mlsanityData to 14 Oct 2026

Human and LLM Biases in Hate Speech Annotations: A Socio-Demographic Analysis of Annotators and Targets

arXiv 2410.07991 ↗First posted 10 Oct 2024cs.CL cs.AI cs.HCcited by 31

Not on the map. Why?

Abstract (arXiv)

The rise of online platforms exacerbated the spread of hate speech, demanding scalable and effective detection. However, the accuracy of hate speech detection systems heavily relies on human-labeled data, which is inherently susceptible to biases. While previous work has examined the issue, the interplay between the characteristics of the annotator and those of the target of the hate are still unexplored. We fill this gap by leveraging an extensive dataset with rich socio-demographic information of both annotators and targets, uncovering how human biases manifest in relation to the target's attributes. Our analysis surfaces the presence of widespread biases, which we quantitatively describe and characterize based on their intensity and prevalence, revealing marked differences. Furthermore, we compare human biases with those exhibited by persona-based LLMs. Our findings indicate that while persona-based LLMs do exhibit biases, these differ significantly from those of human annotators. Overall, our work offers new and nuanced results on human biases in hate speech annotations, as well as fresh insights into the design of AI-driven hate speech detection systems.

Key facts. Cited by 31 papers in the map (data through 14 Oct 2026). Builds on 25 of its 44 references that are in the map. First posted on arXiv 10 Oct 2024.

Builds on

25 of 44 references.

References · 25

  1. 2022Training language models to follow instructions with human feedbackcited by 15,971
  2. 2022Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language modelscited by 1,998
  3. 2022Discovering Language Model Behaviors with Model-Written Evaluationscited by 836
  4. 2022Out of One, Many: Using Language Models to Simulate Human Samplescited by 752
  5. 2022Using Large Language Models to Simulate Multiple Humans and Replicate Human Subject Studiescited by 515
  6. 2021Annotators with Attitudes: How Annotator Beliefs And Identities Bias Toxic Language Detectioncited by 264
  7. 2024Two Tales of Persona in LLMs: A Survey of Role-Playing and Personalizationcited by 208
  8. 2023Bias Runs Deep: Implicit Reasoning Biases in Persona-Assigned LLMscited by 199
  9. 2019Are We Modeling the Task or the Annotator? An Investigation of Annotator Bias in Natural Language Understanding Datasetscited by 175
  10. 2021On Releasing Annotator-Level Labels and Information in Datasetscited by 126
  11. 2023Generative Language Models Exhibit Social Identity Biasescited by 111
  12. 2023Sensitivity, Performance, Robustness: Deconstructing the Effect of Sociodemographic Promptingcited by 81
13 more
  1. 2022Moral Mimicry: Large Language Models Produce Moral Rationalizations Tailored to Political Identitycited by 74
  2. 2023Can ChatGPT Assess Human Personalities? A General Evaluation Frameworkcited by 70
  3. 2023Simulating Social Media Using Large Language Models to Evaluate Alternative News Feed Algorithmscited by 65
  4. 2022Handling Bias in Toxic Speech Detection: A Surveycited by 61
  5. 2018Spread of hate speech in online social mediacited by 59
  6. 2023BiasAsker: Measuring the Bias in Conversational AI Systemcited by 53
  7. 2022Don't Blame the Annotator: Bias Already Starts in the Annotation Instructionscited by 47
  8. 2024Y Social: an LLM-powered Social Media Digital Twincited by 44
  9. 2023Can Large Language Models Capture Dissenting Human Voices?cited by 33
  10. 2020Intersectional Bias in Hate Speech and Abusive Language Datasetscited by 22
  11. 2021The Origin and Value of Disagreement Among Data Labelers: A Case Study of the Individual Difference in Hate Speech Annotationcited by 15
  12. 2023How Crowd Worker Factors Influence Subjective Annotations: A Study of Tagging Misogynistic Hate Speech in Tweetscited by 10
  13. 2024Beyond Trial-and-Error: Predicting User Abandonment After a Moderation Interventioncited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 3; 2025: 17; 2026: 11.

31 papers.

Sort:

By area:

  1. 2025Towards a Design Guideline for RPA Evaluation: A Survey of Large Language Model-Based Role-Playing Agentscited by 22
  2. 2024Contextualized Counterspeech: Strategies for Adaptation, Personalization, and Evaluationcited by 15
  3. 2025Is LLM an Overconfident Judge? Unveiling the Capabilities of LLMs in Detecting Offensive Language with Annotation Disagreementcited by 14
  4. 2026Assessing the Reliability of Persona-Conditioned LLMs as Synthetic Survey Respondentscited by 10
  5. 2025Guardians and Offenders: A Survey on Harmful Content Generation and Safety Mitigation of LLMcited by 9
  6. 2024Robustness and Confounders in the Demographic Alignment of LLMs with Human Perceptions of Offensivenesscited by 7
  7. 2025Hallucination Detection: A Probabilistic Framework Using Embeddings Distance Analysiscited by 6
  8. 2025What is Stigma Attributed to? A Theory-Grounded, Expert-Annotated Interview Corpus for Demystifying Mental-Health Stigmacited by 6
  9. 2025Source Coverage and Citation Bias in LLM-based vs. Traditional Search Enginescited by 6
  10. 2025Distilling Knowledge from Large Language Models: A Concept Bottleneck Model for Hate and Counter Speech Recognitioncited by 5
  11. 2024SubData: Bridging Heterogeneous Datasets to Enable Theory-Driven Evaluation of Political and Demographic Perspectives in LLMscited by 3
  12. 2025Evaluating the Simulation of Human Personality-Driven Susceptibility to Misinformation with LLMscited by 3
19 more
  1. 2025Algorithmic Fairness in NLP: Persona-Infused LLMs for Human-Centric Hate Speech Detectioncited by 2
  2. 2025AlignSurvey: A Comprehensive Benchmark for Human Preferences Alignment in Social Surveyscited by 2
  3. 2026When to Invoke: Refining LLM Fairness with Toxicity Assessmentcited by 2
  4. 2026xList-Hate: A Checklist-Based Framework for Interpretable and Generalizable Hate Speech Detectioncited by 2
  5. 2025Aligning NLP Models with Target Population Perspectives using PAIR: Population-Aligned Instance Replicationcited by 1
  6. 2026Attribute-Based Diagnosis of LLM Alignment with Hate Speech Annotationscited by 1
  7. 2026Rating the Raters: Rasch Measurement Theory for LLM Evaluationcited by 1
  8. 2025Bias in Decision-Making for AI's Ethical Dilemmas: A Comparative Study of ChatGPT and Claudecited by 0
  9. 2025Embracing Dialectic Intersubjectivity: Coordination of Different Perspectives in Content Analysis with LLM Persona Simulationcited by 0
  10. 2025"Not in My Backyard": LLMs Uncover Online and Offline Social Biases Against Homelessnesscited by 0
  11. 2025From Ground Trust to Truth: Disparities in Offensive Language Judgments on Contemporary Korean Political Discoursecited by 0
  12. 2025PRISM: Phase-enhanced Radial-based Image Signature Mapping framework for fingerprinting AI-generated imagescited by 0
  13. 2025Quantifying Feature Importance for Online Content Moderationcited by 0
  14. 2026Are we chasing ghosts? Quantifying unattributable polarization, and attributing the rest to annotator groupscited by 0
  15. 2026Toward Generalized Cross-Lingual Hateful Language Detection with Web-Scale Data and Ensemble LLM Annotationscited by 0
  16. 2026Sociodemographic Biases in Educational Counselling by Large Language Modelscited by 0
  17. 2026Semantic Gradients Interactions in SSD: A Case Study in Racial Identity and Hate Speechcited by 0
  18. 2026Two Wrongs, No Right: Auditing Social-Desirability Bias in LLM Annotators for Computational Social Sciencecited by 0
  19. 2026Contextualized Counterspeech Can Be More Persuasive Than Generic Counterspeechcited by 0