# SLIM: Sparsified Late Interaction for Multi-Vector Retrieval with Inverted Indexes

arXiv: https://arxiv.org/abs/2302.06587 · HTML: https://mlsanity.com/paper/2302.06587/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 13 Feb 2023.

## Built on by (most cited first, 12 of 12)

- [Resources for Brewing BEIR: Reproducible Reference Models and an Official Leaderboard](https://mlsanity.com/paper/2306.07471/) (2023)
- [CSPLADE: Learned Sparse Retrieval with Causal Language Models](https://mlsanity.com/paper/2504.10816/) (2025)
- [Sculpting the Vector Space: Towards Efficient Multi-Vector Visual Document Retrieval via Prune-then-Merge Framework](https://mlsanity.com/paper/2602.19549/) (2026)
- [Generative Retrieval as Multi-Vector Dense Retrieval](https://mlsanity.com/paper/2404.00684/) (2024)
- [Efficient Document Ranking with Learnable Late Interactions](https://mlsanity.com/paper/2406.17968/) (2024)
- [Exp4Fuse: A Rank Fusion Framework for Enhanced Sparse Retrieval using Large Language Model-based Query Expansion](https://mlsanity.com/paper/2506.04760/) (2025)
- [Distillation for Multilingual Information Retrieval](https://mlsanity.com/paper/2405.00977/) (2024)
- [SPRINT: A Unified Toolkit for Evaluating and Demystifying Zero-shot Neural Sparse Retrieval](https://mlsanity.com/paper/2307.10488/) (2023)
- [Towards Lossless Token Pruning in Late-Interaction Retrieval Models](https://mlsanity.com/paper/2504.12778/) (2025)
- [An Efficient Embedding Based Ad Retrieval with GPU-Powered Feature Interaction](https://mlsanity.com/paper/2511.22460/) (2025)
- [ViSAR: Training-Free Adaptive-k Retrieval for Visual Document Question Answering](https://mlsanity.com/paper/2609.02486/) (2026)
- [Beyond Relevance: Structured Semantic Supervision for Product Search with LLM-Augmented Annotations](https://mlsanity.com/paper/2609.23646/) (2026)

## Abstract (arXiv)

This paper introduces Sparsified Late Interaction for Multi-vector (SLIM) retrieval with inverted indexes. Multi-vector retrieval methods have demonstrated their effectiveness on various retrieval datasets, and among them, ColBERT is the most established method based on the late interaction of contextualized token embeddings of pre-trained language models. However, efficient ColBERT implementations require complex engineering and cannot take advantage of off-the-shelf search libraries, impeding their practical use. To address this issue, SLIM first maps each contextualized token vector to a sparse, high-dimensional lexical space before performing late interaction between these sparse token embeddings. We then introduce an efficient two-stage retrieval architecture that includes inverted index retrieval followed by a score refinement module to approximate the sparsified late interaction, which is fully compatible with off-the-shelf lexical search libraries such as Lucene. SLIM achieves competitive accuracy on MS MARCO Passages and BEIR compared to ColBERT while being much smaller and faster on CPUs. To our knowledge, we are the first to explore using sparse token representations for multi-vector retrieval. Source code and data are integrated into the Pyserini IR toolkit.