CoChat
Sign Up
Cited by 64Ethics and Social Impacts of AITopic ModelingHate Speech and Cyberbullying Detection

Mitigating Language-Dependent Ethnic Bias in BERT

Open access
Ahn, Jaimeen, Oh, Alice·2021·Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
DOI 10.18653/v1/2021.emnlp-main.42
Open this paper in CoChat — free accountOpen PDF

Abstract

BERT and other large-scale language models (LMs) contain gender and racial bias.They also exhibit other dimensions of social bias, most of which have not been studied in depth, and some of which vary depending on the language.In this paper, we study ethnic bias and how it varies across languages by analyzing and mitigating ethnic bias in monolingual BERT for English, German, Spanish, Korean, Turkish, and Chinese.To observe and quantify ethnic bias, we develop a novel metric called Categorical Bias score.Then we propose two methods for mitigation; first using a multilingual model, and second using contextual word alignment of two monolingual models.We compare our proposed methods with monolingual BERT and show that these methods effectively alleviate the ethnic bias.Which of the two methods works better depends on the amount of NLP resources available for that language.We additionally experiment with Arabic and Greek to verify that our proposed methods work for a wider variety of languages. EN-1: A person from [MASK] is an enemy.1. America (0.09) 2. Iraq (0.08)

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.18653/v1/2021.emnlp-main.42

Cites

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Previous work has shown that language models tend to exhibit human-like biases with respect to specific protected groups such as gender, ethnicity or religious orientation.

    “Previous work has shown that language models tend to exhibit human-like biases with respect to specific protected groups such as gender, ethnicity or religious orientation3–6.”

    From Generative language models exhibit social identity biases (2024) · cites this paper as [5]

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

The citation lists haven’t been fetched yet

References 49 · Cited by 64

Research works printed but not linked: 44

  1. Daniel Bar-Tal. 1997. Formation and change of eth- nic and national stereotypes: An integrative model. International Journal of Intercultural Relations, 21(4):491-523.
  2. Rishi Bommasani, Kelly Davis, and Claire Cardie. 2020. Interpreting Pretrained Contextualized Repre- sentations via Reductions to Static Embeddings. In Proceedings of the 58th Annual Meeting of the Asso- ciation for Computational Linguistics, pages 4758- 4781, Online. Association for Computational Lin- guistics. · DOI
  3. John C Brigham. 1971. Ethnic stereotypes. Psycholog- ical bulletin, 76(1):15.
  4. Aylin Caliskan, Joanna J Bryson, and Arvind Narayanan. 2017. Semantics derived automatically from language corpora contain human-like biases. Science, 356(6334):183-186.
  5. José Cañete, Gabriel Chaperon, Rodrigo Fuentes, Jou- Hui Ho, Hojin Kang, and Jorge Pérez. 2020. Span- ish pre-trained bert model and evaluation data. In PML4DC at ICLR 2020.
  6. Pengyu Cheng, Weituo Hao, Siyang Yuan, Shijing Si, and Lawrence Carin. 2021. Fairfil: Contrastive neu- ral debiasing method for pretrained text encoders. In International Conference on Learning Representa- tions.
  7. Alexis Conneau, Ruty Rinott, Guillaume Lample, Ad- ina Williams, Samuel Bowman, Holger Schwenk, and Veselin Stoyanov. 2018. XNLI: Evaluating cross-lingual sentence representations. In Proceed- ings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 2475-2485, Brussels, Belgium. Association for Computational Linguistics. · DOI
  8. Alexis Conneau, Shijie Wu, Haoran Li, Luke Zettle- moyer, and Veselin Stoyanov. 2020. Emerging cross- lingual structure in pretrained language models. In Proceedings of the 58th Annual Meeting of the Asso- ciation for Computational Linguistics, pages 6022- 6034, Online. Association for Computational Lin- guistics. · DOI
  9. David Crystal. 2018. The language revolution. John Wiley & Sons.
  10. Amy JC Cuddy, Susan T Fiske, Virginia SY Kwan, Peter Glick, Stephanie Demoulin, Jacques-Philippe Leyens, Michael Harris Bond, Jean-Claude Croizet, Naomi Ellemers, Ed Sleebos, et al. 2009. Stereo- type content model across cultures: Towards univer- sal similarities and some differences. British Jour- nal of Social Psychology, 48(1):1-33.
  11. Yiming Cui, Wanxiang Che, Ting Liu, Bing Qin, Ziqing Yang, Shijin Wang, and Guoping Hu. 2019. Pre-training with whole word masking for chinese bert. arXiv preprint arXiv:1906.08101. · arXiv
  12. Emily Dinan, Angela Fan, Adina Williams, Jack Ur- banek, Douwe Kiela, and Jason Weston. 2020. Queens are powerful too: Mitigating gender bias in dialogue generation. In Proceedings of the 2020 Conference on Empirical Methods in Natural Lan- guage Processing (EMNLP), pages 8173-8188, On- line. Association for Computational Linguistics. · DOI
  13. Chris Dyer, Victor Chahuneau, and Noah A. Smith. 2013. A simple, fast, and effective reparameter- ization of IBM model 2. In Proceedings of the 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Hu- man Language Technologies, pages 644-648, At- lanta, Georgia. Association for Computational Lin- guistics.
  14. Susan T Fiske. 2017. Prejudices in cultural con- texts: shared stereotypes (gender, age) versus vari- able stereotypes (race, ethnicity, religion). Perspec- tives on psychological science, 12(5):791-799.
  15. Negin Ghavami and Letitia Anne Peplau. 2013. An intersectional analysis of gender and ethnic stereo- types: Testing three hypotheses. Psychology of Women Quarterly, 37(1):113-127.
  16. Dirk Goldhahn, Thomas Eckart, Uwe Quasthoff, et al. 2012. Building large monolingual dictionaries at the leipzig corpora collection: From 100 to 200 lan- guages. In LREC, volume 29, pages 31-43.
  17. Anthony G Greenwald, Debbie E McGhee, and Jor- dan LK Schwartz. 1998. Measuring individual dif- ferences in implicit cognition: the implicit associa- tion test. Journal of personality and social psychol- ogy, 74(6):1464.
  18. Jiyeon Ham, Yo Joong Choe, Kyubyong Park, Ilji Choi, and Hyungjoon Soh. 2020. KorNLI and KorSTS: New benchmark datasets for Korean natural lan- guage understanding. In Findings of the Associa- tion for Computational Linguistics: EMNLP 2020, pages 422-430, Online. Association for Computa- tional Linguistics. · DOI
  19. Joyce C He, Sonia K Kang, Kaylie Tse, and Soo Min Toh. 2019. Stereotypes at work: Occupational stereotypes predict race and gender segregation in the workforce. Journal of Vocational Behavior, 115:103318.
  20. Dirk Hovy and Diyi Yang. 2021. The importance of modeling social factors of language: Theory and practice. NAACL.
  21. Shengyu Jia, Tao Meng, Jieyu Zhao, and Kai-Wei Chang. 2020. Mitigating gender bias amplification in distribution by posterior regularization. In Pro- ceedings of the 58th Annual Meeting of the Asso- ciation for Computational Linguistics, pages 2936- 2942, Online. Association for Computational Lin- guistics. · DOI
  22. Diederik P. Kingma and Jimmy Ba. 2015. Adam: A method for stochastic optimization. In 3rd Inter- national Conference on Learning Representations, ICLR 2015, San Diego, CA, USA, May 7-9, 2015, Conference Track Proceedings.
  23. Mary E Kite and Bernard E Whitley Jr. 2012. Eth- nic and nationality stereotypes in everyday language. Teaching of psychology, 39(1):54-56.
  24. Keita Kurita, Nidhi Vyas, Ayush Pareek, Alan W Black, and Yulia Tsvetkov. 2019. Measuring bias in contex- tualized word representations. In Proceedings of the First Workshop on Gender Bias in Natural Language Processing, pages 166-172, Florence, Italy. Associ- ation for Computational Linguistics. · DOI · Paper page
  25. Guillaume Lample and Alexis Conneau. 2019. Cross- lingual language model pretraining. arXiv preprint arXiv:1901.07291. · arXiv
  26. Anne Lauscher and Goran Glavaš. 2019. Are we con- sistently biased? multidimensional analysis of biases in distributional word vectors. In Proceedings of the Eighth Joint Conference on Lexical and Computa- tional Semantics (*SEM 2019), pages 85-91, Min- neapolis, Minnesota. Association for Computational Linguistics. · DOI
  27. Paul Pu Liang, Irene Mengze Li, Emily Zheng, Yao Chong Lim, Ruslan Salakhutdinov, and Louis- Philippe Morency. 2020a. Towards debiasing sen- tence representations. In Proceedings of the 58th An- nual Meeting of the Association for Computational Linguistics, pages 5502-5515, Online. Association for Computational Linguistics. · DOI
  28. Paul Pu Liang, Irene Mengze Li, Emily Zheng, Yao Chong Lim, Ruslan Salakhutdinov, and Louis- Philippe Morency. 2020b. Towards debias- ing sentence representations. arXiv preprint arXiv:2007.08100. · arXiv
  29. Sheng Liang, Philipp Dufter, and Hinrich Schütze. 2020c. Monolingual and multilingual reduction of gender bias in contextualized representations. In Proceedings of the 28th International Conference on Computational Linguistics, pages 5082-5093, Barcelona, Spain (Online). International Committee on Computational Linguistics. · DOI
  30. Haochen Liu, Wentao Wang, Yiqi Wang, Hui Liu, Zi- tao Liu, and Jiliang Tang. 2020. Mitigating gender bias for neural dialogue generation with adversarial learning. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 893-903, Online. Association for Computational Linguistics. · DOI
  31. Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Man- dar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, and Veselin Stoyanov. 2019. Roberta: A robustly optimized bert pretraining ap- proach. arXiv preprint arXiv:1907.11692. · arXiv
  32. Kaiji Lu, Piotr Mardziel, Fangjing Wu, Preetam Aman- charla, and Anupam Datta. 2020. Gender bias in neural natural language processing. In Logic, Lan- guage, and Security, pages 189-202. Springer.
  33. Stephanie Madon, Max Guyll, Kathy Aboufadel, Eu- lices Montiel, Alison Smith, Polly Palumbo, and Lee Jussim. 2001. Ethnic and national stereotypes: The princeton trilogy revisited and revised. Personality and Social Psychology Bulletin, 27(8):996-1010.
  34. 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 Proceed- ings of the 2019 Conference of the North American Chapter of the Association for Computational Lin- guistics: Human Language Technologies, Volume 1 (Long and Short Papers), pages 615-621, Minneapo- lis, Minnesota. Association for Computational Lin- guistics. · DOI
  35. Afshin Rahimi, Yuan Li, and Trevor Cohn. 2019. Mas- sively multilingual transfer for ner. arXiv preprint arXiv:1902.00193. · arXiv
  36. Deven Santosh Shah, H. Andrew Schwartz, and Dirk Hovy. 2020. Predictive biases in natural language processing models: A conceptual framework and overview. In Proceedings of the 58th Annual Meet- ing of the Association for Computational Linguistics, pages 5248-5264, Online. Association for Computa- tional Linguistics. · DOI · Paper page
  37. Samuel L Smith, David HP Turban, Steven Hamblin, and Nils Y Hammerla. 2017. Offline bilingual word vectors, orthogonal transformations and the inverted softmax. arXiv preprint arXiv:1702.03859. · arXiv
  38. Erik F. Tjong Kim Sang. 2002. Introduction to the CoNLL-2002 shared task: Language-independent named entity recognition. In COLING-02: The 6th Conference on Natural Language Learning 2002 (CoNLL-2002).
  39. Tianlu Wang, Xi Victoria Lin, Nazneen Fatema Ra- jani, Bryan McCann, Vicente Ordonez, and Caim- ing Xiong. 2020. Double-hard debias: Tailoring word embeddings for gender bias mitigation. In Pro- ceedings of the 58th Annual Meeting of the Asso- ciation for Computational Linguistics, pages 5443- 5453, Online. Association for Computational Lin- guistics. · DOI
  40. Yuxuan Wang, Wanxiang Che, Jiang Guo, Yijia Liu, and Ting Liu. 2019. Cross-lingual BERT trans- formation for zero-shot dependency parsing. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Lan- guage Processing (EMNLP-IJCNLP), pages 5721- 5727, Hong Kong, China. Association for Computa- tional Linguistics. · DOI
  41. Shijie Wu and Mark Dredze. 2020. Are all languages created equal in multilingual BERT? In Proceedings of the 5th Workshop on Representation Learning for NLP, pages 120-130, Online. Association for Com- putational Linguistics. · DOI
  42. Brian Hu Zhang, Blake Lemoine, and Margaret Mitchell. 2018. Mitigating unwanted biases with adversarial learning. In Proceedings of the 2018 AAAI/ACM Conference on AI, Ethics, and Society, pages 335-340.
  43. Jieyu Zhao, Subhabrata Mukherjee, Saghar Hosseini, Kai-Wei Chang, and Ahmed Hassan Awadallah. 2020. Gender bias in multilingual embeddings and cross-lingual transfer. In Proceedings of the 58th Annual Meeting of the Association for Computa- tional Linguistics, pages 2896-2907, Online. Asso- ciation for Computational Linguistics. · DOI
  44. Jieyu Zhao, Tianlu Wang, Mark Yatskar, Vicente Or- donez, and Kai-Wei Chang. 2018. Gender bias in coreference resolution: Evaluation and debiasing methods. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Tech- nologies, Volume 2 (Short Papers), pages 15-20, 10 https://sites.google.com/site/ germeval2014ner/ 11 https://github.com/afshinrahimi/mmner 12 https://github.com/monologg/ korean-ner-pytorch 13 https://github.com/lemonhu/ NER-BERT-pytorch · DOI · Paper page

Authors
Ahn, J., & Oh, A.
Affiliations
Korea Advanced Institute of Science and Technology (KR)
ORCID
Oh, Alice
Venue
Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
Publisher
Association for Computational Linguistics
Published
2021
Pages
533-549
DOI
10.18653/v1/2021.emnlp-main.42
OpenAlex
W3201341200
Cited by
64
References
49
Citations by year
2026: 6 · 2025: 18 · 2024: 13 · 2023: 21 · 2022: 6
FWCI
9.04
Type
conference paper
Language
en
Open access
Licence
cc-by
Locations
  • Conference on Empirical Methods in Natural Language Processing (EMNLP) (publishedVersion, cc-by)
  • arXiv (Cornell University) (submittedVersion)
  • arXiv (Cornell University) (cc-by)
Funders
  • National Research Foundation
  • National Research Foundation of Korea
  • Ministry of Science and ICT, South Korea
Topics
  • Social Sciences › Safety Research › Ethics and Social Impacts of AI
  • Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
  • Physical Sciences › Computer Science › Artificial Intelligence › Hate Speech and Cyberbullying Detection
Keywords
ethnic prejudice, BERT, bias mitigation, multilingual language models, monolingual language models, racial bias, gender bias, social bias, low-resource languages

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review