QLU-NLP/BianCang-Qwen2.5-7B-Instruct
BianCang-Qwen2.5-7B-Instruct is a 7.6 billion parameter instruction-tuned large language model developed by QLU-NLP, built upon the Qwen2.5 architecture. It is specifically designed and optimized for Traditional Chinese Medicine (TCM) applications, excelling in tasks such as TCM syndrome differentiation, disease diagnosis, and medical license examinations. The model achieves state-of-the-art performance in specialized TCM tasks, making it suitable for assisting in diagnosis and patient self-assessment within the TCM domain.
Loading preview...
BianCang-Qwen2.5-7B-Instruct: A Specialized TCM LLM
BianCang-Qwen2.5-7B-Instruct is a 7.6 billion parameter instruction-tuned model developed by QLU-NLP, built on the Qwen2.5 base architecture. It is part of the BianCang series, specifically engineered to advance the application of large language models in Traditional Chinese Medicine (TCM).
Key Capabilities and Training:
- TCM Specialization: The model is designed to assist with TCM disease diagnosis, syndrome differentiation, and patient self-assessment.
- Two-Stage Training: It utilizes a unique two-stage training method involving domain knowledge injection, followed by knowledge activation and alignment.
- Superior TCM Performance: BianCang-Qwen2.5-7B-Instruct demonstrates advanced performance in core TCM tasks and achieves high scores in various medical license examinations, outperforming its base model (Qwen2.5-7B-Instruct) and other general-purpose LLMs like GPT-4 and DeepSeek in specific TCM benchmarks.
Benchmarked Performance Highlights:
- Achieves 82.10% in TCM Syndrome Differentiation (CoT) and 89.40% in TCM Disease Diagnosis (CoT).
- Scores 90.57% in the MLEC-TCM (FS) and 90.62% in the MLEC-CWM (FS) examinations.
- Notably, it reaches 93.43% in MLEC-Clinic (FS) and 89.91% in MLEC-PublicHealth (FS).
Use Cases:
- TCM Diagnosis Assistance: Aids medical professionals in diagnosing TCM-related conditions.
- Patient Self-Assessment: Supports individuals in understanding their health from a TCM perspective.
- TCM Knowledge Query: Provides answers to queries related to Traditional Chinese Medicine.
This model is intended for academic research and as an intelligent assistant, but users are advised that it cannot replace professional medical advice or diagnosis.