LaWGPT

LaWGPT is a series of open source language models based on Chinese legal knowledge.

This series of models expands the domain-specific vocabulary and large-scale Chinese legal corpus pre-training on the basis of general Chinese base models (such as Chinese-LLaMA, ChatGLM, etc.), enhancing the basic semantic understanding of the large model in the legal field. On this basis, the legal domain dialogue Q&A dataset and the Chinese judicial examination dataset are instructed and fine-tuned to enhance the model's understanding and execution ability of legal content.

For more detailed information, please refer to the technical report.