mlsanityData to 7 Oct 2026

Measuring the Reliability of Hate Speech Annotations: The Case of the European Refugee Crisis

arXiv 1701.08118 ↗First posted 27 Jan 2017cs.CLcited by 95

Not on the map. Why?

Abstract (arXiv)

Some users of social media are spreading racist, sexist, and otherwise hateful content. For the purpose of training a hate speech detection system, the reliability of the annotations is crucial, but there is no universally agreed-upon definition. We collected potentially hateful messages and asked two groups of internet users to determine whether they were hate speech or not, whether they should be banned or not and to rate their degree of offensiveness. One of the groups was shown a definition prior to completing the survey. We aimed to assess whether hate speech can be annotated reliably, and the extent to which existing definitions are in accordance with subjective ratings. Our results indicate that showing users a definition caused them to partially align their own opinion with the definition but did not improve reliability, which was very low overall. We conclude that the presence of hate speech should perhaps not be considered a binary yes-or-no decision, and raters need more detailed instructions for the annotation.

Key facts. Cited by 95 papers in the map (data through 7 Oct 2026). First posted on arXiv 27 Jan 2017.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 26; 2022: 14; 2023: 16; 2024: 22; 2025: 10; 2026: 7.

95 papers.

Sort:

By area:

  1. 2022ToxiGen: A Large-Scale Machine-Generated Dataset for Adversarial and Implicit Hate Speech Detectioncited by 419
  2. 2021Annotators with Attitudes: How Annotator Beliefs And Identities Bias Toxic Language Detectioncited by 258
  3. 2023Large Language Model Alignment: A Surveycited by 241
  4. 2021Challenges in Detoxifying Language Modelscited by 169
  5. 2021Challenges in Automated Debiasing for Toxic Language Detectioncited by 95
  6. 2021Human-in-the-Loop for Data Collection: a Multi-Target Counter Narrative Dataset to Fight Online Hate Speechcited by 81
  7. 2022Language Generation Models Can Cause Harm: So What Can We Do About It? An Actionable Surveycited by 80
  8. 2022Handling Bias in Toxic Speech Detection: A Surveycited by 60
  9. 2024CultureLLM: Incorporating Cultural Differences into Large Language Modelscited by 49
  10. 2021Hate Towards the Political Opponent: A Twitter Corpus Study of the 2020 US Elections on the Basis of Offensive Speech and Stance Detectioncited by 45
  11. 2023Facilitating Fine-grained Detection of Chinese Toxic Language: Hierarchical Taxonomy, Resources, and Benchmarkscited by 45
  12. 2023Exploring Cross-Cultural Differences in English Hate Speech Annotations: From Dataset Construction to Analysiscited by 44
38 more
  1. 2021Whose Opinions Matter? Perspective-aware Models to Identify Opinions of Hate Speech Victims in Abusive Language Detectioncited by 38
  2. 2021A systematic review of Hate Speech automatic detection using Natural Language Processingcited by 26
  3. 2024"They are uncultured": Unveiling Covert Harms and Social Threats in LLM Generated Conversationscited by 24
  4. 2024CulturePark: Boosting Cross-cultural Understanding in Large Language Modelscited by 24
  5. 2024From Languages to Geographies: Towards Evaluating Cultural Bias in Hate Speech Datasetscited by 23
  6. 2024Hate Personified: Investigating the role of LLMs in content moderationcited by 23
  7. 2021Multilingual Offensive Language Identification for Low-resource Languagescited by 22
  8. 2022Listening to Affected Communities to Define Extreme Speech: Dataset and Experimentscited by 21
  9. 2024The Colorful Future of LLMs: Evaluating and Improving LLMs as Emotional Supporters for Queer Youthcited by 20
  10. 2023Vicarious Offense and Noise Audit of Offensive Speech Classifiers: Unifying Human and Machine Disagreement on What is Offensivecited by 16
  11. 2021"Short is the Road that Leads from Fear to Hate": Fear Speech in Indian WhatsApp Groupscited by 13
  12. 2021Hate Speech Classifiers Learn Human-Like Social Stereotypescited by 12
  13. 2024Towards Comprehensive Detection of Chinese Harmful Memescited by 12
  14. 2024Multilingual and Explainable Text Detoxification with Parallel Corporacited by 12
  15. 2021Dataset of Fake News Detection and Fact Verification: A Surveycited by 11
  16. 2022Explainable Abuse Detection as Intent Classification and Slot Fillingcited by 11
  17. 2021An Information Retrieval Approach to Building Datasets for Hate Speech Detectioncited by 9
  18. 2024Subjective textitIsms? On the Danger of Conflating Hate and Offence in Abusive Language Detectioncited by 9
  19. 2023Multi-modal Hate Speech Detection using Machine Learningcited by 8
  20. 2025Operationalizing Pluralistic Values in Large Language Model Alignment Reveals Trade-offs in Safety, Inclusivity, and Model Behaviorcited by 8
  21. 2022Efficient, Uncertainty-based Moderation of Neural Networks Text Classifierscited by 7
  22. 2023BiasX: "Thinking Slow" in Toxic Content Moderation with Explanations of Implied Social Biasescited by 7
  23. 2024Multi3Hate: Multimodal, Multilingual, and Multicultural Hate Speech Detection with Vision-Language Modelscited by 7
  24. 2021Introducing an Abusive Language Classification Framework for Telegram to Investigate the German Hater Communitycited by 6
  25. 2021Toxicity Detection can be Sensitive to the Conversational Contextcited by 6
  26. 2022Codes, Patterns and Shapes of Contemporary Online Antisemitism and Conspiracy Narratives -- an Annotation Guide and Labeled German-Language Dataset in the Context of COVID-19cited by 6
  27. 2023BAN-PL: a Novel Polish Dataset of Banned Harmful and Offensive Content from Wykop.pl web servicecited by 6
  28. 2023Modeling subjectivity (by Mimicking Annotator Annotation) in toxic comment identification across diverse communitiescited by 6
  29. 2024DefVerify: Do Hate Speech Models Reflect Their Dataset's Definition?cited by 6
  30. 2025LLM Content Moderation and User Satisfaction: Evidence from Response Refusals in Chatbot Arenacited by 6
  31. 2025A Modular Taxonomy for Hate Speech Definitions and Its Impact on Zero-Shot LLM Classification Performancecited by 6
  32. 2021Retweet communities reveal the main sources of hate speechcited by 5
  33. 2021Mitigation of Diachronic Bias in Fake News Detection Datasetcited by 5
  34. 2021Practical Benefits of Feature Feedback Under Distribution Shiftcited by 5
  35. 2021Hate, Obscenity, and Insults: Measuring the Exposure of Children to Inappropriate Comments in YouTubecited by 4
  36. 2021Modeling Users and Online Communities for Abuse Detection: A Position on Ethics and Explainabilitycited by 4
  37. 2021Who Blames or Endorses Whom? Entity-to-Entity Directed Sentiment Extraction in News Textcited by 4
  38. 2021The ComMA Dataset V0.2: Annotating Aggression and Bias in Multilingual Social Media Discoursecited by 4

50 most cited of 95 All 95 (JSON)