TOEFL

该数据集旨在评估检测器对于非母语作者撰写英文文本的偏见程度。它有助于我们更好地理解人工智能文本检测中存在的偏见问题。该数据集的规模是可变的，其任务是对非母语文本中的偏见进行检测。

This dataset is designed to evaluate the degree of bias exhibited by detectors toward English texts written by non-native speakers. It facilitates a deeper understanding of the bias-related issues inherent in artificial intelligence (AI) text detection. The size of this dataset is adjustable, and its primary task is to detect bias present in texts authored by non-native speakers.