The paper introduces LM2LM, a method that trains large language models (LLMs) to propose improvements for other LLMs using reinforcement learning (RL). It employs a frozen research world model (RWM) to predict the gains of proposals, providing a cheaper surrogate reward, and periodically validates high‑scoring proposals to keep feedback reliable. Experiments across eight LM development environments show that LM2LM enables a Qwen3.8‑27B model to outperform GPT‑6‑sol on held‑out tasks while reducing data‑generation cost by 33%, and it generalizes to unseen model families.
By Minhua Lin, Zijun Wang, Zewen Liu, Zhaotian Weng, Bing He, Zhan Shi, Benoit Dumoulin, Xiang Zhang, Suhang Wang, Hanqing Lu
arXiv:2610.26342v2 Announce Type: replace
Abstract: Recursive self-improvement (RSI) requires agents to diagnose system bottlenecks and coordinate improvements across training and inference. We refer...
By Ruotian Peng, Yuanyi Ren, Shuaiqi Liu, Xin Guo, Fang Zhu, Yangyi Huang, Yue Wu, Zheng Wang, Yandong Wen, Zhihang Fu
EvalResearchBench (ERB) is a benchmark that tests whether AI agents can autonomously design evaluation tasks for recursive self‑improvement. In ERB, a researcher agent selects or synthesizes tasks, builds graders, and refines them through pilot tests before producing a final evaluator that runs within fixed time and API budgets. Experiments with nine researchers and thirteen candidate models show that the best automated evaluators achieve about 75% concordance with target benchmarks, below the 91% ceiling set by target disagreements, and that human‑designed task samples still outperform many automated approaches.
By Yaolun Zhang, Tianyi Xu, Yujie Zhao, Jishen Zhao, Qingyun Wu, Huazheng Wang
RSI-Opt introduces Recursive Self-Improvement Optimizer Agents that evolve their own training policies by accumulating knowledge from previous runs, including prompts, playbooks, optimizer code, and decision cadences. A meta‑agent rewrites and archives these policies, enabling reusable procedures and code that adapt to new tasks. Across multiple generations and tasks, RSI‑Opt improves optimizer performance, achieving higher z‑scores on evolution suites, better cross‑model transfer, and significant error reductions on PDE benchmarks compared to fixed‑prompt baselines.
By Binghang Lu, Yige Sun, Guang Lin
RSIGym is a new agent-native research environment built on Everything as a Service (EaaS) that exposes training, inference, rollout, evaluation, and sandbox execution through reusable services. It supports shared budget and permission controls for Data, Harness, and Joint improvement tracks, allowing agents to investigate individual interventions and jointly optimize data, training settings, and execution harnesses within the same environment. Using RSIGym, the Opus 5 model achieved the highest RSI-Index of 0.4809 across five benchmarks, improving performance on software engineering, terminal interaction, mathematics, scientific reasoning, and skill-based tasks.
By Fanqing Meng, Lingxiao Du, Haocheng Lu, Qiguang Chen, Ziqi Zhao, Zijian Wu, Jiayuan Zhuo, Mengkang Hu, Michael Qizhe Shieh
arXiv:2610.14224v1 Announce Type: new
Abstract: LLM agents' capabilities depend not only on their underlying models but also on their execution harnesses. Recursive self-improvement (RSI) offers a pr...
By Ruiyao Xu, Yanfei Chen, Zhongying CuiZhu, Bhavana Dalvi Mishra, Yifei Ming, Han Yu, Rujun Han, Chen-Yu Lee, Tomas Pfister
Agentick is a unified benchmark for sequential decision‑making agents that evaluates RL, LLM, VLM, hybrid, and human agents on 37 procedurally generated tasks across six capability categories, four difficulty levels, and five observation modalities via a single Gymnasium‑compatible interface. It includes a Coding API, oracle reference policies, pre‑built SFT datasets, a composable agent harness, and a live leaderboard. An evaluation of 27 configurations and over 90,000 episodes shows no single approach dominates, with GPT‑5 mini leading overall, PPO excelling in planning and multi‑agent tasks, and the reasoning harness boosting LLM performance by 3‑10×, while ASCII observations outperform natural language.
By Roger Creus Castanyer, Pablo Samuel Castro, Glen Berseth
arXiv:2606. 20657v2 Announce Type: replace Abstract: Post-training a frontier model is normally weeks of human work: proposing data and recipe changes, launching runs, reading evals, deciding what to keep.
By Zhan Shi, Bing He, Yisi Sang, Hanqing Lu, Benoit Dumoulin
The paper introduces iCoder-27B, a 27‑billion‑parameter model for RTL design and GPU kernel optimization that is developed through a recursive AI‑led process with minimal human input. Human experts provide high‑level objectives and reusable research skills, while the agent autonomously selects experiments, diagnoses outcomes, and refines training strategies, coordinating SFT, self‑distillation, and reinforcement learning. iCoder outperforms GPT‑5.5 and Claude‑Opus‑4.8 on several benchmarks, demonstrating the feasibility of building frontier‑competitive models with largely automated development.
By Cheng Yang, Jiayang Lyu, Shangyuan Liu, Guibin Zhang, Jiong Lin, Xinlei Yu, Junchi Yan, Shuicheng Yan, Weinan E, Linfeng Zhang, Linfeng Zhang, Qibing Ren
arXiv:2601. 18197v2 Announce Type: replace Abstract: While Large Vision-Language Models (LVLMs) have significantly advanced GUI agents' capabilities in parsing textual instructions, interpreting screen content, and executing tasks, a critical challenge persists: the irreversibility of agent operations-where a single erroneous action can trigger catastrophic deviations.
By Shaokang Wang, Pei Fu, Ruoceng Zhang, Shaojie Zhang, Xiuwen Xi, Jiahui Yang, Bin Qin, Ying Huang, Zhenbo Luo, Jian Luan
arXiv:2608.15763v2 Announce Type: replace
Abstract: AI-powered digital avatar streamers must answer product questions, engage viewers, and execute marketing strategies in real time, demanding low lat...
By TaoLive AIGC LLM Team, Yuhan Sun, Wenhao Lin, Yongdong Luo, Yibo Hu, Meiguang Jin, Junfeng Ma, Weihang Pan, Jiaxin Zhao, Zulong Chen
The paper introduces a self‑evolving harness framework where a frozen language‑model agent first solves tasks and then edits its own harness based on run records. Using a 49‑line seed harness, the evolved harness improves average scores on in‑distribution benchmarks by 4.48 points and on out‑of‑distribution benchmarks by 12.64 points, surpassing Codex on the former and matching it on the latter. Continued evolution on a specific out‑of‑distribution benchmark further raises performance, and the study analyzes emergent mechanisms such as output truncation and history compaction.
By Qiankai Xu