mlsanityData to 7 Oct 2026

Agent memory systems

718 papers19 this weekMedium confidence in this grouping: mostly one topic, with blurry edgesMedium confidence in this grouping: mostly one topic, with blurry edges · About

Keywords: memory, agents, retrieval, context, llm

Key facts. Agent memory systems: 718 papers on arXiv since 2023 (data through 7 Oct 2026). 300 of its 718 mapped papers (42%) cite “A-MEM: Agentic Memory for LLM Agents” (2025, arXiv 2502.12110). 276 of its 718 mapped papers (38%) cite “Evaluating Very Long-Term Conversational Memory of LLM Agents” (2024, arXiv 2402.17753). 90 new papers 17 Sep–7 Oct.

What’s new

17 Sep – 7 Oct 2026

90 new papers, 17 Sep – 7 Oct; 19 this week.

  1. This week · 19
  2. Curating Always-Loaded Context for LLM Agents: A Capacitated Assortment Model with Censored Feedback also fits: Agent skill learning and evolution
  3. VideoEvolve: Co-Evolving Memory and Retrieval for Long Video Understanding
  4. HGP:An on-device personalized agent memory via hybrid graph storage
  5. Learning Situation-Conditioned Thinking Policies for Long-Term LLM Agents
  6. Relevance Is Not Sufficiency: What Actually Closes the Evidence Gap in Long-Term Memory QA
  7. Whose Memory Is It? Scope-Aware Commit Rules for Long-Term LLM Memory
  8. Towards In-Parameter Memory Augmentation for Large Language Models
  9. MINDSET: Energy-based Schema Evolution for Long Conversational Agent Memory
  10. How Much Evidence Should a Coding Agent's Self-Correction Carry? Adaptive Dirichlet Evidence for Self-Distillation also fits: Software engineering coding agents
  11. DSV-Mem: Evaluating Multimodal Memory in Professional Workflows for MLLM Agents also fits: Multimodal and embodied agents
All 90
  1. ReFold: Training-Free Reversible Inter-Turn Context Folding for Long-Horizon Agents also fits: Multimodal and embodied agents
  2. Persistent Memory in Multi-Agent LLM Inference: What It Costs, What It Buys, and When You Can Tell
  3. MemCo: Memory-Centric Collaboration for Generalizing LLM Agents to Unseen Environments
  4. Understanding and Mitigating Inference-Time Overreliance Using Agentic Memory
  5. AMBER: Training Long-Horizon Web Agents through Append-Only Memory
  6. MemPilot: Orchestrating On-Demand Multimodal Memory Curation for LLM Agents
  7. VideoTapestry: Query-Adaptive Memory Refinement for Multi-Agent Long-Video Understanding
  8. Request Order Matters: Cache-History Sensitivity in Selective KV-Cache Reuse for Rolling Agents
  9. PACMI: Provenance-Aware Cascading Memory Invalidation for Long-Term LLM Agents
  10. Earlier · 71
  11. CASE: Cost-Aware Stopping for Efficient Long-Video Agents
  12. Look Before You Leap: Thermodynamic Arbitration of Parametric and Non-Parametric Knowledge in LLM Agents via Self-Regulating Memory Architectures
  13. AECG: Asymmetric Experience Consolidation and Governance In Multi-Agent Systems also fits: Reinforcement learning for LLM agents
  14. Memadapter: Counterfactual Adaptation Against Memory-induced Sycophancy
  15. Memory Canonicalization: A Framework and Benchmark for Cross-Model Drift in Persistent LLM Memory
  16. MemTrace: State-Consistent Memory for Long-Horizon Coding Agents
  17. Knowing the Store: What a Memory Backend Must Write Down Before an Agent Can Read It
  18. When Evidence Changes: Evaluating Memory Repair and Re-reading in Language-Model Agents also fits: Medical and scientific domain agents
  19. DyadMem: A Long-Term Memory Benchmark of How Agents Work with Users
  20. Continual Graph Memory for Mathematical Research Agents also fits: Educational agents and math problem solving
  21. Harnessing LLMs as Agents: What Does It Cost? also fits: General LLM agent frameworks and benchmarks
  22. APDMem: Agent-Controlled Progressive Disclosure for Query-Adaptive Long-Term Memory
  23. Trained Agentic Context Management also fits: General LLM agent frameworks and benchmarks
  24. SEDIMA: Cross-Run Hierarchical Insight Memory for Evolutionary Search Agents also fits: Web, search, and research agents
  25. Mem++: Non-Destructive Memory for Long-Term Organizational LLM Agents
  26. Beyond Memory: Harnessing Long-Horizon Agents with Explicit Belief States
  27. MemFit: Efficient Long-Term Agentic Memory
  28. TAGGRAPH: Tag-Augmented Graphs for Graph Retrieval of Agent Persistent Histories
  29. Persistent Context Graphs for Efficient Memory Compaction in LLM Agents
  30. SparseEngine: Sparse-First Inference Engine
  31. When Context Changes: Understanding Update Failures in LLMs
  32. Action Conditioned Bisimulation For GUI Agent Memory also fits: GUI and mobile agents
  33. Coding Agent Memory Post-training: Unlocking the Memory Potential of Pre-trained File Operations for Long-Horizon Tasks via Reinforcement Learning
  34. When Correct Memory Goes Wrong: Fuzzing Persistent Memory Use in LLM Agents
  35. FOCUS: Training-Free Decision-Preserving Context Compression for LLM Agents
  36. LatCom: Cross-Agent Latent Compression for Efficient Multi-Agent Collaboration also fits: General LLM agent frameworks and benchmarks
  37. UpliftMem: Learning Set-Level Uplift for Agent Memory Retrieval
  38. Simple Agentic Memory for Generalist Robot Policies
  39. Decision-Sufficient State Representations: Measuring and Reducing Write-Time Regret also fits: Reinforcement learning for LLM agents
  40. Memory Is a Derivation: The Distributed-Evidence Paradox in Long-Term Agents
  41. Sprout: Building Dynamic Memory While Reasoning for Agentic Video Understanding
  42. EP-Mem: Elastic Privacy Memory for Social Relationship-Aware LLM Agents
  43. GenMem: Generative Symbolic Memory for Self-Evolving Harness
  44. FlowState: Execution State as Memory for Long-Horizon LLM Agents
  45. Remember Before You're Asked: MemDream for Self-Probing Memory Evolution
  46. Remember by Asking: Retrieval-Induced Memory Evolution for LLM Agents
  47. Just-In-Time Agent Memory with Runtime Agentic Research
  48. PersMem: Internalizing Personality into Dual-Pathway Memory for LLM Agents also fits: Social simulation and multi-agent debate
  49. When Does Selection Replace Extraction? A Pre-Registered Test of Agent Memory with a Typed Decision Model
  50. From Attack Success to Attack Severity: Counterfactual Memory Attacks on LLM Agents also fits: Agent threats, guardrails and governance
  51. Self-Designed Evaluators and Warm Memory for Long-Horizon Agents also fits: Software engineering coding agents
  52. ActiveMem: Dynamic Latent Memory Trees for Long-Horizon Agents
  53. Beyond Memory Construction: Rethinking Memory Access for LLM-based Conversational Agents
  54. The Epistemics of Agent Memory: Measuring, and Governing, the Consolidation Decision in Long-Horizon LLM Agents
  55. ARSM: Auto-Regressive State Machine for Agentic Reasoning Compression
  56. Contract Memory Compiler: Resolve, Then Traverse
  57. Beyond Dyadic Memory: Interaction-Aware Multimodal Memory with Adaptive Agentic Retrieval for Multi-Party Spoken Conversations
  58. MemAgent: Learning to Manage Heterogeneous Memory Providers for LLM Agents
  59. Learning from Others, Acting for You: Cross-User Memory Sharing for LLM Agents
  60. Shared Worlds, Private Minds: Structured Memory for Long-Form Writing as World Creation
  61. Black-Box Auditing of Epistemic Reliability in Multi-Agent Debate Distillation also fits: Social simulation and multi-agent debate
  62. Agentsensus: Consensus-Compressed Shared Memory for Multi-Agent Story Worlds
  63. LAM: Efficient Lossy Agent Memory Framework With A Retrieval-Score Error Bound
  64. EngramRAG: Dynamic Usage-Weighted Topology and Synaptic Consolidation for Multi-Hop Agentic Memory
  65. COUNTERMEM: World-Model Verified Counter-Factual Memory for Language Agents
  66. ActKV: Efficient LLM Agents through Action-Guided KV Cache Management
  67. HasMem: Hard-Origin Adaptively Softened Memory for Long-Term LLM Agents
  68. When Can Agents Forget Their Reasoning? ICLR for Long-Horizon Agent Context Compression
  69. Just-in-Time Memory: Learning to Curate Task-Adaptive Memory for LLM Agents
  70. StateComp: Learning When to Compress History in Long Horizon Agents
  71. Memory Control Signals Emerge Before Action in Long Horizon Agents
  72. DRSR: Learning Set-Level Deletion Risk for Efficient Long-Horizon Agents
  73. MemCalib: Benchmarking and Optimizing Memory Use in LLM Agents
  74. Jev-Mem: System-One-Controlled Agentic Memory for Efficient AI Agents
  75. When Successful Memories Mislead Embodied Agents:Memory Adaption For Task-Conditioned Execution
  76. Impact Is Not Invalidation: Ask About the Claim, Not the Diff also fits: Software engineering coding agents
  77. RPMem: Learning Long-Term Recurrent Parametric Memory Across Sessions for LLM Agents
  78. PSD: Pseudo Self-Distillation of Memory Representation Capabilities for LLM Agents
  79. An Interpretable Memory Decision Controller for LLM Agents Based on Three-Signal Complementarity: Decoupling Confidence and Consistency
  80. MACE: Memory-Agent Co-Evolution with Adaptive Memory Graphs for Multi-Agent Systems
  81. On-Demand Attention: Language Models Know When to Recall

Foundations

Most cited by papers here

  1. A-MEM: Agentic Memory for LLM Agents2025cited by 300 here
  2. Evaluating Very Long-Term Conversational Memory of LLM Agents2024cited by 276 here
  3. Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory2025cited by 210 here
  4. A Survey on the Memory Mechanism of Large Language Model based Agents2024cited by 151 here
  5. Zep: A Temporal Knowledge Graph Architecture for Agent Memory2025cited by 150 here

Lineage

The papers this direction builds on, oldest first.

Each paper on the line cites the one before.

  1. 2023Generative Agents: Interactive Simulacra of Human Behaviorcited by 214 here
  2. 2023Voyager: An Open-Ended Embodied Agent with Large Language Modelscited by 92 here
  3. 2023WebArena: A Realistic Web Environment for Building Autonomous Agentscited by 42 here
  4. 2023MetaGPT: Meta Programming for A Multi-Agent Collaborative Frameworkcited by 42 here
  5. 2023AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversationcited by 46 here
  6. 2023A Survey on Large Language Model based Autonomous Agentscited by 71 here
  7. 2023Cognitive Architectures for Language Agentscited by 42 here
  8. 2024SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineeringcited by 49 here
16 more it builds on, not in this chain
  1. 2023Toolformer: Language Models Can Teach Themselves to Use Toolscited by 76 here
  2. 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 163 here
  3. 2023ExpeL: LLM Agents Are Experiential Learnerscited by 81 here
  4. 2024A Human-Inspired Reading Agent with Gist Memory of Very Long Contextscited by 60 here
  5. 2024Evaluating Very Long-Term Conversational Memory of LLM Agentscited by 276 here
  6. 2024A Survey on the Memory Mechanism of Large Language Model based Agentscited by 151 here
  7. 2025Zep: A Temporal Knowledge Graph Architecture for Agent Memorycited by 150 here
  8. 2025A-MEM: Agentic Memory for LLM Agentscited by 300 here
  9. 2025Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memorycited by 210 here
  10. 2025Memory OS of AI Agentcited by 56 here
  11. 2025MEM1: Learning to Synergize Memory and Reasoning for Efficient Long-Horizon Agentscited by 66 here
  12. 2025Evaluating Memory in LLM Agents via Incremental Multi-Turn Interactionscited by 85 here
  13. 2025MIRIX: Multi-Agent Memory System for LLM-Based Agentscited by 67 here
  14. 2025Memory-R1: Enhancing Large Language Model Agents to Manage and Utilize Memories via Reinforcement Learningcited by 73 here
  15. 2025Memory in the Age of AI Agentscited by 95 here
  16. 2026SimpleMem: Efficient Lifelong Memory for LLM Agentscited by 48 here

The 24 most cited of 1,604.

Browse

30 of 718, ranked 6–35

  1. 6Memory in the Age of AI Agents202595
  2. 7Evaluating Memory in LLM Agents via Incremental Multi-Turn Interactions202585
  3. 8Memory-R1: Enhancing Large Language Model Agents to Manage and Utilize Memories via Reinforcement Learning202573
  4. 9MIRIX: Multi-Agent Memory System for LLM-Based Agents202567
  5. 10MEM1: Learning to Synergize Memory and Reasoning for Efficient Long-Horizon Agents202566
  6. 11A Human-Inspired Reading Agent with Gist Memory of Very Long Contexts202460
  7. 12Memory OS of AI Agent202556
  8. 13SimpleMem: Efficient Lifelong Memory for LLM Agents202648
  9. 14In Prospect and Retrospect: Reflective Memory Management for Long-term Personalized Dialogue Agents202541
  10. 15Hello Again! LLM-powered Personalized Agent for Long-term Dialogue202439
All 30
  1. 16HiAgent: Hierarchical Working Memory Management for Solving Long-Horizon Agent Tasks with Large Language Model202438
  2. 17G-Memory: Tracing Hierarchical Memory for Multi-Agent Systems202537
  3. 18ReasoningBank: Scaling Agent Self-Evolving with Reasoning Memory202537
  4. 19How Memory Management Impacts LLM Agents: An Empirical Study of Experience-Following Behavior202532
  5. 20MemGen: Weaving Generative Latent Memory for Self-Evolving Agents202532
  6. 21MemoryArena: Benchmarking Agent Memory in Interdependent Multi-Session Agentic Tasks202630
  7. 22Evo-Memory: Benchmarking LLM Agent Test-time Learning with Self-Evolving Memory202530
  8. 23MemBench: Towards More Comprehensive Evaluation on the Memory of LLM-based Agents202529
  9. 24ReSum: Unlocking Long-Horizon Search Intelligence via Context Summarization202529
  10. 25AriGraph: Learning Knowledge Graph World Models with Episodic Memory for LLM Agents202429
  11. 26ACON: Optimizing Context Compression for Long-horizon LLM Agents202528
  12. 27MemInsight: Autonomous Memory Augmentation for LLM Agents202527
  13. 28Agentic Memory: Learning Unified Long-Term and Short-Term Memory Management for Large Language Model Agents202626
  14. 29Scaling Long-Horizon LLM Agent via Context-Folding202525
  15. 30Memory as Action: Autonomous Context Curation for Long-Horizon Agentic Tasks202525
  16. 31General Agentic Memory Via Deep Research202525
  17. 32PersonaMem-v2: Towards Personalized Intelligence via Learning Implicit User Personas and Agentic Memory202524
  18. 33Memory for Autonomous LLM Agents:Mechanisms, Evaluation, and Emerging Frontiers202623
  19. 34Hierarchical Memory for High-Efficiency Long-Term Reasoning in LLM Agents202522
  20. 35What Deserves Memory: Adaptive Memory Distillation for LLM Agents202522