Paper page - SPLADE-v3: New baselines for SPLADE (original) (raw)

Published on Mar 11, 2024

Abstract

SPLADE-v3 demonstrates superiority over BM25 and SPLADE++ with performance improvements and competitive results on cross-encoder re-rankers.

A companion to the release of the latest version of the SPLADE library. We describe changes to the training structure and present our latest series of models -- SPLADE-v3. We compare this new version to BM25, SPLADE++, as well as re-rankers, and showcase its effectiveness via a meta-analysis over more than 40 query sets. SPLADE-v3 further pushes the limit of SPLADE models: it is statistically significantly more effective than both BM25 and SPLADE++, while comparing well to cross-encoder re-rankers. Specifically, it gets more than 40MRR@10 on the MS MARCO dev set, and improves by 2% the out-of-domain results on the BEIR benchmark.