twitter-roberta-base-hate

Model overview

Twitter-roberta-base-hate is published by cardiffnlp as a text-classification model. The captured configuration identifies RobertaForSequenceClassification with a roberta model type and a transformers library tag. According to the model card, it is a roBERTa-base model trained on about 58M tweets and fine-tuned for hate-speech detection with the TweetEval benchmark.

Recorded capabilities

Use cases in the source record

• Hate-speech screening for conduct targeting women and immigrants, as described by the publisher as the model's specialization.
• TweetEval-style text-classification research consistent with the card's benchmark reference.

Limitations and unknowns

• According to the model card, the publisher points to a more recent and robust successor at cardiffnlp/twitter-roberta-base-hate-latest.
• No parameter count was extracted from this record.
• No license value was extracted from this record.
• No evaluation results were extracted from this record.
• Provider state is historical snapshot data, not independently refreshed current availability.

Source and provenance

Source: cardiffnlp/twitter-roberta-base-hate

Captured: Unknown. Processed: 2026-09-07T19:34:42.237767+00:00.

Twitter-roBERTa-base for Hate Speech Detection This is a roBERTa-base model trained on ~58M tweets and finetuned for hate speech detection with the TweetEval benchmark. This model is specialized to detect hate speech against women and immigrants. NEW! We have made available a more recent and robust hate speech detection model here: https://huggingface.co/cardiffnlp/twitter-roberta-base-hate-latest Paper: TweetEval benchmark (Findings of EMNLP 2020) . Git Repo: Tweeteval official repository . Example of classification from transformers import AutoModelForSequenceClassification from transformers import TFAutoModelForSequenceClassifica…