SylvanL/ChatTCM-7B-SFT: A Specialized TCM LLM
SylvanL/ChatTCM-7B-SFT is a 7.6 billion parameter model developed by SylvanL, representing China's first fully open-source large language model dedicated to Traditional Chinese Medicine (TCM). It was fine-tuned on the SylvanL/Traditional-Chinese-Medicine-Dataset-SFT dataset over two epochs using the llamafactory framework, building upon the SylvanL/ChatTCM-7B-Pretrain base model.
Key Capabilities
- Classical Text Translation: Translates ancient Chinese medical texts into modern Chinese, aiding in the understanding of TCM classics.
- Clinical Diagnosis & Prescription: Emulates the diagnostic logic and prescription capabilities of mainstream TCM practitioners, analyzing patient cases to provide judgments and recommendations.
- TCM Knowledge Q&A: Offers comprehensive and reliable answers to questions across various TCM knowledge domains.
- Enhanced NLP for TCM: Improves fundamental natural language processing capabilities for TCM terminology, supporting tasks like named entity recognition, relation extraction, and synonym disambiguation.
Use Cases
This model is particularly well-suited for applications requiring deep understanding and generation of content related to Traditional Chinese Medicine. It can assist in:
- Translating historical medical documents.
- Providing diagnostic insights based on patient records.
- Answering complex TCM-related queries.
- Developing advanced NLP tools for the TCM sector.