ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERT

今天分享的是 2020 SIGIR 会议论文:ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERT,论文地址:https://arxiv.org/pdf/2004.12832.pdf

 

在召回模块常用到的模型结构为双塔模型,即DSSM模型,该模型对query和condidate分别建模,可以在离线阶段很方便的获取query和condidate的Embedding表征。结构为

 

你可能感兴趣的:(深度学习)