[Paper Review] A Short Survey of Viewing Large Language Models in Legal Aspect

Zhongxiang Sun|arXiv (Cornell University)|Mar 16, 2023

Artificial Intelligence in Law37 citations

TL;DR

This survey reviews how large language models (LLMs) are applied in legal tasks, discusses associated legal challenges, and examines data resources for domain adaptation, offering future directions for responsible use in law.

ABSTRACT

Large language models (LLMs) have transformed many fields, including natural language processing, computer vision, and reinforcement learning. These models have also made a significant impact in the field of law, where they are being increasingly utilized to automate various legal tasks, such as legal judgement prediction, legal document analysis, and legal document writing. However, the integration of LLMs into the legal field has also raised several legal problems, including privacy concerns, bias, and explainability. In this survey, we explore the integration of LLMs into the field of law. We discuss the various applications of LLMs in legal tasks, examine the legal challenges that arise from their use, and explore the data resources that can be used to specialize LLMs in the legal domain. Finally, we discuss several promising directions and conclude this paper. By doing so, we hope to provide an overview of the current state of LLMs in law and highlight the potential benefits and challenges of their integration.

Motivation & Objective

• Summarize how LLMs are applied to legal tasks such as judgment prediction, document analysis, and writing.
• Analyze legal problems arising from LLM use, including privacy, bias, and explainability.
• Identify data resources and domain-specific datasets to specialize LLMs for law.
• Discuss future directions and recommendations for responsible integration of LLMs in legal practice.

Proposed method

• Classify existing literature on LLM applications in law into applications, legal problems, and data resources.
• Compare with prior surveys to highlight focus on LLMs in legal contexts.
• Discuss case studies and prompting techniques that impact performance in legal tasks.

Experimental results

Research questions

• RQ1 What are the main applications of LLMs in legal tasks such as judgment prediction, document analysis, and writing?
• RQ2 What legal problems (privacy, bias, explainability) arise from deploying LLMs in law, and how can they be mitigated?
• RQ3 What data resources and domain-pretraining approaches exist to specialize LLMs for the legal domain?
• RQ4 What future research directions are needed to responsibly integrate LLMs into legal practice?

Key findings

• LLMs show promise in legal judgment prediction, statutory reasoning, and legal education tasks.
• Prompting techniques (e.g., few-shot, Chain-of-Thought) enhance legal reasoning and task performance.
• Legal challenges include privacy, bias, and explainability, requiring collaboration between researchers and policymakers.
• Domain-specific datasets (e.g., CAIL2018, CaseHOLD, LeCaRD) enable better legal specialization of LLMs.
• The paper highlights the need for guidelines, transparency, and ethical considerations in deploying LLMs in law.