{"ID":2873617,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.07163","arxiv_id":"2509.07163","title":"Beyond Sequential Reranking: Reranker-Guided Search Improves Reasoning Intensive Retrieval","abstract":"The widely used retrieve-and-rerank pipeline faces two critical limitations: they are constrained by the initial retrieval quality of the top-k documents, and the growing computational demands of LLM-based rerankers restrict the number of documents that can be effectively processed. We introduce Reranker-Guided-Search (RGS), a novel approach that bypasses these limitations by directly retrieving documents according to reranker preferences rather than following the traditional sequential reranking method. Our method uses a greedy search on proximity graphs generated by approximate nearest neighbor algorithms, strategically prioritizing promising documents for reranking based on document similarity. Experimental results demonstrate substantial performance improvements across multiple benchmarks: 3.5 points on BRIGHT, 2.9 on FollowIR, and 5.1 on M-BEIR, all within a constrained reranker budget of 100 documents. Our analysis suggests that, given a fixed pair of embedding and reranker models, strategically selecting documents to rerank can significantly improve retrieval accuracy under limited reranker budget.","short_abstract":"The widely used retrieve-and-rerank pipeline faces two critical limitations: they are constrained by the initial retrieval quality of the top-k documents, and the growing computational demands of LLM-based rerankers restrict the number of documents that can be effectively processed. We introduce Reranker-Guided-Search...","url_abs":"https://arxiv.org/abs/2509.07163","url_pdf":"https://arxiv.org/pdf/2509.07163v1","authors":"[\"Haike Xu\",\"Tong Chen\"]","published":"2025-09-08T19:24:09Z","proceeding":"cs.IR","tasks":"[\"cs.IR\",\"cs.CL\",\"cs.LG\"]","methods":"[\"Large Language Model\"]","has_code":false}