The Craft of Selective Prediction: Towards Reliable Case Outcome Classification -- An Empirical Study on European Court of Human Rights Cases

TL;DR

This empirical study investigates how different design choices affect the reliability of legal case outcome classification models, emphasizing confidence estimation and model calibration in high-stakes legal NLP tasks.

Contribution

It systematically explores the impact of pre-training data, model size, and confidence estimation methods on model reliability in legal NLP, a novel focus in this domain.

Findings

Diverse, domain-specific pre-training improves calibration.

Larger models tend to be overconfident.

Monte Carlo dropout provides reliable confidence estimates.

Abstract

In high-stakes decision-making tasks within legal NLP, such as Case Outcome Classification (COC), quantifying a model's predictive confidence is crucial. Confidence estimation enables humans to make more informed decisions, particularly when the model's certainty is low, or where the consequences of a mistake are significant. However, most existing COC works prioritize high task performance over model reliability. This paper conducts an empirical investigation into how various design choices including pre-training corpus, confidence estimator and fine-tuning loss affect the reliability of COC models within the framework of selective prediction. Our experiments on the multi-label COC task, focusing on European Court of Human Rights (ECtHR) cases, highlight the importance of a diverse yet domain-specific pre-training corpus for better calibration. Additionally, we demonstrate that larger models tend to exhibit overconfidence, Monte Carlo dropout methods produce reliable confidence estimates, and confident error regularization effectively mitigates overconfidence. To our knowledge, this is the first systematic exploration of selective prediction in legal NLP. Our findings underscore the need for further research on enhancing confidence measurement and improving the trustworthiness of models in the legal domain.