发表机构
Nanjing University of Aeronautics and Astronautics; Pengcheng Laboratory; Hefei University; Microsoft（南京航空航天大学; 鹏城实验室; 合肥学院; 微软公司）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出结构感知夏普利式技能估值框架SkillSV，通过编译技能结构、分离内容与上下文价值，在四个智能体基准上验证其可恢复单元交互、指导技能压缩等特性。
AI 中文摘要
智能体技能正通过自动化反馈循环不断优化，生成具有长期结构的产物，但其内在价值仍不明确。本文研究技能估值问题：在固定智能体与保留任务分布下，为固定技能的内部单元（如规则、示例、脚本、启发式方法）分配信用。技能估值与数据或提示片段估值不同，因为技能单元具有结构性：它们可能依赖其他单元、属于文档层级、触发智能体行为并消耗有限的提示上下文。本文提出SkillSV，一种结构感知的夏普利式技能估值框架。SkillSV将技能编译为单元、依赖关系和层级，仅评估有效反事实技能；它采用成对删除和长度中性填充，将内容价值与上下文成本分离，并使用带展开预算的估计器对有噪声的智能体评估进行估值。在四个智能体基准测试中，本文评估了SkillSV的忠实度、可操作性和解释性：它能恢复单元间的交互、保留技能的整体提升，并指导安全的剪枝与压缩。
英文摘要
Agent skills are increasingly optimized by automated feedback loops, producing long structured artifacts whose internal value remains unclear. We study skill valuation: assigning credit to the internal units of a fixed skill, such as rules, examples, scripts, and heuristics, under a fixed agent and held-out task distribution. Skill valuation differs from data or prompt-span valuation because skill units are structured: they may depend on other units, belong to a document hierarchy, trigger agent behavior, and consume limited prompt context. We introduce SkillSV, a structure-aware Shapley-style framework for skill valuation. SkillSV compiles a skill into units, dependencies, and hierarchy, so that only valid counterfactual skills are evaluated. It uses paired deletion and length-neutral padding to separate content value from context cost, and estimates the resulting values with a rollout-budgeted estimator for noisy agent evaluations. On four agentic benchmarks, we assess the faithfulness, actionability, and explanation of SkillSV: it recovers unit interactions, preserves aggregate skill lift, and guides safe pruning and compression.
