STRONG -- Structure Controllable Legal Opinion Summary Generation

TL;DR

This paper introduces STRONG, a method for generating structured legal opinion summaries by leveraging argument role predictions to produce coherent, structure-controlled summaries that outperform existing baselines.

Contribution

It presents a novel approach for structure controllable summarization of legal opinions using argument role guidance, improving coherence and structure adherence.

Findings

Outperforms strong baselines on ROUGE and BERTScore metrics

Achieves higher structure similarity in summaries

Effective in long legal opinion summarization

Abstract

We propose an approach for the structure controllable summarization of long legal opinions that considers the argument structure of the document. Our approach involves using predicted argument role information to guide the model in generating coherent summaries that follow a provided structure pattern. We demonstrate the effectiveness of our approach on a dataset of legal opinions and show that it outperforms several strong baselines with respect to ROUGE, BERTScore, and structure similarity.