发表机构
National Key Laboratory for Novel Software Technology, Nanjing University; School of Intelligence Science and Technology, Nanjing University（南京大学计算机软件新技术全国重点实验室; 南京大学智能科学与技术学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
SkillAA提出归因引导的技能图优化框架，通过对比成功与失败执行定位可编辑节点，仅更新局部结构并用门控验证，在多个基准上取得最优性能。
AI 中文摘要
外部技能无需参数更新即可提供领域流程，但现有方法往往直接从失败的执行中编辑技能，缺乏从观察到的失败到可编辑位置的结构化路由；现有技能图也未能充分利用语义边界、对象地址和拓扑依赖来进行技能检索、定向更新和范围化验证。我们提出SkillAA（技能溯因归因），一种针对冻结语言模型的结构化技能优化框架。它将技能适用性、执行和组合统一表示在一个图中，使得同一结构能够支持技能选择、归因引导修复和更新验证。SkillAA对比成功与失败的执行，将候选修复路由到特定的图对象，仅更新选定的局部结构，并使用局部门和全局门在提交前筛选候选更改。使用gpt-5.6-sol，SkillAA在SearchQA、LiveMath和DocVQA上分别达到81.5%、66.7%和91.2%，并在每个主要设置中取得最高观测均值。这些结果支持了归因引导的图编辑和图范围验证的有效性。
英文摘要
External skills provide domain knowledge and procedures without updating model parameters, but flat collections obscure skill applicability, dependencies, and composition. Graphs organize skills into addressable nodes and explicit relations, supporting selection and composition. Yet existing skill-graph methods use this structure mainly for retrieval, rather than to localize updates, scope retesting, or precisely roll back rejected changes. We introduce GRAPHSKILLAA (GraphSkill Abductive Attribution), which uses one addressable graph for skill selection, execution, failure attribution, targeted updating, validation, and rollback. Nodes separate applicability, execution, and exclusion conditions; typed edges encode prerequisite and enhancement relations. The frozen student records used nodes and edges, while the teacher contrasts related successes and failures to route each supported repair to the smallest relevant field or relation; execution lapses or insufficient evidence leave the graph unchanged. A Local Gate retests affected examples, while a Big Gate evaluates the merged graph on the complete update pool; rejected changes are rolled back. With GPT-5.6-sol, GRAPHSKILLAA reaches 81.5%, 66.7%, and 91.2% on SearchQA, LiveMath, and DocVQA, respectively, and attains the highest observed mean in every main setting. These results show that object-level attribution and graph-scoped validation make a skill graph a locally optimizable, testable, and reversible external state.
