Elamur Research
☰
home / papers / 2021 / Latent Hatred: A Benchmark for Understanding Implicit Hate Speech

Latent Hatred: A Benchmark for Understanding Implicit Hate Speech

Written for the Research library (claude-opus-5-5) from the paper text. Original — aclanthology.org/2021.emnlp-main.29.pdf

Summary

Most hate speech research and datasets target overt abuse marked by slurs or profanity. Much online hostility is instead veiled in sarcasm, euphemism, stereotypes or coded threats, and it slips past keyword-based filters. ElSherief et al. address this by proposing a six-category taxonomy of implicit hate grounded in social science work. They use it to annotate a Twitter corpus drawn from the most prevalent U.S. hate groups and their followers. Each implicitly hateful tweet receives a fine-grained category plus free-text descriptions of the targeted group and the implied message. The authors then build baselines for two tasks: classifying implicit hate and generating explanations of it, using fine-tuned GPT and GPT-2. Neural classifiers handle coarse hate detection reasonably well but struggle with the fine-grained types. Generation results look encouraging, though overlap metrics likely underrate semantically correct outputs. The released resource offers a benchmark for subtle, deniable hate, which is what extremist communities increasingly rely on. It also supports explainable moderation tools.

Key points

  • The taxonomy has six classes: White Grievance, Incitement to Violence, Inferiority Language, Irony, Stereotypes and Misinformation, and Threatening and Intimidation. When applied to a sample chosen to be representative, these classes jointly cover 98.6% of the implicitly hateful content.
  • The 6,346 implicit hate tweets each carry two free-text annotations of target group and implied statement. They are split 75/12.5/12.5 for training, validation and testing, with no post appearing in more than one split.
  • Explanations were generated with GPT and GPT-2 and scored with BLEU and ROUGE-L, both averaged over references and as maxima (BLEU∗, ROUGE-L∗). GPT-2 outperformed GPT, probably because its web-text pretraining is closer to the tweet domain.
  • Generating the target group is easier than generating the implied statement (e.g., 83.9 vs 75.3 BLEU). Beam search was the best decoding strategy, followed by top-p sampling with p = 0.92.
  • The data is public at github.com/GT-SALT/implicit-hate. Compared with most earlier corpora, it contains far fewer expletives in the hate class.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

Fields: Natural language processing

Cited by

174 papers of our library cite it (of the 140673 whose bibliographies we read); 3 of them have a page here, 171 not yet.

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2021.emnlp-main.29

@inproceedings{elsherief2021latent,
  title = {{Latent Hatred: A Benchmark for Understanding Implicit Hate Speech}},
  author = {ElSherief, Mai and Ziems, Caleb and Muchlinski, David and Anupindi, Vaishnavi and Seybolt, Jordyn and Choudhury, Munmun De and Yang, Diyi},
  booktitle = {Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing},
  year = {2021},
  doi = {10.18653/v1/2021.emnlp-main.29},
  url = {https://doi.org/10.18653/v1/2021.emnlp-main.29}
}

References

78 references, 14 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Additionally, we identified eight challenges in implicit hate speech detection: coded hate symbols, discourse relations, entity framing, commonsense, metaphorical language, colloquial speech, irony, and identity term bias. To mitigate these challenges, future work could explore deciphering models fo… find
  2. We demonstrate that our corpus can serve as a useful research benchmark for understanding implicit hate speech online. Our work also has implications towards the emerging directions of countering online hate speech (Citron and Norton, 2011; Mathew et al., 2019), detecting online radicalization (Ferr… find
  3. Pinkesh Badjatiya, Shashank Gupta, Manish Gupta, and Vasudeva Varma. 2017. Deep learning for hate speech detection in tweets. In Proceedings of the 26th International Conference on World Wide Web Companion, pages 759–760. find
  4. Valerio Basile, Cristina Bosco, Elisabetta Fersini, Debora Nozza, Viviana Patti, Francisco Manuel Rangel Pardo, Paolo Rosso, and Manuela Sanguinetti. 2019. SemEval-2019 task 5: Multilingual detection of hate speech against immigrants and women in Twitter. In Proceedings of the 13th International Wor… find
  5. Uladzislau Belavusau. 2017. Hate speech. Max Planck Encyclopedia of Comparative Constitutional Law (Oxford University Press, 2017 Forthcoming). find
  6. Mitch Berbrier. 2000. The victim ideology of white supremacists and white separatists in the united states. Sociological Focus, 33(2):175–191. find
  7. Katrina Rebecca Bloch, Tiffany Taylor, and Karen Martinez. 2020. Playing the race card: White injury, white victimhood and the paradox of colour-blind ideology in anti-immigrant discourse. Ethnic and Racial Studies, 43(7):1130–1148. find
  8. This study has been approved by the Institutional Review Board (IRB) at the researchers’ institution. For the annotation process, we included a warning in the instructions that the content might be offensive or upsetting. Annotators were also encouraged to stop the labeling process if they were over… find
  9. Su Lin Blodgett, Solon Barocas, Hal Daumé III, and Hanna Wallach. 2020. Language (technology) is power: A critical survey of “bias” in NLP. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 5454– 5476, Online. Association for Computational Linguistics.… find our card
  10. The authors would like to thank the members of SALT lab and the anonymous reviewers for their thoughtful feedback. The annotation process was funded through the GaTech School of Interactive Computing (SIC) and Institute for Data Engineering and Science (IDEAS) data curation award to ElSherief and Ya… find
  11. Antoine Bosselut, Hannah Rashkin, Maarten Sap, Chaitanya Malaviya, Asli Celikyilmaz, and Yejin Choi. 2019. COMET: Commonsense transformers for automatic knowledge graph construction. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 4762–4779, Florence… find our card
  12. Miranda Bryant. 2019. Instagram’s anti-bullying ai asks users: ‘are you sure you want to post this?’. The Guardian. find
  13. Peter Burnap and Matthew Leighton Williams. 2014. Hate speech, machine classification and statistical modelling of information flows on twitter: Interpretation and communication for policy decision making. find
  14. Tommaso Caselli, Valerio Basile, Jelena Mitrović, Inga Kartoziya, and Michael Granitzer. 2020. I feel offended, don’t be abusive! implicit/explicit messages in offensive and abusive language. In Proceedings of the 12th Language Resources and Evaluation Conference, pages 6193–6202, Marseille, France… find
  15. Danielle Keats Citron and Helen Norton. 2011. Intermediaries and hate speech: Fostering digital citizenship for our information age. Boston University Law Review, 91:1435. find
  16. Raphael Cohen-Almagor. 2009. Holocaust denial is a form of hate speech. In Amsterdam Law Forum, volume 2, pages 33–42. find
  17. Thomas Davidson, Debasmita Bhattacharya, and Ingmar Weber. 2019. Racial bias in hate speech and abusive language detection datasets. In Proceedings of the Third Workshop on Abusive Language Online, pages 25–35, Florence, Italy. Association for Computational Linguistics. find our card
  18. Thomas Davidson, Dana Warmsley, Michael Macy, and Ingmar Weber. 2017. Automated hate speech detection and the problem of offensive language. ArXiv preprint, abs/1703.04009. find our card
  19. Quentin Dénigot and Heather Burnett. 2020. Dogwhistles as identity-based interpretative variation. In Proceedings of the Probability and Meaning Conference (PaM 2020), pages 17–25, Gothenburg. Association for Computational Linguistics. find
  20. Empirical Methods in Natural Language Processing (EMNLP), pages 653–670, Online. Association for Computational Linguistics. find
  21. Paula Fortuna and Sérgio Nunes. 2018. A survey on automatic detection of hate speech in text. ACM Computing Surveys (CSUR), 51(4):1–30. find
  22. Antigoni-Maria Founta, Constantinos Djouvas, Despoina Chatzakou, Ilias Leontiadis, Jeremy Blackburn, Gianluca Stringhini, Athena Vakali, Michael Sirivianos, and Nicolas Kourtellis. 2018. Large scale crowdsourcing and characterization of twitter abusive behavior. Proceedings of the 12th International… find
  23. Lei Gao and Ruihong Huang. 2017. Detecting online hate speech using context aware models. In Proceedings of the International Conference Recent Advances in Natural Language Processing, RANLP 2017, pages 260–266, Varna, Bulgaria. INCOMA Ltd. find
  24. Lei Gao, Alexis Kuppersmith, and Ruihong Huang. 2017. Recognizing explicit and implicit hate speech using a weakly supervised two-path bootstrapping approach. In Proceedings of the Eighth International Joint Conference on Natural Language Processing (Volume 1: Long Papers), pages 774–782, Taipei, Ta… find
  25. Jonas Gehring, Michael Auli, David Grangier, Denis Yarats, and Yann N. Dauphin. 2017. Convolutional sequence to sequence learning. In Proceedings of the 34th International Conference on Machine Learning, ICML 2017, Sydney, NSW, Australia, 6-11 August 2017, volume 70 of Proceedings of Machine Learnin… find our card
  26. Lucas Dixon, John Li, Jeffrey Sorensen, Nithum Thain, and Lucy Vasserman. 2018. Measuring and mitigating unintended bias in text classification. In Proceedings of the 2018 AAAI/ACM Conference on AI, Ethics, and Society, pages 67–73. find
  27. Ona de Gibert, Naiara Perez, Aitor García-Pablos, and Montse Cuadros. 2018. Hate speech dataset from a white supremacy forum. In Proceedings of the 2nd Workshop on Abusive Language Online (ALW2), pages 11–20, Brussels, Belgium. Association for Computational Linguistics. find our card
  28. Nemanja Djuric, Jing Zhou, Robin Morris, Mihajlo Grbovic, Vladan Radosavljevic, and Narayan Bhamidipati. 2015. Hate speech detection with comment embeddings. In Proceedings of the 24th international conference on world wide web, pages 29–30. find
  29. Joshua R Gubler and Nathan P Kalmoe. 2015. Violent rhetoric in protracted group conflicts: Experimental evidence from israel and india. Political Research Quarterly, 68(4):651–664. find
  30. Tom Dreisbach. 2021. irony to spread hate. find
  31. How extremists weaponize Emilio Ferrara, Wen-Qiang Wang, Onur Varol, Alessandro Flammini, and Aram Galstyan. 2016. Predicting online extremism, content adopters, and interaction reciprocity. In International conference on social informatics, pages 22–39. Springer. find
  32. Maxwell Forbes, Jena D. Hwang, Vered Shwartz, Maarten Sap, and Yejin Choi. 2020. Social chemistry 101: Learning to reason about social and moral norms. In Proceedings of the 2020 Conference on Nitin Indurkhya and Fred J Damerau. 2010. Handbook of natural language processing, volume 2. CRC Press. find
  33. Md Rafiqul Islam, Shaowu Liu, Xianzhi Wang, and Guandong Xu. 2020. Deep learning for misinformation detection on online social networks: a survey and new perspectives. Social Network Analysis and Mining, 10(1):1–20. find
  34. Ryan Jones. 2020. List of swear words, bad words, curse words - starting with a. find
  35. David Jurgens, Libby Hemphill, and Eshwar Chandrasekharan. 2019. A just and comprehensive strategy for using NLP to address online abuse. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 3658– 3666, Florence, Italy. Association for Computational Lingu… find
  36. Raquel Justo, Thomas Corcoran, Stephanie M Lukin, Marilyn Walker, and M Inés Torres. 2014. Extracting relevant knowledge for the detection of sarcasm and nastiness in the social web. Knowledge-Based Systems, 69:124–133. find
  37. Nishant Kambhatla, Anahita Mansouri Bigvand, and Anoop Sarkar. 2018. Decipherment of substitution ciphers with neural language models. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 869–874, Brussels, Belgium. Association for Computational Linguistic… find
  38. Brendan Kennedy, Mohammad Atari, Aida M Davani, Leigh Yeh, Ali Omrani, Yehsong Kim, Kris Coombs Jr, Shreya Havaldar, Gwenyth Portillo-Wightman, Elaine Gonzalez, et al. 2018. The gab hate corpus: A collection of 27k posts annotated for hate speech. PsyArXiv. July, 18. find
  39. David Klepper. 2020. Facebook removes nearly 200 accounts tied to hate groups. find
  40. Jonathan Leader Maynard and Susan Benesch. 2016. Dangerous speech and dangerous ideology: An integrated model for monitoring and prevention. Genocide Studies and Prevention. find
  41. Chin-Yew Lin. 2004. ROUGE: A package for automatic evaluation of summaries. In Text Summarization Branches Out, pages 74–81, Barcelona, Spain. Association for Computational Linguistics. find
  42. Rijul Magu and Jiebo Luo. 2018. Determining code words in euphemistic hate speech using word embedding networks. In Proceedings of the 2nd Workshop on Abusive Language Online (ALW2), pages 93–100, Brussels, Belgium. Association for Computational Linguistics. find
  43. Thomas Manzini, Lim Yao Chong, Alan W Black, and Yulia Tsvetkov. 2019. Black is to criminal as caucasian is to police: Detecting and removing multiclass bias in word embeddings. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Huma… find
  44. Binny Mathew, Punyajoy Saha, Hardik Tharad, Subham Rajgaria, Prajwal Singhania, Suman Kalyan Maity, Pawan Goyal, and Animesh Mukherjee. 2019. Thou shalt not hate: Countering online hate speech. find
  45. In Proceedings of the International AAAI Conference on Web and Social Media, volume 13, pages 369–380. find
  46. Cynthia Miller-Idriss. 2020. Hate in the homeland: The new global far right. Princeton University Press. find
  47. Burt L Monroe, Michael P Colaresi, and Kevin M Quinn. 2008. Fightin’words: Lexical feature selection and evaluation for identifying the content of political conflict. Political Analysis, 16(4):372–403. find
  48. Rhiannon S Neilsen. 2015. ‘toxification’as a more precise early warning sign for genocide than dehumanization? an emerging research agenda. Genocide Studies and Prevention: An International Journal, 9(1):9. find
  49. Laura Beth Nielsen. 2002. Subtle, pervasive, harmful: Racist and sexist remarks in public as hate speech. Journal of Social Issues, 58(2):265–280. find
  50. John T Nockleby. 2000. Hate speech. Encyclopedia of the American constitution, 3(2):1277–1279. find
  51. Kishore Papineni, Salim Roukos, Todd Ward, and Wei-Jing Zhu. 2002. Bleu: a method for automatic evaluation of machine translation. In Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics, pages 311–318, Philadelphia, Pennsylvania, USA. Association for Computational… find our card
  52. Jeffrey Pennington, Richard Socher, and Christopher Manning. 2014. GloVe: Global vectors for word representation. In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 1532–1543, Doha, Qatar. Association for Computational Linguistics. find
  53. Shruti Phadke and Tanushree Mitra. 2020. Many faced hate: A cross platform study of content framing and information sharing by online hate groups. In CHI ’20: CHI Conference on Human Factors in Computing Systems, Honolulu, HI, USA, April 25-30, 2020, pages 1–13. ACM. find
  54. James A Piazza. 2020. Politician hate speech and domestic terrorism. International Interactions, pages 1–23. find
  55. Jing Qian, Mai ElSherief, Elizabeth Belding, and William Yang Wang. 2019. Learning to decipher hate symbols. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), pages 3006… find
  56. Jing Qian, Hong Wang, Mai ElSherief, and Xifeng Yan. 2021. Lifelong learning of hate speech classification on social media. In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 2304–2314, Online. Asso… find
  57. International AAAI Conference on Web and Social Media, pages 687–690. AAAI. find
  58. Alec Radford, Karthik Narasimhan, Tim Salimans, and Ilya Sutskever. 2018. Improving language understanding with unsupervised learning. Technical report, OpenAI. find
  59. Keith Somerville. 2011. Violence, hate speech and inflammatory broadcasting in kenya: The problems of definition and identification. Ecquid Novi: African Journalism Studies, 32(1):82–101. find
  60. Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei, and Ilya Sutskever. 2019. Language models are unsupervised multitask learners. OpenAI blog, 1(8):9. find
  61. Robyn Speer, Joshua Chin, and Catherine Havasi. 2017. Conceptnet 5.5: An open multilingual graph of general knowledge. In Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence, February 4-9, 2017, San Francisco, California, USA, pages 4444–4451. AAAI Press. find
  62. Punyajoy Saha, Binny Mathew, Pawan Goyal, and Animesh Mukherjee. 2018. Hateminers: detecting hate speech against women. ArXiv preprint, abs/1812.06700. — [arXiv:1812.06700](https://arxiv.org/abs/1812.06700) find
  63. Manuela Sanguinetti, Fabio Poletto, Cristina Bosco, Viviana Patti, and Marco Stranisci. 2018. An italian twitter corpus of hate speech against immigrants. In Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018). find
  64. Maarten Sap, Dallas Card, Saadia Gabriel, Yejin Choi, and Noah A. Smith. 2019. The risk of racial bias in hate speech detection. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 1668–1678, Florence, Italy. Association for Computational Linguistics. find our card
  65. Maarten Sap, Saadia Gabriel, Lianhui Qin, Dan Jurafsky, Noah A. Smith, and Yejin Choi. 2020. Social bias frames: Reasoning about social and power implications of language. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 5477–5490, Online. Association… find our card
  66. Anna Schmidt and Michael Wiegand. 2017. A survey on hate speech detection using natural language processing. In Proceedings of the Fifth International Workshop on Natural Language Processing for Social Media, pages 1–10, Valencia, Spain. Association for Computational Linguistics. find our card
  67. Ashish Sharma, Adam Miner, David Atkins, and Tim Althoff. 2020. A computational approach to understanding empathy expressed in text-based mental health support. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 5263–5276, Online. Association for… find our card
  68. Elisa Shearer and Jeffrey Gottfried. 2017. News use across social media platforms 2017. Pew Research Center, 7(9). find
  69. Kim Bartel Sheehan. 2018. Crowdsourcing research: data collection with amazon’s mechanical turk. Communication Monographs, 85(1):140–156. find
  70. Leandro Silva, Mainack Mondal, Denzil Correa, Fabrício Benevenuto, and Ingmar Weber. 2016. Analyzing the targets of hate in online social media. In 10th Derald Wing Sue. 2010. Microaggressions in everyday life: Race, gender, and sexual orientation. John Wiley & Sons. find
  71. Twitter. 2021. Twitter’s policy on hateful conduct \| twitter help. find
  72. Denny Vrandečić and Markus Krötzsch. 2014. Wikidata: a free collaborative knowledgebase. Communications of the ACM, 57(10):78–85. find
  73. William Warner and Julia Hirschberg. 2012. Detecting hate speech on the world wide web. In Proceedings of the Second Workshop on Language in Social Media, pages 19–26, Montréal, Canada. Association for Computational Linguistics. find
  74. Zeerak Waseem, Thomas Davidson, Dana Warmsley, and Ingmar Weber. 2017. Understanding abuse: A typology of abusive language detection subtasks. In Proceedings of the First Workshop on Abusive Language Online, pages 78–84, Vancouver, BC, Canada. Association for Computational Linguistics. find our card
  75. Zeerak Waseem and Dirk Hovy. 2016. Hateful symbols or hateful people? predictive features for hate speech detection on Twitter. In Proceedings of the NAACL Student Research Workshop, pages 88–93, San Diego, California. Association for Computational Linguistics. find
  76. Michael Wiegand, Josef Ruppenhofer, and Thomas Kleinbauer. 2019. Detection of Abusive Language: the Problem of Biased Datasets. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short… find our card
  77. Mengzhou Xia, Anjalie Field, and Yulia Tsvetkov. 2020. Demoting racial bias in hate speech detection. In Proceedings of the Eighth International Workshop on Natural Language Processing for Social Media, pages 7–14, Online. Association for Computational Linguistics. find
  78. Marcos Zampieri, Shervin Malmasi, Preslav Nakov, Sara Rosenthal, Noura Farra, and Ritesh Kumar. 2019. Predicting the type and target of offensive posts in social media. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Languag… find our card