Recursive Self-Improvement refers to the process by which an AI system, particularly LLMs, improves its own architecture, knowledge, or reasoning abilities without human intervention. This repository serves as a hub for researchers, developers, and enthusiasts interested in advancing the frontier of self-evolving AI systems.
⭐️ Want to contribute? Send us a pull request! New papers are welcome.
- Alpha_evolve
- Darwin Godel Machine: Open-Ended Evolution of Self-Improving Agents
- Self-Evolving Multi-Agent Collaboration Networks for Software Development
- Self-Adapting Language Models
- Gödel Agent: A Self-Referential Framework for Agents Recursively Self-Improvement
- UI-Genie: A Self-Improving Approach for Iteratively Boosting MLLM-based Mobile GUI Agents
- Can Large Reasoning Models Self-Train?
- Agents of Change: Self-Evolving LLM Agents for Strategic Planning
- ML-Agent: Reinforcing LLM Agents for Autonomous Machine Learning Engineering
- Self-Improving Language Models for Evolutionary Program Synthesis: A Case Study on ARC-AGI
- OpenAlpha_Evolve
- https://github.com/codelion/openevolve
- https://github.com/puripuriprince/xEvolve
- https://github.com/google-deepmind/funsearch
- No Priors Ep. 118 | With Anthropic Co-Founder Ben Mann 18:08 Human Feedback and AI Self-Improvement
- The Gentle Singularity - Sam Altman