onion212/hateful_memes

The Hateful Memes Challenge README The Hateful Memes Challenge is a dataset and benchmark created by Facebook AI to drive and measure progress on multimodal reasoning and understanding. The task focuses on detecting hate speech in multimodal memes. Please see the paper for further details: The Hateful Memes Challenge: Detecting Hate Speech in Multimodal Memes D. Kiela, H. Firooz, A. Mohan, V. Goswami, A. Singh, P. Ringshia, D. Testuggine For more details, see also the website:… See the full description on the dataset page: https://huggingface.co/datasets/onion212/hateful_memes.