NAMEXACT

该数据集包含了一系列名字，通过遮蔽代词的方式将这些名字与性别关联起来。它被用于训练和评估过程中，特别用于将名字与句子配对。该数据集的任务是针对Transformer模型中的性别去偏见。

This dataset comprises a collection of names that are linked to genders through pronoun masking. It is utilized in training and evaluation procedures, especially for the task of pairing names with sentences. The dataset is designed for gender debiasing in Transformer models.