Interpretable Charge Predictions for Criminal Cases: Learning to Generate Court Views from Fact Descriptions
This addresses the need for interpretable legal AI systems, though it is incremental by adapting existing methods to a specific domain.
The paper tackles the problem of generating court views from fact descriptions in criminal cases to improve interpretability of charge prediction systems, achieving effective results with a label-conditioned Seq2Seq model.
In this paper, we propose to study the problem of COURT VIEW GENeration from the fact description in a criminal case. The task aims to improve the interpretability of charge prediction systems and help automatic legal document generation. We formulate this task as a text-to-text natural language generation (NLG) problem. Sequenceto-sequence model has achieved cutting-edge performances in many NLG tasks. However, due to the non-distinctions of fact descriptions, it is hard for Seq2Seq model to generate charge-discriminative court views. In this work, we explore charge labels to tackle this issue. We propose a label-conditioned Seq2Seq model with attention for this problem, to decode court views conditioned on encoded charge labels. Experimental results show the effectiveness of our method.