Archimedes-AUEB at SemEval-2024 Task 5: LLM explains Civil Procedure

TL;DR

This paper presents a novel approach using a teacher-student LLM framework with human-grounded explanations and data augmentation to improve legal argument reasoning in civil procedure, achieving competitive results.

Contribution

It introduces a method to generate human-grounded explanations and synthetic data for fine-tuning small LLMs in legal reasoning tasks, surpassing previous approaches.

Findings

Outperforms its own teacher in explanation quality

Achieved 15th place in SemEval-2024 Task 5

Generated explanations aligned with human legal analyses

Abstract

The SemEval task on Argument Reasoning in Civil Procedure is challenging in that it requires understanding legal concepts and inferring complex arguments. Currently, most Large Language Models (LLM) excelling in the legal realm are principally purposed for classification tasks, hence their reasoning rationale is subject to contention. The approach we advocate involves using a powerful teacher-LLM (ChatGPT) to extend the training dataset with explanations and generate synthetic data. The resulting data are then leveraged to fine-tune a small student-LLM. Contrary to previous work, our explanations are not directly derived from the teacher's internal knowledge. Instead they are grounded in authentic human analyses, therefore delivering a superior reasoning signal. Additionally, a new `mutation' method generates artificial data instances inspired from existing ones. We are publicly releasing the explanations as an extension to the original dataset, along with the synthetic dataset and the prompts that were used to generate both. Our system ranked 15th in the SemEval competition. It outperforms its own teacher and can produce explanations aligned with the original human analyses, as verified by legal experts.