QLU-NLP/BianCang-Qwen2-7B
BianCang-Qwen2-7B is a 7.6 billion parameter Traditional Chinese Medicine (TCM) large language model developed by QLU-NLP, built upon the Qwen2 base architecture. It is specifically fine-tuned for TCM tasks, excelling in syndrome differentiation, disease diagnosis, and medical license examinations. The model utilizes a two-stage training method, injecting domain knowledge and then activating and aligning it, to achieve state-of-the-art performance in specialized TCM applications.
Loading preview...
Overview
QLU-NLP/BianCang-Qwen2-7B is a 7.6 billion parameter Traditional Chinese Medicine (TCM) large language model, part of the BianCang series, developed by the Natural Language Processing and Cognitive Computing team at Qilu University of Technology (Shandong Academy of Sciences). Built on the Qwen2 base model, it employs a two-stage training approach: initial injection of TCM domain knowledge followed by knowledge activation and alignment. This specialized training enables BianCang to achieve advanced performance in core TCM tasks and medical license examinations.
Key Capabilities
- TCM Syndrome Differentiation: Demonstrates high accuracy in identifying TCM syndromes.
- TCM Disease Diagnosis: Excels in diagnosing diseases based on TCM principles.
- Medical License Examinations: Shows strong performance in various medical license exams, including MLEC-TCM and MLEC-CWM.
- Dialogue and Consultation: Designed to assist with TCM dialectics, prescription recommendations, knowledge Q&A, and general consultation.
Good for
- Assisting Medical Professionals: Aids doctors in disease diagnosis within the TCM framework.
- Patient Self-Assessment: Supports patients in understanding their health conditions from a TCM perspective.
- TCM Research and Development: Provides a robust foundation for further research and application of large language models in traditional Chinese medicine.