KaLM-Reranker-V1：用于压缩文档重新排序的快速但非延迟交互方法
KaLM-Reranker-V1: Fast but Not Late Interaction for Compressed Document Reranking
摘要
随着检索系统的规模扩大，高质量重新排序变得愈发重要。不过，目前大多数重新排序算法——无论是基于编码器还是解码器的算法——都会同时对查询和文档进行编码处理，这种处理方式会限制算法的效率以及灵活性。我们提出了KaLM-Reranker-V1这一快速但非延迟交互的重新排序算法。该算法能够将查询和文档的编码过程分离开来，同时仍然能够实现有效的相关性建模。KaLM-Reranker-V1采用编码器-解码器架构：编码器用于通过Matryoshka嵌入池化技术对文档进行预编码，而解码器则负责处理系统指令、用户指令以及查询意图；交叉注意力机制则用于捕捉查询上下文与文档表示之间的相关性。这种设计使得KaLM-Reranker-V1在保持丰富的相关性建模能力的同时，又能实现高效的运行。我们提供了三种不同规模的KaLM-Reranker-V1模型：Nano、Small和Large，对应的激活参数分别为0.27B、1B和4B。在BEIR、MIRACL和LMEB数据集上的大量实验表明，KaLM-Reranker-V1具有出色的重新排序性能，且效率也相当高。在BEIR数据集上，KaLM-Reranker-V1的性能达到了最先进水平，与Qwen3-Reranker系列等优秀工业模型相当；在MIRACL数据集上，尽管它没有经过多语言数据的充分训练，但KaLM-Reranker-V1仍然表现出优秀的重新排序性能。而在LMEB数据集上，重新排序模型具有明显优势，即使是0.27B参数的Nano模型也能与7-12B参数的模型相媲美。
English Abstract
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.