面向RAG系统的预检索查询聚类自适应Top-k文档检索
原标题：Pre-retrieval Query Clustering for Adaptive Top-k Document Retrieval in RAG Systems
AI 摘要
该论文提出一种面向RAG系统的预检索查询聚类框架，用于自适应选择top-k文档检索深度。方法离线通过NDCG-k曲线估计每个查询的饱和点k*，再在嵌入空间聚类查询并为每个簇推荐检索深度；运行时将查询分配到簇并常数时间选择top-k。相比依赖检索后文档聚类的CAR方法，该方案是预检索且以查询为中心，在异构语料中更稳健。作者称在全流量查询测试中效果提升超36%，低复杂度簇token用量减少14%且无准确率损失。
正文节选
Pre-retrieval Query Clustering for Adaptive Top-k Document Retrieval in RAG Systems Abstract. RAG systems commonly retrieve a fixed number of documents (top-k) to ground generation, but this static approach is brittle: simple queries suffer over-retrieval (adding noise and cost) while complex queries are under-retrieved, causing recall failures that cascade into incorrect answers. Motivated by the question of how many documents must be retrieved to answer an arbitrary query reliably, we propose