jina-reranker-v3：文档重排序中的“最后但非最晚”交互机制

jina-reranker-v3: Last but Not Late Interaction for Document Reranking

September 29, 2025

作者: Feng Wang, Yuqing Li, Han Xiao

cs.AI

摘要

jina-reranker-v3 是一款拥有 0.6B 参数的多语言文档重排序模型，它引入了一种新颖的“最后但非延迟交互”机制。与 ColBERT 等延迟交互模型不同，后者先进行独立编码再进行多向量匹配，而我们的方法则在同一上下文窗口内对查询和文档执行因果自注意力，从而在从每个文档的最后一个令牌提取上下文嵌入之前，实现丰富的跨文档交互。这一紧凑架构在 BEIR 基准测试中取得了 61.94 nDCG@10 的顶尖性能，同时其规模仅为生成式列表重排序模型的十分之一。

English

jina-reranker-v3 is a 0.6B parameter multilingual document reranker that introduces a novel last but not late interaction. Unlike late interaction models such as ColBERT that perform separate encoding followed by multi-vector matching, our approach conducts causal self-attention between query and documents within the same context window, enabling rich cross-document interactions before extracting contextual embeddings from the last token of each document. This compact architecture achieves state-of-the-art BEIR performance with 61.94 nDCG@10 while being ten times smaller than generative listwise rerankers.