Tsinghua's SkillMaster Trains LLM Agents to Create and Refine Skills via RL
Researchers from Tsinghua University and collaborators present SkillMaster, a NeurIPS 2026 paper that trains LLM agents to autonomously create, update, and retain skills through reinforcement learning, integrating skill management into the agent's own policy. The method uses two learning signals—making previously failed similar tasks easier and previously successful tasks more efficient—to drive skill editing, enabling skill evolution and task execution to be learned jointly.
Coverage timeline
机器之心机器之心
随着递归自我改进(Recursive Self-Improvement,RSI)研究的兴起,智能体的长期能力演进受到更多关注。技能(Skill)能够将交互经验提炼为可复用的操作流程与任务策略,使一次任务的收获服务于后续任务。让智能体学会自主创建和改进这些技能,为其持续进化提供了一条具体路径。 图 1 左:外部模块负责管理技能,智能体检索并使用技能。右:SkillMaster 将技能管理纳入智能体自身的策略,与任务执行共同学习。 然而,现有方法中的技能创建与更新,通常依赖外部教师模型、预设规则或辅助模块(图 1 左)。执行任务的智能体学会了使用技能,却很少被直接训练去改进技能。要沿着这条路径实现自主改进,还需要让 “如何进化技能” 成为智能体自身可以学习的能力(图 1 右)。 清华大学李勇教授团队提出了 SkillMaster,相关论文已被 NeurIPS 2026 录用,论文第一作者为北京中关村学院博士生杨敏。 研究以一个直观的洞察为出发点:有价值的技能改进,应当让原来失败的相似任务在后续更容易成功,让原来已经成功的任务以更少的步骤完成。SkillMaster 将这两类效果转化为技能编辑的学习信号,训练同一个智能体根据交互经验创建、更新和保留技能,使技能进化与任务执行共同进入强化学习过程。 论文题目:SKILLMASTER: Toward Autonomous Skill Mastery in LLM Agents Author: Min Yang, Jinghua Piao, Xu Xia, Xiaochong Lan, Jiaju Chen, Yongshun Gong, Yong Li Affiliation: Shandong University, Zhongguancun Academy, Tsinghua University, Southeast University, University of Science and Technology of China arXiv: https://arxiv.org/abs/2605.08693 GitHub: https://github.com/sduyangmin/Skill-Master 核心洞察:让失败的任务更容易成功 让成功的任务完成得更高效 要训练智能体改进技能,首先需要知道一次修改给后续任务

Related stories
UCSB and Microsoft Propose ScholarEvolve for AI Agents to Read Papers and Self-Modify Code
Researchers from UC Santa Barbara and Microsoft introduced ScholarEvolve, a method that automates the research process of reading literature, implementing algorithms, and evaluating results to enable AI agents to evolve their harness code without modifying underlying model parameters. The approach addresses the limitation of prior coding-agent-based harness optimization, which relied on the agent's existing knowledge and observed failures, by allowing agents to draw on the latest academic findings. The paper and code are publicly available.
2 reports · 1 sources

METR: Anthropic models improving twice as fast, AI use drives ~80% of acceleration
A METR report, based on an 11-week internal assessment at Anthropic, finds model capability progress roughly doubled since October 2025, with AI-assisted research driving about 80% of the acceleration. The report also reveals three unreleased internal models (Model 2, 3, 4), with the strongest exceeding Opus 5.5 by about three months of progress. Anthropic redacted much of the most informative evidence from the public version, citing confidentiality concerns.
6 reports · 5 sources47 points · 6 comments

Pitt Researchers Uncover 3D Scene Topology in VLM Latent Spaces
University of Pittsburgh researchers analyzed the latent representations of vision-language models (VLMs) and found that intermediate layers encode spatial structures such as rings, grids, and random layouts, corresponding to 3D scene topology. Their paper, accepted as a NeurIPS 2026 spotlight, introduces methods to separate spatial signals from appearance attributes in high-dimensional activations.
1 reports · 1 sources

Alibaba Qwen outlines Personal Agent framework at CNCC 2026
At CNCC 2026, Alibaba Qwen's AI lead Jiang Guanjun presented a technical framework for Personal Agent, emphasizing the coordination of model, memory, and ecosystem to enable proactive services. He noted that AI applications are moving from dialogue to agent and assistant stages, with Personal Agent potentially solving nearly 100% of daily problems if system capabilities align with problem-solving paths. Qwen has built a user demand map covering 30+ industries, identified 8,000+ atomic model capabilities, and reported a 1.6x increase in average token consumption for complex tasks over the past three months.
1 reports · 1 sources

News updates every hour. Subscribe in your RSS reader to get new stories automatically.
Subscribe