GuRE:Generative Query REwriter for Legal Passage Retrieval

2025-05-19

AI Summary

Legal paragraph retrieval (LPR) suffers from severe lexical mismatch between queries and paragraphs. To address this, we propose GuRE, a generative query rewriting method that introduces lightweight large language models (LLMs) to legal retrieval for the first time. GuRE employs supervised query rewriting coupled with instruction tuning—achieving retrieval-agnostic performance gains without end-to-end retriever fine-tuning. It leverages a retrieval-augmented evaluation framework and consistently outperforms state-of-the-art baselines across multiple legal retrieval benchmarks, demonstrating strong generalization and practical deployability. Our key contributions are: (1) pioneering the application of LLM-driven generative rewriting to LPR; (2) revealing how distinct training objectives differentially influence retrieval behavior; and (3) releasing an open-source, efficient, plug-and-play query rewriting solution.

Technology Category

• Natural Language Processing: Generation
• Machine Learning: Large Multimodal Models (LMMs)
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Large language models for search
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Legal Passage Retrieval (LPR) systems are crucial as they help practitioners save time when drafting legal arguments. However, it remains an underexplored avenue. One primary reason is the significant vocabulary mismatch between the query and the target passage. To address this, we propose a simple yet effective method, the Generative query REwriter (GuRE). We leverage the generative capabilities of Large Language Models (LLMs) by training the LLM for query rewriting."Rewritten queries"help retrievers to retrieve target passages by mitigating vocabulary mismatch. Experimental results show that GuRE significantly improves performance in a retriever-agnostic manner, outperforming all baseline methods. Further analysis reveals that different training objectives lead to distinct retrieval behaviors, making GuRE more suitable than direct retriever fine-tuning for real-world applications. Codes are avaiable at github.com/daehuikim/GuRE.

Problem

• Addresses vocabulary mismatch in legal retrieval queries
• Improves legal passage retrieval using generative query rewriting
• Enhances retriever performance without direct fine-tuning

Innovation

• Generative query rewriting using LLMs
• Mitigates vocabulary mismatch in retrieval
• Retriever-agnostic performance improvement