RELexED: Retrieval-Enhanced Legal Summarization with Exemplar Diversity

TL;DR

RELexED is a retrieval-augmented legal summarization framework that uses diverse exemplar summaries to improve the quality and coherence of generated summaries, addressing issues of content deviation and style inconsistency.

Contribution

It introduces a novel two-stage exemplar selection strategy using determinantal point processes and influence functions for legal summarization.

Findings

RELexED outperforms baseline models without exemplars.

The exemplar selection method improves summary coherence.

Significant performance gains on legal summarization datasets.

Abstract

This paper addresses the task of legal summarization, which involves distilling complex legal documents into concise, coherent summaries. Current approaches often struggle with content theme deviation and inconsistent writing styles due to their reliance solely on source documents. We propose RELexED, a retrieval-augmented framework that utilizes exemplar summaries along with the source document to guide the model. RELexED employs a two-stage exemplar selection strategy, leveraging a determinantal point process to balance the trade-off between similarity of exemplars to the query and diversity among exemplars, with scores computed via influence functions. Experimental results on two legal summarization datasets demonstrate that RELexED significantly outperforms models that do not utilize exemplars and those that rely solely on similarity-based exemplar selection.