arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2603.04448cs.AIcs.CLcs.CVcs.LGcs.MA
收藏

SkillNet: 创建、评估和连接AI技能

SkillNet: Create, Evaluate, and Connect AI Skills

Zhejiang University(浙江大学);Tongji University(同济大学);Southeast University(东南大学);Alibaba Group(阿里巴巴集团);Tencent(腾讯);Fudan University(复旦大学);The University of Edinburgh(爱丁堡大学);Monash University(墨尔本大学);National University of Singapore(新加坡国立大学);Nanyang Technological University(南洋理工大学);Huzhou University(湖州大学);Hornor Device Co., Ltd(Hornor设备有限公司);Hangzhou Institute for Advanced Study, UCAS(杭州先进研究所,UCAS)
Yuan Liang, Ruobin Zhong, Haoming Xu, Chen Jiang, Yi Zhong, Runnan Fang, Jia-Chen Gu, Shumin Deng, Yunzhi Yao, Mengru Wang, Shuofei Qiao, Yida Xue, Xin Xu, Tong… 展开作者

Yuan Liang, Ruobin Zhong, Haoming Xu, Chen Jiang, Yi Zhong, Runnan Fang, Jia-Chen Gu, Shumin Deng, Yunzhi Yao, Mengru Wang, Shuofei Qiao, Yida Xue, Xin Xu, Tongtong Wu, Kun Wang, Yang Liu, Zhen Bi, Jungang Lou, Yuchen Eleanor Jiang, Hangcheng Zhu, Gang Yu, Haiwen Hong, Longtao Huang, Hui Xue, Chenxi Wang, Yijun Wang, Zifei Shan, Xi Chen, Zhaopeng Tu, Feiyu Xiong, Xin Xie, Peng Zhang, Zhengke Gui, Lei Liang, Jun Zhou, Chiyu Wu, Jin Shang, Yu Gong, Junyu Lin, Changliang Xu, Hongjie Deng, Wen Zhang, Keyan Ding, Qiang Zhang, Fei Huang, Ningyu Zhang, Jeff Z. Pan, Guilin Qi, Haofen Wang, Huajun Chen

AI总结:

SkillNet通过统一的本体和多维评估机制,大规模创建、评估和连接AI技能,提升代理性能并促进技能的持久掌握。

摘要:

当前的AI代理可以灵活地调用工具并执行复杂任务,但其长期发展受到缺乏系统积累和转移技能的限制。没有统一的机制来整合技能,代理经常"重新发明轮子",在孤立的上下文中重新发现解决方案,而无法利用先前的策略。为克服这一限制,我们引入了SkillNet,一个开放的基础设施,旨在大规模创建、评估和组织AI技能。SkillNet在统一的本体中结构化技能,支持从异质来源创建技能,建立丰富的关系连接,并在安全性、完整性、可执行性、可维护性和成本意识等方面进行多维评估。我们的基础设施集成了超过20万项技能的仓库、交互平台和多功能的Python工具包。在ALFWorld、WebShop和ScienceWorld上的实验评估表明,SkillNet显著提高了代理的性能,使多个骨干模型的平均奖励提高了40%,执行步骤减少了30%。通过将技能正式化为演进、可组合的资产,SkillNet为代理从短暂的经验转向持久的掌握提供了坚实的基础。

英文摘要:

Current AI agents can flexibly invoke tools and execute complex tasks, yet their long-term advancement is hindered by the lack of systematic accumulation and transfer of skills. Without a unified mechanism for skill consolidation, agents frequently ``reinvent the wheel'', rediscovering solutions in isolated contexts without leveraging prior strategies. To address this challenge, we introduce SkillNet, an open infrastructure for creating, evaluating, and organizing AI skills at scale. SkillNet structures skills within a unified ontology that supports creating skills from heterogeneous sources, establishing rich relational connections, and performing multi-dimensional evaluation across Safety, Completeness, Executability, Maintainability, and Cost-awareness. Our infrastructure integrates a repository of over 600,000 skills, an interactive platform, and a versatile Python toolkit. Experiments on ALFWorld, WebShop, and ScienceWorld show 40% higher average rewards and 30% fewer execution steps across multiple backbone models. Furthermore, SkillNet-Gym benchmarks skill retrieval, utilization, and composition, while SkillNet-Fabric enables task-specific skill routing through lightweight Wikis. By formalizing skills as evolving, composable assets, SkillNet provides a robust foundation for agents to move from transient experience to durable mastery.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑