CoRECT: A Framework for Evaluating Embedding Compression Techniques at Scale

TL;DR

This paper introduces CoRECT, a comprehensive framework for evaluating embedding compression techniques at scale, addressing the impact of corpus complexity on dense retrieval performance and benchmarking eight compression methods.

Contribution

The paper presents CoRECT, a new large-scale evaluation framework with a curated dataset collection, enabling consistent comparison of embedding compression techniques across diverse corpus complexities.

Findings

Non-learned compression reduces index size significantly with minimal performance loss.

Performance of compression methods varies across models and datasets.

CoRECT facilitates informed selection of compression techniques.

Abstract

Dense retrieval systems have proven to be effective across various benchmarks, but require substantial memory to store large search indices. Recent advances in embedding compression show that index sizes can be greatly reduced with minimal loss in ranking quality. However, existing studies often overlook the role of corpus complexity -- a critical factor, as recent work shows that both corpus size and document length strongly affect dense retrieval performance. In this paper, we introduce CoRECT (Controlled Retrieval Evaluation of Compression Techniques), a framework for large-scale evaluation of embedding compression methods, supported by a newly curated dataset collection. To demonstrate its utility, we benchmark eight representative types of compression methods. Notably, we show that non-learned compression achieves substantial index size reduction, even on up to 100M passages, with statistically insignificant performance loss. However, selecting the optimal compression method remains challenging, as performance varies across models. Such variability highlights the necessity of CoRECT to enable consistent comparison and informed selection of compression methods. All code, data, and results are available on GitHub and HuggingFace.