alexandrainst/scala
收藏Hugging Face2024-07-17 更新2024-07-22 收录
下载链接：
https://hf-mirror.com/datasets/alexandrainst/scala
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含文档及其是否语法正确的信息。它通过自动生成的脚本从通用依存树库中生成，旨在评估语言可接受性（二元分类，判断文本是否正确）。数据集支持多种语言，包括丹麦语、瑞典语、挪威语、冰岛语、法罗语、德语、荷兰语和英语。数据集的结构包括文本、错误类型和标签三个字段。
The ScaLA dataset is designed for evaluating linguistic acceptability, containing texts and their grammatical correctness labels. The dataset is automatically generated using a script to corrupt documents from a universal dependencies treebank. It supports multiple languages including Danish, Swedish, Norwegian, Icelandic, Faroese, German, Dutch, and English. Each sample includes text, corruption type, and label (correct or incorrect).
提供机构：
alexandrainst
