让智能体系统边用边进化，技能与结构协同优化。
SkillMAS: Skill Co-Evolution with LLM-based Multi-Agent System
- 用可信执行记录分配功劳，动态调整技能与系统结构。
- 保留失败证据触发结构重组，避免无效技能堆积。
- 适合部署后需持续优化的复杂任务系统，如机器人操作。
大语言模型（LLM）智能体系统在部署后常需自我改进，但现有方法往往将技能演化与多智能体系统（MAS）重构割裂，导致组织瓶颈、上下文压力和角色错配。本文提出SkillMAS，一种非参数化框架，实现技能演化与系统重构的耦合。该框架通过可信执行轨迹进行效用学习以分配责任，采用有界技能演化机制，在不引发无序库增长的前提下精炼可复用流程，并在保留失败证据且执行者效用不足时触发证据门控的系统结构重排。在具身操作、命令行执行和零售工作流三个场景中，SkillMAS在报告基准下表现具有竞争力，同时清晰揭示了部署后专业化行为的归因、更新与应用方式。
原文摘要 · Abstract (English)
Large language model (LLM) agent systems are increasingly expected to improve after deployment, but existing work often decouples two adaptation targets: skill evolution and multi-agent system (MAS) restructuring. This separation can create organization bottlenecks, context pressure, and mis-specialization. We present SkillMAS, a non-parametric framework for adaptive specialization in multi-agent systems that couples skill evolution with MAS restructuring. SkillMAS uses Utility Learning to assign credit from verified execution traces, bounded skill evolution to refine reusable procedures without unfiltered library growth, and evidence-gated MAS restructuring when retained failures and Executor Utility indicate a structural mismatch. Across embodied manipulation, command-line execution, and retail workflows, SkillMAS is competitive under the reported harnesses while clarifying how post-deployment specialization is attributed, updated, and applied.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。