{"ID":2868595,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.15588","arxiv_id":"2509.15588","title":"CFDA \u0026 CLIP at TREC iKAT 2025: Enhancing Personalized Conversational Search via Query Reformulation and Rank Fusion","abstract":"The 2025 TREC Interactive Knowledge Assistance Track (iKAT) featured both interactive and offline submission tasks. The former requires systems to operate under real-time constraints, making robustness and efficiency as important as accuracy, while the latter enables controlled evaluation of passage ranking and response generation with pre-defined datasets. To address this, we explored query rewriting and retrieval fusion as core strategies. We built our pipelines around Best-of-$N$ selection and Reciprocal Rank Fusion (RRF) strategies to handle different submission tasks. Results show that reranking and fusion improve robustness while revealing trade-offs between effectiveness and efficiency across both tasks.","short_abstract":"The 2025 TREC Interactive Knowledge Assistance Track (iKAT) featured both interactive and offline submission tasks. The former requires systems to operate under real-time constraints, making robustness and efficiency as important as accuracy, while the latter enables controlled evaluation of passage ranking and respons...","url_abs":"https://arxiv.org/abs/2509.15588","url_pdf":"https://arxiv.org/pdf/2509.15588v1","authors":"[\"Yu-Cheng Chang\",\"Guan-Wei Yeo\",\"Quah Eugene\",\"Fan-Jie Shih\",\"Yuan-Ching Kuo\",\"Tsung-En Yu\",\"Hung-Chun Hsu\",\"Ming-Feng Tsai\",\"Chuan-Ju Wang\"]","published":"2025-09-19T04:42:31Z","proceeding":"cs.IR","tasks":"[\"cs.IR\",\"cs.AI\"]","methods":"[]","has_code":false}