Beyond Borders: Investigating Cross-Jurisdiction Transfer in Legal Case Summarization

TL;DR

The paper tackles cross-jurisdiction transfer in legal case summarization, focusing on zero-shot transfer to target jurisdictions lacking reference summaries. It compares unsupervised extractive methods, supervised transfer with various backbones, adversarial domain adaptation, and silver-summary augmentation across three jurisdictional datasets, revealing that jurisdictional similarity and pre-training drive transfer success. Adversarial training improves generalization for general-purpose backbones but may cause representation erasure in legal pre-trained models, while incorporating silver summaries mitigates this effect and boosts performance, especially for extractive datasets. The findings guide practical design of adaptable legal summarizers across jurisdictions, suggesting when to rely on source similarity, pre-trained models, or target-scope augmentation to achieve robust cross-jurisdiction performance.

Abstract

Legal professionals face the challenge of managing an overwhelming volume of lengthy judgments, making automated legal case summarization crucial. However, prior approaches mainly focused on training and evaluating these models within the same jurisdiction. In this study, we explore the cross-jurisdictional generalizability of legal case summarization models.Specifically, we explore how to effectively summarize legal cases of a target jurisdiction where reference summaries are not available. In particular, we investigate whether supplementing models with unlabeled target jurisdiction corpus and extractive silver summaries obtained from unsupervised algorithms on target data enhances transfer performance. Our comprehensive study on three datasets from different jurisdictions highlights the role of pre-training in improving transfer performance. We shed light on the pivotal influence of jurisdictional similarity in selecting optimal source datasets for effective transfer. Furthermore, our findings underscore that incorporating unlabeled target data yields improvements in general pre-trained models, with additional gains when silver summaries are introduced. This augmentation is especially valuable when dealing with extractive datasets and scenarios featuring limited alignment between source and target jurisdictions. Our study provides key insights for developing adaptable legal case summarization systems, transcending jurisdictional boundaries.