SkillPager: 通过语义节点检索实现查询自适应的技能内导航
SkillPager: Query-Adaptive Intra-Skill Navigation via Semantic Node Retrieval
AI总结：
针对基于技能的LLM代理在长过程文档中需要高效检索的问题，提出SkillPager两阶段框架，通过离线解析Markdown技能为类型化语义节点并在线利用MMR进行查询条件节点选择，在保持高上下文充分性的同时显著减少提示令牌。
摘要：
基于技能的LLM代理越来越依赖长过程文档，但全文档提示浪费令牌并稀释对执行至关重要的信息。我们将此设置研究为技能内检索，其目标是从已知技能文档中为给定查询选择最小且执行充分的上下文。我们提出SkillPager，一个两阶段框架，离线将每个Markdown技能解析为类型化语义节点，并在线利用最大边际相关性（MMR）进行全局的、查询条件的节点选择。在包含395个技能和1,975个查询的基准测试中，SkillPager实现了78.89%的LLM判断上下文充分性，而穷举全文档基线为82.23%，同时减少了47.04%的提示令牌。粒度消融实验表明，将相同的检索算法应用于原始固定长度块可达到可比的81.77%充分性，但令牌成本增加了28.81%，证明效率提升源于类型化语义粒度而非检索算法本身。在基于图的基线中，SkillPager以12.16%的幅度优于最强基线。进一步的消融实验表明，支持内容在候选池中保留并通过自适应选择而非静态启发式移除时最为有效。这些结果将类型化文档内检索确定为基于技能的代理的一个独特访问问题。
英文摘要：
Skill-based LLM agents increasingly rely on long procedural documents, but full-document prompting wastes tokens and dilutes information critical to execution. We study this setting as intra-skill retrieval, where the goal is to select a minimal, execution-sufficient context from a known skill document given a query. We present SkillPager, a two-stage framework that parses each Markdown skill into typed semantic nodes offline and leverages Maximal Marginal Relevance (MMR) to perform global, query-conditioned node selection online. On a benchmark of 395 skills and 1,975 queries, SkillPager achieves 78.89% LLM-judged context sufficiency, compared to 82.23% for the exhaustive full-document baseline, while reducing prompt tokens by 47.04%. A granularity ablation shows that applying the same retrieval algorithm to raw fixed-length chunks reaches a comparable 81.77% sufficiency but increases token cost by 28.81%, demonstrating that efficiency gains are driven by typed semantic granularity rather than the retrieval algorithm alone. Among graph-based baselines, SkillPager outperforms the strongest baseline by a margin of 12.16%. Further ablations show that supporting content is most effective when retained in the candidate pool and selected adaptively rather than removed by static heuristics. These results identify typed intra-document retrieval as a distinct access problem for skill-based agents.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
