timonziegenbein/human-like-edit-sequences

该数据集包含了文本序列（sequence）、标签（label）、文档ID（doc_id）、句子索引（sent_idx）、编辑索引（edit_idx）、原始句子（original_sentence）、不适当部分（inappropriate_part）和重写部分（rewritten_part）等字段。数据集被划分为训练集（train）、测试集（test）和验证集（dev），分别用于模型的训练、测试和验证。The dataset includes fields such as text sequence (sequence), label (label), document ID (doc_id), sentence index (sent_idx), edit index (edit_idx), original sentence (original_sentence), inappropriate part (inappropriate_part), and rewritten part (rewritten_part). The dataset is divided into training set (train), test set (test), and validation set (dev) for model training, testing, and validation respectively.