{"url":"/dataset/toxigen","name":"ToxiGen","full_name":null,"description_markdown":"A large-scale and machine-generated dataset of 274,186 toxic and benign statements about 13 minority groups. \r\n\r\nThis dataset uses a demonstration-based prompting framework and an adversarial classifier-in-the-loop decoding method to generate subtly toxic and\r\nbenign text with a massive pre-trained language model (GPT-3). Controlling machine generation in this way allows TOXIGEN to cover implicitly toxic text at a larger scale,\r\nand about more demographic groups, than previous resources of human-written text. TOXIGEN can be used to fight human-written and machine-generated\r\ntoxicity.","description_withheld":null,"homepage":"https://github.com/microsoft/TOXIGEN","introduced_date":"2022-03-17","introduced_date_note":null,"introduced_by":{"paper":"/paper/toxigen-a-large-scale-machine-generated","title":"ToxiGen: A Large-Scale Machine-Generated Dataset for Adversarial and Implicit Hate Speech Detection","first_author":"Thomas Hartvigsen","url":null},"license":null,"modalities":[{"name":"Texts","url":"/datasets/modality/texts"}],"tasks":[{"name":"Hate Speech Detection","url":"/task/hate-speech-detection","datasets_with_task":"/datasets/task/hate-speech-detection"}],"languages":[{"name":"English","url":"/datasets/language/english"}],"variants":["ToxiGen"],"data_loaders":[{"repo":"https://github.com/microsoft/toxigen","url":"https://huggingface.co/datasets/skg/toxigen-data","frameworks":[]}],"num_papers_in_archive":85,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[],"papers_with_a_benchmark_row":[],"syntology_totals":{"read_at":"2026-09-24T18:15:14+00:00","papers_with_samples":0,"samples_harvested":0,"samples_ran":0,"samples_unverified":0,"pointer_only_for_licence":0,"papers_with_no_sample_that_ran":0,"note":"the per-paper counts above, summed; not a rate"},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks."}