1.Darwin Among the Machines(Samuel Butler, 1863):https://en.wikipedia.org/wiki/Darwin_among_the_Machines
2. EURISKO:Lenat, D. B. (1982). The Nature of Heuristics. Artificial Intelligence, 19(2), 189–249.3. STaR(NeurIPS 2022):https://arxiv.org/abs/2203.144654. Constitutional AI(Anthropic, 2022):https://arxiv.org/abs/2212.080735. Reflexion(NeurIPS 2023):https://arxiv.org/abs/2303.113666. OPRO(ICLR 2024):https://arxiv.org/abs/2309.034097. SPIN(ICML 2024):https://arxiv.org/abs/2401.013358. Self-Rewarding Language Models(ICML 2024):https://arxiv.org/abs/2401.100209. Meta-Rewarding(2024):https://arxiv.org/abs/2407.1959410. ADAS(ICLR 2025):https://arxiv.org/abs/2408.0843511. AFlow(ICLR 2025):https://arxiv.org/abs/2410.1076212. Darwin Gödel Machine(DGM, 2025):https://arxiv.org/abs/2505.2295413. Self-Harness(2026):https://arxiv.org/abs/2606.0949814. WebEvolver(EMNLP 2025):https://arxiv.org/abs/2504.2102415. SEAL(NeurIPS 2025):https://arxiv.org/abs/2506.1094316. MetaClaw(arXiv 2026):https://arxiv.org/abs/2603.1718717. AgentX(快手, 2026):https://arxiv.org/abs/2606.2685918. Bilevel Autoresearch(2026):https://arxiv.org/abs/2603.2342019. Recursive Harness Self-Improvement(2026):https://arxiv.org/abs/2607.1552420. Motus2(2026):https://arxiv.org/abs/2608.3023721. AI4AI-Bench(2026):https://arxiv.org/abs/2608.2031822. AutoResearchEval(2026):https://arxiv.org/abs/2608.1490523. Self-Improvements in Modern Agentic Systems(综述, 2026):https://arxiv.org/abs/2607.1310424. Recursive Self-Improvement in AI(综述, 2026):https://arxiv.org/abs/2607.0766325. Recursive Criticality of AI Self-Improvement(2026):https://arxiv.org/abs/2609.001371. OpenAI GPT-5.6 与 RSI Index:https://openai.com/index/gpt-5-6/2. OpenAI 内部研究加速报告:https://openai.com/index/research-acceleration-view-inside-openai/3. Jakub Pachocki《An Alien Mind》:https://openai.com/index/an-alien-mind/4. Google Gemini 3.8 Flash 与递归模型精炼循环:https://blog.google/innovation-and-ai/models-and-research/gemini-models/3-8-flash-and-3-8-flash-cyber/5. AIDE²:Level 1 RSI 的团队实验报告(Weco AI):https://www.weco.ai/blog/first-evidence-of-recursive-self-improvement6. 快手 AgentX:推荐系统自我迭代:https://mp.weixin.qq.com/s/7aW9zaGkwq-hzv25U3aN8A7. Lilian Weng:Harness Engineering for Self-Improvement:https://lilianweng.github.io/posts/2026-07-04-harness/