Forward Index Compression for Learned Sparse Retrieval

TL;DR

This paper investigates efficient compression methods for the forward index in learned sparse text retrieval, introducing DotVByte to reduce storage without sacrificing retrieval speed or accuracy.

Contribution

It presents a novel compression algorithm, DotVByte, optimized for inner product calculations, improving space efficiency in sparse retrieval systems.

Findings

StreamVByte offers the best trade-off among memory, accuracy, and latency.

DotVByte achieves significant space savings.

The proposed methods maintain retrieval efficiency on MsMarco.

Abstract

Text retrieval using learned sparse representations of queries and documents has, over the years, evolved into a highly effective approach to search. It is thanks to recent advances in approximate nearest neighbor search-with the emergence of highly efficient algorithms such as the inverted index-based Seismic and the graph-based Hnsw-that retrieval with sparse representations became viable in practice. In this work, we scrutinize the efficiency of sparse retrieval algorithms and focus particularly on the size of a data structure that is common to all algorithmic flavors and that constitutes a substantial fraction of the overall index size: the forward index. In particular, we seek compression techniques to reduce the storage footprint of the forward index without compromising search quality or inner product computation latency. In our examination with various integer compression techniques, we report that StreamVByte achieves the best trade-off between memory footprint, retrieval accuracy, and latency. We then improve StreamVByte by introducing DotVByte, a new algorithm tailored to inner product computation. Experiments on MsMarco show that our improvements lead to significant space savings while maintaining retrieval efficiency.