IRCLApr 2, 2025

Efficient Constant-Space Multi-Vector Retrieval

arXiv:2504.01818v115 citationsh-index: 11ECIR
Originality Incremental advance
AI Analysis

This addresses storage efficiency for retrieval systems, but it is incremental as it builds on existing multi-vector methods.

The paper tackles the high storage cost of multi-vector retrieval methods like ColBERT by encoding documents into a fixed number of vectors, reducing storage and improving OS paging management, and experiments on MSMARCO and BEIR show it retains most effectiveness.

Multi-vector retrieval methods, exemplified by the ColBERT architecture, have shown substantial promise for retrieval by providing strong trade-offs in terms of retrieval latency and effectiveness. However, they come at a high cost in terms of storage since a (potentially compressed) vector needs to be stored for every token in the input collection. To overcome this issue, we propose encoding documents to a fixed number of vectors, which are no longer necessarily tied to the input tokens. Beyond reducing the storage costs, our approach has the advantage that document representations become of a fixed size on disk, allowing for better OS paging management. Through experiments using the MSMARCO passage corpus and BEIR with the ColBERT-v2 architecture, a representative multi-vector ranking model architecture, we find that passages can be effectively encoded into a fixed number of vectors while retaining most of the original effectiveness.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes