ko-ged-mc-elementary is a benchmark dataset designed to evaluate the comprehension and reasoning capabilities of large language models (LLMs) in the Korean language. The dataset comprises multiple-choice questions extracted from the General Equivalency Diploma (GED) exams. Specifically, this dataset contains questions from the second GED exam of 2024. All exam items are sourced from the official archives managed by the Korean Ministry of Education.… See the full description on the dataset page:
https://huggingface.co/datasets/davidkim205/ko-ged-mc-elementary.
