tbilisi-ai-lab/ai2_arc-ka

ARC数据集包含ARC-Challenge和ARC-Easy两个配置。该数据集的特征包括标识符、问题、带有标签和文本的多个选择项、答案键以及另一种形式的问题和选择项（question_ka和choices_ka）。数据集分为训练集、验证集和测试集，包括示例数量和字节大小。该数据集由第比利斯人工智能实验室开发，并遵循odc-by许可证。

The ARC dataset consists of two configurations, ARC-Challenge and ARC-Easy. The features of the dataset include an identifier, a question, multiple choice options with labels and texts, an answer key, and an alternative form of the question and choices (question_ka and choices_ka). The dataset is split into training, validation, and test sets, detailing the number of examples and size in bytes. The dataset was developed by Tbilisi AI Lab and is licensed under odc-by.