Au sujet de ce document
- sl:arxiv_author :
- sl:arxiv_firstAuthor : Thibault Formal
- sl:arxiv_num : 2404.13950
- sl:arxiv_published : 2024-04-22T07:51:13Z
- sl:arxiv_summary : The late interaction paradigm introduced with ColBERT stands out in the
neural Information Retrieval space, offering a compelling
effectiveness-efficiency trade-off across many benchmarks. Efficient late
interaction retrieval is based on an optimized multi-step strategy, where an
approximate search first identifies a set of candidate documents to re-rank
exactly. In this work, we introduce SPLATE, a simple and lightweight adaptation
of the ColBERTv2 model which learns an ``MLM adapter'', mapping its frozen
token embeddings to a sparse vocabulary space with a partially learned SPLADE
module. This allows us to perform the candidate generation step in late
interaction pipelines with traditional sparse retrieval techniques, making it
particularly appealing for running ColBERT in CPU environments. Our SPLATE
ColBERTv2 pipeline achieves the same effectiveness as the PLAID ColBERTv2
engine by re-ranking 50 documents that can be retrieved under 10ms.@en
- sl:arxiv_title : SPLATE: Sparse Late Interaction Retrieval@en
- sl:arxiv_updated : 2024-04-22T07:51:13Z
- sl:bookmarkOf : https://arxiv.org/abs/2404.13950
- sl:creationDate : 2024-04-23
- sl:creationTime : 2024-04-23T23:13:59Z
Documents with similar tags (experimental)
Préférences...