"ZhongYiBench_V1"

"A comprehensive benchmark for evaluating TCM-oriented LLMs. It integrates two components: objective evaluation and subjective evaluation. The objective evaluation draws on real questions from national authoritative TCM examinations. The subjective evaluation consists of two categories. One is comprehensive classical reasoning evaluation, which covers term interpretation and discussion questions. The other is clinical case analysis evaluation, which focuses on syndrome differentiation, prescription formulation and related reasoning tasks. All subjective responses are scored by certified TCM physicians on a standardized scale. This benchmark provides a unified framework to measure the accuracy, logical rigor, and clinical competence of TCM-specific LLMs. "