Optimizing Test-Time Query Representations for Dense Retrieval

TL;DR

This paper introduces TOUR, a method that optimizes query representations at test time for dense retrieval, significantly enhancing retrieval accuracy and efficiency by leveraging test-time signals and pseudo-labels.

Contribution

TOUR is a novel test-time optimization technique that refines query representations using pseudo-labels from a re-ranker, generalizing the Rocchio algorithm for improved dense retrieval.

Findings

Improves open-domain question answering accuracy.

Enhances passage retrieval performance.

Increases re-ranking speed by 1.3-2.4x.

Abstract

Recent developments of dense retrieval rely on quality representations of queries and contexts from pre-trained query and context encoders. In this paper, we introduce TOUR (Test-Time Optimization of Query Representations), which further optimizes instance-level query representations guided by signals from test-time retrieval results. We leverage a cross-encoder re-ranker to provide fine-grained pseudo labels over retrieval results and iteratively optimize query representations with gradient descent. Our theoretical analysis reveals that TOUR can be viewed as a generalization of the classical Rocchio algorithm for pseudo relevance feedback, and we present two variants that leverage pseudo-labels as hard binary or soft continuous labels. We first apply TOUR on phrase retrieval with our proposed phrase re-ranker, and also evaluate its effectiveness on passage retrieval with an off-the-shelf re-ranker. TOUR greatly improves end-to-end open-domain question answering accuracy, as well as passage retrieval performance. TOUR also consistently improves direct re-ranking by up to 2.0% while running 1.3-2.4x faster with an efficient implementation.