A Brief Comparison of Training-Free Multi-Vector Sequence Compression Methods
This work addresses a practical deployment issue for multi-vector retrieval systems, but it is incremental as it compares existing methods without introducing new techniques.
The paper tackled the problem of large index sizes in multi-vector retrieval models by evaluating training-free compression methods, finding that token merging outperforms token pruning in reducing index size while maintaining retrieval effectiveness.
While multi-vector retrieval models outperform single-vector models of comparable size in retrieval quality, their practicality is limited by substantially larger index sizes, driven by the additional sequence-length dimension in their document embeddings. Because document embedding size dictates both memory overhead and query latency, compression is essential for deployment. In this work, we present an evaluation of training-free methods targeting the token sequence length, a dimension unique to multi-vector retrieval. Our findings suggest that token merging is strictly superior to token pruning for reducing index size while maintaining retrieval effectiveness.