observed difference when triggered strategic skills are enabled: the no-store baseline wins 3/10 games and adding the skill layer 6/10. At this sample size the comparison is directional rather than statistically decisive (Fisher exact p\\approx0.37); a cross-backbone probe and public accumulating-context baselines are reported as operational comparisons rather than controlled tests of the contract variable itself. We release a reproducible testbed: 298 completed trajectories with condition tags, frozen memory/skill snapshots, prompt records, and analysis scripts -- an agent design and a validated, reusable methodology for studying how explicit memory layers shape long-horizon LLM-agent decisions.",
"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"
],
"url": "https://arxiv.org/abs/2607.02255",
"pdf_url": "https://arxiv.org/pdf/2607.02255",
"path": "papers/items/2026-2607-02255-agenticsts-a-bounded-memory-testbed-for-long-horizon-llm-agents.md",
"local_title": "\"AgenticSTS: A Bounded-Memory Testbed for Long-Horizon LLM Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"planning",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "llm-agent",
"collection_score": "23"
},
{
"arxiv_id": "2607.02689",
"title": "S-EMBER: A Large-Scale Benchmark for Streaming Egocentric Memory Retrieval",
"abstract": "As wearable devices enable continuous first-person recording, AI assistants must reason across long time horizons to recall past experiences-a capability known as episodic memory. Current benchmarks often rely on offline evaluation with access to entire video files, failing to simulate the streaming reality of wearable intelligence. We introduce S-EMBER (Streaming Egocentric Memory Benchmark for Episodic Retrieval), a large-scale benchmark comprising 3,141 videos totaling 388 hours of organic activity captured via Ray-Ban Meta smart glasses. S-EMBER formalizes grounded streaming episodic retrieval, a paradigm shift from global offline search to causal, active recall triggered by visual events in a continuous stream. We provide 9,448 QA pairs requiring manual visual proof through precise temporal localization and supporting flexible response lengths to simulate natural human-AI interaction. Our extensive benchmarking of frontier models uncovers a localization paradox: while semantic reasoning improves with parameter scale, temporal grounding precision remains a stagnant architectural bottleneck that does not benefit from brute-force increases in model size, resolution, or frame density. S-EMBER establishes a hardware-authentic foundation for developing grounded, reliable episodic memory in the next generation of wearable AI agents.",
"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"
],
"url": "https://arxiv.org/abs/2607.02689",
"pdf_url": "https://arxiv.org/pdf/2607.02689",
"path": "papers/items/2026-2607-02689-s-ember-a-large-scale-benchmark-for-streaming-egocentric-memory-retrieval.md",
"local_title": "\"S-EMBER: A Large-Scale Benchmark for Streaming Egocentric Memory Retrieval\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "ai-agent",
"collection_score": "16"
},
{
"arxiv_id": "2607.02716",
"title": "Evaluating Large Language Models for Decision-Making in Agent-Based Urban Mobility Simulations",
"abstract": "Urban mobility modeling faces challenges in representing decision-making in dynamic environments. Although Multi-Agent Systems are widely used, rule-based approaches rely on fixed heuristics that limit adaptive behavior. This work investigates the integration of Large Language Models (LLMs) as decision-making components in multi-agent simulations. We propose a hybrid architecture that connects the GAMA platform to an external LLM-based module through an API, enabling agents to determine whether route replanning is necessary. Rather than replacing routing algorithms, the LLM serves as a decision layer that guides replanning behavior. The approach incorporates persistent memory, allowing past interactions to influence future decisions and promote behavioral consistency. We compare rule-based and LLM-assisted approaches across multiple road-blockage scenarios and population scales. Results indicate that LLM-enabled agents exhibit greater adaptability and contextual awareness, particularly in scenarios with higher route flexibility. Memory influences performance and behavioral consistency, with effects varying across configurations. Overall, LLMs serve as complementary cognitive layers that enrich behavioral representations in urban mobility simulations and hold potential for modeling complex decision-making in spatial multi-agent systems.",
"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"
],
"url": "https://arxiv.org/abs/2607.02716",
"pdf_url": "https://arxiv.org/pdf/2607.02716",
"path": "papers/items/2026-2607-02716-evaluating-large-language-models-for-decision-making-in-agent-based-urban-mobili.md",
"local_title": "Evaluating Large Language Models for Decision-Making in Agent-Based Urban Mobility Simulations",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"multi-agent",
"planning",
"tool-use",
"world-model"
],
"collection_queries": "multi-agent-llm",
"collection_score": "16"
},
{
"arxiv_id": "2607.03510",
"title": "CAGE-1: Control, Assurance, and Governance Evaluation for Enterprise Agentic AI",
"abstract": "Enterprise artificial intelligence is moving from experimentation into operational workflows. Early programs focused on model access and retrieval-augmented generation, but enterprises are now beginning to deploy agents that plan, retrieve, remember, call tools, update systems, and coordinate work across applications. This changes the evaluation problem. Leaders are no longer asking only whether an answer is accurate or fluent. They need to know who authorized an action, which policy applied, whether evidence was current, whether memory was valid, whether a tool call was permitted, whether the decision can be replayed, and whether the agent can be stopped before it creates business impact. This paper introduces CAGE-1: Control, Assurance, and Governance Evaluation for Enterprise Agentic AI. CAGE-1 is an evaluation framework for deciding whether enterprise agents are ready for deployment. It evaluates authority, policy enforcement, retrieval quality, memory integrity, tool safety, auditability, human oversight, conflict handling, safe failure, Prebind Assurance, operational readiness, and business fitness. CAGE-1 introduces Prebind Assurance to describe the evaluated ability to prove that an agentic action is controlled before it becomes binding, effective, or operationally consequential. The framework tests whether a proposed action is admitted, held, narrowed, refused, escalated, quarantined, or made non-effective before protected consequence forms.",
"published": "2026-07-03",
"updated": "2026-07-03",
"authors": [
"Roopam W. Sure"
],
"categories": [
"cs.SE",
"cs.AI",
"cs.CY"
],
"url": "https://arxiv.org/abs/2607.03510",
"pdf_url": "https://arxiv.org/pdf/2607.03510",
"path": "papers/items/2026-2607-03510-cage-1-control-assurance-and-governance-evaluation-for-enterprise-agentic-ai.md",
"local_title": "\"CAGE-1: Control, Assurance, and Governance Evaluation for Enterprise Agentic AI\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"planning",
"rag",
"tool-use",
"workflow-agent"
],
"collection_queries": "agentic-ai",
"collection_score": "19"
},
{
"arxiv_id": "2607.03525",
"title": "GameEngineBench: Evaluating Coding Agents on Real C++ Runtime Environments",
"abstract": "Game engines provide real-time simulation, rendering, physics, interaction, networking, and asset pipelines, making them valuable not only for games but also for 3D applications in healthcare, robotics, architecture, manufacturing, and related domains. Because game development is where these systems are most mature and publicly available, it offers a practical testbed for evaluating coding agents that must modify C++ code within stateful, interactive, real-time systems. We present GameEngineBench, a benchmark for evaluating coding agents on scoped C++ implementation tasks inside Unreal Engine 5 projects, built from nine real-world game repositories. The evaluation set consists of 110 tasks spanning gameplay mechanics, multiplayer behavior, AI and world orchestration, animation and movement, UI and session code, loading behavior, online-service integration, persistence, data serialization, XR behavior, and rendering-oriented plugins. These tasks require models to make native C++ changes that compile and satisfy behavioral tests within executable Unreal Engine projects. Across twelve evaluated configurations, the strongest model reaches 55.5\\% pass@1, while 31 tasks remain unsolved by every configuration. Our results demonstrate that frontier coding agents continue to struggle with deeply integrated C++ development for real-time interactive software, highlighting game-engine benchmarks as a valuable complement to existing software engineering evaluations.",
"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"
],
"url": "https://arxiv.org/abs/2607.03525",
"pdf_url": "https://arxiv.org/pdf/2607.03525",
"path": "papers/items/2026-2607-03525-gameenginebench-evaluating-coding-agents-on-real-c-runtime-environments.md",
"local_title": "\"GameEngineBench: Evaluating Coding Agents on Real C++ Runtime Environments\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"coding-agent",
"embodied-agent",
"memory",
"tool-use",
"world-model"
],
"collection_queries": "coding-agent",
"collection_score": "16"
},
{
"arxiv_id": "2607.03601",
"title": "ArchEval: Measuring AI Agents as Computer Architects",
"abstract": "Computer architecture has long used benchmarks to make progress measurable. LLM agents create a different measurement problem: success is not merely writing code or tuning parameters. The agent must interpret workloads, choose mechanisms, use simulators, predict performance, satisfy hard constraints, and decide which feasible design is worth evaluating. This paper introduces ArchEval, a benchmark and platform for evaluating LLM agents on computer architecture design and optimization. It contains 20 challenges across CPU core mechanisms, system architecture, memory systems, accelerators, and compute-in-memory, backed by eight simulators. Each challenge is posed under three settings: L1 full harness, with repeated simulator feedback; L2 simulator-code container, where simulator source is available but the agent must assemble its own workflow; and L3 agent-only, with no runnable feedback before submission. Each run reports baseline-normalized verifier performance and records the full trajectory, connecting results to workload analysis, simulator-tool use, prediction, constraint handling, and artifact integrity. Initial results show a sharp boundary in current agents. With L1 support, all four evaluated agents reach or exceed baseline and improve real designs across diverse simulators. Removing support exposes weaknesses: many agents fail to turn simulator source into useful experiments, and L3 predictions often disagree with verifier results. In L3, only GPT-5.5 + Codex remains above baseline, reaching 1.21x geomean performance and a 65% win rate; the other three fall below baseline. Even GPT-5.5 + Codex has only a 15% performance-modeling pass rate. ArchEval frames today's agents as useful optimization assistants rather than autonomous architects, and identifies capabilities needed next: simulator-tool use, calibrated prediction, pre-feedback judgment, and useful mechanism discovery.",
"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"
],
"url": "https://arxiv.org/abs/2607.03601",
"pdf_url": "https://arxiv.org/pdf/2607.03601",
"path": "papers/items/2026-2607-03601-archeval-measuring-ai-agents-as-computer-architects.md",
"local_title": "\"ArchEval: Measuring AI Agents as Computer Architects\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"reasoning",
"tool-use",
"workflow-agent"
],
"collection_queries": "ai-agent, llm-agent, tool-use",
"collection_score": "21"
},
{
"arxiv_id": "2607.03726",
"title": "SelfMem: Self-Optimizing Memory for AI Agents",
"abstract": "While current AI agents support increasingly long context windows, tool use, and skill execution for long-horizon tasks, they still require memory systems to effectively leverage historical experience. Existing memory frameworks typically rely on fixed storage, retrieval, and summarization mechanisms, which can be rigid across different tasks and often require manual tuning. To address this limitation, we propose SelfMem, a self-optimizing memory framework. Inspired by prior work on self-improving AI, we follow the principle of \"teaching an agent to fish rather than giving it a fish.\" Instead of forcing the model to follow a predefined memory strategy or format, SelfMem provides an environment with memory tools and feedback signals that allow the agent to explore, evaluate, and refine its own memory strategy. Our results show that SelfMem consistently outperforms retrieval, compression, and agent-memory baselines on BEAM across conversation scales from 100K to 1M tokens. Compared with the strongest baseline, SelfMem improves the official score by 48.7%, 40.8%, and 41.9% at 100K, 500K, and 1M, respectively. Further question-type analysis shows broad robustness across diverse memory demands, and our optimization study shows that model-guided strategy refinement further improves performance.",
"published": "2026-07-04",
"updated": "2026-07-04",
"authors": [
"Shu Yang",
"Junchao Wu",
"Derek F. Wong",
"Di Wang"
],
"categories": [
"cs.CL"
],
"url": "https://arxiv.org/abs/2607.03726",
"pdf_url": "https://arxiv.org/pdf/2607.03726",
"path": "papers/items/2026-2607-03726-selfmem-self-optimizing-memory-for-ai-agents.md",
"local_title": "\"SelfMem: Self-Optimizing Memory for AI Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"planning",
"rag",
"tool-use"
],
"collection_queries": "agent-memory, ai-agent, tool-use",
"collection_score": "20"
},
{
"arxiv_id": "2607.04009",
"title": "PhysMiner: An Agentic AI Framework for Discovering Turbulence Physics",
"abstract": "Uncovering the physical mechanisms of turbulent flows remains a fundamental challenge in fluid mechanics. In particular, conventional velocity-gradient analysis methods suffer from shear contamination, which hinders accurate identification of the dominant physical mechanisms. This study presents PhysMiner, an automated framework integrating the triple decomposition method of the velocity gradient tensor with large language model-driven reasoning for turbulence-physics discovery. The triple decomposition module automatically decomposes flow fields into rigid rotation, pure shearing, and normal straining components, enabling statistical analysis, contour visualization, vortex-line extraction, and threshold-insensitive vortex identification while eliminating shear contamination. These automated capabilities are validated across five benchmarks, ranging from canonical configurations to complex engineering flows. A discover-physics agent combines flow statistics, spatial structures, and literature-derived knowledge to perform pattern recognition and physical inference, while a review Agent iteratively validates physical consistency to ensure reliable conclusions. A continuously evolving Triple Decomposition Library accumulates statistical knowledge from successfully analyzed flows, enabling cross-case comparison and progressive enhancement of inductive capability. The complete PhysMiner pipeline is validated end-to-end on the periodic hill flow, where the framework autonomously generates turbulence modeling recommendations and derives an improved subgrid-scale model with superior Reynolds-stress predictions. PhysMiner is open to the public and establishes a foundation for long-term collaborative advancement in automated turbulence-physics discovery.",
"published": "2026-07-04",
"updated": "2026-07-04",
"authors": [
"Jiawei Chen",
"Han Gao",
"Ping He"
],
"categories": [
"physics.flu-dyn"
],
"url": "https://arxiv.org/abs/2607.04009",
"pdf_url": "https://arxiv.org/pdf/2607.04009",
"path": "papers/items/2026-2607-04009-physminer-an-agentic-ai-framework-for-discovering-turbulence-physics.md",
"local_title": "\"PhysMiner: An Agentic AI Framework for Discovering Turbulence Physics\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"reasoning",
"tool-use"
],
"collection_queries": "agentic-ai",
"collection_score": "15"
},
{
"arxiv_id": "2607.04089",
"title": "PLACEMEM: Toward a Compute-Aware Memory Plane for Lifelong Agents",
"abstract": "Lifelong agents need more than larger context windows and better retrieval. They need memories that can persist, evolve, and be corrected without forcing the serving stack to recompute the same history on every turn or silently reuse stale runtime state. We present PLACEMEM as a systems position on lifelong-agent memory, instantiated by an executable control-plane prototype. The central claim is that agent memory should be represented as versioned capsules that unify semantics, provenance, validity, and reusable runtime state under one correction-aware identity. In the current prototype, capsules drive prompt-level text retrieval, KV-aware routing, and cascading invalidation over live streamed backends; prospective layer-frontier replay is intentionally framed as a deeper integration agenda rather than a claimed engine feature. We describe a vLLM-first prototype with persistent capsule state, concurrency-safe invalidation, an OpenAI-compatible routing sidecar, a typed metadata contract, and a benchmark harness that measures live first-token latency, reuse, and post-correction behavior. The result is both an executable artifact that demonstrates correction-aware control-plane behavior today and a concrete roadmap for replay-aware serving integration in future lifelong-agent systems.",
"published": "2026-07-05",
"updated": "2026-07-05",
"authors": [
"Sukanta Ganguly"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2607.04089",
"pdf_url": "https://arxiv.org/pdf/2607.04089",
"path": "papers/items/2026-2607-04089-placemem-toward-a-compute-aware-memory-plane-for-lifelong-agents.md",
"local_title": "\"PLACEMEM: Toward a Compute-Aware Memory Plane for Lifelong Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"planning",
"rag"
],
"collection_queries": "agent-memory",
"collection_score": "15"
},
{
"arxiv_id": "2607.04162",
"title": "ACE: Agentic Control for Embodied Manipulation via Zero-shot Workflow Reasoning",
"abstract": "Open-ended tabletop manipulation requires agents to not only understand natural language but also adapt to dynamic environments and execution failures. We present ACE (Agentic Control for Embodied Manipulation), a zero-shot workflow reasoning framework for tabletop pick-and-place from natural language. Rather than relying on direct low-level action mapping, ACE combines agentic workflow reasoning with two robot-facing executable skills: a visual grounding interface and a reusable pick-and-place primitive. To bridge semantic reasoning and physical control, the active sub-goal is grounded into a mask-mediated vision-action interface. This unified mask specifies the target object and destination, is tracked over time, exposed for human verification, and ultimately passed to a task-agnostic downstream policy for execution. Crucially, ACE operates in a closed loop supported by a multi-timescale memory. After an action is executed, the system automatically verifies whether the intended sub-goal succeeded, using the outcome to advance, retry, repair, or replan. This enables online adaptation to user corrections, scene changes, and physical failures. We evaluate ACE on logically complex, long-horizon tasks, including zero-shot multi-step equation formation with number cubes and constraint-based object retrieval. ACE demonstrates task-level zero-shot generalization on novel semantic constraints and randomized tabletop scenes without task-specific retraining. Specifically, while standard end-to-end baselines struggle to complete these logically demanding tasks, ACE achieves a 50% success rate in equation formation and a 70% success rate in constraint retrieval. This contrast demonstrates that explicit workflow reasoning and mask-mediated control offer a robust, practical route toward adaptable robotic manipulation.",
"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"
],
"url": "https://arxiv.org/abs/2607.04162",
"pdf_url": "https://arxiv.org/pdf/2607.04162",
"path": "papers/items/2026-2607-04162-ace-agentic-control-for-embodied-manipulation-via-zero-shot-workflow-reasoning.md",
"local_title": "\"ACE: Agentic Control for Embodied Manipulation via Zero-shot Workflow Reasoning\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"embodied-agent",
"memory",
"planning",
"rag",
"reasoning",
"tool-use",
"workflow-agent"
],
"collection_queries": "agentic-ai",
"collection_score": "17"
},
{
"arxiv_id": "2607.04391",
"title": "Memory-Orchestrated Semantic System (MOSS): An Auditable Agentic Memory Architecture",
"abstract": "Long-term memory remains a structural weakness of AI agents. The dominant approach, retrieval-augmented generation (RAG), relies on embedding-based similarity search, which is opaque by construction, difficult to audit, and bounded by the theoretical limits of vector representations. We present the Memory-Orchestrated Semantic System (MOSS), an agentic memory architecture in which the agent drives retrieval over a structured relational database. MOSS is model-agnostic, storage-agnostic, and API-agnostic: it runs on any relational engine, connects to any LLM provider (or to deterministic non-LLM processes), and deploys on any infrastructure, local or cloud. Its retrieval execution is symbolic and reproducible (once a query is formulated, no LLM participates in the retrieval loop) and every step of the system, from indexing to answer formulation, is logged and inspectable, making MOSS auditable by construction. Rather than imposing an external ontology, MOSS derives its conceptual vocabulary from the corpus itself. We report on a longitudinal deployment unique in the agentic-memory literature: a year of continuous production over an individual scholar's working corpus--a conversational corpus reaching back to October 2024 (some 44 million tokens, retroactively indexed) comprising 110,183 segments, alongside 163,494 catalogued documents, 569 inductively derived concepts, 322,662 concept annotations, and eleven metadata graphs totaling approximately five million relations--across four successive infrastructure generations. While the present case is that of a single researcher, the architecture is in no way specific to one person: it serves a team, an institution, or any entity that accumulates knowledge over time. We argue that auditable, sovereign, structurally unbounded memory is a precondition for AI agents intended to accompany a person or an organization over years rather than sessions.",
"published": "2026-07-05",
"updated": "2026-07-05",
"authors": [
"Serge Lacasse",
"Jérémie Hatier",
"Alex Baker"
],
"categories": [
"cs.CL"
],
"url": "https://arxiv.org/abs/2607.04391",
"pdf_url": "https://arxiv.org/pdf/2607.04391",
"path": "papers/items/2026-2607-04391-memory-orchestrated-semantic-system-moss-an-auditable-agentic-memory-architectur.md",
"local_title": "\"Memory-Orchestrated Semantic System (MOSS): An Auditable Agentic Memory Architecture\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"tool-use"
],
"collection_queries": "agent-memory, ai-agent, rag-agent",
"collection_score": "20"
},
{
"arxiv_id": "2607.04395",
"title": "NKI-Agent: Domain-Specific Fine-Tuning and Agentic Tool Use for Neuron Kernel Generation",
"abstract": "Recent agentic approaches to LLM-based kernel generation have achieved impressive results on CUDA. For emerging AI accelerators such as AWS Trainium and Inferentia, automated kernel generation and optimization remain largely unaddressed. Writing kernels for these chips via the Neuron Kernel Interface (NKI) is particularly challenging: developers must navigate a multi-engine architecture, tile-based programming, and explicit data movement across multi-level memory hierarchy. Moreover, no publicly-available training data, benchmarks, or tool-augmented agents exist for this domain. We introduce NKI-Agent, the first system combining domain-specific supervised fine-tuning (SFT) with a compile-verify-fix agent loop for NKI kernel generation. We adapt the existing CUDA-Agent framework to Neuron hardware, curate 6,000 NKI kernel generation tasks for training, and construct NKIBench, a 250-task benchmark across three difficulty levels. Evaluated on real Trn1 hardware, NKI-Agent with Claude Opus 4.8 and a rank-aware system prompt achieves a 77.3% pass rate on the 150-task NKIBench. We show that tool use is critical: Opus 4.8 scores 6% in single-shot mode without agent tools. On a 60-task subset, we show that an SFT-trained Qwen3-Coder-30B-A3B achieves 25.0% pass rate at 1/100th the cost, outperforming Claude Sonnet 4 (15.0%). We also report that Group Relative Policy Optimization (GRPO) with binary compilation reward fails to improve over SFT, providing guidance on reward design for RL-based kernel generation.",
"published": "2026-07-05",
"updated": "2026-07-05",
"authors": [
"Junjie Tang",
"Jun Huan",
"Hao Zhou",
"Yuhao Zhang",
"Lin Wang"
],
"categories": [
"cs.LG"
],
"url": "https://arxiv.org/abs/2607.04395",
"pdf_url": "https://arxiv.org/pdf/2607.04395",
"path": "papers/items/2026-2607-04395-nki-agent-domain-specific-fine-tuning-and-agentic-tool-use-for-neuron-kernel-gen.md",
"local_title": "\"NKI-Agent: Domain-Specific Fine-Tuning and Agentic Tool Use for Neuron Kernel Generation\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"tool-use"
],
"collection_queries": "tool-use",
"collection_score": "19"
},
{
"arxiv_id": "2607.04426",
"title": "ACE-Brain-0.5: A Unified Embodied Foundational Model for Physical Agentic AI",
"abstract": "Embodied AI is moving from isolated perception or action modules toward physical agents that understand, plan under goals, act through robot bodies, monitor progress, and improve from experience. Existing systems address this loop only in parts: end-to-end policies generate actions but often lack spatial reasoning, planning, and execution assessment, while robot-agent systems orchestrate tools or specialists but do not learn a shared representation. This fragmentation limits general Physical Agentic AI. We present ACE-Brain-0.5, a unified embodied foundation model that organizes robot intelligence into five coupled functions: spatial perception, decision making, embodied interaction, self-monitoring, and self-improvement. Built on ACE-Brain-0, which established spatial intelligence as a shared scaffold across robot platforms, ACE-Brain-0.5 extends an understanding-centric model into a closed-loop foundation model. A single 8B backbone instantiates the first four functions: grounding objects and affordances, reasoning over 3D and egocentric spatial relations, decomposing instructions into subgoals, generating navigation and manipulation actions, and estimating progress for verification and recovery. To unify these capabilities without cross-task interference, we introduce SSR+, which extends Scaffold-Specialize-Reconcile with a Reactivate stage after task-vector merging. The fifth function, self-improvement, is realized by a companion framework that updates external execution state, including task schemas, spatial memory, and failure-recovery cases, from rollouts. Across fifteen benchmarks, ACE-Brain-0.5 improves over ACE-Brain-0 on 14 of 18 spatial perception and grounding benchmarks, achieves competitive navigation and manipulation performance, and provides strong progress estimation in ID and OOD settings. Together, these results mark an early step toward general Physical Agentic AI.",
"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"
],
"url": "https://arxiv.org/abs/2607.04426",
"pdf_url": "https://arxiv.org/pdf/2607.04426",
"path": "papers/items/2026-2607-04426-ace-brain-0-5-a-unified-embodied-foundational-model-for-physical-agentic-ai.md",
"local_title": "\"ACE-Brain-0.5: A Unified Embodied Foundational Model for Physical Agentic AI\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"embodied-agent",
"memory",
"planning",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "agentic-ai",
"collection_score": "17"
},
{
"arxiv_id": "2607.04433",
"title": "Autonomous Information Seeking: A Roadmap for Agentic Recommender Systems",
"abstract": "The rapid integration of large language model-based agents into recommender systems has driven a shift from static, ranking-based pipelines toward autonomous and interactive systems that can reason, plan, and act. This survey provides a comprehensive overview of this emerging landscape by introducing a unified taxonomy grounded in the level of autonomy and three core paradigms of agentic recommender systems: agent-assisted recommendation, agent-as-recommender, and agent-as-user-simulator. The autonomy framework organizes existing methods along increasing capabilities in proactivity, context awareness, interaction flexibility, and adaptivity. Building on this framework, the survey analyzes how each paradigm adopts different agentic architectures and how agents enhance key components such as profiles, memory, tool use, workflows, and optimization mechanisms. We further examine evaluation methodologies for agentic recommendation, covering automated metrics, LLM-based judging, and simulation-based assessment, and discuss their limitations in capturing reasoning quality, user experience, and system behavior. Beyond existing evaluation protocols, we further discuss unresolved issues in evaluating agentic recommender systems, including trajectory-level assessment, agent contribution analysis, and calibration of user simulation. Lastly, the survey outlines open challenges in lifelong user modeling, contextual abstraction, multimodal alignment, controllability, trustworthiness, privacy, scalability, and efficiency. Together, these analyses establish a unified foundation for understanding the current progress of agentic recommender systems and highlight promising opportunities for developing more autonomous, reliable, and human-aligned recommendation agents.",
"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"
],
"url": "https://arxiv.org/abs/2607.04433",
"pdf_url": "https://arxiv.org/pdf/2607.04433",
"path": "papers/items/2026-2607-04433-autonomous-information-seeking-a-roadmap-for-agentic-recommender-systems.md",
"local_title": "\"Autonomous Information Seeking: A Roadmap for Agentic Recommender Systems\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"planning",
"reasoning",
"tool-use",
"workflow-agent",
"world-model"
],
"collection_queries": "tool-use",
"collection_score": "21"
},
{
"arxiv_id": "2607.04617",
"title": "MRMS: A Multi-Resolution Memory Substrate for Long-Lived AI Agents",
"abstract": "Long-lived AI agents require continuity across interactions, but continuity cannot be obtained by simply extending the prompt window. An agent must preserve useful prior experience, retrieve it selectively, distinguish personal context from external evidence, and revise memory when the underlying situation changes. We propose an architectural memory substrate organized along two orthogonal axes: a representational axis spanning structured records, vector representations, and graph relations; and a temporal axis spanning short-term traces, medium-term abstractions, and long-term semantic commitments. Its key design constraint is synchronized structured-vector-graph memory: structured records govern eligibility, vector representations support recall, and graph relations adjudicate support, contradiction, and supersession before gated context projection. Its central claim is that reliable personalization is a memory design problem: useful memory is structured, selectively exposed, continuously consolidated, and epistemically labeled rather than stored as undifferentiated conversation history. Beyond the framework, we instantiate MRMS as a lightweight prototype implementing structured records, vector retrieval, temporal policies, and graph-based revision. The prototype exercises the core substrate mechanisms through pre-generation memory selection, revision, boundary enforcement, and evidence attribution under controlled long-lived interaction scenarios with explicit evidence requirements.",
"published": "2026-07-06",
"updated": "2026-07-06",
"authors": [
"Jizhizi Li",
"Amy Shi-Nash"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2607.04617",
"pdf_url": "https://arxiv.org/pdf/2607.04617",
"path": "papers/items/2026-2607-04617-mrms-a-multi-resolution-memory-substrate-for-long-lived-ai-agents.md",
"local_title": "\"MRMS: A Multi-Resolution Memory Substrate for Long-Lived AI Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"rag",
"tool-use"
],
"collection_queries": "ai-agent",
"collection_score": "16"
},
{
"arxiv_id": "2607.05029",
"title": "Your Agent's Memories Are Not Its Own: Forged Reasoning Attacks on LLM Agent Memory and Defenses",
"abstract": "Persistent memory has enabled large language model (LLM) agents to store factual knowledge, prior decisions, reasoning histories, tool usage information, and context. While this has improved the agent's functionality and continuity across tasks, it has also introduced a new attack surface: the agent's own reasoning history. In this paper, we introduce the Forged Amplifying Rationale Memory Attack (FARMA), which poisons an agent's remembered reasoning rather than its factual knowledge. It inserts forged reasoning traces using evasive language that bypasses keyword-based defenses, then amplifies them through self-referential reinforcement that defeats consensus-based defenses. To address FARMA, we introduce SENTINEL, a layered defense pipeline to detect forged reasoning entries. Its central component is the Reasoning Guard that structurally analyzes candidate entries for forgery using five weighted signals. We evaluate FARMA and SENTINEL across multiple agents and different LLM models with 50 trials and show that FARMA achieves an attack success rate of up to 100% under baseline conditions and is capable of defeating defense mechanisms like keyword filter and A-MemGuard. Our evaluation also shows that SENTINEL reduces FARMA's attack success rate to as low as 0% with no false positives observed across 326 benign agent traces. Our work demonstrates the need to protect not only an agent's retrieved content but also the integrity of its reasoning history.",
"published": "2026-07-06",
"updated": "2026-07-06",
"authors": [
"Neeraj Karamchandani",
"Piyush Nagasubramaniam",
"Sencun Zhu",
"Dinghao Wu"
],
"categories": [
"cs.CR",
"cs.AI"
],
"url": "https://arxiv.org/abs/2607.05029",
"pdf_url": "https://arxiv.org/pdf/2607.05029",
"path": "papers/items/2026-2607-05029-your-agent-s-memories-are-not-its-own-forged-reasoning-attacks-on-llm-agent-memo.md",
"local_title": "\"Your Agent's Memories Are Not Its Own: Forged Reasoning Attacks on LLM Agent Memory and Defenses\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"reasoning",
"tool-use"
],
"collection_queries": "agent-memory, llm-agent",
"collection_score": "19"
},
{
"arxiv_id": "2607.05202",
"title": "EvoAgentBench: Benchmarking Agent Self-Evolution via Ability Transfer",
"abstract": "Agent self-evolution in long-horizon LLM systems is largely procedural: useful experience is not merely stored information, but reusable procedures for searching, debugging, and verification. Yet current evaluations do not isolate this form of transfer. Agent benchmarks test single-episode task solving; memory benchmarks target information retention rather than procedural reuse. We introduce EvoAgentBench, a benchmark for agent self-evolution via Ability-guided transfer across four agentic domains: web research, algorithmic reasoning, software engineering, and knowledge work. EvoAgentBench extracts trace-grounded Abilities from agent executions, canonicalizes them into operational units, and builds domain-specific Ability Graphs linking tasks that share procedural overlap. By design, every test task is backed by verified training-side Ability support. Across a 528/267 train/test split, two scaffolds, and three backbones, curated Ability content transfers reliably across model families, but no current automatic method sustains positive gain in all settings. EvoAgentBench shifts self-evolution evaluation from aggregate accuracy comparison to fine-grained diagnosis of experience encoding, routing, and uptake. The benchmark is publicly available at https://huggingface.co/datasets/EverMind-AI/EvoAgentBench.",
"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"
],
"url": "https://arxiv.org/abs/2607.05202",
"pdf_url": "https://arxiv.org/pdf/2607.05202",
"path": "papers/items/2026-2607-05202-evoagentbench-benchmarking-agent-self-evolution-via-ability-transfer.md",
"local_title": "\"EvoAgentBench: Benchmarking Agent Self-Evolution via Ability Transfer\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"coding-agent",
"computer-use",
"memory",
"planning",
"reasoning"
],
"collection_queries": "agent-evaluation",
"collection_score": "19"
},
{
"arxiv_id": "2607.05318",
"title": "PiSAs: Benchmarking Contextual Integrity in Multi-User Agentic Systems",
"abstract": "As LLM agents evolve from single-user assistants into shared organizational infrastructure, new privacy risks emerge: inappropriate information may not only be exposed through outputs for external recipients, but also internally across users through inter-agent messages, shared memory and agents. These data spillage risks are not captured by existing privacy benchmarks grounded in contextual integrity (CI) as they focus primarily on either single-user settings or interactions between independently owned agents. We introducePiSAs (Privacy in Shared Agentic systems), a benchmark for assessing unintentional leaks with dual CI annotations: whether an information is appropriate for the task, and which users may legitimately access it. This enables direct measurement of cross-user spillage across agentic system components and interfaces, such as outputs, inter-agent communication, and memory. PiSAsis system-agnostic and supports evaluation across different agent topologies and memory regimes. We find that, although system design improves CI compliance, results are bottlenecked by incorrect LLM judgment calls: even state-of-the-art models fail to reliably filter inappropriate content or restrict transmission to authorized users. Our findings underscore the need for privacy-preserving strategies, beyond those studied in this work.",
"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"
],
"url": "https://arxiv.org/abs/2607.05318",
"pdf_url": "https://arxiv.org/pdf/2607.05318",
"path": "papers/items/2026-2607-05318-pisas-benchmarking-contextual-integrity-in-multi-user-agentic-systems.md",
"local_title": "\"PiSAs: Benchmarking Contextual Integrity in Multi-User Agentic Systems\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"memory",
"tool-use"
],
"collection_queries": "llm-agent",
"collection_score": "20"
},
{
"arxiv_id": "2607.05363",
"title": "SovereignPA-Bench: Evaluating User-Owned Personal Agents under Evolving Intent, Platform Mediation, and Consent Constraints",
"abstract": "Personal agents are becoming persistent user-owned intermediaries: they remember preferences, filter platform-mediated information, use tools, and negotiate with services. Existing benchmarks evaluate tool use, web navigation, desktop control, personalization, recommendation, and evolving context, but rarely ask whether an agent preserves user sovereignty: advancing the user's current interests while respecting privacy, consent, evidence, user burden, and resistance to manipulative incentives. We introduce SovereignPA-Bench, an executable benchmark for evaluating user-owned personal agents under evolving intent, platform mediation, privacy boundaries, consent constraints, evidence requirements, and burden tradeoffs. The benchmark separates agent-visible ObservableState from evaluator-only HiddenLabels, reports component metrics for task success, alignment, privacy, consent, evidence, manipulation, burden, and auditability, and preserves paired scenario ordering for model and policy comparisons. We evaluate 120 sovereignty stress scenarios across 4 model families and 8 policy baselines, yielding 3,840 frozen-prompt trajectories with raw prompts, outputs, provider-form responses, parsed actions, recomputable metrics, hard-set analyses, qualitative cases, and a blinded 3-annotator audit over 240 items. Full-sovereign scaffolding improves sovereignty score over direct, memory-only, consent-only, evidence-only, ReAct/tool-use, safety-prompt, and judge-guard baselines while reducing privacy leakage, consent violation, over-concession, and manipulation capture. Human audit shows high agreement on privacy and consent and lower agreement on manipulation, identifying the subjective frontier of platform-persuasion judgments. These results show that personal-agent evaluation must move beyond task completion toward representative, consent-aware, evidence-grounded action.",
"published": "2026-07-06",
"updated": "2026-07-06",
"authors": [
"Dylan Zongmin Liu"
],
"categories": [
"cs.AI"
],
"url": "https://arxiv.org/abs/2607.05363",
"pdf_url": "https://arxiv.org/pdf/2607.05363",
"path": "papers/items/2026-2607-05363-sovereignpa-bench-evaluating-user-owned-personal-agents-under-evolving-intent-pl.md",
"local_title": "\"SovereignPA-Bench: Evaluating User-Owned Personal Agents under Evolving Intent, Platform Mediation, and Consent Constraints\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"agent-safety",
"computer-use",
"embodied-agent",
"memory",
"tool-use"
],
"collection_queries": "agent-evaluation, tool-use",
"collection_score": "16"
},
{
"arxiv_id": "2607.05690",
"title": "Memory in the Loop: In-Process Retrieval as ExtendedWorking Memory for Language Agents",
"abstract": "Language agents run a loop - observe, reason, act - but the memory they reason over sits outside it: a store queried at most once per turn. We study the regime where memory moves inside the loop, read and written on every step. The obstacle has always been latency: networked stores answer in tens to hundreds of milliseconds, and in-loop retrieval can inflate end-to-end latency by up to 83x when retrieval is expensive. Prior work manages that cost rather than questioning it: serving-layer scheduling hides it, \"memory-first\" designs ration retrieval to once per turn. We argue latency is a property of where the store lives, not the in-loop pattern: an in-process store answers in ~100us, three orders of magnitude below the network regime, and at that speed the per-step tax collapses. By the extended-mind thesis's parity principle, a store fast enough to be constantly and directly available becomes extended working memory, not a tool the agent merely consults. The premise is causal: holding a fixed per-turn memory-latency budget and varying only the store's answer speed, redundant actions rise monotonically with latency - 0.0 of 12 at in-process speed, 7.2 of 12 at a 110ms cloud round trip (gpt-5-nano, gpt-5-mini; exact permutation p=0.0079). We demonstrate the regime end-to-end: across four GPT-5-class models under a bounded window, recall improves from 0/5 to 3.6-4.8/5 with in-loop memory, store ops at p50 80-165us - though an instructed restate-every-reply baseline also solves it perfectly, at a token cost that grows with the working set. The store never lost a fact in any run (244 of 244 writes kept); every miss traces to the agent's read policy, not the store. Our measurements also relocate the bottleneck: the dominant per-step cost is embedding (~200-400ms over the network); pairing the in-process store with a small local embedder returns the complete operation to a measured ~40us.",
"published": "2026-07-06",
"updated": "2026-07-06",
"authors": [
"Yusuf Khan",
"Carlo Lipizzi"
],
"categories": [
"cs.AI",
"cs.CL"
],
"url": "https://arxiv.org/abs/2607.05690",
"pdf_url": "https://arxiv.org/pdf/2607.05690",
"path": "papers/items/2026-2607-05690-memory-in-the-loop-in-process-retrieval-as-extendedworking-memory-for-language-a.md",
"local_title": "\"Memory in the Loop: In-Process Retrieval as ExtendedWorking Memory for Language Agents\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"memory",
"rag",
"tool-use"
],
"collection_queries": "language-agent",
"collection_score": "15"
},
{
"arxiv_id": "2607.05794",
"title": "From Passive Retrieval to Active Memory Navigation: Learning to Use Memory as a Structured Action Space",
"abstract": "Long-term user memory is essential for personalized conversational agents, yet many memory systems still expose memory through passive retrieval interfaces, making the model a consumer of pre-selected evidence. We introduce NapMem, a framework for learning to use long-term user memory as a structured action space rather than passively retrieved context. NapMem organizes user history into a linked multi-granularity memory pyramid, where raw conversations, typed memory records, topic tracks, and user profiles are connected through provenance relations, and exposes these levels through memory tools. The agent is trained to select memory according to the query and intermediate evidence, allowing it to inspect different memory granularities before answering. Experiments on PersonaMem-v2, LongMemEval, and LoCoMo show that a NapMem agent trained with memory-tool reinforcement learning is competitive across diverse memory-intensive tasks, while evaluations on non-memory tasks suggest that the learned policy largely preserves general reasoning and tool-use abilities. Additional analyses examine storage, inference cost, tool-use behavior, and ablations over navigation, memory granularity, and RL training. Our results suggest that long-term user memory benefits from coupling structured storage with a learned policy for using memory at the appropriate granularity.",
"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"
],
"url": "https://arxiv.org/abs/2607.05794",
"pdf_url": "https://arxiv.org/pdf/2607.05794",
"path": "papers/items/2026-2607-05794-from-passive-retrieval-to-active-memory-navigation-learning-to-use-memory-as-a-s.md",
"local_title": "\"From Passive Retrieval to Active Memory Navigation: Learning to Use Memory as a Structured Action Space\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"embodied-agent",
"memory",
"rag",
"reasoning",
"tool-use"
],
"collection_queries": "tool-use",
"collection_score": "13"
},
{
"arxiv_id": "2607.06140",
"title": "CurateEvo: Data-Curation Evolving for Agentic Post-Training",
"abstract": "Large language model (LLM) agents require post-training methods that can improve long-horizon decision making from environment feedback. However, existing agentic post-training pipelines often treat data curation as a fixed preprocessing step, focusing mainly on data augmentation while neglecting filtering, refinement, and adaptation to downstream failures. We propose CurateEvo, a failure-driven dynamic evolution framework for agentic post-training data curation. CurateEvo represents the curation strategy as executable code and iteratively rewrites it using failed trajectories from a held-out development set. At each epoch, the evolved strategy transforms a fixed raw corpus into supervised fine-tuning data, reinforcement learning data, and an inference-time memory bank. The evolution process first improves effectiveness by diagnosing recurring failure modes and augmenting, filtering, or refining data accordingly, and then improves efficiency by pruning redundant or low-utility training turns under a cost-aware objective. Experiments on ACEBench-Agent, BFCL-V4, and τ^2-Bench under both labeled and wild-data settings show that CurateEvo consistently outperforms prior curation methods, improving average scores by 3.2 and 2.7 points, respectively. Further analyses demonstrate that CurateEvo is compatible with different post-training recipes and substantially reduces curation overhead.",
"published": "2026-07-07",
"updated": "2026-07-07",
"authors": [
"Dingzirui Wang",
"Xuanliang Zhang",
"Keyan Xu",
"Qingfu Zhu",
"Wanxiang Che"
],
"categories": [
"cs.CL"
],
"url": "https://arxiv.org/abs/2607.06140",
"pdf_url": "https://arxiv.org/pdf/2607.06140",
"path": "papers/items/2026-2607-06140-curateevo-data-curation-evolving-for-agentic-post-training.md",
"local_title": "\"CurateEvo: Data-Curation Evolving for Agentic Post-Training\"",
"local_status": "queued",
"local_topics": [
"memory",
"planning",
"rag"
],
"collection_queries": "llm-agent",
"collection_score": "14"
},
{
"arxiv_id": "2607.06195",
"title": "LogicHunter: Testing LLM Agent Frameworks with an Agentic Oracle",
"abstract": "Large Language Model (LLM) agent frameworks such as LangChain, LlamaIndex, and CrewAI have become critical infrastructure powering production AI systems, yet they remain severely under-tested due to fundamental challenges in automated testing. Unlike traditional software, where crashes serve as reliable oracles, defects in these pure Python frameworks manifest as ordinary exceptions or silent semantic failures, creating profound oracle ambiguity. This problem is exacerbated by strict type governance through Pydantic schemas and complex protocol requirements that cause existing fuzzers to generate overwhelming invalid inputs, while traditional test generators produce only trivial cases with weak regression assertions. We present LogicHunter, a fuzzing framework that addresses both the generation and oracle challenges through active specification-aware testing. LogicHunter employs specification-driven generation that systematically fuses formal type constraints with authentic usage patterns from real-world repositories, synthesizing inputs that are valid by construction yet semantically extreme, equipped with behavioral probes to expose silent failures. To resolve oracle ambiguity, we introduce the Agentic Oracle, which transcends passive classification by actively retrieving documentation, navigating source code, and inspecting runtime states through a ReAct-based architecture with Dual-Layer State Management and Dual-Stream Memory. Evaluated on three widely deployed frameworks, LogicHunter discovered 40 previously unknown bugs with 30 confirmed and 26 fixed by developers, while state-of-the-art baselines reported no bugs as final findings. The Agentic Oracle achieves 91.17% precision, surpassing the best passive approach at 29.27% by 61 percentage points.",
"published": "2026-07-07",
"updated": "2026-07-07",
"authors": [
"Minghui Long",
"Yanjie Zhao",
"Haoyu Wang"
],
"categories": [
"cs.SE"
],
"url": "https://arxiv.org/abs/2607.06195",
"pdf_url": "https://arxiv.org/pdf/2607.06195",
"path": "papers/items/2026-2607-06195-logichunter-testing-llm-agent-frameworks-with-an-agentic-oracle.md",
"local_title": "\"LogicHunter: Testing LLM Agent Frameworks with an Agentic Oracle\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory"
],
"collection_queries": "llm-agent",
"collection_score": "17"
},
{
"arxiv_id": "2607.06273",
"title": "AgentTether: Graph-Guided Diagnosis and Runtime Intervention for Reliable LLM Agent Operation",
"abstract": "Large language model (LLM) agents are increasingly used for multi-step, stateful tool-use tasks, yet production reliability remains limited. Unlike static software repair, agent repair must recover dynamic trajectories whose early decisions can propagate into later errors and external state changes. Existing automatic remedies address only part of this problem: blind retry adds no diagnosis, outcome feedback says whether a run failed but not where or why, and self-reflection often lacks grounded evidence to prevent the same failure from recurring. We present AgentTether, a run-time repair framework that automates post-run diagnosis and guided recovery without modifying the underlying agent or environment. AgentTether abstracts each run into Transition Units, links them through a dependency-aware Critical Transition Graph, and localizes failure-critical subtrajectories by combining an offline normal-behavior model with a run-local graph detector. It then converts the localized cause into behavior-scoped guidance backed by cross-iteration Repair Memory, and can optionally apply guarded run-time intervention to keep the correction active during re-execution. The same design can be deployed as an offline diagnostic-and-guidance tool or as an online repair layer. We evaluate AgentTether on 261 tau-bench tasks across three domains with Qwen3.7-max, and test cross-model transfer on Banking with GPT-5.4. On the hardest Banking domain, AgentTether repairs 59.04% (49/83) of initially failed Qwen3.7-max tasks and 65.12% (56/86) of initially failed GPT-5.4 tasks. Overall, AgentTether improves repair effectiveness while reducing agent turns and end-to-end approach tokens, suggesting a practical reliability layer that can wrap existing agent deployments, reduce wasted re-execution, and improve recovery without retraining the agent.",
"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"
],
"url": "https://arxiv.org/abs/2607.06273",
"pdf_url": "https://arxiv.org/pdf/2607.06273",
"path": "papers/items/2026-2607-06273-agenttether-graph-guided-diagnosis-and-runtime-intervention-for-reliable-llm-age.md",
"local_title": "\"AgentTether: Graph-Guided Diagnosis and Runtime Intervention for Reliable LLM Agent Operation\"",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"computer-use",
"memory",
"reasoning",
"tool-use"
],
"collection_queries": "llm-agent, tool-use",
"collection_score": "17"
},
{
"arxiv_id": "2607.06341",
"title": "Harnessing Code Agents for Automatic Software Verification",
"abstract": "Formal verification offers the strongest guarantee of software correctness, but it does not scale: the proofs demanded by interactive theorem provers such as Coq require enormous expert effort. Large language models (LLMs) promise to generate these proofs automatically, yet existing approaches wire a fixed, human-designed proof strategy into the system and constrain the model to follow it (retrieving premises and predicting tactics one step at a time, or splitting goals by divide-and-conquer), and still prove only a fraction of their target theorems. We show that imposing such a strategy is unnecessary and limiting. Handing the whole lemma to a general LLM code agent (for example, Claude Code), free to choose its own approach, and wrapping it in a verification harness is both simpler and more effective, achieving full coverage: every targeted lemma proved, with no failures and no Coq expert intervention. The agent writes the proofs under feedback and hard constraints from the harness that keep each one sound (accepted only when the prover's kernel closes it), complete (no obligation left unproved or silently dropped), and terminating (no divergent tactics). We evaluate this harness plus code agent along three dimensions. (1) Core logic: on Iris, the state-of-the-art separation logic for concurrent and memory-manipulating programs, Aria proves all 4,257 lemmas of the four core modules and the 217 lemmas verifying Rust's standard libraries built on it, fully automatically. (2) Comparison with prior LLM provers: on reglang, where prior provers manage barely one in eight, Aria proves all 318. (3) Generality: on iris-lean, the unfinished Lean 4 port of Iris, it proves 72 not-yet-ported lemmas, showing the approach is not specific to Coq. A state-of-the-art model (Claude Opus 4.7) can write proofs for verified software development fully and automatically.",
"published": "2026-07-07",
"updated": "2026-07-07",
"authors": [
"Shuangxiang Kan",
"Shuanglong Kan",
"Sebastian Ertel"
],
"categories": [
"cs.FL",
"cs.AI",
"cs.SE"
],
"url": "https://arxiv.org/abs/2607.06341",
"pdf_url": "https://arxiv.org/pdf/2607.06341",
"path": "papers/items/2026-2607-06341-harnessing-code-agents-for-automatic-software-verification.md",
"local_title": "Harnessing Code Agents for Automatic Software Verification",
"local_status": "queued",
"local_topics": [
"agent-evaluation",
"coding-agent",
"memory",
"rag",
"tool-use"
],
"collection_queries": "coding-agent",
"collection_score": "13"
}
]
}
Reference in New Issue View Git Blame Copy Permalink