datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
enwiki-image-content-moderationThis dataset is composed of scores of images taken from English Wikipedia and Wikimedia Commons. The scores are the outputs of the models
https://github.com/bumble-tech/private-detector
https://huggingface.co/Freepik/nsfw_image_detector
https://huggingface.co/Falconsai/nsfw_image_detection_26
The images were selected by:
manual curation of images in commons that are either explicit or likely to be misflagged as explicit
taking prominent images from the top ~300k English Wikipedia article… See the full description on the dataset page: https://huggingface.co/datasets/derenrich/enwiki-image-content-moderation.content-moderation
Note:
This dataset contains the EVAL portion of the Jigsaw Toxic Comment Dataset.
It should be used for model evaluation. For training, one can use the original Jigsaw dataset: https://huggingface.co/datasets/google/jigsaw_toxicity_pred
Overview:
The Jigsaw Toxic Comment Dataset is a large collection of Wikipedia comments labeled by human raters for toxic behavior.
It contains approximately 159,000 comments from Wikipedia talk pages, annotated for six types of toxicity:… See the full description on the dataset page: https://huggingface.co/datasets/GuardrailsAI/content-moderation.content-moderation-input-dataset
Access Guidelines - READ THIS BEFORE REQUESTING ACCESS!
Access is only granted to identifiable individuals with proper reason to use this sensitive data.
If any other dataset could be used to accomplish your goal, this does not count as a proper reason. Half sentences and bullet points do not suffice and will be declined. Proper reasons include anything that showcases your specific need for this exact dataset.
Content Moderation Dataset
Overview
This… See the full description on the dataset page: https://huggingface.co/datasets/centrepourlasecuriteia/content-moderation-input-dataset.merged_content_moderation_and_prompt_injection_newContent-Moderation-and-Safety
🇰🇿 Content Moderation and Safety, Kazakh Context
Dataset Summary
Content Moderation and Safety (Profanity) Kazakh Context is a comprehensive dataset designed specifically to train Large Language Models (LLMs) in detecting, classifying, and mitigating toxic, aggressive, or unsafe text in the Kazakh language.
📊 Dataset Statistics
General Metrics
Metric
Count
Total Samples
17,827
Total Words (approx.)
1,674,638
Avg.… See the full description on the dataset page: https://huggingface.co/datasets/farabi-lab/Content-Moderation-and-Safety.merged_content_moderation_and_prompt_injectioncontent-moderation-output-dataset
Access Guidelines - READ THIS BEFORE REQUESTING ACCESS!
Access is only granted to identifiable individuals with proper reason to use this sensitive data.
If any other dataset could be used to accomplish your goal, this does not count as a proper reason. Half sentences and bullet points do not suffice and will be declined. Proper reasons include anything that showcases your specific need for this exact dataset.
Content Moderation Output Dataset
Overview… See the full description on the dataset page: https://huggingface.co/datasets/centrepourlasecuriteia/content-moderation-output-dataset.Content_Moderation_and_Safety_Kazakh_Context
🇰🇿 Content Moderation and Safety Kazakh Context
Dataset Summary
Toxic Speech Analysis and Mitigation, Kazakh Context is an advanced AI Safety dataset designed to train Large Language Models (LLMs) to detect, deeply analyze, and constructively rewrite toxic or harmful speech in the Kazakh language.
📊 Dataset Statistics
General Metrics
Metric
Count
Total Samples
12,063
Total Words (approx.)
5,869,718
Avg. Words per… See the full description on the dataset page: https://huggingface.co/datasets/farabi-lab/Content_Moderation_and_Safety_Kazakh_Context.content-moderationTheCulture_content_moderationgenerative_ai_content_moderationqwen25-content-moderationContent_Moderationcontent_moderation_bells_v2