SkillOS: 为自演化代理学习技能编目
SkillOS: Learning Skill Curation for Self-Evolving Agents
AI总结：
SkillOS通过经验驱动的强化学习方法，解决自演化代理中复杂长期技能编目的学习问题，优于记忆-free和强记忆基线，在效果和效率上均表现优异，技能库逐步演变成更丰富的Markdown文件。
摘要：
基于大语言模型的代理越来越多地被用于处理流式任务，但它们通常仍是一次性问题解决者，无法从过去交互中学习。从经验中提取的可重用技能为自演化提供了自然基质，其中高质量的技能编目是关键瓶颈。现有方法要么依赖手动技能编目，要么规定启发式技能操作，或训练短期地平线技能操作。然而，它们仍然难以从间接和延迟反馈中学习复杂长期编目策略。为解决这一挑战，我们提出了SkillOS，一种基于经验的强化学习训练配方，用于学习自演化代理中的技能编目。SkillOS将一个冻结的代理执行器（检索和应用技能）与可训练的技能编目器（更新外部SkillRepo）配对。为了为编目提供学习信号，我们设计了复合奖励，并基于技能相关的任务依赖性在分组任务流上进行训练，其中早期轨迹更新SkillRepo，而后期相关任务评估这些更新。在多轮代理任务和单轮推理任务中，SkillOS在效果和效率上均优于无记忆和强记忆基线，所学的技能编目器能跨不同执行器backbone和任务领域泛化。进一步分析显示，所学的编目器产生更针对性的技能使用，而SkillRepo中的技能逐步演变成更丰富的Markdown文件，编码更高层次的元技能。
英文摘要：
LLM-based agents are increasingly deployed to handle streaming tasks, yet they often remain one-off problem solvers that fail to learn from past interactions. Reusable skills distilled from experience provide a natural substrate for self-evolution, where high-quality skill curation serves as the key bottleneck. Existing approaches either rely on manual skill curation, prescribe heuristic skill operations, or train for short-horizon skill operations. However, they still struggle to learn complex long-term curation policies from indirect and delayed feedback. To tackle this challenge, we propose SkillOS, an experience-driven RL training recipe for learning skill curation in self-evolving agents. SkillOS pairs a frozen agent executor that retrieves and applies skills with a trainable skill curator that updates an external SkillRepo from accumulated experience. To provide learning signals for curation, we design composite rewards and train on grouped task streams based on skill-relevant task dependencies, where earlier trajectories update the SkillRepo, and later related tasks evaluate these updates. Across multi-turn agentic tasks and single-turn reasoning tasks, SkillOS consistently outperforms memory-free and strong memory-based baselines in both effectiveness and efficiency, with the learned skill curator generalizing across different executor backbones and task domains. Further analyses show that the learned curator produces more targeted skill use, while the skills in SkillRepo evolve into more richly structured Markdown files that encode higher-level meta-skills over time.
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。
