mlsanityData to 7 Oct 2026

MemSkill: Learning and Evolving Memory Skills for Self-Evolving Agents

arXiv 2602.02474 ↗First posted 2 Feb 2026cs.CL cs.AI cs.LGcited by 129

On the map: LLM agents and tool use → Agent memory systems

also in 1 more
Abstract (arXiv)

Most Large Language Model (LLM) agent memory systems rely on a small set of static, hand-designed operations for extracting memory. These fixed procedures hard-code human priors about what to store and how to revise memory, making them rigid under diverse interaction patterns and inefficient on long histories. To this end, we present MemSkill, which reframes these operations as learnable and evolvable memory skills, structured and reusable routines for extracting, consolidating, and pruning information from interaction traces. Inspired by the design philosophy of agent skills, MemSkill employs a controller that learns to select a small set of relevant skills, paired with an LLM-based executor that produces skill-guided memories. Beyond learning skill selection, MemSkill introduces a designer that periodically reviews hard cases where selected skills yield incorrect or incomplete memories, and evolves the skill set by proposing refinements and new skills. Together, MemSkill forms a closed-loop procedure that improves both the skill-selection policy and the skill set itself. Experiments on LoCoMo, LongMemEval, HotpotQA, and ALFWorld demonstrate that MemSkill improves task performance over strong baselines and generalizes well across settings. Further analyses shed light on how skills evolve, offering insights toward more adaptive, self-evolving memory management for LLM agents.

Key facts. Cited by 129 papers in the map (data through 7 Oct 2026). Placed in Agent memory systems, one of 16 directions in LLM agents and tool use. Builds on 33 of its 38 references that are in the map. First posted on arXiv 2 Feb 2026.

Builds on

33 of 38 references.

Closest ancestors · 11

Same direction, newest first

  1. 2025MemEvolve: Meta-Evolution of Agent Memory Systems21 Dec
  2. 2025Memory in the Age of AI Agents15 Dec
  3. 2025Evo-Memory: Benchmarking LLM Agent Test-time Learning with Self-Evolving Memory25 Nov
  4. 2025ReasoningBank: Scaling Agent Self-Evolving with Reasoning Memory29 Sep
  5. 2025MemGen: Weaving Generative Latent Memory for Self-Evolving Agents29 Sep
  6. 2025Memory-R1: Enhancing Large Language Model Agents to Manage and Utilize Memories via Reinforcement Learning27 Aug
  7. 2025Memory OS of AI Agent30 May
  8. 2025Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory28 Apr
  9. 2025A-MEM: Agentic Memory for LLM Agents17 Feb
  10. 2024Evaluating Very Long-Term Conversational Memory of LLM Agents27 Feb
  11. 2024A Human-Inspired Reading Agent with Gist Memory of Very Long Contexts15 Feb

Other references · 22

  1. 2017Proximal Policy Optimization Algorithmscited by 14,441
  2. 2024The Llama 3 Herd of Modelscited by 14,192
  3. 2025Qwen3 Technical Reportcited by 8,195
  4. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,686
  5. 2021Unsupervised Dense Information Retrieval with Contrastive Learningcited by 1,203
  6. 2020ALFWorld: Aligning Text and Embodied Environments for Interactive Learningcited by 1,143
  7. 2023ExpeL: LLM Agents Are Experiential Learnerscited by 814
  8. 2025AlphaEvolve: A coding agent for scientific and algorithmic discoverycited by 763
  9. 2024LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memorycited by 656
  10. 2024Automated Design of Agentic Systemscited by 360
  11. 2024AppWorld: A Controllable World of Apps and People for Benchmarking Interactive Coding Agentscited by 359
  12. 2025RAGEN: Understanding Self-Evolution in LLM Agents via Multi-Turn Reinforcement Learningcited by 357
10 more
  1. 2024Agent Workflow Memorycited by 342
  2. 2025Absolute Zero: Reinforced Self-play Reasoning with Zero Datacited by 323
  3. 2025MemAgent: Reshaping Long-Context LLM with Multi-Conv RL-based Memory Agentcited by 235
  4. 2025R-Zero: Self-Evolving Reasoning LLM from Zero Datacited by 210
  5. 2019Stochastic Beams and Where to Find Them: The Gumbel-Top-k Trick for Sampling Sequences Without Replacementcited by 205
  6. 2023Chain-of-Note: Enhancing Robustness in Retrieval-Augmented Language Modelscited by 180
  7. 2025SkillWeaver: Web Agents can Self-Improve by Discovering and Honing Skillscited by 174
  8. 2025EvolveR: Self-Evolving LLM Agents through an Experience-Driven Lifecyclecited by 129
  9. 2025AgentEvolver: Towards Efficient Self-Evolving Agent Systemcited by 77
  10. 2025Multi-Agent Evolve: LLM Self-Improve through Co-evolutioncited by 58

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 129.

129 papers.

Sort:

By area:

  1. 2026SkillClaw: Let Skills Evolve Collectively with Agentic Evolvercited by 76
  2. 2026SkillX: Automatically Constructing Skill Knowledge Bases for Agentscited by 69
  3. 2026Externalization in LLM Agents: A Unified Review of Memory, Skills, Protocols and Harness Engineeringcited by 68
  4. 2026SkillOS: Learning Skill Curation for Self-Evolving Agentscited by 45
  5. 2026SkillForge: Forging Domain-Specific, Self-Evolving Agent Skills in Cloud Technical Supportcited by 43
  6. 2026MetaClaw: Just Talk -- An Agent That Meta-Learns and Evolves in the Wildcited by 35
  7. 2026Skill1: Unified Evolution of Skill-Augmented Agents via Reinforcement Learningcited by 28
  8. 2026SkillRevise: Improving LLM-Authored Agent Skills via Trace-Conditioned Skill Revisioncited by 21
  9. 2026Anatomy of Agentic Memory: Taxonomy and Empirical Analysis of Evaluation and System Limitationscited by 20
  10. 2026SkillFlow:Benchmarking Lifelong Skill Discovery and Evolution for Autonomous Agentscited by 19
  11. 2026From Context to Skills: Can Language Models Learn from Context Skillfully?cited by 19
  12. 2026From Storage to Experience: A Survey on the Evolution of LLM Agent Memory Mechanismscited by 19
38 more
  1. 2026A Comprehensive Survey on Agent Skills: Taxonomy, Techniques, and Applicationscited by 17
  2. 2026LongMemEval-V2: Evaluating Long-Term Agent Memory Toward Experienced Colleaguescited by 17
  3. 2026Experience Compression Spectrum: Unifying Memory, Skills, and Rules in LLM Agentscited by 16
  4. 2026MemoryCD: Benchmarking Long-Context User Memory of LLM Agents for Lifelong Cross-Domain Personalizationcited by 15
  5. 2026Co-Evolving LLM Decision and Skill Bank Agents for Long-Horizon Taskscited by 13
  6. 2026Evaluating Memory Structure in LLM Agentscited by 12
  7. 2026Skill-R1: Agent Skill Evolution via Reinforcement Learningcited by 12
  8. 2026MetaSkill-Evolve: Recursive Self-Improvement of LLM Agents via Two-Timescale Meta-Skill Evolutioncited by 12
  9. 2026Evo-Harness: Context-to-Harness Skill Compilation for Self-Evolving Agentscited by 12
  10. 2026EvolveMem:Self-Evolving Memory Architecture via AutoResearch for LLM Agentscited by 11
  11. 2026SkillAudit: Ground-Truth-Free Skill Evolution via Paired Trajectory Auditingcited by 11
  12. 2026Evolving Medical Imaging Agents via Experience-driven Self-skill Discoverycited by 10
  13. 2026M^star: Every Task Deserves Its Own Memory Harnesscited by 9
  14. 2026From Procedural Skills to Strategy Genes: Towards Experience-Driven Test-Time Evolutioncited by 9
  15. 2026Belief Memory: Agent Memory Under Partial Observabilitycited by 9
  16. 2026Group of Skills: Group-Structured Skill Retrieval for Agent Skill Librariescited by 9
  17. 2026MementoGUI: Learning Agentic Multimodal Memory Control for Long-Horizon GUI Agentscited by 9
  18. 2026Combee: Scaling Prompt Learning for Self-Improving Language Model Agentscited by 8
  19. 2026SkillGraph: Self-Evolving Multi-Agent Collaboration with Multimodal Graph Topologycited by 8
  20. 2026Generate, Filter, Control, Replay: A Comprehensive Survey of Rollout Strategies for LLM Reinforcement Learningcited by 8
  21. 2026SkillMaster: Toward Autonomous Skill Mastery in LLM Agentscited by 8
  22. 2026AutoPyVerifier: Learning Compact Executable Verifiers for Large Language Model Outputscited by 7
  23. 2026When Continual Learning Moves to Memory: A Study of Experience Reuse in LLM Agentscited by 7
  24. 2026Harnessing LLM Agents with Skill Programscited by 7
  25. 2026SkillCoach: Self-Evolving Rubrics for Evaluating and Enhancing Agentic Skill-Usecited by 7
  26. 2026SKILLC: Learning Autonomous Skill Internalization in LLM Agents via Contrastive Credit Assignmentcited by 6
  27. 2026FederatedSkill: Federated Learning for Agentic Skill Evolutioncited by 6
  28. 2026Agent Skill Evaluation and Evolution: Frameworks and Benchmarkscited by 6
  29. 2026Dynamic Agent Skills: A Lifecycle Survey and Taxonomy of Evolving Skill Librariescited by 6
  30. 2026SmartSearch: How Ranking Beats Structure for Conversational Memory Retrievalcited by 5
  31. 2026VideoWeaver: Evaluating and Evolving Skills for Agentic Long Video Generationcited by 5
  32. 2026SkillCAT: Contrastive, Assessment-Augmented and Topology-AwareSkill Self-Evolution for LLM Agentscited by 5
  33. 2026Memory as a Controlled Process: Learned Adaptive Memory Management for LLM Agentscited by 5
  34. 2026RSIAgent: Autonomous Exploration for Recursive Self-improvement in New Environmentscited by 5
  35. 2026Ask Only When Needed: Proactive Retrieval from Memory and Skills for Experience-Driven Lifelong Agentscited by 4
  36. 2026MemQ: Integrating Q-Learning into Self-Evolving Memory Agents over Provenance DAGscited by 4
  37. 2026HAGE: Harnessing Agentic Memory via RL-Driven Weighted Graph Evolutioncited by 4
  38. 2026Role-Agent: Bootstrapping LLM Agents via Dual-Role Evolutioncited by 4

50 most cited of 129 All 129 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026Memento-Skills: Let Agents Design Agents
  2. 2026Compiled Memory: Not More Information, but More Precise Instructions for Language Agents
  3. 2025Memp: Exploring Agent Procedural Memory
  4. 2026AdMem: Advanced Memory for Task-solving Agents
  5. 2026MemFactory: Unified Inference & Training Framework for Agent Memory
  6. 2026MemGym: a Long-Horizon Memory Environment for LLM Agents
  7. 2026Mem-π: Adaptive Memory through Learning When and What to Generate
  8. 2026MemReader: From Passive to Active Extraction for Long-Term Agent Memory