Hybrid Inverted Index Is a Robust Accelerator for Dense Retrieval

TL;DR

The paper introduces the Hybrid Inverted Index (HI$^2$), a novel approach combining embedding clustering and salient term matching to enhance dense retrieval speed and accuracy without quality loss.

Contribution

It proposes a collaborative framework using embedding clusters and salient terms, with learned selectors, to improve dense retrieval efficiency and effectiveness.

Findings

Achieves lossless retrieval quality with competitive efficiency

Effectively combines clustering and lexical matching for dense retrieval

Demonstrates superior performance on popular benchmarks

Abstract

Inverted file structure is a common technique for accelerating dense retrieval. It clusters documents based on their embeddings; during searching, it probes nearby clusters w.r.t. an input query and only evaluates documents within them by subsequent codecs, thus avoiding the expensive cost of exhaustive traversal. However, the clustering is always lossy, which results in the miss of relevant documents in the probed clusters and hence degrades retrieval quality. In contrast, lexical matching, such as overlaps of salient terms, tends to be strong feature for identifying relevant documents. In this work, we present the Hybrid Inverted Index (HI$^2$), where the embedding clusters and salient terms work collaboratively to accelerate dense retrieval. To make best of both effectiveness and efficiency, we devise a cluster selector and a term selector, to construct compact inverted lists and efficiently searching through them. Moreover, we leverage simple unsupervised algorithms as well as end-to-end knowledge distillation to learn these two modules, with the latter further boosting the effectiveness. Based on comprehensive experiments on popular retrieval benchmarks, we verify that clusters and terms indeed complement each other, enabling HI$^2$ to achieve lossless retrieval quality with competitive efficiency across various index settings. Our code and checkpoint are publicly available at https://github.com/namespace-Pt/Adon/tree/HI2.