arXiv cs.AI· Yewei Zhao, Derong Xu, Wei Chen, Yin Xu, Peilun Zhou, Xiaohan Wang, Guojun Yin, Wei Lin, Mingjun Xiao·· 13 小时前AI 评分34
CDSE：面向自我改进智能体的技能认知分化方法
From Trajectories to Differentiated Skills: Cognitive Differentiation for Self-Improving Agents
AI 导读
研究者提出技能演化中的认知分化方法 CDSE，将统一技能拆分为 Meta、Experience、Tool、Script 四个功能组件，从成功与失败轨迹中提取语义梯度并按功能角色整合更新。
正文 · 原文
Abstract:Large language model (LLM) agents tackle complex tasks through reasoning and interaction with external environments. Agent skills provide persistent, transferable capabilities that guide task planning, inform contextual decisions, support tool use, and enable executable procedures. Yet existing skill-evolution methods largely organize knowledge around skill artifacts, overlooking the functional heterogeneity within individual skills and its implications for knowledge maintenance and use. Inspired by cognitive science, we propose Cognitive Differentiation in Skill Evolution (CDSE), which evolves a unified skill through four functionally distinct yet coordinated components: Meta, Experience, Tool, and Script. CDSE extracts evidence-grounded semantic gradients from successful and failed trajectories, consolidates them by functional role, and applies component-specific updates and validation with cross-component coordination and adaptive invocation. An analysis of 500 top-ranked skill packages shows that 96.8% contain knowledge spanning at least two functional roles. Across six benchmarks, three agent models, and two execution harnesses, CDSE achieves the highest scores in all 36 settings, improving over the Base agent by 7.6 percentage points and the strongest competing method in each setting by 2.6 points on average. Ablation studies further support the importance of functional differentiation and adaptive invocation.
| Comments: | 22 pages, 7 figures, 17 tables, including appendices |
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.23599 [cs.AI] |
| (or arXiv:2610.23599v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.23599 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yewei Zhao [view email]
[v1]
Fri, 16 Oct 2026 15:16:56 UTC (1,161 KB)
来源：arXiv cs.AI · arxiv.org