ShizhenGPT
收藏资源简介:
ShizhenGPT是一个专门为传统中医学(TCM)定制的大型多模态语言模型。该数据集是迄今为止最大的TCM数据集,包含超过100GB的文本和200GB的多模态数据,包括120万张图像、200多个小时的音频和多种生理信号。该数据集通过领域特定的预训练和指令调整,使ShizhenGPT能够获得深度的TCM知识和多模态推理能力。该数据集的应用领域包括临床决策、医学教育和传统医学知识的保存。
ShizhenGPT is a large multimodal language model specifically customized for Traditional Chinese Medicine (TCM). This dataset is the largest TCM dataset to date, comprising over 100 GB of text data and 200 GB of multimodal data, including 1.2 million images, more than 200 hours of audio, and various physiological signals. Through domain-specific pre-training and instruction tuning, this dataset enables ShizhenGPT to acquire in-depth TCM knowledge and multimodal reasoning abilities. The application fields of this dataset cover clinical decision-making, medical education, and the preservation of traditional medical knowledge.

- 1通过香港中文大学(深圳) · 2025年



