SkillQR：面向推理密集型查询改写的智能体技能发现
原标题：SkillQR: Agentic Skill Discovery for Reason-Intensive Query Rewriting
AI 摘要
该论文提出 SkillQR，一个无需训练的查询改写框架，能从目标检索语料中自动发现推理策略集，并在推理时动态将查询路由到合适的改写策略。作者指出最优改写策略取决于检索目标而非仅取决于模型，现有方法依赖昂贵专有模型、大量训练数据或迭代检索改写。在 BRIGHT 和 R2MED 两个推理密集型检索基准上，SkillQR 无需训练数据即达到与微调改写模型相当的性能，且推理延迟低于迭代改写方法。
正文节选
SkillQR: Agentic Skill Discovery for Reason-Intensive Query Rewriting Abstract Dense retrievers provide grounded evidence that mitigates hallucinations in outputs from Large Language Models (LLMs). Real-world queries are often verbose, ambiguously framed, or require multi-step reasoning, introducing irrelevant details that weaken semantic matching for dense retrievers. Query rewriting is a widely adopted pre-retrieval technique that leverages LLMs to reformulate queries and bridge the semantic g