{"ID":2896584,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2507.06838","arxiv_id":"2507.06838","title":"Shifting from Ranking to Set Selection for Retrieval Augmented Generation","abstract":"Retrieval in Retrieval-Augmented Generation(RAG) must ensure that retrieved passages are not only individually relevant but also collectively form a comprehensive set. Existing approaches primarily rerank top-k passages based on their individual relevance, often failing to meet the information needs of complex queries in multi-hop question answering. In this work, we propose a set-wise passage selection approach and introduce SETR, which explicitly identifies the information requirements of a query through Chain-of-Thought reasoning and selects an optimal set of passages that collectively satisfy those requirements. Experiments on multi-hop RAG benchmarks show that SETR outperforms both proprietary LLM-based rerankers and open-source baselines in terms of answer correctness and retrieval quality, providing an effective and efficient alternative to traditional rerankers in RAG systems. The code is available at https://github.com/LGAI-Research/SetR","short_abstract":"Retrieval in Retrieval-Augmented Generation(RAG) must ensure that retrieved passages are not only individually relevant but also collectively form a comprehensive set. Existing approaches primarily rerank top-k passages based on their individual relevance, often failing to meet the information needs of complex queries...","url_abs":"https://arxiv.org/abs/2507.06838","url_pdf":"https://arxiv.org/pdf/2507.06838v2","authors":"[\"Dahyun Lee\",\"Yongrae Jo\",\"Haeju Park\",\"Moontae Lee\"]","published":"2025-07-09T13:35:36Z","proceeding":"cs.CL","tasks":"[\"cs.CL\",\"cs.IR\"]","methods":"[\"RAG\",\"Large Language Model\"]","has_code":false,"code_links":[{"ID":612294,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2896584,"paper_url":"https://arxiv.org/abs/2507.06838","paper_title":"Shifting from Ranking to Set Selection for Retrieval Augmented Generation","repo_url":"https://github.com/LGAI-Research/SetR","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}