Incorporating Precedents for Legal Judgement Prediction on European Court of Human Rights Cases
This work addresses legal judgement prediction for European Court of Human Rights cases, offering an incremental improvement by incorporating precedents to enhance model accuracy.
The paper tackled the problem of legal judgement prediction (LJP) on European Court of Human Rights cases by integrating precedents, resulting in improved performance, especially for sparser articles, through methods like joint training and a precedent fusion module.
Inspired by the legal doctrine of stare decisis, which leverages precedents (prior cases) for informed decision-making, we explore methods to integrate them into LJP models. To facilitate precedent retrieval, we train a retriever with a fine-grained relevance signal based on the overlap ratio of alleged articles between cases. We investigate two strategies to integrate precedents: direct incorporation at inference via label interpolation based on case proximity and during training via a precedent fusion module using a stacked-cross attention model. We employ joint training of the retriever and LJP models to address latent space divergence between them. Our experiments on LJP tasks from the ECHR jurisdiction reveal that integrating precedents during training coupled with joint training of the retriever and LJP model, outperforms models without precedents or with precedents incorporated only at inference, particularly benefiting sparser articles.