eagerlearners at SemEval2024 Task 5: The Legal Argument Reasoning Task in Civil Procedure

TL;DR

This paper evaluates the effectiveness of zero-shot large language models in classifying complex legal case data from U.S. civil procedure, achieving a maximum F1 score of 64%.

Contribution

It provides a comparative analysis of zero-shot and fine-tuned models on legal datasets, highlighting their capabilities and limitations.

Findings

Zero-shot models achieved up to 64% F1 score.

Large input token models perform competitively on legal data.

Legal datasets pose significant challenges for language models.

Abstract

This study investigates the performance of the zero-shot method in classifying data using three large language models, alongside two models with large input token sizes and the two pre-trained models on legal data. Our main dataset comes from the domain of U.S. civil procedure. It includes summaries of legal cases, specific questions, potential answers, and detailed explanations for why each solution is relevant, all sourced from a book aimed at law students. By comparing different methods, we aimed to understand how effectively they handle the complexities found in legal datasets. Our findings show how well the zero-shot method of large language models can understand complicated data. We achieved our highest F1 score of 64% in these experiments.