somosnlp-hackathon-2025/exam_zh_multitopic_dialect_culture

这是一个包含300个选择题的数据集，来源于多种普通话为基础的评估，包括地区方言理解和文化和一般知识评估。数据覆盖了客家话、长沙话、闽南话、陕北话、四川话、南昌话、上海话和广东话等多种方言，以及互联网俚语测试、歇后语识别、古汉语词汇意义理解等内容。

This dataset contains 300 multiple-choice questions from a variety of Mandarin-based assessments, covering both regional dialect comprehension and cultural/general knowledge. It spans dialects such as Hakka, Changsha, Hokkien, Northern Shaanxi, Sichuan, Nanchang, Shanghainese, and Cantonese, as well as topics like Chinese Internet slang tests, Xiehouyu recognition, and classical Chinese lexical meaning comprehension.