{ "generated_at": "2026-07-10", "source_candidates": 516, "title_central": 379, "screened": 516, "windows": { "previous_six_months": { "from_inclusive": "2025-07-09", "to_exclusive": "2026-01-10", "candidates": 94, "title_central": 61, "title_theme_counts": { "personal_shared": 10, "retrieval_representation": 11, "experience_skill_policy": 19, "execution_multimodal": 9, "update_forgetting": 2, "benchmark_diagnosis": 6, "systems_efficiency": 4, "security_governance": 4 } }, "recent_six_months": { "from_inclusive": "2026-01-10", "to_exclusive": "2026-07-11", "candidates": 380, "title_central": 292, "title_theme_counts": { "retrieval_representation": 65, "execution_multimodal": 58, "benchmark_diagnosis": 56, "experience_skill_policy": 65, "systems_efficiency": 37, "update_forgetting": 22, "personal_shared": 27, "security_governance": 32 } } }, "period_counts": { "2025-H1": { "candidates": 37, "title_central": 24, "screen_core": 31, "screen_supporting": 5, "screen_peripheral": 1 }, "baseline-2025H2-Jan2026": { "candidates": 130, "title_central": 85, "screen_core": 115, "screen_peripheral": 3, "screen_supporting": 12 }, "transition-2026-Feb-Mar": { "candidates": 87, "screen_core": 78, "title_central": 64, "screen_supporting": 8, "screen_peripheral": 1 }, "recent-2026-Apr-Jul": { "candidates": 262, "screen_core": 217, "title_central": 206, "screen_supporting": 45 } }, "theme_counts_title_central": { "2025-H1": { "retrieval_access": 17, "representation_organization": 13, "experience_skill_learning": 9, "evaluation_diagnosis": 18, "systems_efficiency": 11, "multimodal_embodied_gui": 14, "execution_state": 5, "update_forgetting": 4, "security_privacy_trust": 6, "multi_agent_shared": 4 }, "baseline-2025H2-Jan2026": { "update_forgetting": 35, "experience_skill_learning": 45, "systems_efficiency": 46, "retrieval_access": 65, "execution_state": 30, "evaluation_diagnosis": 59, "representation_organization": 51, "security_privacy_trust": 15, "multi_agent_shared": 16, "multimodal_embodied_gui": 36 }, "transition-2026-Feb-Mar": { "retrieval_access": 51, "representation_organization": 45, "update_forgetting": 23, "multimodal_embodied_gui": 23, "systems_efficiency": 39, "experience_skill_learning": 24, "evaluation_diagnosis": 50, "multi_agent_shared": 10, "security_privacy_trust": 12, "execution_state": 20 }, "recent-2026-Apr-Jul": { "retrieval_access": 166, "experience_skill_learning": 72, "evaluation_diagnosis": 161, "multimodal_embodied_gui": 73, "multi_agent_shared": 10, "execution_state": 64, "representation_organization": 101, "security_privacy_trust": 74, "systems_efficiency": 126, "update_forgetting": 96 } }, "screen_counts": { "relevance": { "core": 441, "supporting": 70, "peripheral": 5 }, "problem": { "update_forgetting": 214, "execution_state": 129, "representation": 50, "security_privacy": 25, "experience_learning": 46, "shared_memory": 6, "systems_cost": 2, "access": 24, "application": 3, "evaluation": 17 }, "source": { "model": 516 } }, "evidence_score_distribution": { "7": 32, "4": 65, "0": 21, "6": 69, "2": 50, "5": 72, "1": 4, "3": 47, "8": 18, "9": 1 }, "top_evidence_candidates": { "retrieval_access": [ { "arxiv_id": "2606.12945", "published": "2026-06-11", "title": "Learning What to Remember: A Cognitively Grounded Multi-Factor Value Model for Agentic Memory", "score": 9, "url": "https://arxiv.org/abs/2606.12945" }, { "arxiv_id": "2606.20954", "published": "2026-06-18", "title": "Learning What Not to Forget: Long-Horizon Agent Memory from a Few Kilobytes of Learning", "score": 8, "url": "https://arxiv.org/abs/2606.20954" }, { "arxiv_id": "2606.15903", "published": "2026-06-14", "title": "Control-Plane Placement Shapes Forgetting: An Architectural Study of Agent Memory Across Thirteen System Configurations", "score": 8, "url": "https://arxiv.org/abs/2606.15903" }, { "arxiv_id": "2605.30690", "published": "2026-05-29", "title": "ElasticMem: Latent Memory as a Learnable Resource for LLM Agents", "score": 8, "url": "https://arxiv.org/abs/2605.30690" }, { "arxiv_id": "2605.13941", "published": "2026-05-13", "title": "EvolveMem:Self-Evolving Memory Architecture via AutoResearch for LLM Agents", "score": 8, "url": "https://arxiv.org/abs/2605.13941" }, { "arxiv_id": "2605.12493", "published": "2026-05-12", "title": "LongMemEval-V2: Evaluating Long-Term Agent Memory Toward Experienced Colleagues", "score": 8, "url": "https://arxiv.org/abs/2605.12493" }, { "arxiv_id": "2605.08538", "published": "2026-05-08", "title": "Human-Inspired Memory Architecture for LLM Agents", "score": 8, "url": "https://arxiv.org/abs/2605.08538" }, { "arxiv_id": "2604.11563", "published": "2026-04-13", "title": "Synthius-Mem: Brain-Inspired Hallucination-Resistant Persona Memory Achieving 94.4% Memory Accuracy and 99.6% Adversarial Robustness on LoCoMo", "score": 8, "url": "https://arxiv.org/abs/2604.11563" }, { "arxiv_id": "2604.01007", "published": "2026-04-01", "title": "Omni-SimpleMem: Autoresearch-Guided Discovery of Lifelong Multimodal Agent Memory", "score": 8, "url": "https://arxiv.org/abs/2604.01007" }, { "arxiv_id": "2603.13017", "published": "2026-03-13", "title": "Structured Distillation for Personalized Agent Memory: 11x Token Reduction with Retrieval Preservation", "score": 8, "url": "https://arxiv.org/abs/2603.13017" }, { "arxiv_id": "2602.22769", "published": "2026-02-26", "title": "AMA-Bench: Evaluating Long-Horizon Memory for Agentic Applications", "score": 8, "url": "https://arxiv.org/abs/2602.22769" }, { "arxiv_id": "2602.21394", "published": "2026-02-24", "title": "MemoPhishAgent: Memory-Augmented Multi-Modal LLM Agent for Phishing URL Detection", "score": 8, "url": "https://arxiv.org/abs/2602.21394" }, { "arxiv_id": "2602.16493", "published": "2026-02-18", "title": "MMA: Multimodal Memory Agent", "score": 8, "url": "https://arxiv.org/abs/2602.16493" }, { "arxiv_id": "2601.06282", "published": "2026-01-09", "title": "Amory: Building Coherent Narrative-Driven Agent Memory through Agentic Reasoning", "score": 8, "url": "https://arxiv.org/abs/2601.06282" }, { "arxiv_id": "2512.11303", "published": "2025-12-12", "title": "Unifying Dynamic Tool Creation and Cross-Task Experience Sharing through Cognitive Memory Architecture", "score": 8, "url": "https://arxiv.org/abs/2512.11303" }, { "arxiv_id": "2512.02425", "published": "2025-12-02", "title": "WorldMM: Dynamic Multimodal Memory Agent for Long Video Reasoning", "score": 8, "url": "https://arxiv.org/abs/2512.02425" }, { "arxiv_id": "2607.03726", "published": "2026-07-04", "title": "SelfMem: Self-Optimizing Memory for AI Agents", "score": 7, "url": "https://arxiv.org/abs/2607.03726" }, { "arxiv_id": "2606.29914", "published": "2026-06-29", "title": "MemDelta: Controlled Baselines and Hidden Confounds in Agent Memory Evaluation", "score": 7, "url": "https://arxiv.org/abs/2606.29914" }, { "arxiv_id": "2606.24535", "published": "2026-06-23", "title": "Governed Shared Memory for Multi-Agent LLM Systems", "score": 7, "url": "https://arxiv.org/abs/2606.24535" }, { "arxiv_id": "2606.22844", "published": "2026-06-22", "title": "RaMem: Contextual Reinstatement for Long-term Agentic Memory", "score": 7, "url": "https://arxiv.org/abs/2606.22844" } ], "representation_organization": [ { "arxiv_id": "2605.23986", "published": "2026-05-16", "title": "MemForest: An Efficient Agent Memory System with Hierarchical Temporal Indexing", "score": 8, "url": "https://arxiv.org/abs/2605.23986" }, { "arxiv_id": "2605.16233", "published": "2026-05-15", "title": "FORGE: Self-Evolving Agent Memory With No Weight Updates via Population Broadcast", "score": 8, "url": "https://arxiv.org/abs/2605.16233" }, { "arxiv_id": "2605.13941", "published": "2026-05-13", "title": "EvolveMem:Self-Evolving Memory Architecture via AutoResearch for LLM Agents", "score": 8, "url": "https://arxiv.org/abs/2605.13941" }, { "arxiv_id": "2605.08538", "published": "2026-05-08", "title": "Human-Inspired Memory Architecture for LLM Agents", "score": 8, "url": "https://arxiv.org/abs/2605.08538" }, { "arxiv_id": "2604.11563", "published": "2026-04-13", "title": "Synthius-Mem: Brain-Inspired Hallucination-Resistant Persona Memory Achieving 94.4% Memory Accuracy and 99.6% Adversarial Robustness on LoCoMo", "score": 8, "url": "https://arxiv.org/abs/2604.11563" }, { "arxiv_id": "2603.13017", "published": "2026-03-13", "title": "Structured Distillation for Personalized Agent Memory: 11x Token Reduction with Retrieval Preservation", "score": 8, "url": "https://arxiv.org/abs/2603.13017" }, { "arxiv_id": "2602.16493", "published": "2026-02-18", "title": "MMA: Multimodal Memory Agent", "score": 8, "url": "https://arxiv.org/abs/2602.16493" }, { "arxiv_id": "2601.06282", "published": "2026-01-09", "title": "Amory: Building Coherent Narrative-Driven Agent Memory through Agentic Reasoning", "score": 8, "url": "https://arxiv.org/abs/2601.06282" }, { "arxiv_id": "2512.11303", "published": "2025-12-12", "title": "Unifying Dynamic Tool Creation and Cross-Task Experience Sharing through Cognitive Memory Architecture", "score": 8, "url": "https://arxiv.org/abs/2512.11303" }, { "arxiv_id": "2512.02425", "published": "2025-12-02", "title": "WorldMM: Dynamic Multimodal Memory Agent for Long Video Reasoning", "score": 8, "url": "https://arxiv.org/abs/2512.02425" }, { "arxiv_id": "2606.29788", "published": "2026-06-29", "title": "MemLeak: Diagnosing Information Leaks in Multimodal Agent Memory", "score": 7, "url": "https://arxiv.org/abs/2606.29788" }, { "arxiv_id": "2606.22844", "published": "2026-06-22", "title": "RaMem: Contextual Reinstatement for Long-term Agentic Memory", "score": 7, "url": "https://arxiv.org/abs/2606.22844" }, { "arxiv_id": "2606.09900", "published": "2026-06-05", "title": "Less Context, More Accuracy: A Bi-Temporal Memory Engine for LLM Agents Where a Lean Retrieved Context Beats the Full History", "score": 7, "url": "https://arxiv.org/abs/2606.09900" }, { "arxiv_id": "2605.14498", "published": "2026-05-14", "title": "GroupMemBench: Benchmarking LLM Agent Memory in Multi-Party Conversations", "score": 7, "url": "https://arxiv.org/abs/2605.14498" }, { "arxiv_id": "2605.09033", "published": "2026-05-09", "title": "ShadowMerge: A Novel Poisoning Attack on Graph-Based Agent Memory via Relation-Channel Conflicts", "score": 7, "url": "https://arxiv.org/abs/2605.09033" }, { "arxiv_id": "2605.06527", "published": "2026-05-07", "title": "STALE: Can LLM Agents Know When Their Memories Are No Longer Valid?", "score": 7, "url": "https://arxiv.org/abs/2605.06527" }, { "arxiv_id": "2605.03312", "published": "2026-05-05", "title": "MemFlow: Intent-Driven Memory Orchestration for Small Language Model Agents", "score": 7, "url": "https://arxiv.org/abs/2605.03312" }, { "arxiv_id": "2603.14597", "published": "2026-03-15", "title": "D-MEM: Dopamine-Gated Agentic Memory via Reward Prediction Error Routing", "score": 7, "url": "https://arxiv.org/abs/2603.14597" }, { "arxiv_id": "2603.13676", "published": "2026-03-14", "title": "TheraAgent: Multi-Agent Framework with Self-Evolving Memory and Evidence-Calibrated Reasoning for PET Theranostics", "score": 7, "url": "https://arxiv.org/abs/2603.13676" }, { "arxiv_id": "2603.04549", "published": "2026-03-04", "title": "Adaptive Memory Admission Control for LLM Agents", "score": 7, "url": "https://arxiv.org/abs/2603.04549" } ], "update_forgetting": [ { "arxiv_id": "2606.12945", "published": "2026-06-11", "title": "Learning What to Remember: A Cognitively Grounded Multi-Factor Value Model for Agentic Memory", "score": 9, "url": "https://arxiv.org/abs/2606.12945" }, { "arxiv_id": "2606.20954", "published": "2026-06-18", "title": "Learning What Not to Forget: Long-Horizon Agent Memory from a Few Kilobytes of Learning", "score": 8, "url": "https://arxiv.org/abs/2606.20954" }, { "arxiv_id": "2606.15903", "published": "2026-06-14", "title": "Control-Plane Placement Shapes Forgetting: An Architectural Study of Agent Memory Across Thirteen System Configurations", "score": 8, "url": "https://arxiv.org/abs/2606.15903" }, { "arxiv_id": "2605.23986", "published": "2026-05-16", "title": "MemForest: An Efficient Agent Memory System with Hierarchical Temporal Indexing", "score": 8, "url": "https://arxiv.org/abs/2605.23986" }, { "arxiv_id": "2605.16233", "published": "2026-05-15", "title": "FORGE: Self-Evolving Agent Memory With No Weight Updates via Population Broadcast", "score": 8, "url": "https://arxiv.org/abs/2605.16233" }, { "arxiv_id": "2605.07242", "published": "2026-05-08", "title": "MEMOREPAIR: Barrier-First Cascade Repair in Agentic Memory", "score": 8, "url": "https://arxiv.org/abs/2605.07242" }, { "arxiv_id": "2605.08538", "published": "2026-05-08", "title": "Human-Inspired Memory Architecture for LLM Agents", "score": 8, "url": "https://arxiv.org/abs/2605.08538" }, { "arxiv_id": "2604.11563", "published": "2026-04-13", "title": "Synthius-Mem: Brain-Inspired Hallucination-Resistant Persona Memory Achieving 94.4% Memory Accuracy and 99.6% Adversarial Robustness on LoCoMo", "score": 8, "url": "https://arxiv.org/abs/2604.11563" }, { "arxiv_id": "2602.16493", "published": "2026-02-18", "title": "MMA: Multimodal Memory Agent", "score": 8, "url": "https://arxiv.org/abs/2602.16493" }, { "arxiv_id": "2601.06282", "published": "2026-01-09", "title": "Amory: Building Coherent Narrative-Driven Agent Memory through Agentic Reasoning", "score": 8, "url": "https://arxiv.org/abs/2601.06282" }, { "arxiv_id": "2512.02425", "published": "2025-12-02", "title": "WorldMM: Dynamic Multimodal Memory Agent for Long Video Reasoning", "score": 8, "url": "https://arxiv.org/abs/2512.02425" }, { "arxiv_id": "2607.01523", "published": "2026-07-01", "title": "Multi-Head Recurrent Memory Agents", "score": 7, "url": "https://arxiv.org/abs/2607.01523" }, { "arxiv_id": "2606.29788", "published": "2026-06-29", "title": "MemLeak: Diagnosing Information Leaks in Multimodal Agent Memory", "score": 7, "url": "https://arxiv.org/abs/2606.29788" }, { "arxiv_id": "2606.29914", "published": "2026-06-29", "title": "MemDelta: Controlled Baselines and Hidden Confounds in Agent Memory Evaluation", "score": 7, "url": "https://arxiv.org/abs/2606.29914" }, { "arxiv_id": "2606.24535", "published": "2026-06-23", "title": "Governed Shared Memory for Multi-Agent LLM Systems", "score": 7, "url": "https://arxiv.org/abs/2606.24535" }, { "arxiv_id": "2606.25161", "published": "2026-06-23", "title": "TRUSTMEM: Learning Trustworthy Memory Consolidation for LLM Agents with Long-Term Memory", "score": 7, "url": "https://arxiv.org/abs/2606.25161" }, { "arxiv_id": "2606.09900", "published": "2026-06-05", "title": "Less Context, More Accuracy: A Bi-Temporal Memory Engine for LLM Agents Where a Lean Retrieved Context Beats the Full History", "score": 7, "url": "https://arxiv.org/abs/2606.09900" }, { "arxiv_id": "2605.17625", "published": "2026-05-17", "title": "Episodic-Semantic Memory Architecture for Long-Horizon Scientific Agents", "score": 7, "url": "https://arxiv.org/abs/2605.17625" }, { "arxiv_id": "2605.14498", "published": "2026-05-14", "title": "GroupMemBench: Benchmarking LLM Agent Memory in Multi-Party Conversations", "score": 7, "url": "https://arxiv.org/abs/2605.14498" }, { "arxiv_id": "2605.09033", "published": "2026-05-09", "title": "ShadowMerge: A Novel Poisoning Attack on Graph-Based Agent Memory via Relation-Channel Conflicts", "score": 7, "url": "https://arxiv.org/abs/2605.09033" } ], "execution_state": [ { "arxiv_id": "2606.20954", "published": "2026-06-18", "title": "Learning What Not to Forget: Long-Horizon Agent Memory from a Few Kilobytes of Learning", "score": 8, "url": "https://arxiv.org/abs/2606.20954" }, { "arxiv_id": "2602.22769", "published": "2026-02-26", "title": "AMA-Bench: Evaluating Long-Horizon Memory for Agentic Applications", "score": 8, "url": "https://arxiv.org/abs/2602.22769" }, { "arxiv_id": "2602.16493", "published": "2026-02-18", "title": "MMA: Multimodal Memory Agent", "score": 8, "url": "https://arxiv.org/abs/2602.16493" }, { "arxiv_id": "2607.03726", "published": "2026-07-04", "title": "SelfMem: Self-Optimizing Memory for AI Agents", "score": 7, "url": "https://arxiv.org/abs/2607.03726" }, { "arxiv_id": "2605.17625", "published": "2026-05-17", "title": "Episodic-Semantic Memory Architecture for Long-Horizon Scientific Agents", "score": 7, "url": "https://arxiv.org/abs/2605.17625" }, { "arxiv_id": "2605.03312", "published": "2026-05-05", "title": "MemFlow: Intent-Driven Memory Orchestration for Small Language Model Agents", "score": 7, "url": "https://arxiv.org/abs/2605.03312" }, { "arxiv_id": "2512.12818", "published": "2025-12-14", "title": "Hindsight is 20/20: Building Agent Memory that Retains, Recalls, and Reflects", "score": 7, "url": "https://arxiv.org/abs/2512.12818" }, { "arxiv_id": "2505.22657", "published": "2025-05-28", "title": "3DLLM-Mem: Long-Term Spatial-Temporal Memory for Embodied 3D Large Language Model", "score": 7, "url": "https://arxiv.org/abs/2505.22657" }, { "arxiv_id": "2505.19436", "published": "2025-05-26", "title": "Task Memory Engine: Spatial Memory for Robust Multi-Step LLM Agents", "score": 7, "url": "https://arxiv.org/abs/2505.19436" }, { "arxiv_id": "2607.08716", "published": "2026-07-09", "title": "Remember When It Matters: Proactive Memory Agent for Long-Horizon Agents", "score": 6, "url": "https://arxiv.org/abs/2607.08716" }, { "arxiv_id": "2606.27499", "published": "2026-06-25", "title": "DMV-Bench: Diagnosing Long-Horizon Multimodal Agents' Visual Memory with Incidental Cue Injection", "score": 6, "url": "https://arxiv.org/abs/2606.27499" }, { "arxiv_id": "2606.24775", "published": "2026-06-23", "title": "Are We Ready For An Agent-Native Memory System?", "score": 6, "url": "https://arxiv.org/abs/2606.24775" }, { "arxiv_id": "2606.22877", "published": "2026-06-22", "title": "DynamicMem: A Long-Horizon Memory Benchmark in Real-World Settings", "score": 6, "url": "https://arxiv.org/abs/2606.22877" }, { "arxiv_id": "2606.05684", "published": "2026-06-04", "title": "AdaMEM: Test-Time Adaptive Memory for Language Agents", "score": 6, "url": "https://arxiv.org/abs/2606.05684" }, { "arxiv_id": "2606.06090", "published": "2026-06-04", "title": "Beyond Semantic Organization: Memory as Execution State Management for Long-Horizon Agents", "score": 6, "url": "https://arxiv.org/abs/2606.06090" }, { "arxiv_id": "2606.03374", "published": "2026-06-02", "title": "eMEM: A Hybrid Spatio-Temporal Memory System For Embodied Agents", "score": 6, "url": "https://arxiv.org/abs/2606.03374" }, { "arxiv_id": "2606.00619", "published": "2026-05-30", "title": "MemPro: Agentic Memory Systems as Evolvable Programs", "score": 6, "url": "https://arxiv.org/abs/2606.00619" }, { "arxiv_id": "2605.18421", "published": "2026-05-18", "title": "EvoMemBench: Benchmarking Agent Memory from a Self-Evolving Perspective", "score": 6, "url": "https://arxiv.org/abs/2605.18421" }, { "arxiv_id": "2605.17641", "published": "2026-05-17", "title": "Causal Intervention-Based Memory Selection for Long-Horizon LLM Agents", "score": 6, "url": "https://arxiv.org/abs/2605.17641" }, { "arxiv_id": "2605.16746", "published": "2026-05-16", "title": "State Contamination in Memory-Augmented LLM Agents", "score": 6, "url": "https://arxiv.org/abs/2605.16746" } ], "experience_skill_learning": [ { "arxiv_id": "2605.30690", "published": "2026-05-29", "title": "ElasticMem: Latent Memory as a Learnable Resource for LLM Agents", "score": 8, "url": "https://arxiv.org/abs/2605.30690" }, { "arxiv_id": "2605.16233", "published": "2026-05-15", "title": "FORGE: Self-Evolving Agent Memory With No Weight Updates via Population Broadcast", "score": 8, "url": "https://arxiv.org/abs/2605.16233" }, { "arxiv_id": "2605.13941", "published": "2026-05-13", "title": "EvolveMem:Self-Evolving Memory Architecture via AutoResearch for LLM Agents", "score": 8, "url": "https://arxiv.org/abs/2605.13941" }, { "arxiv_id": "2605.12493", "published": "2026-05-12", "title": "LongMemEval-V2: Evaluating Long-Term Agent Memory Toward Experienced Colleagues", "score": 8, "url": "https://arxiv.org/abs/2605.12493" }, { "arxiv_id": "2605.07242", "published": "2026-05-08", "title": "MEMOREPAIR: Barrier-First Cascade Repair in Agentic Memory", "score": 8, "url": "https://arxiv.org/abs/2605.07242" }, { "arxiv_id": "2604.11563", "published": "2026-04-13", "title": "Synthius-Mem: Brain-Inspired Hallucination-Resistant Persona Memory Achieving 94.4% Memory Accuracy and 99.6% Adversarial Robustness on LoCoMo", "score": 8, "url": "https://arxiv.org/abs/2604.11563" }, { "arxiv_id": "2604.01007", "published": "2026-04-01", "title": "Omni-SimpleMem: Autoresearch-Guided Discovery of Lifelong Multimodal Agent Memory", "score": 8, "url": "https://arxiv.org/abs/2604.01007" }, { "arxiv_id": "2602.21394", "published": "2026-02-24", "title": "MemoPhishAgent: Memory-Augmented Multi-Modal LLM Agent for Phishing URL Detection", "score": 8, "url": "https://arxiv.org/abs/2602.21394" }, { "arxiv_id": "2512.11303", "published": "2025-12-12", "title": "Unifying Dynamic Tool Creation and Cross-Task Experience Sharing through Cognitive Memory Architecture", "score": 8, "url": "https://arxiv.org/abs/2512.11303" }, { "arxiv_id": "2607.03726", "published": "2026-07-04", "title": "SelfMem: Self-Optimizing Memory for AI Agents", "score": 7, "url": "https://arxiv.org/abs/2607.03726" }, { "arxiv_id": "2606.25161", "published": "2026-06-23", "title": "TRUSTMEM: Learning Trustworthy Memory Consolidation for LLM Agents with Long-Term Memory", "score": 7, "url": "https://arxiv.org/abs/2606.25161" }, { "arxiv_id": "2606.22844", "published": "2026-06-22", "title": "RaMem: Contextual Reinstatement for Long-term Agentic Memory", "score": 7, "url": "https://arxiv.org/abs/2606.22844" }, { "arxiv_id": "2606.23127", "published": "2026-06-22", "title": "Managing Procedural Memory in LLM Agents: Control, Adaptation, and Evaluation", "score": 7, "url": "https://arxiv.org/abs/2606.23127" }, { "arxiv_id": "2605.14498", "published": "2026-05-14", "title": "GroupMemBench: Benchmarking LLM Agent Memory in Multi-Party Conversations", "score": 7, "url": "https://arxiv.org/abs/2605.14498" }, { "arxiv_id": "2603.13676", "published": "2026-03-14", "title": "TheraAgent: Multi-Agent Framework with Self-Evolving Memory and Evidence-Calibrated Reasoning for PET Theranostics", "score": 7, "url": "https://arxiv.org/abs/2603.13676" }, { "arxiv_id": "2512.12967", "published": "2025-12-15", "title": "QwenLong-L1.5: Post-Training Recipe for Long-Context Reasoning and Memory Management", "score": 7, "url": "https://arxiv.org/abs/2512.12967" }, { "arxiv_id": "2512.12818", "published": "2025-12-14", "title": "Hindsight is 20/20: Building Agent Memory that Retains, Recalls, and Reflects", "score": 7, "url": "https://arxiv.org/abs/2512.12818" }, { "arxiv_id": "2512.10696", "published": "2025-12-11", "title": "Remember Me, Refine Me: A Dynamic Procedural Memory Framework for Experience-Driven Agent Evolution", "score": 7, "url": "https://arxiv.org/abs/2512.10696" }, { "arxiv_id": "2505.22657", "published": "2025-05-28", "title": "3DLLM-Mem: Long-Term Spatial-Temporal Memory for Embodied 3D Large Language Model", "score": 7, "url": "https://arxiv.org/abs/2505.22657" }, { "arxiv_id": "2501.13956", "published": "2025-01-20", "title": "Zep: A Temporal Knowledge Graph Architecture for Agent Memory", "score": 7, "url": "https://arxiv.org/abs/2501.13956" } ], "evaluation_diagnosis": [ { "arxiv_id": "2606.12945", "published": "2026-06-11", "title": "Learning What to Remember: A Cognitively Grounded Multi-Factor Value Model for Agentic Memory", "score": 9, "url": "https://arxiv.org/abs/2606.12945" }, { "arxiv_id": "2606.20954", "published": "2026-06-18", "title": "Learning What Not to Forget: Long-Horizon Agent Memory from a Few Kilobytes of Learning", "score": 8, "url": "https://arxiv.org/abs/2606.20954" }, { "arxiv_id": "2606.15903", "published": "2026-06-14", "title": "Control-Plane Placement Shapes Forgetting: An Architectural Study of Agent Memory Across Thirteen System Configurations", "score": 8, "url": "https://arxiv.org/abs/2606.15903" }, { "arxiv_id": "2605.30690", "published": "2026-05-29", "title": "ElasticMem: Latent Memory as a Learnable Resource for LLM Agents", "score": 8, "url": "https://arxiv.org/abs/2605.30690" }, { "arxiv_id": "2605.23986", "published": "2026-05-16", "title": "MemForest: An Efficient Agent Memory System with Hierarchical Temporal Indexing", "score": 8, "url": "https://arxiv.org/abs/2605.23986" }, { "arxiv_id": "2605.16233", "published": "2026-05-15", "title": "FORGE: Self-Evolving Agent Memory With No Weight Updates via Population Broadcast", "score": 8, "url": "https://arxiv.org/abs/2605.16233" }, { "arxiv_id": "2605.13941", "published": "2026-05-13", "title": "EvolveMem:Self-Evolving Memory Architecture via AutoResearch for LLM Agents", "score": 8, "url": "https://arxiv.org/abs/2605.13941" }, { "arxiv_id": "2605.12493", "published": "2026-05-12", "title": "LongMemEval-V2: Evaluating Long-Term Agent Memory Toward Experienced Colleagues", "score": 8, "url": "https://arxiv.org/abs/2605.12493" }, { "arxiv_id": "2605.07242", "published": "2026-05-08", "title": "MEMOREPAIR: Barrier-First Cascade Repair in Agentic Memory", "score": 8, "url": "https://arxiv.org/abs/2605.07242" }, { "arxiv_id": "2605.08538", "published": "2026-05-08", "title": "Human-Inspired Memory Architecture for LLM Agents", "score": 8, "url": "https://arxiv.org/abs/2605.08538" }, { "arxiv_id": "2604.11563", "published": "2026-04-13", "title": "Synthius-Mem: Brain-Inspired Hallucination-Resistant Persona Memory Achieving 94.4% Memory Accuracy and 99.6% Adversarial Robustness on LoCoMo", "score": 8, "url": "https://arxiv.org/abs/2604.11563" }, { "arxiv_id": "2604.01007", "published": "2026-04-01", "title": "Omni-SimpleMem: Autoresearch-Guided Discovery of Lifelong Multimodal Agent Memory", "score": 8, "url": "https://arxiv.org/abs/2604.01007" }, { "arxiv_id": "2603.13017", "published": "2026-03-13", "title": "Structured Distillation for Personalized Agent Memory: 11x Token Reduction with Retrieval Preservation", "score": 8, "url": "https://arxiv.org/abs/2603.13017" }, { "arxiv_id": "2602.22769", "published": "2026-02-26", "title": "AMA-Bench: Evaluating Long-Horizon Memory for Agentic Applications", "score": 8, "url": "https://arxiv.org/abs/2602.22769" }, { "arxiv_id": "2602.21394", "published": "2026-02-24", "title": "MemoPhishAgent: Memory-Augmented Multi-Modal LLM Agent for Phishing URL Detection", "score": 8, "url": "https://arxiv.org/abs/2602.21394" }, { "arxiv_id": "2602.16493", "published": "2026-02-18", "title": "MMA: Multimodal Memory Agent", "score": 8, "url": "https://arxiv.org/abs/2602.16493" }, { "arxiv_id": "2601.06282", "published": "2026-01-09", "title": "Amory: Building Coherent Narrative-Driven Agent Memory through Agentic Reasoning", "score": 8, "url": "https://arxiv.org/abs/2601.06282" }, { "arxiv_id": "2512.11303", "published": "2025-12-12", "title": "Unifying Dynamic Tool Creation and Cross-Task Experience Sharing through Cognitive Memory Architecture", "score": 8, "url": "https://arxiv.org/abs/2512.11303" }, { "arxiv_id": "2512.02425", "published": "2025-12-02", "title": "WorldMM: Dynamic Multimodal Memory Agent for Long Video Reasoning", "score": 8, "url": "https://arxiv.org/abs/2512.02425" }, { "arxiv_id": "2607.03726", "published": "2026-07-04", "title": "SelfMem: Self-Optimizing Memory for AI Agents", "score": 7, "url": "https://arxiv.org/abs/2607.03726" } ], "security_privacy_trust": [ { "arxiv_id": "2606.15903", "published": "2026-06-14", "title": "Control-Plane Placement Shapes Forgetting: An Architectural Study of Agent Memory Across Thirteen System Configurations", "score": 8, "url": "https://arxiv.org/abs/2606.15903" }, { "arxiv_id": "2605.16233", "published": "2026-05-15", "title": "FORGE: Self-Evolving Agent Memory With No Weight Updates via Population Broadcast", "score": 8, "url": "https://arxiv.org/abs/2605.16233" }, { "arxiv_id": "2605.07242", "published": "2026-05-08", "title": "MEMOREPAIR: Barrier-First Cascade Repair in Agentic Memory", "score": 8, "url": "https://arxiv.org/abs/2605.07242" }, { "arxiv_id": "2605.08538", "published": "2026-05-08", "title": "Human-Inspired Memory Architecture for LLM Agents", "score": 8, "url": "https://arxiv.org/abs/2605.08538" }, { "arxiv_id": "2604.11563", "published": "2026-04-13", "title": "Synthius-Mem: Brain-Inspired Hallucination-Resistant Persona Memory Achieving 94.4% Memory Accuracy and 99.6% Adversarial Robustness on LoCoMo", "score": 8, "url": "https://arxiv.org/abs/2604.11563" }, { "arxiv_id": "2602.21394", "published": "2026-02-24", "title": "MemoPhishAgent: Memory-Augmented Multi-Modal LLM Agent for Phishing URL Detection", "score": 8, "url": "https://arxiv.org/abs/2602.21394" }, { "arxiv_id": "2607.01523", "published": "2026-07-01", "title": "Multi-Head Recurrent Memory Agents", "score": 7, "url": "https://arxiv.org/abs/2607.01523" }, { "arxiv_id": "2606.29788", "published": "2026-06-29", "title": "MemLeak: Diagnosing Information Leaks in Multimodal Agent Memory", "score": 7, "url": "https://arxiv.org/abs/2606.29788" }, { "arxiv_id": "2606.24535", "published": "2026-06-23", "title": "Governed Shared Memory for Multi-Agent LLM Systems", "score": 7, "url": "https://arxiv.org/abs/2606.24535" }, { "arxiv_id": "2606.25161", "published": "2026-06-23", "title": "TRUSTMEM: Learning Trustworthy Memory Consolidation for LLM Agents with Long-Term Memory", "score": 7, "url": "https://arxiv.org/abs/2606.25161" }, { "arxiv_id": "2606.09900", "published": "2026-06-05", "title": "Less Context, More Accuracy: A Bi-Temporal Memory Engine for LLM Agents Where a Lean Retrieved Context Beats the Full History", "score": 7, "url": "https://arxiv.org/abs/2606.09900" }, { "arxiv_id": "2605.29630", "published": "2026-05-28", "title": "Entity-Collision: A Stratified Protocol for Attributing Retrieval Lift in Agent Memory", "score": 7, "url": "https://arxiv.org/abs/2605.29630" }, { "arxiv_id": "2605.14498", "published": "2026-05-14", "title": "GroupMemBench: Benchmarking LLM Agent Memory in Multi-Party Conversations", "score": 7, "url": "https://arxiv.org/abs/2605.14498" }, { "arxiv_id": "2605.09033", "published": "2026-05-09", "title": "ShadowMerge: A Novel Poisoning Attack on Graph-Based Agent Memory via Relation-Channel Conflicts", "score": 7, "url": "https://arxiv.org/abs/2605.09033" }, { "arxiv_id": "2605.08442", "published": "2026-05-08", "title": "Defense effectiveness across architectural layers: a mechanistic evaluation of persistent memory attacks on stateful LLM agents", "score": 7, "url": "https://arxiv.org/abs/2605.08442" }, { "arxiv_id": "2603.14597", "published": "2026-03-15", "title": "D-MEM: Dopamine-Gated Agentic Memory via Reward Prediction Error Routing", "score": 7, "url": "https://arxiv.org/abs/2603.14597" }, { "arxiv_id": "2603.13676", "published": "2026-03-14", "title": "TheraAgent: Multi-Agent Framework with Self-Evolving Memory and Evidence-Calibrated Reasoning for PET Theranostics", "score": 7, "url": "https://arxiv.org/abs/2603.13676" }, { "arxiv_id": "2512.12818", "published": "2025-12-14", "title": "Hindsight is 20/20: Building Agent Memory that Retains, Recalls, and Reflects", "score": 7, "url": "https://arxiv.org/abs/2512.12818" }, { "arxiv_id": "2505.19436", "published": "2025-05-26", "title": "Task Memory Engine: Spatial Memory for Robust Multi-Step LLM Agents", "score": 7, "url": "https://arxiv.org/abs/2505.19436" }, { "arxiv_id": "2607.05029", "published": "2026-07-06", "title": "Your Agent's Memories Are Not Its Own: Forged Reasoning Attacks on LLM Agent Memory and Defenses", "score": 6, "url": "https://arxiv.org/abs/2607.05029" } ], "multi_agent_shared": [ { "arxiv_id": "2512.11303", "published": "2025-12-12", "title": "Unifying Dynamic Tool Creation and Cross-Task Experience Sharing through Cognitive Memory Architecture", "score": 8, "url": "https://arxiv.org/abs/2512.11303" }, { "arxiv_id": "2606.24535", "published": "2026-06-23", "title": "Governed Shared Memory for Multi-Agent LLM Systems", "score": 7, "url": "https://arxiv.org/abs/2606.24535" }, { "arxiv_id": "2603.13676", "published": "2026-03-14", "title": "TheraAgent: Multi-Agent Framework with Self-Evolving Memory and Evidence-Calibrated Reasoning for PET Theranostics", "score": 7, "url": "https://arxiv.org/abs/2603.13676" }, { "arxiv_id": "2601.20352", "published": "2026-01-28", "title": "AMA: Adaptive Memory via Multi-Agent Collaboration", "score": 7, "url": "https://arxiv.org/abs/2601.20352" }, { "arxiv_id": "2512.10166", "published": "2025-12-10", "title": "Emergent Collective Memory in Decentralized Multi-Agent AI Systems", "score": 7, "url": "https://arxiv.org/abs/2512.10166" }, { "arxiv_id": "2605.22721", "published": "2026-05-21", "title": "Self-Evolving Multi-Agent Systems via Decentralized Memory", "score": 6, "url": "https://arxiv.org/abs/2605.22721" }, { "arxiv_id": "2605.16746", "published": "2026-05-16", "title": "State Contamination in Memory-Augmented LLM Agents", "score": 6, "url": "https://arxiv.org/abs/2605.16746" }, { "arxiv_id": "2602.21477", "published": "2026-02-25", "title": "Pancake: Hierarchical Memory System for Multi-Agent LLM Serving", "score": 6, "url": "https://arxiv.org/abs/2602.21477" }, { "arxiv_id": "2507.07957", "published": "2025-07-10", "title": "MIRIX: Multi-Agent Memory System for LLM-Based Agents", "score": 6, "url": "https://arxiv.org/abs/2507.07957" }, { "arxiv_id": "2502.16090", "published": "2025-02-22", "title": "Echo: A Large Language Model with Temporal Episodic Memory", "score": 6, "url": "https://arxiv.org/abs/2502.16090" }, { "arxiv_id": "2605.04811", "published": "2026-05-06", "title": "Tree-based Credit Assignment for Multi-Agent Memory System", "score": 5, "url": "https://arxiv.org/abs/2605.04811" }, { "arxiv_id": "2603.12631", "published": "2026-03-13", "title": "Joint Optimization of Multi-agent Memory System", "score": 5, "url": "https://arxiv.org/abs/2603.12631" }, { "arxiv_id": "2603.02240", "published": "2026-02-17", "title": "SuperLocalMemory: Privacy-Preserving Multi-Agent Memory with Bayesian Trust Defense Against Memory Poisoning", "score": 5, "url": "https://arxiv.org/abs/2603.02240" }, { "arxiv_id": "2603.04428", "published": "2026-02-17", "title": "Agent Memory Below the Prompt: Persistent Q4 KV Cache for Multi-Agent LLM Inference on Edge Devices", "score": 5, "url": "https://arxiv.org/abs/2603.04428" }, { "arxiv_id": "2602.03036", "published": "2026-02-03", "title": "LatentMem: Customizing Latent Memory for Multi-Agent Systems", "score": 5, "url": "https://arxiv.org/abs/2602.03036" }, { "arxiv_id": "2601.12771", "published": "2026-01-19", "title": "Who Does This Name Remind You of ? Nationality Prediction via Large Language Model Associative Memory", "score": 5, "url": "https://arxiv.org/abs/2601.12771" }, { "arxiv_id": "2601.02732", "published": "2026-01-06", "title": "Agentic Memory Enhanced Recursive Reasoning for Root Cause Localization in Microservices", "score": 5, "url": "https://arxiv.org/abs/2601.02732" }, { "arxiv_id": "2606.28349", "published": "2026-06-03", "title": "HMARS: A Hierarchical Multi-Agent Memory System for Long-Context Reasoning", "score": 4, "url": "https://arxiv.org/abs/2606.28349" }, { "arxiv_id": "2604.02623", "published": "2026-04-03", "title": "Poison Once, Exploit Forever: Environment-Injected Memory Poisoning Attacks on Web Agents", "score": 4, "url": "https://arxiv.org/abs/2604.02623" }, { "arxiv_id": "2604.01560", "published": "2026-04-02", "title": "DeltaMem: Towards Agentic Memory Management via Reinforcement Learning", "score": 4, "url": "https://arxiv.org/abs/2604.01560" } ], "multimodal_embodied_gui": [ { "arxiv_id": "2605.30690", "published": "2026-05-29", "title": "ElasticMem: Latent Memory as a Learnable Resource for LLM Agents", "score": 8, "url": "https://arxiv.org/abs/2605.30690" }, { "arxiv_id": "2605.12493", "published": "2026-05-12", "title": "LongMemEval-V2: Evaluating Long-Term Agent Memory Toward Experienced Colleagues", "score": 8, "url": "https://arxiv.org/abs/2605.12493" }, { "arxiv_id": "2604.01007", "published": "2026-04-01", "title": "Omni-SimpleMem: Autoresearch-Guided Discovery of Lifelong Multimodal Agent Memory", "score": 8, "url": "https://arxiv.org/abs/2604.01007" }, { "arxiv_id": "2602.21394", "published": "2026-02-24", "title": "MemoPhishAgent: Memory-Augmented Multi-Modal LLM Agent for Phishing URL Detection", "score": 8, "url": "https://arxiv.org/abs/2602.21394" }, { "arxiv_id": "2602.16493", "published": "2026-02-18", "title": "MMA: Multimodal Memory Agent", "score": 8, "url": "https://arxiv.org/abs/2602.16493" }, { "arxiv_id": "2512.02425", "published": "2025-12-02", "title": "WorldMM: Dynamic Multimodal Memory Agent for Long Video Reasoning", "score": 8, "url": "https://arxiv.org/abs/2512.02425" }, { "arxiv_id": "2607.03726", "published": "2026-07-04", "title": "SelfMem: Self-Optimizing Memory for AI Agents", "score": 7, "url": "https://arxiv.org/abs/2607.03726" }, { "arxiv_id": "2606.29788", "published": "2026-06-29", "title": "MemLeak: Diagnosing Information Leaks in Multimodal Agent Memory", "score": 7, "url": "https://arxiv.org/abs/2606.29788" }, { "arxiv_id": "2606.25161", "published": "2026-06-23", "title": "TRUSTMEM: Learning Trustworthy Memory Consolidation for LLM Agents with Long-Term Memory", "score": 7, "url": "https://arxiv.org/abs/2606.25161" }, { "arxiv_id": "2606.23127", "published": "2026-06-22", "title": "Managing Procedural Memory in LLM Agents: Control, Adaptation, and Evaluation", "score": 7, "url": "https://arxiv.org/abs/2606.23127" }, { "arxiv_id": "2605.14498", "published": "2026-05-14", "title": "GroupMemBench: Benchmarking LLM Agent Memory in Multi-Party Conversations", "score": 7, "url": "https://arxiv.org/abs/2605.14498" }, { "arxiv_id": "2605.08442", "published": "2026-05-08", "title": "Defense effectiveness across architectural layers: a mechanistic evaluation of persistent memory attacks on stateful LLM agents", "score": 7, "url": "https://arxiv.org/abs/2605.08442" }, { "arxiv_id": "2512.12818", "published": "2025-12-14", "title": "Hindsight is 20/20: Building Agent Memory that Retains, Recalls, and Reflects", "score": 7, "url": "https://arxiv.org/abs/2512.12818" }, { "arxiv_id": "2508.12630", "published": "2025-08-18", "title": "Semantic Anchoring in Agentic Memory: Leveraging Linguistic Structures for Persistent Conversational Context", "score": 7, "url": "https://arxiv.org/abs/2508.12630" }, { "arxiv_id": "2505.22657", "published": "2025-05-28", "title": "3DLLM-Mem: Long-Term Spatial-Temporal Memory for Embodied 3D Large Language Model", "score": 7, "url": "https://arxiv.org/abs/2505.22657" }, { "arxiv_id": "2607.08716", "published": "2026-07-09", "title": "Remember When It Matters: Proactive Memory Agent for Long-Horizon Agents", "score": 6, "url": "https://arxiv.org/abs/2607.08716" }, { "arxiv_id": "2606.27499", "published": "2026-06-25", "title": "DMV-Bench: Diagnosing Long-Horizon Multimodal Agents' Visual Memory with Incidental Cue Injection", "score": 6, "url": "https://arxiv.org/abs/2606.27499" }, { "arxiv_id": "2606.05684", "published": "2026-06-04", "title": "AdaMEM: Test-Time Adaptive Memory for Language Agents", "score": 6, "url": "https://arxiv.org/abs/2606.05684" }, { "arxiv_id": "2606.06090", "published": "2026-06-04", "title": "Beyond Semantic Organization: Memory as Execution State Management for Long-Horizon Agents", "score": 6, "url": "https://arxiv.org/abs/2606.06090" }, { "arxiv_id": "2606.03374", "published": "2026-06-02", "title": "eMEM: A Hybrid Spatio-Temporal Memory System For Embodied Agents", "score": 6, "url": "https://arxiv.org/abs/2606.03374" } ], "systems_efficiency": [ { "arxiv_id": "2606.12945", "published": "2026-06-11", "title": "Learning What to Remember: A Cognitively Grounded Multi-Factor Value Model for Agentic Memory", "score": 9, "url": "https://arxiv.org/abs/2606.12945" }, { "arxiv_id": "2606.20954", "published": "2026-06-18", "title": "Learning What Not to Forget: Long-Horizon Agent Memory from a Few Kilobytes of Learning", "score": 8, "url": "https://arxiv.org/abs/2606.20954" }, { "arxiv_id": "2606.15903", "published": "2026-06-14", "title": "Control-Plane Placement Shapes Forgetting: An Architectural Study of Agent Memory Across Thirteen System Configurations", "score": 8, "url": "https://arxiv.org/abs/2606.15903" }, { "arxiv_id": "2605.30690", "published": "2026-05-29", "title": "ElasticMem: Latent Memory as a Learnable Resource for LLM Agents", "score": 8, "url": "https://arxiv.org/abs/2605.30690" }, { "arxiv_id": "2605.23986", "published": "2026-05-16", "title": "MemForest: An Efficient Agent Memory System with Hierarchical Temporal Indexing", "score": 8, "url": "https://arxiv.org/abs/2605.23986" }, { "arxiv_id": "2605.16233", "published": "2026-05-15", "title": "FORGE: Self-Evolving Agent Memory With No Weight Updates via Population Broadcast", "score": 8, "url": "https://arxiv.org/abs/2605.16233" }, { "arxiv_id": "2605.12493", "published": "2026-05-12", "title": "LongMemEval-V2: Evaluating Long-Term Agent Memory Toward Experienced Colleagues", "score": 8, "url": "https://arxiv.org/abs/2605.12493" }, { "arxiv_id": "2605.07242", "published": "2026-05-08", "title": "MEMOREPAIR: Barrier-First Cascade Repair in Agentic Memory", "score": 8, "url": "https://arxiv.org/abs/2605.07242" }, { "arxiv_id": "2605.08538", "published": "2026-05-08", "title": "Human-Inspired Memory Architecture for LLM Agents", "score": 8, "url": "https://arxiv.org/abs/2605.08538" }, { "arxiv_id": "2604.11563", "published": "2026-04-13", "title": "Synthius-Mem: Brain-Inspired Hallucination-Resistant Persona Memory Achieving 94.4% Memory Accuracy and 99.6% Adversarial Robustness on LoCoMo", "score": 8, "url": "https://arxiv.org/abs/2604.11563" }, { "arxiv_id": "2603.13017", "published": "2026-03-13", "title": "Structured Distillation for Personalized Agent Memory: 11x Token Reduction with Retrieval Preservation", "score": 8, "url": "https://arxiv.org/abs/2603.13017" }, { "arxiv_id": "2601.06282", "published": "2026-01-09", "title": "Amory: Building Coherent Narrative-Driven Agent Memory through Agentic Reasoning", "score": 8, "url": "https://arxiv.org/abs/2601.06282" }, { "arxiv_id": "2512.11303", "published": "2025-12-12", "title": "Unifying Dynamic Tool Creation and Cross-Task Experience Sharing through Cognitive Memory Architecture", "score": 8, "url": "https://arxiv.org/abs/2512.11303" }, { "arxiv_id": "2607.03726", "published": "2026-07-04", "title": "SelfMem: Self-Optimizing Memory for AI Agents", "score": 7, "url": "https://arxiv.org/abs/2607.03726" }, { "arxiv_id": "2607.01523", "published": "2026-07-01", "title": "Multi-Head Recurrent Memory Agents", "score": 7, "url": "https://arxiv.org/abs/2607.01523" }, { "arxiv_id": "2606.29914", "published": "2026-06-29", "title": "MemDelta: Controlled Baselines and Hidden Confounds in Agent Memory Evaluation", "score": 7, "url": "https://arxiv.org/abs/2606.29914" }, { "arxiv_id": "2606.24535", "published": "2026-06-23", "title": "Governed Shared Memory for Multi-Agent LLM Systems", "score": 7, "url": "https://arxiv.org/abs/2606.24535" }, { "arxiv_id": "2606.09900", "published": "2026-06-05", "title": "Less Context, More Accuracy: A Bi-Temporal Memory Engine for LLM Agents Where a Lean Retrieved Context Beats the Full History", "score": 7, "url": "https://arxiv.org/abs/2606.09900" }, { "arxiv_id": "2605.29630", "published": "2026-05-28", "title": "Entity-Collision: A Stratified Protocol for Attributing Retrieval Lift in Agent Memory", "score": 7, "url": "https://arxiv.org/abs/2605.29630" }, { "arxiv_id": "2605.17625", "published": "2026-05-17", "title": "Episodic-Semantic Memory Architecture for Long-Horizon Scientific Agents", "score": 7, "url": "https://arxiv.org/abs/2605.17625" } ] } }