DISC-LawLLM: Fine-tuning Large Language Models for Intelligent Legal Services

Shengbin Yue,Wei Chen,Siyuan Wang,Bingxuan Li,Chenchen Shen,Shujun Liu,Yuxuan Zhou,Yao Xiao,Song Yun,Xuanjing Huang,Zhongyu Wei

2023-09-24

Abstract:We propose DISC-LawLLM, an intelligent legal system utilizing large language models (LLMs) to provide a wide range of legal services. We adopt legal syllogism prompting strategies to construct supervised fine-tuning datasets in the Chinese Judicial domain and fine-tune LLMs with legal reasoning capability. We augment LLMs with a retrieval module to enhance models' ability to access and utilize external legal knowledge. A comprehensive legal benchmark, DISC-Law-Eval, is presented to evaluate intelligent legal systems from both objective and subjective dimensions. Quantitative and qualitative results on DISC-Law-Eval demonstrate the effectiveness of our system in serving various users across diverse legal scenarios. The detailed resources are available at <a class="link-external link-https" href="https://github.com/FudanDISC/DISC-LawLLM" rel="external noopener nofollow">this https URL</a>.

Computation and Language

What problem does this paper attempt to address?

The paper aims to address the need for intelligent legal systems in the legal field, particularly by leveraging large language models (LLMs) to provide a wide range of legal services. Specifically, the paper proposes DISC-LawLLM, a large language model optimized for legal reasoning and knowledge retrieval capabilities. The research team attempts to solve the problem through the following points:

1. **Constructing a high-quality supervised fine-tuning dataset**: By adopting a legal syllogism prompting strategy, a supervised fine-tuning dataset, DISC-Law-SFT, was constructed within the Chinese judicial domain to train the model with legal reasoning capabilities.
2. **Enhancing retrieval and reasoning capabilities**: A retrieval module was introduced, enabling the model to access and utilize external legal knowledge, thereby improving its reliability and accuracy in practical applications.
3. **Designing a comprehensive evaluation benchmark**: A comprehensive legal benchmark test, DISC-Law-Eval, was proposed to evaluate intelligent legal systems from both objective and subjective dimensions.

Through these methods, the paper demonstrates the effectiveness of DISC-LawLLM in handling legal issues of varying difficulty levels and significantly outperforms existing large legal language models in multiple tasks.