Robust Deep Reinforcement Learning for Extractive Legal Summarization

Abstract

Automatic summarization of legal texts is an important and still a challenging task since legal documents are often long and complicated with unusual structures and styles. Recent advances of deep models trained end-to-end with differentiable losses can well-summarize natural text, yet when applied to legal domain, they show limited results. In this paper, we propose to use reinforcement learning to train current deep summarization models to improve their performance on the legal domain. To this end, we adopt proximal policy optimization methods and introduce novel reward functions that encourage the generation of candidate summaries satisfying both lexical and semantic criteria. We apply our method to training different summarization backbones and observe a consistent and significant performance gain across 3 public legal datasets.

References in corpus (7)

• Adam: A Method for Stochastic Optimization
• Fine-Tuning Language Models from Human Preferences
• Text Summarization with Pretrained Encoders
• BillSum: A Corpus for Automatic Summarization of US Legislation
• Learning to summarize from human feedback
• Plain English Summarization of Contracts
• Deep Reinforcement Learning with Distributional Semantic Rewards for Abstractive Summarization