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

• DOI: 10.48550/arxiv.2309.11325
• Publication Date: 2023-01-01
• Authors: Shengbin Yue; Wei Chen; Siyuan Wang; Bingxuan Li; Chenchen Shen; Shujun Liu; Yuxuan Zhou; Yao Xiao; Song Yun; Wei Lin; Xuanjing Huang; Zhongyu Wei
• Keywords: Benchmark (surveying); Legal case; Plain language

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