Hate Speech Detection - Groups

FHM Dataset

The FHM dataset, a multimodal framework for detecting hateful memes on social media.

The Dynamic dataset, a collection of tweets generated using a human-and-model-in-the-loop process, containing challenging perturbations.

The Waseem dataset, a collection of tweets containing hate speech, with a focus on sexist and racist language.

The Hate Speech Tweets dataset contains over 24,000 English tweets labeled as non-offensive, hate speech, and profanity.

The HateXplain dataset, containing 20,000 posts from Gab and Twitter, annotated with hate/offensive/normal labels.

The implicit hate dataset is a specialized collection of data aimed at detecting hate speech.

The HateXplain dataset is a benchmark dataset for explainable hate speech detection.

The dataset used for probing LLMs for hate speech detection, including HateXplain, implicit hate, and ToxicSpans datasets.

The Hateful Memes dataset aims to help develop models that more eﬀectively detect multimodal hateful content.

The Hateful Memes dataset is a multimodal dataset containing 10,000+ new examples of multimodal content.