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

Feng Wang, Yuqing Li, Han Xiao

jina-reranker-v3: Last but Not Late Interaction for Document Reranking: 13 upvotes on Hugging Face Daily Papers, #35 of 61 papers on 2025-10-01. Day-by-day upvote history.

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.

Paper page on Hugging Face · arXiv

Data: hysts-bot-data/daily-papers-stats and the Daily Papers API. Open data: tardellirs/paper-pulse-data. Sister project: Model Pulse, the download history of every model on the Hub.