RELexED: Retrieval-Enhanced Legal Summarization with Exemplar Diversity
This work addresses legal summarization for practitioners needing concise summaries, but it is incremental as it builds on retrieval-augmented methods with a focus on diversity.
The paper tackled the problem of content theme deviation and inconsistent writing styles in legal summarization by proposing RELexED, a retrieval-augmented framework that uses a two-stage exemplar selection strategy with diversity balancing, which significantly outperformed models without exemplars or with similarity-only selection on two legal datasets.
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.