GuRE:Generative Query REwriter for Legal Passage Retrieval

TL;DR

GuRE leverages large language models to rewrite legal queries, reducing vocabulary mismatch and significantly improving passage retrieval performance without retriever fine-tuning.

Contribution

Introduces GuRE, a novel query rewriting method using LLMs to enhance legal passage retrieval, outperforming baseline methods and offering versatile training objectives.

Findings

GuRE outperforms all baseline retrieval methods.

Different training objectives lead to distinct retrieval behaviors.

GuRE is more suitable than direct retriever fine-tuning for real-world use.

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.