method · first seen 4 Oct 2026
recursive self-improvement
Also: RSI, LLM-driven code evolution, self-evolving training pipeline, self-improvement via synthetic data
Agent autonomously improving its own capabilities through iterative learning cycles that revise its own knowledge or models.
Recursive self-improvement, long a theoretical idea about AI systems improving themselves, is showing up here as a concrete empirical method: agents that rewrite their own harnesses, workflows or training data across rounds using execution feedback. A formal-verification paper proposes diagnosing root causes instead of patching symptoms. One paper defines an 'RSI tax' to count the token cost of self-improvement. Another adds memory so evolved harnesses carry over to new benchmarks or models. Applications include visuomotor policies, image restoration, video generation and automated RL. Tooling is appearing too: a gym environment, an evaluation SDK, and a research atlas of evolved agent harnesses.
If the field settles on shared environments and cost-aware evaluation, claims about agents improving themselves become measurable and comparable rather than speculative. Work like the RSI-tax analysis may also show where these loops stop paying off.
Analyst note (emerging), 17 Nov 2026.
Mentions per week
PapersCodeDiscussion
Rank 3 on 16 Nov 2026 · 65 mentions in 28 days · 64 independent groups · papers → code → discussion · first seen 6 weeks ago · acceleration +0.11
Earliest sources
- Recursive Self-Improvement of Visuomotor Policies through Local Recovery SupervisionPapers · 4 Oct 2026 · Yuzhi Zhang, Xinyu Liu, Yu Zhang
- LucasXinYu/harness-rsiCode · 4 Oct 2026 · LucasXinYu · 25 stars
- Second-Order Problem Solving for Recursive Self-Improvement in Formal VerificationPapers · 5 Oct 2026 · Yuxuan Jiang, Aditya Vempaty, Ashish Jagmohan
- evolvent-ai/RSIGymCode · 5 Oct 2026 · evolvent-ai · 146 stars
- EmbodiedSmith: Scaling Embodied Data through Recursive Self-Improvement Flywheel in SimulationPapers · 6 Oct 2026 · Yikai Qin, Yifei Deng, Mingjian Liang, Wenxuan Song, Zepeng Lin, Zhiyi Jiang, Ji…
- RIWANav: Recursive World-Action Models with Self-Improvement for Urban NavigationPapers · 6 Oct 2026 · Jing Xie, Shouwei Ruan, Yubin Wang, Yuxiang Zhang, Haitao Yang, Songchang Jin, D…
Latest sources
- Evaluation-Free Recursive Self-Improvement for Coding AgentsPapers · 16 Nov 2026 · Zhe Ren, Guowei Rong, Yimeng Chen, Nanbo Li, Piotr Piękos, Mateusz Ostaszewski,…
- The avalanche that hit math is coming for AI developmentDiscussion · 13 Nov 2026 · crescit_eundo · 68 points
- Self-Improvement Memory Enables Continual Recursive Self-ImprovementPapers · 13 Nov 2026 · Xin Yu, Fei Xia, Jiaqi Pan, Youbao Tang, Chengzhi Liu, Adams Yu, Sicong Jiang, Y…
- RecurIR: Learning When and How to Revise Restoration Policies via Recursive Self-ImprovementPapers · 13 Nov 2026 · Hu Gao, Haoran Zhang, Yulong Chen, Lizhuang Ma
- AutoRL: A Multi-Agent System towards Autonomous Reinforcement LearningPapers · 13 Nov 2026 · Simon Tang, Wenjie Shi, Yongjing Yin, Keheng Wang, Jialin Liu, Zheng Zhang, Linv…
- PhysLoop: Harnessing LLMs' Physics Awareness for Self-Improving Video Generation ModelsPapers · 12 Nov 2026 · Yuxuan Chen, Xiaojie Li, Hao Chen, Juncheng Ma, Kewei Zhang, Zhongjing Du, Guoqi…
- The RSI Tax: When Recursive Self-Improvement Is Not Worth the CostPapers · 10 Nov 2026 · Yitong Zhang, Ang Li, Yisen Wang
- NCP: Verifier-Guided Optimization of Weight-Space Control PoliciesPapers · 9 Nov 2026 · Ekrem Aksoy
- ForgeSystem: An Architecture for Demand-Specific Artefact Production with AI AgentsPapers · 9 Nov 2026 · Qingfeng He, Zhui Zhu, Yaojian Chen, Shangzhan Li, Leshan Li, Tingjian Zhang, Qi…
- TokenRhythm/InfraRSICode · 8 Nov 2026 · TokenRhythm · 98 stars
- InfraRSI: Harnessing Recursive Self-Improvement for Training InfrastructurePapers · 8 Nov 2026 · InfraRSI Team, Yuchuan Tian, Haocheng Mei, Ye Yuan, Chao Xu, Boxun Li, Guohao Da…
- JoMoHa: Joint Model--Harness Optimization with Reinforcement LearningPapers · 8 Nov 2026 · Xudong Wu, Guangchen Lan, Mingqi Yuan, Jiayu Chen
Mentioned together with
In the ledger
- 2026-W47 · call #4 · already mainstream · Wikipedia article "Recursive self-improvement" averages 688 views a day