{"ID":2868290,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.19376","arxiv_id":"2509.19376","title":"Solving Freshness in RAG: A Simple Recency Prior and the Limits of Heuristic Trend Detection","abstract":"We address temporal failures in RAG systems using two methods on cybersecurity data. A simple recency prior achieved an accuracy of 1.00 on freshness tasks. In contrast, a clustering heuristic for topic evolution failed (0.08 F1-score), showing trend detection requires methods beyond simple heuristics.","short_abstract":"We address temporal failures in RAG systems using two methods on cybersecurity data. A simple recency prior achieved an accuracy of 1.00 on freshness tasks. In contrast, a clustering heuristic for topic evolution failed (0.08 F1-score), showing trend detection requires methods beyond simple heuristics.","url_abs":"https://arxiv.org/abs/2509.19376","url_pdf":"https://arxiv.org/pdf/2509.19376v1","authors":"[\"Matthew Grofsky\"]","published":"2025-09-20T00:19:37Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"cs.AI\"]","methods":"[]","has_code":false}