cardiffnlp/twitter-roberta-base-hate-latest-st

This is a RoBERTa-base model trained on 154M tweets until the end of December 2022 and finetuned for hate speech detection (multiclass classification) on the TweetHate dataset of SuperTweetEval. The original Twitter-based RoBERTa model can be found here.

Labels

"id2label": { "0": "hate_gender", "1": "hate_race", "2": "hate_sexuality", "3": "hate_religion", "4": "hate_origin", "5": "hate_disability", "6": "hate_age", "7": "not_hate" }

Example

from transformers import pipeline
text = 'Eid Mubarak Everyone!!! ❤ May Allah unite all Muslims, show us the right path, and bless us with good health.❣'

pipe = pipeline('text-classification', model="cardiffnlp/twitter-roberta-base-hate-latest-st")
pipe(text)
>> [{'label': 'not_hate', 'score': 0.9997966885566711}]

Citation Information

Please cite the reference paper if you use this model.

@inproceedings{antypas2023supertweeteval,
  title={SuperTweetEval: A Challenging, Unified and Heterogeneous Benchmark for Social Media NLP Research},
  author={Dimosthenis Antypas and Asahi Ushio and Francesco Barbieri and Leonardo Neves and Kiamehr Rezaee and Luis Espinosa-Anke and Jiaxin Pei and Jose Camacho-Collados},
  booktitle={Findings of the Association for Computational Linguistics: EMNLP 2023},
  year={2023}
}