ylkhayat/CUAD-generation-workshop

该数据集包含用于文本生成、引用分析或信息检索等任务的结构化数据。数据集特征包括文档ID（docid）、前文（previous_text）、目标文本（gold_text）、引用（citations）、参考文档段落（oracle_documents_passages）和顶部相关段落（top_k_passages）。数据集分为训练集和测试集，分别包含1000和4182个样本。

This dataset contains structured data for tasks such as text generation, citation analysis, or information retrieval. The features of the dataset include document ID (docid), previous text (previous_text), target text (gold_text), citations, reference document passages (oracle_documents_passages), and top relevant passages (top_k_passages). The dataset is divided into training and test sets, containing 1000 and 4182 samples, respectively.