LegiLM: A Fine-Tuned Legal Language Model for Data Compliance
Linkai Zhu, Lu Yang, Chaofan Li, Shanwen Hu, Lu Liu, Bin Yin
TL;DR
LegiLM tackles GDPR compliance detection for data-sharing and privacy policies by fine-tuning a base English legal LLM derived from SaulLM-7B with GDPR texts, contracts, and policy data. It integrates instruction-based fine-tuning, retrieval-augmented reasoning, and contrastive learning to enhance legal accuracy and output diversity. In custom benchmarking, LegiLM achieves an LQA accuracy of 68.05% and justification quality of 68.21%, surpassing several baselines and demonstrating reliable breach detection with sound legal justifications. The work provides public resources and a concrete dataset to support AI-assisted legal consulting and automated compliance analysis in practice.
Abstract
Ensuring compliance with international data protection standards for privacy and data security is a crucial but complex task, often requiring substantial legal expertise. This paper introduces LegiLM, a novel legal language model specifically tailored for consulting on data or information compliance. LegiLM leverages a pre-trained GDPR Fines dataset and has been fine-tuned to automatically assess whether particular actions or events breach data security and privacy regulations. By incorporating a specialized dataset that includes global data protection laws, meticulously annotated policy documents, and relevant privacy policies, LegiLM is optimized for addressing data compliance challenges. The model integrates advanced legal reasoning methods and information retrieval enhancements to enhance accuracy and reliability in practical legal consulting scenarios. Our evaluation using a custom benchmark dataset demonstrates that LegiLM excels in detecting data regulation breaches, offering sound legal justifications, and recommending necessary compliance modifications, setting a new benchmark for AI-driven legal compliance solutions. Our resources are publicly available at https://github.com/DAOLegalAI/LegiLM