datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
hatecheck
Dataset Card for HateCheck
Dataset Description
HateCheck is a suite of functional test for hate speech detection models.
The dataset contains 3,728 validated test cases in 29 functional tests.
19 functional tests correspond to distinct types of hate. The other 11 functional tests cover challenging types of non-hate.
This allows for targeted diagnostic insights into model performance.
In our ACL paper, we found critical weaknesses in all commercial and academic hate… See the full description on the dataset page: https://huggingface.co/datasets/Paul/hatecheck.multilingual-hatespeech-dataset
[!NOTE]
Dataset origin: https://www.kaggle.com/datasets/wajidhassanmoosa/multilingual-hatespeech-dataset
Description
This dataset contains hate speech text with labels where 0 represents non-hate and 1 shows hate
texts also the data from different languages needed to be identified as a corresponding
correct language. The following are the languages in the dataset with the numbers corresponding to that language.
(1 Arabic)(2 English)(3 Chinese)(4 French) (5 German) (6 Russian)(7… See the full description on the dataset page: https://huggingface.co/datasets/FrancophonIA/multilingual-hatespeech-dataset.HateBR
HateBR: The Evaluation Benchmark for Brazilian Portuguese Hate Speech Detection
HateBR is the first large-scale, expert-annotated dataset of Brazilian Instagram comments specifically designed for hate speech detection on the web and social media. The dataset was collected from Brazilian Instagram comments made by politicians and manually annotated by specialists.
It contains 7,000 documents, annotated across three distinct layers:
Binary classification (offensive vs. non-offensive… See the full description on the dataset page: https://huggingface.co/datasets/franciellevargas/HateBR.Hate-Speech-Tweetshatecheck-german
Dataset Card for Multilingual HateCheck
Dataset Description
Multilingual HateCheck (MHC) is a suite of functional tests for hate speech detection models in 10 different languages: Arabic, Dutch, French, German, Hindi, Italian, Mandarin, Polish, Portuguese and Spanish.
For each language, there are 25+ functional tests that correspond to distinct types of hate and challenging non-hate.
This allows for targeted diagnostic insights into model performance.
For more details… See the full description on the dataset page: https://huggingface.co/datasets/Paul/hatecheck-german.hatecheck-polish
Dataset Card for Multilingual HateCheck
Dataset Description
Multilingual HateCheck (MHC) is a suite of functional tests for hate speech detection models in 10 different languages: Arabic, Dutch, French, German, Hindi, Italian, Mandarin, Polish, Portuguese and Spanish.
For each language, there are 25+ functional tests that correspond to distinct types of hate and challenging non-hate.
This allows for targeted diagnostic insights into model performance.
For more details… See the full description on the dataset page: https://huggingface.co/datasets/Paul/hatecheck-polish.hatecheck-mandarin
Dataset Card for Multilingual HateCheck
Dataset Description
Multilingual HateCheck (MHC) is a suite of functional tests for hate speech detection models in 10 different languages: Arabic, Dutch, French, German, Hindi, Italian, Mandarin, Polish, Portuguese and Spanish.
For each language, there are 25+ functional tests that correspond to distinct types of hate and challenging non-hate.
This allows for targeted diagnostic insights into model performance.
For more details… See the full description on the dataset page: https://huggingface.co/datasets/Paul/hatecheck-mandarin.hatecheck-italian
Dataset Card for Multilingual HateCheck
Dataset Description
Multilingual HateCheck (MHC) is a suite of functional tests for hate speech detection models in 10 different languages: Arabic, Dutch, French, German, Hindi, Italian, Mandarin, Polish, Portuguese and Spanish.
For each language, there are 25+ functional tests that correspond to distinct types of hate and challenging non-hate.
This allows for targeted diagnostic insights into model performance.
For more details… See the full description on the dataset page: https://huggingface.co/datasets/Paul/hatecheck-italian.turkish-hate-speech-superset
Turkish Hate Speech Superset
This dataset is a superset (N=41,423) of posts annotated as hateful or not. It results from the preprocessing and merge of all available Turkish hate speech datasets in April 2024. These datasets were identified through a systematic survey of hate speech datasets conducted in early 2024. We only kept datasets that:
are documented
are publicly available
focus on hate speech, defined broadly as "any kind of communication in speech, writing or behavior… See the full description on the dataset page: https://huggingface.co/datasets/manueltonneau/turkish-hate-speech-superset.hatecheck-dutch
Dataset Card for Multilingual HateCheck
Dataset Description
Multilingual HateCheck (MHC) is a suite of functional tests for hate speech detection models in 10 different languages: Arabic, Dutch, French, German, Hindi, Italian, Mandarin, Polish, Portuguese and Spanish.
For each language, there are 25+ functional tests that correspond to distinct types of hate and challenging non-hate.
This allows for targeted diagnostic insights into model performance.
For more details… See the full description on the dataset page: https://huggingface.co/datasets/Paul/hatecheck-dutch.hatecheck-spanish
Dataset Card for Multilingual HateCheck
Dataset Description
Multilingual HateCheck (MHC) is a suite of functional tests for hate speech detection models in 10 different languages: Arabic, Dutch, French, German, Hindi, Italian, Mandarin, Polish, Portuguese and Spanish.
For each language, there are 25+ functional tests that correspond to distinct types of hate and challenging non-hate.
This allows for targeted diagnostic insights into model performance.
For more details… See the full description on the dataset page: https://huggingface.co/datasets/Paul/hatecheck-spanish.Automated_Hate_Speech_Detection_and_the_Problem_of_Offensive_Languagehatecheck-portuguese
Dataset Card for Multilingual HateCheck
Dataset Description
Multilingual HateCheck (MHC) is a suite of functional tests for hate speech detection models in 10 different languages: Arabic, Dutch, French, German, Hindi, Italian, Mandarin, Polish, Portuguese and Spanish.
For each language, there are 25+ functional tests that correspond to distinct types of hate and challenging non-hate.
This allows for targeted diagnostic insights into model performance.
For more details… See the full description on the dataset page: https://huggingface.co/datasets/Paul/hatecheck-portuguese.hatecheck-french
Dataset Card for Multilingual HateCheck
Dataset Description
Multilingual HateCheck (MHC) is a suite of functional tests for hate speech detection models in 10 different languages: Arabic, Dutch, French, German, Hindi, Italian, Mandarin, Polish, Portuguese and Spanish.
For each language, there are 25+ functional tests that correspond to distinct types of hate and challenging non-hate.
This allows for targeted diagnostic insights into model performance.
For more details… See the full description on the dataset page: https://huggingface.co/datasets/Paul/hatecheck-french.Dynamically-Generated-Hate-Speech-Dataset
Dataset Card for dynamically generated hate speech dataset
Dataset Summary
This is a copy of the Dynamically-Generated-Hate-Speech-Dataset, presented in this paper by
Bertie Vidgen, Tristan Thrush, Zeerak Waseem and Douwe Kiela
Original README from GitHub
Dynamically-Generated-Hate-Speech-Dataset
ReadMe for v0.2 of the Dynamically Generated Hate Speech Dataset from Vidgen et al. (2021). If you use the dataset, please cite our paper in the… See the full description on the dataset page: https://huggingface.co/datasets/LennardZuendorf/Dynamically-Generated-Hate-Speech-Dataset.spanish-hate-speech-superset
Spanish Hate Speech Superset
This dataset is a superset (N=29,855) of posts annotated as hateful or not. It results from the preprocessing and merge of all available Spanish hate speech datasets in April 2024. These datasets were identified through a systematic survey of hate speech datasets conducted in early 2024. We only kept datasets that:
are documented
are publicly available or could be retrieved with the Twitter API
focus on hate speech, defined broadly as "any kind of… See the full description on the dataset page: https://huggingface.co/datasets/manueltonneau/spanish-hate-speech-superset.hatecheck-hindi
Dataset Card for Multilingual HateCheck
Dataset Description
Multilingual HateCheck (MHC) is a suite of functional tests for hate speech detection models in 10 different languages: Arabic, Dutch, French, German, Hindi, Italian, Mandarin, Polish, Portuguese and Spanish.
For each language, there are 25+ functional tests that correspond to distinct types of hate and challenging non-hate.
This allows for targeted diagnostic insights into model performance.
For more details… See the full description on the dataset page: https://huggingface.co/datasets/Paul/hatecheck-hindi.hatecheck-arabic
Dataset Card for Multilingual HateCheck
Dataset Description
Multilingual HateCheck (MHC) is a suite of functional tests for hate speech detection models in 10 different languages: Arabic, Dutch, French, German, Hindi, Italian, Mandarin, Polish, Portuguese and Spanish.
For each language, there are 25+ functional tests that correspond to distinct types of hate and challenging non-hate.
This allows for targeted diagnostic insights into model performance.
For more details… See the full description on the dataset page: https://huggingface.co/datasets/Paul/hatecheck-arabic.not_hate_dataset
Not Dataset — Controlled Negation Harm–Safe Pairs
⚠️ WARNING: This dataset contains potentially sensitive, harmful, and offensive content. ⚠️
Paper
Polarity-Aware Probing for Quantifying Latent Alignment in Language Models
Code/Github
https://github.com/SadSabrina/polarity-probing
Abstract
Recent developments in unsupervised probing, notably Contrast-Consistent Search (CCS), have opened the possibility of evaluating latent model beliefs without… See the full description on the dataset page: https://huggingface.co/datasets/SabrinaSadiekh/not_hate_dataset.arabic-hate-speech-superset
Arabic Hate Speech Superset
This dataset is a superset (N=449,078) of posts annotated as hateful or not. It results from the preprocessing and merge of all available Arabic hate speech datasets in April 2024. These datasets were identified through a systematic survey of hate speech datasets conducted in early 2024. We only kept datasets that:
are documented
are publicly available or could be retrieved with the Twitter API
focus on hate speech, defined broadly as "any kind of… See the full description on the dataset page: https://huggingface.co/datasets/manueltonneau/arabic-hate-speech-superset.large-scale-hate-speech-turkish-v1The dataset published in the LREC 2022 paper "Large-Scale Hate Speech Detection with Cross-Domain Transfer".
This is Dataset v1 (Turkish):
The original dataset that includes 100,000 tweets in Turkish. The annotations with more than 60% agreement are included.
TweetID: Tweet ID from Twitter API
LangID: 0 (Turkish)
TopicID: Domain of the topic 0-Religion, 1-Gender, 2-Race, 3-Politics, 4-Sports
HateLabel: Final hate label decision 0-Normal, 1-Offensive, 2-Hate
GitHub Repo:… See the full description on the dataset page: https://huggingface.co/datasets/ctoraman/large-scale-hate-speech-turkish-v1.large-scale-hate-speech-turkish-v2The dataset published in the LREC 2022 paper "Large-Scale Hate Speech Detection with Cross-Domain Transfer".
This is Dataset v2 (Turkish):
The modified dataset that includes 60,310 tweets in Turkish. The annotations with more than 80% agreement are included.
TweetID: Tweet ID from Twitter API
LangID: 0 (Turkish)
TopicID: Domain of the topic 0-Religion, 1-Gender, 2-Race, 3-Politics, 4-Sports
HateLabel: Final hate label decision 0-Normal, 1-Offensive, 2-Hate
GitHub Repo:… See the full description on the dataset page: https://huggingface.co/datasets/ctoraman/large-scale-hate-speech-turkish-v2.multilabel-tagalog-hate-speechfrench-hate-speech-superset
French Hate Speech Superset
This dataset is a superset (N=18,071) of posts annotated as hateful or not. It results from the preprocessing and merge of all available French hate speech datasets in April 2024. These datasets were identified through a systematic survey of hate speech datasets conducted in early 2024. We only kept datasets that:
are documented
are publicly available
focus on hate speech, defined broadly as "any kind of communication in speech, writing or behavior, that… See the full description on the dataset page: https://huggingface.co/datasets/manueltonneau/french-hate-speech-superset.large-scale-hate-speech-v1The dataset published in the LREC 2022 paper "Large-Scale Hate Speech Detection with Cross-Domain Transfer".
This is Dataset v1:
The original dataset that includes 100,000 tweets in English. The annotations with more than 60% agreement are included.
TweetID: Tweet ID from Twitter API
LangID: 1 (English)
TopicID: Domain of the topic 0-Religion, 1-Gender, 2-Race, 3-Politics, 4-Sports
HateLabel: Final hate label decision 0-Normal, 1-Offensive, 2-Hate
GitHub Repo:
NOTE:… See the full description on the dataset page: https://huggingface.co/datasets/ctoraman/large-scale-hate-speech-v1.german-hate-speech-superset
German Hate Speech Superset
This dataset is a superset (N=50,545) of posts annotated as hateful or not. It results from the preprocessing and merge of all available German hate speech datasets in April 2024. These datasets were identified through a systematic survey of hate speech datasets conducted in early 2024. We only kept datasets that:
are documented
are publicly available
focus on hate speech, defined broadly as "any kind of communication in speech, writing or behavior, that… See the full description on the dataset page: https://huggingface.co/datasets/manueltonneau/german-hate-speech-superset.hate_speech_open_data_original_class_test_setHatEval_Relabled_with_Author_Featuresindonesian-hate-speech-superset
Indonesian Hate Speech Superset
This dataset is a superset (N=14,306) of posts annotated as hateful or not. It results from the preprocessing and merge of all available Indonesian hate speech datasets in April 2024. These datasets were identified through a systematic survey of hate speech datasets conducted in early 2024. We only kept datasets that:
are documented
are publicly available
focus on hate speech, defined broadly as "any kind of communication in speech, writing or… See the full description on the dataset page: https://huggingface.co/datasets/manueltonneau/indonesian-hate-speech-superset.Hateful_Memes_in_VLMThis dataset contains the response of VLMs (InstructBlip, ShareGPT4V, LLaVA and CogVLM) to hateful memes and the annotation to these responses. For more information, please refer to paper "From Meme to Threat: On the Hateful Meme Understanding and Induced Hateful Content Generation in Open-Source Vision Language Models."