TweepFake

该数据集是一个专注于检测虚假、由人工智能生成的推文的目标短文本数据集。它被用于展示所提出的多粒度预训练单元（MPU）方法的有效性。该数据集主要包含短文本，通常是单句或者几句话，其任务是识别短推文中的由人工智能生成的文本。

This dataset is a targeted short-text dataset focused on detecting fake and AI-generated tweets. It is utilized to demonstrate the effectiveness of the proposed Multi-Granularity Pre-training Unit (MPU) method. This dataset mainly consists of short texts, typically single sentences or a small number of sentences, and its task is to identify AI-generated texts within short tweets.