STRONG -- Structure Controllable Legal Opinion Summary Generation
This addresses the need for structured legal document summarization, which is incremental as it builds on existing summarization methods with domain-specific structural control.
The paper tackles the problem of generating structured summaries for long legal opinions by using predicted argument roles to guide summary generation according to provided structure patterns, demonstrating improved performance over baselines on metrics including ROUGE, BERTScore, and structure similarity.
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.