XBMU Chinese–Tibetan Multi-Department Medical QA Dataset

The XBMU Chinese–Tibetan Medical QA Dataset is the first large-scale bilingual medical question-answering dataset constructed by Northwest Minzu University. It contains a total of 40,274 parallel question-answer pairs, covering six major clinical fields: otorhinolaryngology, ophthalmology, internal medicine, neurology, surgery, and nutrition and healthcare. The data are derived from real medical consultation texts and have undergone multiple rounds of cleaning, de-identification, standardization, and expert review to ensure privacy compliance and semantic accuracy. Each sample includes the fields question_zh, answer_zh, and the corresponding question_bo, answer_bo, maintaining a consistent structure. The dataset supports both Chinese medical question-answering tasks and Chinese-Tibetan machine translation research and can be used for multilingual QA generation, terminology alignment, cross-lingual knowledge transfer, and model robustness evaluation.