Unified and Efficient Approach for Multi-Vector Similarity Search

TL;DR

This paper introduces MV-HNSW, a native hierarchical graph index for multi-vector similarity search, significantly improving efficiency and recall in semantic retrieval tasks.

Contribution

The paper presents MV-HNSW, the first native multi-vector index with a novel edge-weight function and search strategy, outperforming existing filter-and-refine methods.

Findings

Achieves over 90% recall on seven datasets.

Reduces search latency by up to 14 times.

Outperforms existing methods in state-of-the-art performance.

Abstract

Multi-Vector Similarity Search is essential for fine-grained semantic retrieval in many real-world applications, offering richer representations than traditional single-vector paradigms. Due to the lack of native multi-vector index, existing methods rely on a filter-and-refine framework built upon single-vector indexes. By treating token vectors within each multi-vector object in isolation and ignoring their correlations, these methods face an inherent dilemma: aggressive filtering sacrifices recall, while conservative filtering incurs prohibitive computational cost during refinement. To address this limitation, we propose MV-HNSW, the first native hierarchical graph index designed for multi-vector data. MV-HNSW introduces a novel edge-weight function that satisfies essential properties (symmetry, cardinality robustness, and query consistency) for graph-based indexing, an accelerated multi-vector similarity computation algorithm, and an augmented search strategy that dynamically discovers topologically disconnected yet relevant candidates. Extensive experiments on seven real-world datasets show that MV-HNSW achieves state-of-the-art search performance, maintaining over 90% recall while reducing search latency by up to 14.0$\times$ compared to existing methods.