SEACrowd/SEADialogues

该数据集包含对话生成的相关信息，其中包括两个人物的语言、主题、名称、性格、性别和角色描述，以及用于生成对话的提示和五种不同模型的生成对话。数据集分为训练集，可用于训练对话生成模型。

The dataset includes information related to dialogue generation, such as language, topics, names, personalities, genders, and persona descriptions for two characters, as well as prompts and generated conversations from five different models. The dataset is split into a training set, which can be used to train dialogue generation models.