ThinkQE: Query Expansion via an Evolving Thinking Process

TL;DR

ThinkQE introduces a novel query expansion framework that enhances web search by promoting semantic exploration and iterative refinement, leading to improved retrieval performance across multiple benchmarks.

Contribution

It presents a test-time expansion method combining a thinking process with corpus feedback, addressing the narrow focus of existing LLM-based approaches.

Findings

Outperforms prior methods on DL19, DL20, and BRIGHT benchmarks.

Enhances diversity and exploration in query expansion.

Demonstrates effectiveness without additional training.

Abstract

Effective query expansion for web search benefits from promoting both exploration and result diversity to capture multiple interpretations and facets of a query. While recent LLM-based methods have improved retrieval performance and demonstrate strong domain generalization without additional training, they often generate narrowly focused expansions that overlook these desiderata. We propose ThinkQE, a test-time query expansion framework addressing this limitation through two key components: a thinking-based expansion process that encourages deeper and comprehensive semantic exploration, and a corpus-interaction strategy that iteratively refines expansions using retrieval feedback from the corpus. Experiments on diverse web search benchmarks (DL19, DL20, and BRIGHT) show ThinkQE consistently outperforms prior approaches, including training-intensive dense retrievers and rerankers.