# Self-evolving LLM agents

Area: [Synthetic data and self-improvement for LLMs](https://mlsanity.com/synth/index.md). HTML: https://mlsanity.com/synth/self-evolving-llm-agents/

## Key facts

- Self-evolving LLM agents: 342 papers on arXiv since 2022 (data through 14 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).
- 63 new papers 17 Sep–14 Oct.

## Most cited here

- [Darwin Godel Machine: Open-Ended Evolution of Self-Improving Agents](https://mlsanity.com/paper/2505.22954/) (2025), cited by 29 papers here
- [Agentic Context Engineering: Evolving Contexts for Self-Improving Language Models](https://mlsanity.com/paper/2510.04618/) (2025), cited by 29 papers here
- [A Comprehensive Survey of Self-Evolving AI Agents: A New Paradigm Bridging Foundation Models and Lifelong Agentic Systems](https://mlsanity.com/paper/2508.07407/) (2025), cited by 27 papers here
- [ReasoningBank: Scaling Agent Self-Evolving with Reasoning Memory](https://mlsanity.com/paper/2509.25140/) (2025), cited by 27 papers here
- [A Survey of Self-Evolving Agents: What, When, How, and Where to Evolve on the Path to Artificial Super Intelligence](https://mlsanity.com/paper/2507.21046/) (2025), cited by 25 papers here
- [EvolveR: Self-Evolving LLM Agents through an Experience-Driven Lifecycle](https://mlsanity.com/paper/2510.16079/) (2025), cited by 16 papers here
- [Evo-Memory: Benchmarking LLM Agent Test-time Learning with Self-Evolving Memory](https://mlsanity.com/paper/2511.20857/) (2025), cited by 16 papers here
- [MemSkill: Learning and Evolving Memory Skills for Self-Evolving Agents](https://mlsanity.com/paper/2602.02474/) (2026), cited by 14 papers here
- [WebRL: Training LLM Web Agents via Self-Evolving Online Curriculum Reinforcement Learning](https://mlsanity.com/paper/2411.02337/) (2024), cited by 13 papers here
- [OS-Copilot: Towards Generalist Computer Agents with Self-Improvement](https://mlsanity.com/paper/2402.07456/) (2024), cited by 12 papers here

## New (17 Sep–14 Oct)

- [Evolving Together, Failing Apart: The Skill-Policy Co-evolution Trap in Self-Improving Agents](https://mlsanity.com/paper/2610.20845/) (2026-10-14)
- [Reflect After Failure: Brain-Inspired Self-Evolving Agents for Embodied Spatial Intelligence](https://mlsanity.com/paper/2610.20312/) (2026-10-14)
- [Evolve Together, Fall into Place: How Social Scaffolding Shapes Self-Evolving Multi-Agent Systems](https://mlsanity.com/paper/2610.19799/) (2026-10-14)
- [TimeRSI: Towards Self-Improved Agentic Deployment for Time-Series Foundation Models](https://mlsanity.com/paper/2610.19725/) (2026-10-14)
- [L3M: A Verifiable Harness and Runtime for Self-Extending and Self-Improving Agents](https://mlsanity.com/paper/2610.19591/) (2026-10-14)
- [XferAgent: Cross-Agent Experience Transfer for Self-Improving LLM Agents](https://mlsanity.com/paper/2610.19581/) (2026-10-14)
- [VideoHarness-SRSI: Structured Recursive Self-Improvement for Video Understanding Harness](https://mlsanity.com/paper/2610.19424/) (2026-10-14)
- [STRATA: Self-Maintaining Structured Experience Graphs for Self-Evolving Language Agents](https://mlsanity.com/paper/2610.19386/) (2026-10-14)
- [SEAR: Self-Evolving Agentic Robotics](https://mlsanity.com/paper/2610.18792/) (2026-10-13)
- [AReaL-RSI: Adaptive Scaling and Coordination of Recursive Self-Improvement](https://mlsanity.com/paper/2610.18083/) (2026-10-13)
- [D-World: An Explicit World-Language-Action Model with Recursive Self-Improvement under Dual Control](https://mlsanity.com/paper/2610.17114/) (2026-10-12)
- [Heterogeneous Agentic Coevolution](https://mlsanity.com/paper/2610.16078/) (2026-10-12)
- [Small but Mighty: Test Suites as Improvement Signals for Self-Evolving Agents](https://mlsanity.com/paper/2610.15964/) (2026-10-11)
- [SpecForge: Recursive Self-Improvement for Coding Agents via Behavioral Specifications](https://mlsanity.com/paper/2610.15708/) (2026-10-11)
- [EvoCowork: A Benchmark for Real-World Office Tasks](https://mlsanity.com/paper/2610.14828/) (2026-10-10)
- [ClawLoop: Self-Improving Agents via Harness-Native Training Loop](https://mlsanity.com/paper/2610.14118/) (2026-10-09)
- [VETR: Execution-Grounded Self-Improving Agent for Symbolic Graphics Reasoning](https://mlsanity.com/paper/2610.13098/) (2026-10-09)
- [EvoBelief: Falsifiable World Models for Self-Improving LLM Agents](https://mlsanity.com/paper/2610.12789/) (2026-10-08)
- [Beyond Binary Verifiable Rewards: Behavioral Self-Rewarding for SWE Agents](https://mlsanity.com/paper/2610.12598/) (2026-10-08)
- [Know What You Don't Know: Agentic Data Synthesis by Evolving Generation Space for Closed-Loop Learning](https://mlsanity.com/paper/2610.12593/) (2026-10-08)
- [Embodied Turing Machines: Stateful Code for Robot Recursive Self-Improvement](https://mlsanity.com/paper/2610.12369/) (2026-10-08)
- [EvoAlloc: A Self-Evolving Resource Allocation Agent for Efficient Program Evolution](https://mlsanity.com/paper/2610.12086/) (2026-10-08)
- [Memento 3: Model-Based Recursive Self-Improvement through Reflective Rulebooks](https://mlsanity.com/paper/2610.11794/) (2026-10-08)
- [Who Verifies the Verifier? Co-Evolving Inspectable Graders with Self-Improving Agents](https://mlsanity.com/paper/2610.11464/) (2026-10-08)
- [SynCo: Data Synthesis Co-Training for Self-Evolving LLMs via Multi-Agent Reinforcement Learning](https://mlsanity.com/paper/2610.11345/) (2026-10-08)
- [The Harness as the Only Mutable Surface: Compliance-Bounded Self-Evolution of LLM Agents in Credit Pipelines, with a Measured Admission Gate](https://mlsanity.com/paper/2610.10629/) (2026-10-07)
- [DAEDALUS: Bootstrapping Agent Memory from Self-Generated Tasks](https://mlsanity.com/paper/2610.08048/) (2026-10-06)
- [Learning from Revision Consequences: Hindsight Meta-Experience Distillation for Self-Improving Agents](https://mlsanity.com/paper/2610.07979/) (2026-10-06)
- [EmbodiedSmith: Scaling Embodied Data through Recursive Self-Improvement Flywheel in Simulation](https://mlsanity.com/paper/2610.07969/) (2026-10-06)
- [SIGMA: Self-Improving Alignment Generalization from a Model Spec](https://mlsanity.com/paper/2610.07935/) (2026-10-06)