ValueConsistency

홈페이지 · 논문 2편

ValueConsistency is a dataset of both controversial and uncontroversial questions in English, Chinese, German, and Japanese for topics from the U.S., China, Germany, and Japan. It was generated via prompting by GPT-4 and validated manually. You can find details about how we made the dataset in the linked paper and in our code base. Curated by: Jared Moore, Tanvi Desphande, Diyi Yang Language(s) (NLP): English, Chinese (Mandarin), German, Japanese License: MIT