{"ID":2843545,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.08505","arxiv_id":"2511.08505","title":"Structured RAG for Answering Aggregative Questions","abstract":"Retrieval-Augmented Generation (RAG) has become the dominant approach for answering questions over large corpora. However, current datasets and methods are highly focused on cases where only a small part of the corpus (usually a few paragraphs) is relevant per query, and fail to capture the rich world of aggregative queries. These require gathering information from a large set of documents and reasoning over them. To address this gap, we propose S-RAG, an approach specifically designed for such queries. At ingestion time, S-RAG constructs a structured representation of the corpus; at inference time, it translates natural-language queries into formal queries over said representation. To validate our approach and promote further research in this area, we introduce two new datasets of aggregative queries: HOTELS and WORLD CUP. Experiments with S-RAG on the newly introduced datasets, as well as on a public benchmark, demonstrate that it substantially outperforms both common RAG systems and long-context LLMs.","short_abstract":"Retrieval-Augmented Generation (RAG) has become the dominant approach for answering questions over large corpora. However, current datasets and methods are highly focused on cases where only a small part of the corpus (usually a few paragraphs) is relevant per query, and fail to capture the rich world of aggregative qu...","url_abs":"https://arxiv.org/abs/2511.08505","url_pdf":"https://arxiv.org/pdf/2511.08505v1","authors":"[\"Omri Koshorek\",\"Niv Granot\",\"Aviv Alloni\",\"Shahar Admati\",\"Roee Hendel\",\"Ido Weiss\",\"Alan Arazi\",\"Shay-Nitzan Cohen\",\"Yonatan Belinkov\"]","published":"2025-11-11T17:39:34Z","proceeding":"cs.CL","tasks":"[\"cs.CL\",\"cs.LG\"]","methods":"[\"RAG\",\"Large Language Model\"]","has_code":false}