mlsanityData to 7 Oct 2026

Self-evolving LLM agents

342 papers6 this weekHigh confidence in this grouping: the papers form one clear topicHigh confidence in this grouping: the papers form one clear topic · About

Keywords: agents, agent, self-evolving, memory, skill

Key facts. Self-evolving LLM agents: 342 papers on arXiv since 2022 (data through 7 Oct 2026). 29 of its 342 mapped papers (8%) cite “Darwin Godel Machine: Open-Ended Evolution of Self-Improving Agents” (2025, arXiv 2505.22954). 29 of its 342 mapped papers (8%) cite “Agentic Context Engineering: Evolving Contexts for Self-Improving Language Models” (2025, arXiv 2510.04618). 38 new papers 17 Sep–7 Oct.

What’s new

17 Sep – 7 Oct 2026

38 new papers, 17 Sep – 7 Oct; 6 this week.

  1. This week · 6
  2. The Harness as the Only Mutable Surface: Compliance-Bounded Self-Evolution of LLM Agents in Credit Pipelines, with a Measured Admission Gate
  3. DAEDALUS: Bootstrapping Agent Memory from Self-Generated Tasks
  4. Learning from Revision Consequences: Hindsight Meta-Experience Distillation for Self-Improving Agents
  5. EmbodiedSmith: Scaling Embodied Data through Recursive Self-Improvement Flywheel in Simulation
  6. SIGMA: Self-Improving Alignment Generalization from a Model Spec also fits: Self-rewarding alignment and self-training
  7. ServeLearnBench: How Well Can Agents Self-Improve from Serving Experience?
  8. Earlier · 32
  9. MESH-Harness: Self-Improving Agent Harnesses via Bandit-Guided Compositional Evolution
  10. ForkPilot: Self-Evolving Policy for Retrospective Search in Long-Horizon Agents
  11. SEIS: Self-Evolving Inference Systems
  12. MobiAgent: Dual-Loop Recursive Policy Self-Improvement for Long-Horizon Mobile Manipulation
All 38
  1. VERSE: Verified Self-Evolving Optimizer for Agent Harnesses
  2. Reconstruct, Practice, Go Real: Guided Self-Improvement for Embodied Agents
  3. My FAULT: Self-Diagnosis as Credit Assignment in Self-Evolving Agentic Reinforcement Learning
  4. GUI-HARVEST: Self-Improving GUI Agents through Evidence-Driven Harness Evolution
  5. Self-Evolving Coding Rules for AI Coding Agents
  6. RSIGame: Autonomous Agentic Game Development with Recursive Self-improvement
  7. Self-Evolving Algorithm-Design Agents: Escaping In-Context Evolutionary Stagnation via Population-Curated Policy Optimization
  8. SelfSearch: Reward-Free Search for Self-Improving Agents
  9. CollabFlow: Recursive Self-Improvement of Agent Collaboration
  10. EvoSteer: Online Self-Evolving Graph Orchestration via Reference-Anchored Credit Assignment
  11. Component-Aware Feedback for Self-Evolving Programs
  12. AREX-2: Advancing Self-Improving Agents through Long-Horizon Reflective Tasks
  13. Video-RSI: Recursive Self-Improvement of Video Understanding Agents via Harness Evolution
  14. AnyAct: Universal Action for Self-Evolving Agents
  15. Harness Evolution as Learning: Approximation, Generalization, and Optimization Limits of Self-Improving Personal Agents
  16. SEABench: Benchmarking Endogenous Misalignment In Self-Evolving Agents
  17. SAGE: A Statistical Acceptance Gate for Self-Evolving Agents
  18. AutoDataBench: Can Agents Write the Data That Feeds the Self-Improvement Loop?
  19. BEHAVE: Functional Behavior Modeling Enables Self-Improving Agents for Hardware Design and Verification
  20. Beyond Skill Evolution: Self-Evolving Context Management Policies for Long-Horizon Agent Harnesses
  21. GenMem: Generative Symbolic Memory for Self-Evolving Harness
  22. R2 Flow: Recursive Self-Improvement via Recursive Skill Evolution
  23. NLPG: Natural-Language Policy Gradients for Self-Evolving Language Agents
  24. Self-Evolving Time-Series Forecasting Agents with Episodic Memory and Online Policy Learning
  25. When Better Gets Worse: Improvement Fidelity for Self-Improving Agents in Adaptive Worlds
  26. Breaking the Environment Wall: Evolving LLM Agent Environments for Recursive Self-Improvement
  27. RRSI: Regularized Recursive Self-Improvement of Agent Harnesses
  28. From Semantic Decisions to Feasible Trajectories: Self-Evolving LLM-Guided Optimal Control for Narrow-Space Parking also fits: Self-play RL for reasoning

Foundations

Most cited by papers here

  1. Darwin Godel Machine: Open-Ended Evolution of Self-Improving Agents2025cited by 29 here
  2. Agentic Context Engineering: Evolving Contexts for Self-Improving Language Models2025cited by 29 here
  3. A Comprehensive Survey of Self-Evolving AI Agents: A New Paradigm Bridging Foundation Models and Lifelong Agentic Systems2025cited by 27 here
  4. ReasoningBank: Scaling Agent Self-Evolving with Reasoning Memory2025cited by 27 here
  5. A Survey of Self-Evolving Agents: What, When, How, and Where to Evolve on the Path to Artificial Super Intelligence2025cited by 25 here

Lineage

The papers this direction builds on, oldest first.

Each paper on the line cites the one before.

  1. 2022STaR: Bootstrapping Reasoning With Reasoningcited by 20 here
  2. 2023DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelinescited by 28 here
  3. 2024Symbolic Learning Enables Self-Evolving Agentscited by 11 here
  4. 2025Darwin Godel Machine: Open-Ended Evolution of Self-Improving Agentscited by 29 here
  5. 2025A Survey of Self-Evolving Agents: What, When, How, and Where to Evolve on the Path to Artificial Super Intelligencecited by 25 here
  6. 2025Evo-Memory: Benchmarking LLM Agent Test-time Learning with Self-Evolving Memorycited by 16 here
  7. 2026MemSkill: Learning and Evolving Memory Skills for Self-Evolving Agentscited by 14 here
17 more it builds on, not in this chain
  1. 2022Large Language Models Can Self-Improvecited by 13 here
  2. 2022Self-Instruct: Aligning Language Models with Self-Generated Instructionscited by 15 here
  3. 2023Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolutioncited by 33 here
  4. 2023Self-Taught Optimizer (STOP): Recursively Self-Improving Code Generationcited by 14 here
  5. 2024Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Modelscited by 13 here
  6. 2024Self-Rewarding Language Modelscited by 14 here
  7. 2024OS-Copilot: Towards Generalist Computer Agents with Self-Improvementcited by 12 here
  8. 2024Recursive Introspection: Teaching Language Model Agents How to Self-Improvecited by 10 here
  9. 2024Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameterscited by 10 here
  10. 2024WebRL: Training LLM Web Agents via Self-Evolving Online Curriculum Reinforcement Learningcited by 13 here
  11. 2025Absolute Zero: Reinforced Self-play Reasoning with Zero Datacited by 19 here
  12. 2025R-Zero: Self-Evolving Reasoning LLM from Zero Datacited by 16 here
  13. 2025A Comprehensive Survey of Self-Evolving AI Agents: A New Paradigm Bridging Foundation Models and Lifelong Agentic Systemscited by 27 here
  14. 2025ReasoningBank: Scaling Agent Self-Evolving with Reasoning Memorycited by 27 here
  15. 2025Agentic Context Engineering: Evolving Contexts for Self-Improving Language Modelscited by 29 here
  16. 2025EvolveR: Self-Evolving LLM Agents through an Experience-Driven Lifecyclecited by 16 here
  17. 2026CoEvoSkills: Self-Evolving Agent Skills via Co-Evolutionary Verificationcited by 12 here

The 24 most cited of 278.

Browse

30 of 342, ranked 6–35

  1. 6EvolveR: Self-Evolving LLM Agents through an Experience-Driven Lifecycle202516
  2. 7Evo-Memory: Benchmarking LLM Agent Test-time Learning with Self-Evolving Memory202516
  3. 8MemSkill: Learning and Evolving Memory Skills for Self-Evolving Agents202614
  4. 9WebRL: Training LLM Web Agents via Self-Evolving Online Curriculum Reinforcement Learning202413
  5. 10CoEvoSkills: Self-Evolving Agent Skills via Co-Evolutionary Verification202612
  6. 11OS-Copilot: Towards Generalist Computer Agents with Self-Improvement202412
  7. 12Symbolic Learning Enables Self-Evolving Agents202411
  8. 13SkillOpt: Executive Strategy for Self-Evolving Agent Skills202610
  9. 14A Self-Improving Coding Agent202510
  10. 15Self-Evolving Multi-Agent Collaboration Networks for Software Development202410
All 30
  1. 16EvoTest: Evolutionary Test-Time Learning for Self-Improving Agentic Systems20259
  2. 17Huxley-G"odel Machine: Human-Level Coding Agent Development by an Approximation of the Optimal Self-Improving Machine20259
  3. 18Reinforcement Learning for Self-Improving Agent with Skill Library20259
  4. 19Self-evolving Agents with reflective and memory-augmented abilities20249
  5. 20WebEvolver: Enhancing Web Agent Self-Improvement with Coevolving World Model20258
  6. 21SEAgent: Self-Evolving Computer Use Agent with Autonomous Learning from Experience20258
  7. 22Multi-Agent Evolve: LLM Self-Improve through Co-evolution20258
  8. 23AgentEvolver: Towards Efficient Self-Evolving Agent System20258
  9. 24Contextual Experience Replay for Self-Improvement of Language Agents20257
  10. 25MemGen: Weaving Generative Latent Memory for Self-Evolving Agents20257
  11. 26Live-SWE-agent: Can Software Engineering Agents Self-Evolve on the Fly?20257
  12. 27Mobile-Agent-E: Self-Evolving Mobile Assistant for Complex Tasks20256
  13. 28Your Agent May Misevolve: Emergent Risks in Self-evolving LLM Agents20256
  14. 29Richelieu: Self-Evolving LLM-Based Agents for AI Diplomacy20246
  15. 30SkillOS: Learning Skill Curation for Self-Evolving Agents20265
  16. 31Harness Updating Is Not Harness Benefit: Disentangling Evolution Capabilities in Self-Evolving LLM Agents20265
  17. 32SEW: Self-Evolving Agentic Workflows for Automated Code Generation20255
  18. 33Agent0: Unleashing Self-Evolving Agents from Zero Data via Tool-Integrated Reasoning20255
  19. 34Toward Training Superintelligent Software Agents through Self-Play SWE-RL20255
  20. 35Large Language Models Can Self-Improve At Web Agent Tasks20245