This paper introduces Sparsified Late Interaction for Multi-vector (SLIM) retrieval with
inverted indexes. Multi-vector retrieval methods have demonstrated their effectiveness on …

The top-$ k $ operator returns a $ k $-sparse vector, where the non-zero values correspond
to the $ k $ largest values of the input. Unfortunately, because it is a discontinuous function …

Multi-vector retrieval models such as ColBERT [Khattab et al., 2020] allow token-level
interactions between queries and documents, and hence achieve state of the art on many …

Multi-vector retrieval methods combine the merits of sparse (eg BM25) and dense (eg DPR)
retrievers and have achieved state-of-the-art performance on various retrieval tasks. These …

Sparse representation learning based on Pre-trained Language Models has seen a growing
interest in Information Retrieval. Such approaches can take advantage of the proven …

For a given query generative retrieval generates identifiers of relevant documents in an end-
to-end manner using a sequence-to-sequence architecture. The relation between …

With the development of pre-trained language models, the dense retrieval models have
become promising alternatives to the traditional retrieval models that rely on exact match …

Dual encoder models are ubiquitous in modern classification and retrieval. Crucial for
training such dual encoders is an accurate estimation of gradients from the partition function …

The late interaction paradigm introduced with ColBERT stands out in the neural Information
Retrieval space, offering a compelling effectiveness-efficiency trade-off across many …

Lexical exact-match systems perform text retrieval efficiently with sparse matching signals
and fast retrieval through inverted lists, but naturally suffer from the mismatch between …