AI systems learning to improve themselves over time
The Last AI Built by Humans: Toward Genuine Recursive Self-Improvement
Machine LearningArtificial IntelligenceComputation and Language
Summary
Many current AI models struggle to improve their own skills without human help. The authors explain a way for AI to get better by learning from experience and feedback on its own, a process called recursive self-improvement (RSI). They describe steps for AI to gain increasing independence in how it improves itself and adapts to new environments. They also discuss how this ability could be useful in different fields, like science and software development, but point out that there are still challenges to making true RSI happen. Overall, the paper connects theoretical ideas with practical efforts to create smarter, self-improving AI.
recursive self-improvementlarge language modelsHeadroom-Closed Indexautonomymachine learningfeedbackexperience acquisitionenvironment adaptationscientific discoverysoftware engineering
Authors
Yi Duan, Ying Liu, Zirui Tang, Haodong Chen, Jun Zhou, Yumou Liu, Bangrui Xu, Yukai Wu, Sidi Chen, Yuhan Zhou, Haoyu Wang, Xiaoyou Yu, Shaokun Han, Xuzhou Zhu, Le Zhou, Bolin Lu, Wei Zhou, Jiachen Liu, Nuozhou Fang, Jiaxin Tian, Ruoyu Chen, Yuxuan Li, Kai Zuo, Kaiyan Zhang, Jiantao Qiu, Conghui He, Guoliang Li, Bowen Zhou, Zhiyuan Liu, Zhoufutu Wen, Jihua Kang, Xuanhe Zhou, Fan Wu
Abstract
Recursive self-improvement (RSI) enables AI systems to turn experience and feedback into persistent changes that improve both their capabilities and the process of future improvement. We first use the Headroom-Closed Index (HCI) to reveal the problems of existing LLMs, then introduce the RSI concept and its development roadmap: from improvement-execution autonomy, improvement-strategy autonomy, experience-acquisition autonomy, and environment-adaptation autonomy, to recursive meta-improvement. Next we examine RSI across scenarios (e.g., scientific discovery, embodied intelligence, software engineering), highlighting their distinct requirements and development speeds. Drawing on diverse industry practices and preliminary empirical evidence, we connect RSI research with practical systems and identify key challenges to achieving genuine RSI.