HateBERT: Retraining BERT for Abusive Language Detection in English

By Tommaso Caselli, Valerio Basile, Jelena Mitrović, Michael Granitzer. 2020

503 citations recorded by the source corpus.

We introduce HateBERT, a re-trained BERT model for abusive language detection in English. The model was trained on RAL-E, a large-scale dataset of Reddit comments in English from communities banned for being offensive, abusive, or hateful that we have curated and made available to the public. We present the results of a detailed comparison between a general pre-trained language model and the retrained version on three English datasets for offensive, abusive language and hate speech detection tasks. In all datasets, HateBERT outperforms the corresponding general BERT model. We also discuss a battery of experiments comparing the portability of the fine-tuned models across the datasets, suggesting that portability is affected by compatibility of the annotated phenomena.

Subfields: Artificial Intelligence, Natural Language Processing, Safety, Robustness & Interpretability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Natural Language Processing

In Safety, Robustness & Interpretability

Citations in the corpus

Towards Safer Generative Language Models: A Survey on Safety Risks, Evaluations, and Improvements (2023, methodology): “To improve the performance of classifiers, Caselli et al. (2021) utilized unsafe content from Reddit 1 to re-train a Bert model called HateBert, allowing Hate-Bert to learn more knowledge about harmful content and thus become more sensitive to it.”
ToxiGen: A Large-Scale Machine-Generated Dataset for Adversarial and Implicit Hate Speech Detection (2022, methodology): “We ﬁne-tune the widely-used HateBERT (Caselli et al., 2021) and ToxDectRoBERTa (Zhou et al., 2021) models on the training portion of T OXI G EN , using the prompt labels as proxies for a true toxicity label.”
ConvAbuse: Data, Analysis, and Benchmarks for Nuanced Abuse Detection in Conversational AI (2021, methodology): “as HateBERT (Caselli et al., 2021) or HurtBERT (Koufakou et al.”
Towards Agile Text Classifiers for Everyone (2023, background): “…have also grown in sophistication, from linear models such as support vector machines (MacAvaney et al., 2019) to more modern approaches that rely on deep learning architectures like CNNs (Gambäck and Sikdar, 2017) and, more recently, transformers (Caselli et al., 2020; Zhou et al., 2021).”
Leveraging Bias in Pre-Trained Word Embeddings for Unsupervised Microaggression Detection (2021, unknown): “Recently, however, there have been studies involving word representations created from tailored collections of social media content aimed at capturing abusive phenomena like verbal aggression (Dynel, 2021) and hate speech (Caselli et al., 2020).”
Orca: Progressive Learning from Complex Explanation Traces of GPT-4 (2023, methodology): “It is important to note that any selected off-the-shelf toxicity detector (including HateBERT) potentially has its own biases and weaknesses depending on the type of data they have been trained on.”
Towards multidomain and multilingual abusive language detection: a survey (2021, methodology): “…more on investigating transferability or generalizability of state-of-the-art transformer-based models such as Bidirectional Encoder Representations from Transformers (BERT) [19, 48, 66, 79, 83, 90, 92, 134] and its variant like RoBERTa [48] in the cross-domain abusive language detection task.”
AMMUS : A Survey of Transformer-based Pretrained Models in Natural Language Processing (2021, methodology): “HateBERT [81] BERT MLM RAL-E (dataset of 1.”
Few-shot Instruction Prompts for Pretrained Language Models to Detect Social Biases (2021, methodology): “Recently, transformers have been fine-tuned for hate-speech detection (Caselli et al., 2020), detection of targeted offensive language (Rosenthal et al., 2021) and bias (Sap et al., 2020).”