mlsanityData to 7 Oct 2026

Darwin Godel Machine: Open-Ended Evolution of Self-Improving Agents

arXiv 2505.22954 ↗First posted 29 May 2025cs.AIcited by 294

On the map: LLM agents and tool use → General LLM agent frameworks and benchmarks

also in 1 more
Abstract (arXiv)

Today's AI systems have human-designed, fixed architectures and cannot autonomously and continuously improve themselves. The advance of AI could itself be automated. If done safely, that would accelerate AI development and allow us to reap its benefits much sooner. Meta-learning can automate the discovery of novel algorithms, but is limited by first-order improvements and the human design of a suitable search space. The G"odel machine proposed a theoretical alternative: a self-improving AI that repeatedly modifies itself in a provably beneficial manner. Unfortunately, proving that most changes are net beneficial is impossible in practice. We introduce the Darwin G"odel Machine (DGM), a self-improving system that iteratively modifies its own code (thereby also improving its ability to modify its own codebase) and empirically validates each change using coding benchmarks. Inspired by Darwinian evolution and open-endedness research, the DGM maintains an archive of generated coding agents. It grows the archive by sampling an agent from it and using a foundation model to create a new, interesting, version of the sampled agent. This open-ended exploration forms a growing tree of diverse, high-quality agents and allows the parallel exploration of many different paths through the search space. Empirically, the DGM automatically improves its coding capabilities (e.g., better code editing tools, long-context window management, peer-review mechanisms), increasing performance on SWE-bench from 20.0% to 50.0%, and on Polyglot from 14.2% to 30.7%. Furthermore, the DGM significantly outperforms baselines without self-improvement or open-ended exploration. All experiments were done with safety precautions (e.g., sandboxing, human oversight). The DGM is a significant step toward self-improving AI, capable of gathering its own stepping stones along paths that unfold into endless innovation.

Key facts. Cited by 294 papers in the map (data through 7 Oct 2026). Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use. Builds on 60 of its 184 references that are in the map. First posted on arXiv 29 May 2025.

Builds on

60 of 184 references.

Closest ancestors · 6

Same direction, newest first

  1. 2025MAS-GPT: Training LLMs to Build LLM-based Multi-Agent Systems5 Mar
  2. 2025MAGELLAN: Metacognitive predictions of learning progress guide autotelic LLM agents in large goal spaces11 Feb
  3. 2025Multi-agent Architecture Search via Agentic Supernet6 Feb
  4. 2024G"odel Agent: A Self-Referential Agent Framework for Recursive Self-Improvement6 Oct
  5. 2024Symbolic Learning Enables Self-Evolving Agents26 Jun
  6. 2024EvoAgent: Towards Automatic Multi-Agent Generation via Evolutionary Algorithms20 Jun

Other references · 54

  1. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,146
  2. 2022Constitutional AI: Harmlessness from AI Feedbackcited by 2,950
  3. 2023Voyager: An Open-Ended Embodied Agent with Large Language Modelscited by 2,081
  4. 2024SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineeringcited by 1,628
  5. 2025s1: Simple test-time scalingcited by 1,311
  6. 2024OpenAI o1 System Cardcited by 1,220
  7. 2024The AI Scientist: Towards Fully Automated Open-Ended Scientific Discoverycited by 1,002
  8. 2023Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debatecited by 995
  9. 2023DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelinescited by 803
  10. 2018Diversity is All You Need: Learning Skills without a Reward Functioncited by 734
  11. 2017Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithmcited by 723
  12. 2022Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learnedcited by 702
38 more
  1. 2022Large Language Models Can Self-Improvecited by 656
  2. 2023Eureka: Human-Level Reward Design via Coding Large Language Modelscited by 542
  3. 2015Illuminating search spaces by mapping elitescited by 406
  4. 2024Agentless: Demystifying LLM-based Software Engineering Agentscited by 400
  5. 2024AFlow: Automating Agentic Workflow Generationcited by 395
  6. 2017Population Based Training of Neural Networkscited by 351
  7. 2023Beyond Human Data: Scaling Self-Training for Problem-Solving with Language Modelscited by 325
  8. 2024Alignment faking in large language modelscited by 316
  9. 2024Quiet-STaR: Language Models Can Teach Themselves to Think Before Speakingcited by 292
  10. 2024Evolution of Heuristics: Towards Efficient Automatic Algorithm Design Using Large Language Modelcited by 258
  11. 2024Debating with More Persuasive LLMs Leads to More Truthful Answerscited by 255
  12. 2024TextGrad: Automatic "Differentiation" via Textcited by 250
  13. 2019Go-Explore: a New Approach for Hard-Exploration Problemscited by 207
  14. 2023RoboGen: Towards Unleashing Infinite Data for Automated Robot Learning via Generative Simulationcited by 206
  15. 2024Foundational Challenges in Assuring Alignment and Safety of Large Language Modelscited by 193
  16. 2019Paired Open-Ended Trailblazer (POET): Endlessly Generating Increasingly Complex and Diverse Learning Environments and Their Solutionscited by 182
  17. 2024The Prompt Report: A Systematic Survey of Prompt Engineering Techniquescited by 115
  18. 2023Mindstorms in Natural Language-Based Societies of Mindcited by 105
  19. 2024GLoRe: When, Where, and How to Improve LLM Reasoning via Global and Local Refinementscited by 101
  20. 2023Motif: Intrinsic Motivation from Artificial Intelligence Feedbackcited by 99
  21. 2019AI-GAs: AI-generating algorithms, an alternate paradigm for producing general artificial intelligencecited by 84
  22. 2024Open-Endedness is Essential for Artificial Superhuman Intelligencecited by 79
  23. 2024Self-Evolving Multi-Agent Collaboration Networks for Software Developmentcited by 73
  24. 2023Unleashing the potential of prompt engineering for large language modelscited by 72
  25. 2023RET-LLM: Towards a General Read-Write Memory for Large Language Modelscited by 64
  26. 2023Think-in-Memory: Recalling and Post-thinking Enable LLMs with Long-Term Memorycited by 64
  27. 2025A Self-Improving Coding Agentcited by 58
  28. 2025FlowReasoner: Reinforcing Query-Level Meta-Agentscited by 56
  29. 2025AutoRedTeamer: Autonomous Red Teaming with Lifelong Attack Integrationcited by 38
  30. 2021Multi-task curriculum learning in a complex, visual, hard-exploration domain: Minecraftcited by 34
  31. 2024Intelligent Go-Explore: Standing on the Shoulders of Giant Foundation Modelscited by 32
  32. 2024Surveying the Effects of Quality, Diversity, and Complexity in Synthetic Data From Large Language Modelscited by 31
  33. 2024MaestroMotif: Skill Design from Artificial Intelligence Feedbackcited by 24
  34. 2025Weak-for-Strong: Training Weak Meta-Agent to Harness Strong Executorscited by 20
  35. 2024Word2World: Generating Stories and Worlds through Large Language Modelscited by 18
  36. 2025GNNs as Predictors of Agentic Workflow Performancescited by 17
  37. 2021Training Learned Optimizers with Randomly Initialized Learned Optimizerscited by 13
  38. 2025DebFlow: Automating Agent Creation via Agent Debatecited by 10

50 most cited of 54 All 54 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 46; 2026: 248.

294 papers.

Sort:

By area:

  1. 2025Memory in the Age of AI Agentscited by 285
  2. 2025The Landscape of Agentic Reinforcement Learning for LLMs: A Surveycited by 195
  3. 2025A Survey of Self-Evolving Agents: What, When, How, and Where to Evolve on the Path to Artificial Super Intelligencecited by 155
  4. 2026Towards End-to-End Automation of AI Researchcited by 150
  5. 2025Live-SWE-agent: Can Software Engineering Agents Self-Evolve on the Fly?cited by 91
  6. 2025MemEvolve: Meta-Evolution of Agent Memory Systemscited by 81
  7. 2026Self-Harness: Harnesses That Improve Themselvescited by 79
  8. 2025SWE-Exp: Experience-Driven Software Issue Resolutioncited by 64
  9. 2025Your Agent May Misevolve: Emergent Risks in Self-evolving LLM Agentscited by 52
  10. 2025SWE-EVO: Benchmarking Coding Agents in Long-Horizon Software Evolution Scenarioscited by 49
  11. 2026PostTrainBench: Can LLM Agents Automate LLM Post-Training?cited by 45
  12. 2025Huxley-G"odel Machine: Human-Level Coding Agent Development by an Approximation of the Optimal Self-Improving Machinecited by 44
38 more
  1. 2025EvoAgentX: An Automated Framework for Evolving Agentic Workflowscited by 43
  2. 2026Meta Context Engineering via Agentic Skill Evolutioncited by 42
  3. 2026EvoScientist: Towards Multi-Agent Evolving AI Scientists for End-to-End Scientific Discoverycited by 41
  4. 2025Toward Training Superintelligent Software Agents through Self-Play SWE-RLcited by 37
  5. 2026HarnessX: A Composable, Adaptive, and Evolvable Agent Harness Foundrycited by 37
  6. 2026Group-Evolving Agents: Open-Ended Self-Improvement via Experience Sharingcited by 32
  7. 2026Tool-R0: Self-Evolving LLM Agents for Tool-Learning from Zero Datacited by 31
  8. 2025CoMAS: Co-Evolving Multi-Agent Systems via Interaction Rewardscited by 27
  9. 2026SpecBench: Measuring Reward Hacking in Long-Horizon Coding Agentscited by 25
  10. 2026Recursive Self-Improvement in AI: From Bounded Self-Refinement to Autonomous Research Loopscited by 25
  11. 2025EvoLMM: Self-Evolving Large Multimodal Models with Continuous Rewardscited by 24
  12. 2026Recursive Harness Self-Improvementcited by 22
  13. 2026From Failed Trajectories to Reliable LLM Agents: Diagnosing and Repairing Harness Flawscited by 21
  14. 2025Rethinking Thinking Tokens: LLMs as Improvement Operatorscited by 20
  15. 2025Searching for Privacy Risks in LLM Agents via Simulationcited by 19
  16. 2025AutoEnv: Automated Environments for Measuring Cross-Environment Agent Learningcited by 18
  17. 2026SIA: Self Improving AI with Harness & Weight Updatescited by 18
  18. 2026RHO: Your Coding Agent is Secretly a Roboticistcited by 18
  19. 2025Sculptor: Empowering LLMs with Cognitive Agency via Active Context Managementcited by 17
  20. 2026Learning to Continually Learn via Meta-learning Agentic Memory Designscited by 16
  21. 2026The Red Queen G"odel Machine: Co-Evolving Agents and Their Evaluatorscited by 16
  22. 2025Neural Architecture Discovery via Autonomous Evolutioncited by 15
  23. 2026Rethinking the Value of Agent-Generated Tests for LLM-Based Software Engineering Agentscited by 13
  24. 2026SimWorld Studio: Automatic Environment Generation with Evolving Coding Agent for Embodied Agent Learningcited by 13
  25. 2026HarnessCompass: Guiding Automatic Harness Evolution toward Generalizable and Effective Agent Harnessescited by 13
  26. 2025Can Language Models Discover Scaling Laws?cited by 12
  27. 2026VeRO: A Harness for Agents to Optimize Agentscited by 12
  28. 2026Can LLMs Beat Classical Hyperparameter Optimization Algorithms? A Study on autoresearchcited by 12
  29. 2026MetaSkill-Evolve: Recursive Self-Improvement of LLM Agents via Two-Timescale Meta-Skill Evolutioncited by 12
  30. 2026TTHE: Test-Time Harness Evolutioncited by 12
  31. 2026SWE-Replay: Efficient Test-Time Scaling for Software Engineering Agentscited by 11
  32. 2026HarnessBank: Semantic Gene-Bank Search with Gated Verification for Agent-Harness Self-Evolutioncited by 11
  33. 2025OpenApps: Simulating Environment Variations to Measure UI-Agent Reliabilitycited by 10
  34. 2026Toward Generalist Autonomous Research via Hypothesis-Tree Refinementcited by 10
  35. 2026Harness-Aware Self-Evolving: Co-Evolving Model Weights, Harness, and Task Solutionscited by 10
  36. 2025Self-Abstraction from Grounded Experience for Plan-Guided Policy Refinementcited by 9
  37. 2026MOSS: Self-Evolution through Source-Level Rewriting in Autonomous Agent Systemscited by 9
  38. 2026EvoTrainer: Co-Evolving LLM Policies and Training Harnesses for Autonomous Agentic Reinforcement Learningcited by 9

50 most cited of 294 All 294 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026Mendel G"odel Machine: Recursive Self-Improving Coding Agents via Comparative Evolution
  2. 2026Teaching LLMs to Self-Evolve: Cultivating Core Meta-Skills with Reinforcement Learning
  3. 2024Automated Design of Agentic Systems
  4. 2026CausalEvolve: Towards Open-Ended Discovery with Causal Scratchpad
  5. 2026FreeEvolve: Learning to Evolve Beyond Fixed Loops
  6. 2026Towards AGI A Pragmatic Approach Towards Self Evolving Agent
  7. 2026What Do Evolutionary Coding Agents Evolve?
  8. 2025Evolving in Tasks: Empowering the Multi-modality Large Language Model as the Computer Use Agent