KaLM-Reranker-V1：快速但非晚期交互的壓縮文檔重排序

摘要

隨著檢索系統規模擴大，高品質的重排序變得越來越重要。然而，現有的重排序模型 — 無論是基於編碼器還是解碼器 — 大多將查詢與段落聯合編碼，使得運算高度耦合，限制了部署效率與靈活性。我們提出 KaLM-Reranker-V1，一種快速但非延遲交互（FBNL）的重排序器，它在解耦查詢與段落運算的同時，仍保留了具表現力的相關性建模。KaLM-Reranker-V1 採用編碼器-解碼器架構，編碼器透過 Matryoshka 嵌入池化（Matryoshka embedding pooling）預先編碼段落，解碼器則負責處理系統指令、使用者指令與查詢意圖；交叉注意力（cross-attention）機制進一步捕捉查詢上下文與段落表徵之間的相關性。這種設計讓 KaLM-Reranker-V1 藉由解耦的段落編碼達到高效率，同時透過交叉注意力保留豐富的相關性建模，因此並非延遲交互。我們以三種規模實例化 KaLM-Reranker-V1，分別為 Nano（0.27B 激活參數）、Small（1B）與 Large（4B）。在 BEIR、MIRACL 與 LMEB 上的大規模實驗顯示，KaLM-Reranker-V1 在具備優異效率的同時，達到了強大的重排序表現。在 BEIR 上，KaLM-Reranker-V1 達到最先進的效能，與 Qwen3-Reranker 系列等強勁工業模型並駕齊驅；在 MIRACL 上，儘管未經大量多語言資料訓練，KaLM-Reranker-V1 仍展現出極佳的重排序性能。此外，在 LMEB 上，重排序模型展現出明顯優勢，即使是 0.27B 的 Nano 模型也能與 7–12B 的嵌入模型競爭。

English

As retrieval systems scale, high-quality reranking becomes increasingly important. However, most existing rerankers, whether encoder-based or decoder-based, jointly encode the query and passage, tightly coupling their computation and limiting deployment efficiency as well as flexibility. We present KaLM-Reranker-V1, a fast but not late-interaction (FBNL) reranker that decouples query and passage computation while retaining expressive relevance modeling. Built on an encoder-decoder architecture, KaLM-Reranker-V1 uses the encoder to pre-encode passages with Matryoshka embedding pooling, while the decoder models the system instruction, user instruction, and query intent; cross-attention then captures relevance between the query context and passage representations. This design makes KaLM-Reranker-V1 efficient through decoupled passage encoding, yet not late interaction, by preserving rich relevance modeling through cross-attention. We instantiate KaLM-Reranker-V1 in three sizes, Nano, Small, and Large, with 0.27B, 1B, and 4B activated parameters, respectively. Extensive experiments on BEIR, MIRACL, and LMEB demonstrate that KaLM-Reranker-V1 achieves strong reranking performance with superior efficiency. On BEIR, KaLM-Reranker-V1 achieves state-of-the-art performance, on par with strong industrial models such as the Qwen3-Reranker series; on MIRACL, despite not being extensively trained on multilingual data, KaLM-Reranker-V1 still shows excellent reranking performance. Moreover, on LMEB, reranking models demonstrate a clear advantage, with even the 0.27B Nano model remaining competitive with 7-12B embedding models.