語彙⼀致(BM25) と 意味(Dense) は相補的 • 順位で合算 → スコアのスケール差に頑健 • k=60 等、⼿堅い初期値から ◦ 最適値はデータ依存 ◦ 他にはスコア正規化+加重和 / 学習型融合 RRF(d) = Σ 1 / (k + rankᵢ(d)) 順位さえあれば任意の検索器を混ぜられる Gordon V. Cormack, Charles L A Clarke, and Stefan Buettcher. 2009. Reciprocal rank fusion outperforms condorcet and individual rank learning methods. In Proceedings of the 32nd international ACM SIGIR conference on Research and development in information retrieval (SIGIR '09). Association for Computing Machinery, New York, NY, USA, 758‒759. https://doi.org/10.1145/1571941.1572114 12
Late Interaction, Cross-encoder の⽐較を⾏いたいと考え た 2. JaColBERTv2.5 を Retriever / Reranker として利⽤可能かを確認したい と考えた Chuan Meng, Litu Ou, Sean MacAvaney, and Jeff Dalton. 2026. Revisiting Text Ranking in Deep Research. In Proceedings of the 49th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR '26). Association for Computing Machinery, New York, NY, USA, 3006–3016. https://doi.org/10.1145/3805712.3808557 18