互补查询改写：在强RAG基线下的协同效应研究
原标题：Better Together: Complementary Query Rewriting Under a Strong RAG Baseline
AI 摘要
该研究在固定强RAG流水线（BGE检索、交叉编码器重排、MMR多样化）下，系统比较了四种查询改写策略与HyDE、Query2Doc两种基线，在HotpotQA、AmbigNQ和企业数据集上发现：单独改写最多与强基线持平，但组合不同改写方法能显著提升检索效果，企业数据上HIT@10提升达13.8个百分点。研究进一步提出成本感知的置信度门控路由器，在模拟中仅用40%的改写成本捕获约一半的合并增益，并避免在AmbigNQ上的性能损失。结论是应将查询改写视为互补覆盖来源，通过成本感知路由应用，而非替代强基线。
正文节选
Better Together: Complementary Query Rewriting Under a Strong RAG Baseline Abstract A popular way to improve Retrieval-Augmented Generation (RAG) is to rewrite the user’s question into several variants and search with all of them. We test whether this actually helps once the underlying search is already strong. Under one fixed, competitive pipeline (BGE dense retrieval, cross-encoder reranking, and MMR diversification), we compare four query-rewriting strategies (S1–S4) against two strong LLM ba