{"id":"arxiv:2410.04444","title":"Gödel Agent: A Self-Referential Agent Framework for Recursive Self-Improvement","date":"2024-10-06","section":"Agents, open-endedness, AGI","quality":-0.050839,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/","source_url":"https://arxiv.org/abs/2410.04444","text_available":true,"authors":["Xunjian Yin","Xinyi Wang","Liangming Pan","Li Lin"],"outgoing":[{"source":"arxiv:2410.04444","target":"arxiv:2203.11171","evidence":"id","id":"arxiv:2203.11171","title":"Self-Consistency Improves Chain of Thought Reasoning in Language Models","url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/02a15b098d96a94d463bb470f59a34723abc34e59196f35c539f9796d8f3c3df/"},{"source":"arxiv:2410.04444","target":"arxiv:2303.17651","evidence":"title","id":"arxiv:2303.17651","title":"Self-Refine: Iterative Refinement with Self-Feedback","url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/accef2157d87cf33be93cb3a9af246528b0a3742a9e2e700a194956fac8f5cf4/"},{"source":"arxiv:2410.04444","target":"arxiv:2407.21783","evidence":"id","id":"arxiv:2407.21783","title":"The Llama 3 Herd of Models","url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/de3cebb4c28bc5d41a843279f219dca14476284991ea5c7c0aeab7068a40e306/"},{"source":"arxiv:2410.04444","target":"arxiv:2408.08435","evidence":"id","id":"arxiv:2408.08435","title":"Automated Design of Agentic Systems","url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/c9890d5668d1c1b33e29569b74626fd586f456af6a97c657b8c99895c90f59a5/"}],"incoming":[{"source":"arxiv:2505.22954","target":"arxiv:2410.04444","evidence":"id","id":"arxiv:2505.22954","title":"Darwin Godel Machine: Open-Ended Evolution of Self-Improving Agents","url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/c29f716871f80314868b5647692de387f26e12166b93bd362915fd272b6e081c/"},{"source":"arxiv:2608.08311","target":"arxiv:2410.04444","evidence":"id","id":"arxiv:2608.08311","title":"Ouroboros: A Self-Developing Frontier Coding Agent with Reviewed Core Evolution","url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/4b8f53ea59f810bf1717d78f1907b7596b91df2bec6751aba12accc8598c78e9/"}],"text_parts":[{"start":0,"end":15650,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-1.html"},{"start":15650,"end":31566,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-2.html"},{"start":31566,"end":40222,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-3.html"}],"contents":[{"title":"Gödel Agent: A Self-Referential Agent Framework for Recursively Self-Improvement","offset":0,"level":1,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-1.html#text"},{"title":"Abstract","offset":84,"level":6,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-1.html#text"},{"title":"1 Introduction","offset":1068,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-1.html#text"},{"title":"2 Related Work","offset":6331,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-1.html#text"},{"title":"3 Self-Referential Gödel Agent","offset":8822,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-1.html#text"},{"title":"4 Gödel Agent Implementation","offset":13620,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-1.html#text"},{"title":"4.1 Implementation Details","offset":14227,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-1.html#text"},{"title":"4.2 Additional Designs","offset":16480,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-2.html#text"},{"title":"5 Experiments","offset":17834,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-2.html#text"},{"title":"5.1 Baseline Methods","offset":18332,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-2.html#text"},{"title":"5.2 Experimental Settings","offset":19652,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-2.html#text"},{"title":"5.3 Experimental Results and Analysis","offset":20829,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-2.html#text"},{"title":"6 Analysis","offset":22410,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-2.html#text"},{"title":"6.1 Analysis of Initial Tools","offset":22873,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-2.html#text"},{"title":"6.2 Robustness Analysis of the Agent","offset":24005,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-2.html#text"},{"title":"6.3 Case Study: Game of 24","offset":24955,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-2.html#text"},{"title":"7 Discussions and Future Directions","offset":27920,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-2.html#text"},{"title":"7.1 Discussions","offset":27960,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-2.html#text"},{"title":"7.2 Future Directions","offset":29176,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-2.html#text"},{"title":"8 Conclusion","offset":31549,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-2.html#text"},{"title":"Limitations","offset":32136,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-3.html#text"},{"title":"Ethics Statement","offset":33440,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-3.html#text"},{"title":"Appendix A Goal Prompt of Gödel Agent","offset":34405,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-3.html#text"},{"title":"Appendix B Experiment Details","offset":34743,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-3.html#text"},{"title":"Appendix C Representative Policies Improved by Gödel Agent","offset":35787,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-3.html#text"},{"title":"C.1 Codes of the Best Policies Found by Gödel Agent Across Four Tasks","offset":35850,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-3.html#text"},{"title":"C.2 Codes in Game of 24 Tasks","offset":36530,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-3.html#text"},{"title":"Appendix D Cost of Experiments","offset":36742,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-3.html#text"},{"title":"Appendix E Additional Novel Policies Designed by Gödel Agent","offset":37414,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-3.html#text"},{"title":"Appendix F Comparison Between Random Sampling and Gödel Agent Performance","offset":38324,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-3.html#text"},{"title":"Instructions for reporting errors","offset":39258,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/part-3.html#text"}],"text_code_points":40222,"source_markdown":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/source.md"}