Dataset for "Prompt Language as an Interface Condition: Multilingual Variation in LLM Translation and Explanation of Traditional Chinese Medicine Terms"

This dataset contains 972 large language model responses generated for 18 traditional Chinese medicine terms across six general-purpose models, three prompt languages (Chinese, English, and Malay), and three independent runs. It includes the generated English terms and explanations, model and prompt-condition information, WHO 2007 and 2022 terminology reference forms, and the final reconciled coding variables for terminological form, Pinyin visibility, and conceptual representation used in the accompanying study.