Document Open Access Logo

Polarizations in Static Word Embeddings: Investigating Bias in Marginalized Dialects of Portuguese

Authors Raimundo Juracy Campos Ferro Junior , Eugénio Ribeiro , Leonardo Sampaio Rocha

PDF

File

Thumbnail PDF
PDF
OASIcs.SLATE.2026.7.pdf
  • Filesize: 0.75 MB
  • 15 pages

Document Identifiers

Subject Classification

ACM Subject Classification
  • Computing methodologies → Natural language processing
Keywords
  • word embeddings
  • bias detection
  • Portuguese
  • LGBTQIAP+
  • pajubá
  • NLP
  • lexical association
  • static embeddings

Metrics

Abstract

This study investigates bias in static word embeddings applied to Pajubá, a dialect spoken within the Brazilian LGBTQIAP+ community. Four models from the NILC repository (Word2Vec, GloVe, FastText, and Wang2Vec) were evaluated across two dimensions: representational capacity and polarity analysis. The vocabulary coverage test revealed that approximately 74% of dialectal terms are present in the models' vector spaces. The RND and SC-WEAT tests consistently point toward negative associations for identity terms across all models, with the SC-WEAT further suggesting that reappropriated dialect terms carry their standard Portuguese positive valence into the embedding space, while explicit identitarian markers remain negatively encoded. An adjectivation test confirms stereotypical associations, including the linkage of travesti with criminality and the pathologisation of gay. These findings suggest that static word embeddings fail to capture the sociolinguistic complexity of Pajubá and systematically reflect the dominant and often discriminatory discourses present in the training corpora.

Cite As Get BibTex

Raimundo Juracy Campos Ferro Junior, Eugénio Ribeiro, and Leonardo Sampaio Rocha. Polarizations in Static Word Embeddings: Investigating Bias in Marginalized Dialects of Portuguese. In 15th Symposium on Languages, Applications and Technologies (SLATE 2026). Open Access Series in Informatics (OASIcs), Volume 144, pp. 7:1-7:15, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2026) https://doi.org/10.4230/OASIcs.SLATE.2026.7

Author Details

Raimundo Juracy Campos Ferro Junior
  • Universidade Federal do Ceará, Fortaleza, Brasil
  • ISTAR, Instituto Universitário de Lisboa (ISCTE-IUL), Portugal
Eugénio Ribeiro
  • INESC-ID Lisboa, Portugal
  • ISTAR, Instituto Universitário de Lisboa (ISCTE-IUL), Portugal
Leonardo Sampaio Rocha
  • Universidade Federal do Ceará, Fortaleza, Brasil

Funding

This work was supported by Portuguese national funds through Fundação para a Ciência e a Tecnologia, I.P. (FCT) under projects UID/50021/2025 (DOI: https://doi.org/10.54499/UID/50021/2025), UID/PRR/50021/2025 (DOI: https://doi.org/10.54499/UID/PRR/50021/2025), and UID/04466/2025 (DOI: https://doi.org/10.54499/UID/04466/2025).
  • Ferro Junior, Raimundo Juracy Campos: Funded by CAPES (PDSE scholarship)

References

  1. Karylleila dos Santos Andrade, Sheila de Carvalho P. Gonçalves, Filipe Porto, and Luciana C. e Silva Andrade. Bajubá: linguagem de grupo lgbtt como representação sócio‑histórica e cultural. Desafios - Revista Interdisciplinar da UFTO, 5(4):36-46, 2018. Google Scholar
  2. Gabriela Costa Araujo. Bajubá: memórias e diálogos das travestis. Paco Editorial, Jundiaí, SP, 1 edition, 2019. Google Scholar
  3. Piotr Bojanowski, Edouard Grave, Armand Joulin, and Tomas Mikolov. Enriching word vectors with subword information. Transactions of the Association for Computational Linguistics, 5:135-146, 2017. URL: https://doi.org/10.1162/tacl_a_00051.
  4. Tolga Bolukbasi, Kai-Wei Chang, James Zou, Venkatesh Saligrama, and Adam Kalai. Man is to computer programmer as woman is to homemaker? debiasing word embeddings. In Advances in Neural Information Processing Systems, volume 29, pages 4349-4357, 2016. URL: https://proceedings.neurips.cc/paper/2016/hash/a486cd07e4ac3d270571622f4f316ec5-Abstract.html.
  5. Aylin Caliskan, Joanna J. Bryson, and Arvind Narayanan. Semantics derived automatically from language corpora contain human-like biases. Science, 356(6334):183-186, April 2017. URL: https://doi.org/10.1126/science.aal4230.
  6. Paula Carvalho, Mário J. Silva, Carlos Costa, and Luís Sarmento. Sentilex‑pt: Principais características e potencialidades. Oslo Studies in Language, 7(1):e1444, 2015. URL: https://doi.org/10.5617/osla.1444.
  7. Fernanda Tiemi de Souza Taso, Valéria Quadros dos Reis, and Fábio Viduani Martinez. Analyzing discourses in Portuguese word embeddings: A case of gender bias outside the English-speaking world. Journal on Interactive Systems, 16(1), 2025. URL: https://doi.org/10.5753/jis.2025.5958.
  8. Fatma Elsafoury, Steven R. Wilson, Stamos Katsigiannis, and Naeem Ramzan. SOS: Systematic offensive stereotyping bias in word embeddings. In Proceedings of the 29th International Conference on Computational Linguistics, pages 1263-1274, Gyeongju, Republic of Korea, 2022. International Committee on Computational Linguistics. URL: https://aclanthology.org/2022.coling-1.108.
  9. Kawin Ethayarajh, David Duvenaud, and Graeme Hirst. Understanding undesirable word embedding associations. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 1696-1705, Florence, Italy, 2019. Association for Computational Linguistics. URL: https://doi.org/10.18653/v1/P19-1166.
  10. Nikhil Garg, Londa Schiebinger, Dan Jurafsky, and James Zou. Word embeddings quantify 100 years of gender and ethnic stereotypes. Proceedings of the National Academy of Sciences, 115(16):E3635-E3644, 2018. URL: https://doi.org/10.1073/pnas.1720347115.
  11. Hila Gonen and Yoav Goldberg. Lipstick on a pig: Debiasing methods cover up systematic gender biases in word embeddings but do not remove them. 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 609-614, Minneapolis, Minnesota, June 2019. Association for Computational Linguistics. URL: https://doi.org/10.18653/v1/N19-1061.
  12. Anthony G. Greenwald, Debbie E. McGhee, and Jordan L. K. Schwartz. Measuring individual differences in implicit cognition: The implicit association test. Journal of Personality and Social Psychology, 74(6):1464-1480, 1998. URL: https://doi.org/10.1037/0022-3514.74.6.1464.
  13. Bruce Pengfei Han. PsychWordVec: Word embedding research framework for psychological science. R package, 2025. URL: https://psychbruce.github.io/PsychWordVec/.
  14. Nathan Hartmann, Erick Fonseca, Christopher Shulby, Marcos Treviso, Jéssica Silva, and Sandra Aluísio. Portuguese word embeddings: Evaluating on word analogies and natural language tasks. In Proceedings of the 11th Brazilian Symposium in Information and Human Language Technology, pages 122-131, Uberlândia, Brazil, October 2017. Sociedade Brasileira de Computação. URL: https://aclanthology.org/W17-6615/.
  15. Svetlana Kiritchenko and Saif M. Mohammad. Examining gender and race bias in two hundred sentiment analysis systems. In Proceedings of the Seventh Joint Conference on Lexical and Computational Semantics (*SEM), pages 43-53, New Orleans, Louisiana, June 2018. Association for Computational Linguistics. URL: https://doi.org/10.18653/v1/S18-2005.
  16. Neeraja Kirtane and Tanvi Anand. Mitigating gender stereotypes in Hindi and Marathi. In Proceedings of the 4th Workshop on Gender Bias in Natural Language Processing (GeBNLP), pages 145-150, Seattle, Washington, July 2022. Association for Computational Linguistics. URL: https://doi.org/10.18653/v1/2022.gebnlp-1.16.
  17. Wang Ling, Chris Dyer, Alan W. Black, and Isabel Trancoso. Two/Too simple adaptations of Word2Vec for syntax problems. In Proceedings of the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 1299-1304, Denver, Colorado, May-June 2015. Association for Computational Linguistics. URL: https://doi.org/10.3115/v1/N15-1142.
  18. Vijit Malik, Sunipa Dev, Akihiro Nishi, Nanyun Peng, and Kai-Wei Chang. Socially aware bias measurements for Hindi language representations. In Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 1041-1052, Seattle, United States, July 2022. Association for Computational Linguistics. URL: https://doi.org/10.18653/v1/2022.naacl-main.76.
  19. Tomáš Mikolov, Ilya Sutskever, Kai Chen, Greg S. Corrado, and Jeffrey Dean. Distributed representations of words and phrases and their compositionality. In Advances in Neural Information Processing Systems, pages 3111-3119, 2013. NIPS 2013. URL: https://papers.nips.cc/paper/5021-distributed-representations-of-words-and-phrases-and-their-compositionality.
  20. Luiz Mott, Eduardo Michels, and Paulinho. Homicídios da população de lésbicas, gays, bissexuais, travestis, transexuais ou transgêneros (LGBT) no Brasil: uma análise espacial. Ciência & Saúde Coletiva, 25(6):2195-2208, 2020. URL: https://doi.org/10.1590/1413-81232020256.19672019.
  21. Orestis Papakyriakopoulos, Simon Hegelich, Juan Carlos Medina Serrano, and Fabienne Marco. Bias in word embeddings. In Proceedings of the 2020 Conference on Fairness, Accountability, and Transparency, pages 446-457, 2020. URL: https://doi.org/10.1145/3351095.3372843.
  22. Jeffrey Pennington, Richard Socher, and Christopher D. Manning. 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, October 2014. Association for Computational Linguistics. URL: https://doi.org/10.3115/v1/D14-1162.
  23. Alexandre Puttick, Catherine Ikae, Carlotta Rigotti, Eduard Fosch-Villaronga, Mark W. Kharas, Roger A. Søraa, and Mascha Kurpicz-Briki. A systematic review of bias detection methods for non-English word embeddings and language models. Artificial Intelligence Review, 58:370, 2025. URL: https://doi.org/10.1007/s10462-025-11375-8.
  24. Saghar Omrani Sabbaghi and Aylin Caliskan. Measuring gender bias in word embeddings of gendered languages requires disentangling grammatical gender signals. In Proceedings of the 2022 AAAI/ACM Conference on AI, Ethics, and Society, pages 518-531, 2022. URL: https://doi.org/10.1145/3514094.3534176.
  25. Fábio Souza, Rodrigo Nogueira, and Roberto Lotufo. BERTimbau: Pretrained BERT models for Brazilian Portuguese. In Intelligent Systems. BRACIS 2020, volume 12319 of Lecture Notes in Computer Science, pages 403-417. Springer, 2020. URL: https://doi.org/10.1007/978-3-030-61377-8_28.
  26. Angelo Vip and Fred Libi. Aurélia, a Dicionária da Língua Afiada. Editora do Bispo, São Paulo, Brazil, 2006. Google Scholar
Any Issues?
X

Feedback on the Current Page

CAPTCHA

Thanks for your feedback!

Feedback submitted to Dagstuhl Publishing

Could not send message

Please try again later or send an E-mail
© 2023-2026 Schloss Dagstuhl – LZI GmbH Schloss Dagstuhl – LZI GmbH About DROPS Imprint Privacy Contact