# Hate Speech Classifiers Learn Human-Like Social Stereotypes

arXiv: https://arxiv.org/abs/2110.14839 · HTML: https://mlsanity.com/paper/2110.14839/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- Builds on 27 of its 83 references that are in the map.
- First posted on arXiv 28 Oct 2021.

## Builds on (most cited first, 27 of 27)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Equality of Opportunity in Supervised Learning](https://mlsanity.com/paper/1610.02413/) (2016)
- [A Survey on Bias and Fairness in Machine Learning](https://mlsanity.com/paper/1908.09635/) (2019)
- [Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings](https://mlsanity.com/paper/1607.06520/) (2016)
- [Language (Technology) is Power: A Critical Survey of "Bias" in NLP](https://mlsanity.com/paper/2005.14050/) (2020)
- [Semantics derived automatically from language corpora contain human-like biases](https://mlsanity.com/paper/1608.07187/) (2016)
- [Automated Hate Speech Detection and the Problem of Offensive Language](https://mlsanity.com/paper/1703.04009/) (2017)
- [Certifying and removing disparate impact](https://mlsanity.com/paper/1412.3756/) (2014)
- [Fair prediction with disparate impact: A study of bias in recidivism prediction instruments](https://mlsanity.com/paper/1610.07524/) (2016)
- [Men Also Like Shopping: Reducing Gender Bias Amplification using Corpus-level Constraints](https://mlsanity.com/paper/1707.09457/) (2017)
- [Dealing with Disagreements: Looking Beyond the Majority Vote in Subjective Annotations](https://mlsanity.com/paper/2110.05719/) (2021)
- [Word Embeddings Quantify 100 Years of Gender and Ethnic Stereotypes](https://mlsanity.com/paper/1711.08412/) (2017)
- [Racial Bias in Hate Speech and Abusive Language Detection Datasets](https://mlsanity.com/paper/1905.12516/) (2019)
- [Social Biases in NLP Models as Barriers for Persons with Disabilities](https://mlsanity.com/paper/2005.00813/) (2020)
- [Reducing Gender Bias in Abusive Language Detection](https://mlsanity.com/paper/1808.07231/) (2018)
- [Black is to Criminal as Caucasian is to Police: Detecting and Removing Multiclass Bias in Word Embeddings](https://mlsanity.com/paper/1904.04047/) (2019)
- [On Releasing Annotator-Level Labels and Information in Datasets](https://mlsanity.com/paper/2110.05699/) (2021)
- [Measuring the Reliability of Hate Speech Annotations: The Case of the European Refugee Crisis](https://mlsanity.com/paper/1701.08118/) (2017)
- [Hate Speech Detection and Racial Bias Mitigation in Social Media based on BERT model](https://mlsanity.com/paper/2008.06460/) (2020)
- [Contextualizing Hate Speech Classifiers with Post-hoc Explanation](https://mlsanity.com/paper/2005.02439/) (2020)
- [Perturbation Sensitivity Analysis to Detect Unintended Model Biases](https://mlsanity.com/paper/1910.04210/) (2019)
- [Racial Disparity in Natural Language Processing: A Case Study of Social Media African-American English](https://mlsanity.com/paper/1707.00061/) (2017)
- [Demoting Racial Bias in Hate Speech Detection](https://mlsanity.com/paper/2005.12246/) (2020)
- [Confronting Abusive Language Online: A Survey from the Ethical and Human Rights Perspective](https://mlsanity.com/paper/2012.12305/) (2020)
- [What are the biases in my word embedding?](https://mlsanity.com/paper/1812.08769/) (2018)
- [Tackling Racial Bias in Automated Online Hate Detection: Towards Fair and Accurate Classification of Hateful Online Users Using Geometric Deep Learning](https://mlsanity.com/paper/2103.11806/) (2021)

## Built on by (most cited first, 12 of 12)

- [Annotators with Attitudes: How Annotator Beliefs And Identities Bias Toxic Language Detection](https://mlsanity.com/paper/2111.07997/) (2021)
- [The Moral Foundations Reddit Corpus](https://mlsanity.com/paper/2208.05545/) (2022)
- [Two Contrasting Data Annotation Paradigms for Subjective NLP Tasks](https://mlsanity.com/paper/2112.07475/) (2021)
- [Justice in Misinformation Detection Systems: An Analysis of Algorithms, Stakeholders, and Potential Harms](https://mlsanity.com/paper/2204.13568/) (2022)
- [The Devil is in the Neurons: Interpreting and Mitigating Social Biases in Pre-trained Language Models](https://mlsanity.com/paper/2406.10130/) (2024)
- [Dealing with Annotator Disagreement in Hate Speech Classification](https://mlsanity.com/paper/2502.08266/) (2025)
- [Controlling Federated Learning for Covertness](https://mlsanity.com/paper/2308.08825/) (2023)
- [Mapping Violence: Developing an Extensive Framework to Build a Bangla Sectarian Expression Dataset from Social Media Interactions](https://mlsanity.com/paper/2404.11752/) (2024)
- [Social-Group-Agnostic Word Embedding Debiasing via the Stereotype Content Model](https://mlsanity.com/paper/2210.05831/) (2022)
- [Learning Who Disagrees: Demographic Importance Weighting for Modeling Annotator Distributions with DiADEM](https://mlsanity.com/paper/2604.08425/) (2026)
- [Understanding Annotator Safety Policy with Interpretability](https://mlsanity.com/paper/2605.05329/) (2026)
- [Epistemic Injustice in Language Models: An Audit of Pretraining Filters and Guardrails](https://mlsanity.com/paper/2606.05936/) (2026)

## Abstract (arXiv)

Social stereotypes negatively impact individuals' judgements about different groups and may have a critical role in how people understand language directed toward minority social groups. Here, we assess the role of social stereotypes in the automated detection of hateful language by examining the relation between individual annotator biases and erroneous classification of texts by hate speech classifiers. Specifically, in Study 1 we investigate the impact of novice annotators' stereotypes on their hate-speech-annotation behavior. In Study 2 we examine the effect of language-embedded stereotypes on expert annotators' aggregated judgements in a large annotated corpus. Finally, in Study 3 we demonstrate how language-embedded stereotypes are associated with systematic prediction errors in a neural-network hate speech classifier. Our results demonstrate that hate speech classifiers learn human-like biases which can further perpetuate social inequalities when propagated at scale. This framework, combining social psychological and computational linguistic methods, provides insights into additional sources of bias in hate speech moderation, informing ongoing debates regarding fairness in machine learning.