English(EN)R$^2$ Flow: Recursive Self-Improvement via Recursive Skill Evolution
AI代理通过演化的外部模型探索递归自我改进 · 跟踪10个来源
作者PulseAugur 编辑部·[27 个来源]·
多篇研究论文探讨了AI代理的递归自我改进(RSI)技术,旨在无需持续的人工干预或核心模型的再训练即可增强其能力。这些方法侧重于演化外部组件,如验证器、约束器或世界模型,然后由这些组件指导代理的学习和决策过程。方法包括用于验证器程序的统一协同演化(UCoE)、用于具身代理的AdaRSI、用于策略和记忆适应的MMLA-RSI,以及用于显式世界-语言-动作模型的D-World。其他研究引入了物理RSI(用于机器人)、AReaL-RSI(用于协调研究团队)和利用预测世界模型的RSI-Anything等框架,所有这些都为更自主和有效的AI开发做出了贡献。
AI
arXiv:2610.23599v1 Announce Type: new Abstract: Large language model (LLM) agents tackle complex tasks through reasoning and interaction with external environments. Agent skills provide persistent, transferable capabilities that guide task planning, inform contextual decisions, s…
arXiv:2610.04184v2 Announce Type: replace Abstract: Recursive self-improvement (RSI) relies on evaluation feedback to assess progress and guide further research, yet repeatedly running complex benchmarks is costly and slows iteration. Human experts reduce this cost by selecting b…
arXiv:2609.33867v2 Announce Type: replace Abstract: LLM-based agents can improve themselves across tasks by reusing and revising the skills they orchestrate into executable procedures. Flow-based training fits this loop: it samples procedures in proportion to reward, and the flow…
arXiv cs.AI
TIER_1English(EN)·Renhao Li, Xinjie Shen, Yang Su, Jianhong Tu, Derek F. Wong, Dayiheng Liu, Min Yang·
arXiv:2610.21492v1 Announce Type: new Abstract: Language-model agents can accumulate interaction feedback, but reliably turning that feedback into reusable experience remains challenging. Existing methods either optimize external experience scaffolds or train experience-related b…
arXiv:2609.08183v2 Announce Type: replace Abstract: Recursive self-improvement (RSI) requires a concrete mechanism through which an AI system observes its capabilities and converts that evidence into the next round of learning. We present NeoHorse-1, a family of agent-native mode…
arXiv cs.AI
TIER_1English(EN)·Huan Zhang, Mingju Chen, Dongxu Zhou, Can Lv, Sen Cui, Faguo Wu, Shiji Zhou·
arXiv:2610.21791v1 Announce Type: new Abstract: Large language model (LLM) agents have demonstrated strong capabilities in complex tasks, including code generation, tool use, and multi-hop question answering. Existing self-improvement methods enhance agent performance through mod…
arXiv cs.AI
TIER_1English(EN)·Brian B. Moser, Stanislav Frolov, Tobias C. Nauen, Federico Raue, Andreas Dengel·
arXiv:2610.21646v1 Announce Type: new Abstract: With the rise of large language models, agentic systems are becoming increasingly autonomous, as they write code, call tools, revise prompts, and propose modifications to their own components. This autonomy raises a fundamental ques…
arXiv:2610.14625v1 Announce Type: new Abstract: Recursive self-improvement (RSI) offers a route to AI systems that participate in their own continued development. Recent methods allow language agents to retain changes to their parameters, memory, skills, and code for subsequent e…
arXiv cs.AI
TIER_1English(EN)·Tianyi Xu, Yaolun Zhang, Yiran Wu, Joe Nguyen, Shengyu Dai, Saba Moeinizade, Zhenwen Shao, Qingyun Wu, Huazheng Wang·
arXiv:2610.17609v1 Announce Type: new Abstract: Training-free self-improvement lets an agent use feedback to revise the response it is about to return and to update the harness state that shapes how it handles later tasks. On open-ended tasks that feedback rarely comes from refer…
arXiv:2610.18083v1 Announce Type: new Abstract: Scaling recursive self-improvement requires additional computation to produce useful research progress in reaching persistent objectives. Intuitively, more researcher agents can explore more hypotheses, but early evaluation scores m…
arXiv cs.AI
TIER_1English(EN)·Hai Zhou, Fan Yang, Zeyi Chen, Lejun Zhang, Chenyang Fu, Jianping Li, Jiayi Zhu, Xianyang Liu, Yingxuan Yang, Lili Meng, Lu Wang, Xinyang Han, Jiankai Sun, Weirui Ye, Guanchu Wang, Wanhua Li, Ming Zhou, Boyi Li, Ming Jin, Kun Shao, Muhao Chen, Guang Chen…·
arXiv:2610.18842v1 Announce Type: new Abstract: Robots are increasingly autonomous in task execution, yet their improvement remains largely human-driven. This study investigates Physical Recursive Self-Improvement (Physical RSI), in which physical experience guides lasting update…
arXiv:2610.17114v1 Announce Type: cross Abstract: Frontier large language models (LLMs) now reason about space well enough to program robots, yet the world models behind robot policies are implicit, held in network weights or predicted video, and cannot be inspected or corrected …
arXiv:2610.20294v1 Announce Type: new Abstract: An agent can improve its behavior within an unresolved problem by changing how it acts and which experience it retains. We propose MMLA-RSI, a recursive self-improvement scheme with two implementations in Memory-Mediated Learning Ar…
arXiv:2610.20474v1 Announce Type: new Abstract: Pretrained vision-language-action models (VLAs) often fail midway through unseen long-horizon tasks. Recursive self-improvement (RSI) offers a remedy, but embodied RSI methods either refine the harness around a frozen executor and s…
arXiv:2610.13912v1 Announce Type: new Abstract: Recursive self-improvement (RSI) aims to transform interaction experience into persistent improvements, yet external feedback is costly and world-model predictions may become unreliable as policies evolve. We introduce RSI-Anything,…
arXiv:2610.12176v1 Announce Type: new Abstract: Recursive self-improvement (RSI) of a model on non-verifiable tasks, such as open-ended research, faces a supervision bottleneck when its outputs exceed what even human experts can reliably assess, leaving the model itself (optimize…
arXiv:2610.11794v1 Announce Type: new Abstract: Learning to act in unfamiliar environments requires agents to infer how the world works and revise that understanding as new evidence arrives. Yet limited observations can support multiple world models that explain past interactions…
Recursive self-improvement (RSI) aims to transform interaction experience into persistent improvements, yet external feedback is costly and world-model predictions may become unreliable as policies evolve. We introduce RSI-Anything, a domain-general framework built on JEPA-Anythi…
arXiv:2610.10310v1 Announce Type: new Abstract: Recursive self-improvement requires carrying accepted changes into later improvement cycles, while studying agent-proposed changes also requires substantial research infrastructure. Existing settings often leave agents to rebuild ro…
Learning to act in unfamiliar environments requires agents to infer how the world works and revise that understanding as new evidence arrives. Yet limited observations can support multiple world models that explain past interactions but predict different outcomes in unseen states…
arXiv:2610.07105v1 Announce Type: cross Abstract: Recommendation recursive self-improvement (Rec-RSI) feeds recommender outputs into subsequent training. Evaluating each round solely through its latest model assumes that the successor consolidates the update, although pre- and po…
Recommendation recursive self-improvement (Rec-RSI) feeds recommender outputs into subsequent training. Evaluating each round solely through its latest model assumes that the successor consolidates the update, although pre- and post-update models may retain complementary ranking …
Sequential recommenders can generate synthetic interaction sequences and retrain on the augmented corpus in a recursive self-improvement loop. To limit error accumulation, current methods verify each generated sequence remains predictive of the user's real interactions and discar…
Hacker News — AI stories ≥50 points
TIER_1English(EN)·hardmaru·
New Sakana AI paper: Recursive Self-Improvement through Multi-Agent Self-Supervision (MASS).
Blog: pub.sakana.ai/mass/
Paper: arxiv.org/abs/2610.12176
How do we keep improving AI when tasks become too complex for humans to reliably evaluate?
The model eventually hits a supervi…
<h2> TL;DR </h2> <p>One self-improving model family now holds eleven public number-one benchmark records at the same time, across math, science, law, structured output, and decisions. The interesting part is not any single score, it is that a recursive self-improvement loop, boun…