Problem
Research questions and friction points this paper is trying to address.
external skills
structured routing
semantic boundaries
SkillAA: Attribution-Guided Skill-Graph Updating with Targeted Validation and Rollback
🤖 AI Summary
为解决外部技能更新不精准的问题，提出SkillAA框架，通过归因引导的图编辑和局部验证方法优化技能图，提高模型性能。
📝 Abstract
External skills provide domain procedures without parameter updates, but existing methods often edit skills directly from failed rollouts without structured routing from an observed failure to an editable location; existing skill graphs also underuse semantic boundaries, object addresses, and topological dependencies for skill retrieval, targeted updating, and scoped validation. We introduce SkillAA (Skill Abductive Attribution), a structured skill-optimization framework for frozen language models. It represents skill applicability, execution, and composition in a unified graph, allowing the same structure to support skill selection, attribution-guided repair, and update validation. SkillAA contrasts successful and failed executions to route candidate repairs to specific graph objects, updates only the selected local structure, and uses Local and Big Gates to screen candidate changes before commitment. With gpt-5.6-sol, SkillAA 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 support the utility of attribution-guided graph editing and graph-scoped validation.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Attribution-Guided
Structured Skill-Optimization
Unified Graph
Local and Big Gates
Targeted Validation
🔎 Similar Papers
