This dataset is part of AIGTBench, a comprehensive benchmark for AI-Generated Text (AIGT) detection research, introduced in the paper "Are We in the AI-Generated Text World Already? Quantifying and Monitoring AIGT on Social Media".
This specific subset contains our custom-generated social media data designed to train and evaluate AIGT detectors on content that closely resembles real social media posts from Medium, Quora, and Reddit platforms.… See the full description on the dataset page:
https://huggingface.co/datasets/tarryzhang/AIGTBench.