permitt/superglue-mne

BalkanBench SuperGLUE - Montenegrin是BalkanBench项目的一部分，专门用于评估黑山语（Montenegrin）的语言模型性能。该数据集包含5个任务配置（boolq, cb, copa, multirc, rte），覆盖文本分类、多项选择、问答等多种任务类型。数据集是通过将原始SuperGLUE英语数据集翻译成黑山语并经过人工验证创建的。数据集分为训练集、验证集和测试集，其中测试集的数据和标签存放在一个受限制的仓库中。数据集的总条目数为65,049条。

BalkanBench SuperGLUE - Montenegrin is part of the BalkanBench project, specifically designed to evaluate the performance of language models in Montenegrin. The dataset includes 5 task configurations (boolq, cb, copa, multirc, rte), covering various task types such as text classification, multiple-choice, and question-answering. The dataset was created by translating the original SuperGLUE English dataset into Montenegrin and undergoing human verification. It is divided into training, validation, and test sets, with the test set data and labels stored in a restricted repository. The total number of entries in the dataset is 65,049.