[2N1-GS-4] Web intelligence:

[2N1-GS-4-02] Evaluating Late-interaction Approaches with Trained Text Embedding Models for Re-ranking

〇Yu Hatsushika¹, Kunihiro Takeoka², Masafumi Oyamada², Chihiro Shibata¹ (1. Hosei University Graduate School of Science and Engineering, 2. NEC Data Science Laboratory)

Keywords:search, reranking, deep learning

Reranking is to rank thousands of documents retrieved by a search engine from a large number of documents
by their relevance to a query. Late interaction, which independently encodes queries and documents before eval-
uating token-level interactions, shows promise in retrieval tasks; however, its application in reranking has been
less explored. This paper investigates whether late-interaction approaches work well on reranking and how much
influence pre-trained models, especially trained sentence-embedding models, are used in the late-interaction. Our
experiments show that late interaction is one of the best options for reranking on accuracy and latency perspec-
tives, and in in-domain settings, the late interaction with trained sentence-embedding models mostly overperforms
it with pre-trained language models.