TCM-Eval: An Expert-Level Dynamic and Extensible Benchmark for Traditional Chinese Medicine
This addresses the problem of limited LLM capabilities in TCM for researchers and practitioners, representing a domain-specific advancement.
The authors tackled the lack of standardized benchmarks and high-quality training data for applying Large Language Models (LLMs) to Traditional Chinese Medicine (TCM) by introducing TCM-Eval, a dynamic and extensible benchmark, and developed ZhiMingTang (ZMT), an LLM that significantly exceeds the passing threshold for human practitioners.
Large Language Models (LLMs) have demonstrated remarkable capabilities in modern medicine, yet their application in Traditional Chinese Medicine (TCM) remains severely limited by the absence of standardized benchmarks and the scarcity of high-quality training data. To address these challenges, we introduce TCM-Eval, the first dynamic and extensible benchmark for TCM, meticulously curated from national medical licensing examinations and validated by TCM experts. Furthermore, we construct a large-scale training corpus and propose Self-Iterative Chain-of-Thought Enhancement (SI-CoTE) to autonomously enrich question-answer pairs with validated reasoning chains through rejection sampling, establishing a virtuous cycle of data and model co-evolution. Using this enriched training data, we develop ZhiMingTang (ZMT), a state-of-the-art LLM specifically designed for TCM, which significantly exceeds the passing threshold for human practitioners. To encourage future research and development, we release a public leaderboard, fostering community engagement and continuous improvement.