Exploiting Contrastive Learning and Numerical Evidence for Confusing Legal Judgment Prediction
This work addresses a core challenge in legal AI by improving prediction accuracy for confusing cases, though it is incremental as it builds on existing contrastive learning and numeracy techniques.
The paper tackles the problem of distinguishing confusing legal cases with subtle text differences in legal judgment prediction by proposing a supervised contrastive learning method and enhancing representations with numerical evidence, achieving new state-of-the-art results on public benchmarks.
Given the fact description text of a legal case, legal judgment prediction (LJP) aims to predict the case's charge, law article and penalty term. A core problem of LJP is how to distinguish confusing legal cases, where only subtle text differences exist. Previous studies fail to distinguish different classification errors with a standard cross-entropy classification loss, and ignore the numbers in the fact description for predicting the term of penalty. To tackle these issues, in this work, first, we propose a moco-based supervised contrastive learning to learn distinguishable representations, and explore the best strategy to construct positive example pairs to benefit all three subtasks of LJP simultaneously. Second, in order to exploit the numbers in legal cases for predicting the penalty terms of certain cases, we further enhance the representation of the fact description with extracted crime amounts which are encoded by a pre-trained numeracy model. Extensive experiments on public benchmarks show that the proposed method achieves new state-of-the-art results, especially on confusing legal cases. Ablation studies also demonstrate the effectiveness of each component.