CoChat
Sign Up
Cited by 15Ethics and Social Impacts of AITopic ModelingAdvanced Neural Network Applications

A comparative study on the impact of model compression techniques on fairness in language models

Open access
Ramesh, Krithika, Chavan, Arnav, Pandit, Shrey, Sitaram, Sunayana·2023·Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
DOI 10.18653/v1/2023.acl-long.878
Open this paper in CoChat — free accountOpen PDF

Abstract

Compression techniques for deep learning have become increasingly popular, particularly in settings where latency and memory constraints are imposed.Several methods, such as pruning, distillation, and quantization, have been adopted for compressing models, each providing distinct advantages.However, existing literature demonstrates that compressing deep learning models could affect their fairness.Our analysis involves a comprehensive evaluation of pruned, distilled, and quantized language models, which we benchmark across a range of intrinsic and extrinsic metrics for measuring bias in text classification.We also investigate the impact of using multilingual models and evaluation measures.Our findings highlight the significance of considering both the pre-trained model and the chosen compression strategy in developing equitable language technologies.The results also indicate that compression strategies can have an adverse effect on fairness measures.

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/v1/2023.acl-long.878

Cites

13 of 13 references verified0 citation errors · 0 not checkable

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?

  • In an analysis of stereotyping and toxicity classification in text, pruning may amplify bias in some cases, but with mixed effects and dependency on the degree of pruning.

    “In an analysis of stereotyping and toxicity classification in text, Ramesh et al. (2023) also find that pruning may amplify bias in some cases, but with mixed effects and dependency on the degree of pruning.”

    From Bias and Fairness in Large Language Models: A Survey (2024) · cites this paper as Ramesh et al. (2023)

Self-check

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

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

Cites

Flagged first, then by citations

Research works printed but not linked: 47

  1. Jaimeen Ahn, Hwaran Lee, Jinhwa Kim, and Alice Oh. 2022. Why knowledge distillation amplifies gender bias and how to mitigate from the perspective of Dis- tilBERT. In Proceedings of the 4th Workshop on Gen- der Bias in Natural Language Processing (GeBNLP), pages 266-272, Seattle, Washington. Association for Computational Linguistics. · DOI · Paper page
  2. Sarah Alnegheimish, Alicia Guo, and Yi Sun. 2022. Using natural sentence prompts for understanding bi- ases in language models. In Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 2824-2830, Seattle, United States. Association for Computational Lin- guistics. · DOI
  3. Vamsi Aribandi, Yi Tay, and Donald Metzler. 2021. How reliable are model diagnostics? In Findings of the Association for Computational Linguistics: ACL- IJCNLP 2021, pages 1778-1785, Online. Association for Computational Linguistics. · DOI · Paper page
  4. Marion Bartl, Malvina Nissim, and Albert Gatt. 2020. Unmasking contextual stereotypes: Measuring and mitigating BERT's gender bias. In Proceedings of the Second Workshop on Gender Bias in Natural Language Processing, pages 1-16, Barcelona, Spain (Online). Association for Computational Linguistics.
  5. António Câmara, Nina Taneja, Tamjeed Azad, Emily Allaway, and Richard Zemel. 2022. Mapping the multilingual margins: Intersectional biases of sen- timent analysis systems in English, Spanish, and Arabic. In Proceedings of the Second Workshop on Language Technology for Equality, Diversity and In- clusion, pages 90-106, Dublin, Ireland. Association for Computational Linguistics. · DOI
  6. Yang Cao, Yada Pruksachatkun, Kai-Wei Chang, Rahul Gupta, Varun Kumar, Jwala Dhamala, and Aram Gal- styan. 2022. On the intrinsic and extrinsic fairness evaluation metrics for contextualized language repre- sentations. In Proceedings of the 60th Annual Meet- ing of the Association for Computational Linguistics (Volume 2: Short Papers), pages 561-570, Dublin, Ireland. Association for Computational Linguistics. · DOI
  7. Alexis Conneau, Kartikay Khandelwal, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettle- moyer, and Veselin Stoyanov. 2019. Unsupervised cross-lingual representation learning at scale. CoRR, abs/1911.02116. · arXiv
  8. Sunipa Dev, Masoud Monajatipoor, Anaelia Ovalle, Ar- jun Subramonian, Jeff Phillips, and Kai-Wei Chang. 2021. Harms of gender exclusivity and challenges in non-binary representation in language technologies. In Proceedings of the 2021 Conference on Empiri- cal Methods in Natural Language Processing, pages 1968-1994, Online and Punta Cana, Dominican Re- public. Association for Computational Linguistics. · DOI
  9. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2018. BERT: pre-training of deep bidirectional transformers for language under- standing. CoRR, abs/1810.04805. · arXiv
  10. Anjalie Field, Su Lin Blodgett, Zeerak Waseem, and Yulia Tsvetkov. 2021. A survey of race, racism, and anti-racism in NLP. In Proceedings of the 59th An- nual Meeting of the Association for Computational Linguistics and the 11th International Joint Confer- ence on Natural Language Processing (Volume 1: Long Papers), pages 1905-1925, Online. Association for Computational Linguistics. · DOI
  11. Tanmay Garg, Sarah Masud, Tharun Suresh, and Tan- moy Chakraborty. 2022. Handling bias in toxic speech detection: A survey. · DOI
  12. Seraphina Goldfarb-Tarrant, Rebecca Marchant, Ri- cardo Muñoz Sánchez, Mugdha Pandya, and Adam Lopez. 2021. Intrinsic bias metrics do not correlate with application bias. In Proceedings of the 59th An- nual Meeting of the Association for Computational Linguistics and the 11th International Joint Confer- ence on Natural Language Processing (Volume 1: Long Papers), pages 1926-1940, Online. Association for Computational Linguistics. · DOI · Paper page
  13. Hila Gonen, Yova Kementchedjhieva, and Yoav Gold- berg. 2019. How does grammatical gender affect noun representations in gender-marking languages? In Proceedings of the 23rd Conference on Computa- tional Natural Language Learning (CoNLL), pages 463-471, Hong Kong, China. Association for Com- putational Linguistics. · DOI
  14. Sophie Groenwold, Lily Ou, Aesha Parekh, Samhita Honnavalli, Sharon Levy, Diba Mirza, and William Yang Wang. 2020a. Investigating african- american vernacular english in transformer-based text generation. In Proceedings of the 2020 Con- ference on Empirical Methods in Natural Language Processing (EMNLP), pages 5877-5883.
  15. Sophie Groenwold, Lily Ou, Aesha Parekh, Samhita Honnavalli, Sharon Levy, Diba Mirza, and William Yang Wang. 2020b. Investigating African- American Vernacular English in transformer-based text generation. In Proceedings of the 2020 Con- ference on Empirical Methods in Natural Language Processing (EMNLP), pages 5877-5883, Online. As- sociation for Computational Linguistics. · DOI
  16. Umang Gupta, Jwala Dhamala, Varun Kumar, Apurv Verma, Yada Pruksachatkun, Satyapriya Krishna, Rahul Gupta, Kai-Wei Chang, Greg Ver Steeg, and Aram Galstyan. 2022. Mitigating gender bias in dis- tilled language models via counterfactual role rever- sal. In Findings of the Association for Computational Linguistics: ACL 2022, pages 658-678, Dublin, Ire- land. Association for Computational Linguistics. · DOI
  17. Marius Hessenthaler, Emma Strubell, Dirk Hovy, and Anne Lauscher. 2022. Bridging fairness and environ- mental sustainability in natural language processing. · DOI
  18. Geoffrey Hinton, Oriol Vinyals, and Jeff Dean. 2015. Distilling the knowledge in a neural network. · DOI · Paper page
  19. Sara Hooker, Nyalleng Moorosi, Gregory Clark, Samy Bengio, and Emily Denton. 2020. Characterising bias in compressed models. · DOI
  20. Dirk Hovy, Anders Johannsen, and Anders Søgaard. 2015. User review sites as a resource for large-scale sociolinguistic studies. In Proceedings of the 24th International Conference on World Wide Web, WWW '15, page 452-461, Republic and Canton of Geneva, CHE. International World Wide Web Conferences Steering Committee. · DOI
  21. Xiaolei Huang. 2022. Easy adaptation to mitigate gender bias in multilingual text classification. In Proceedings of the 2022 Conference of the North American Chapter of the Association for Computa- tional Linguistics: Human Language Technologies, pages 717-723, Seattle, United States. Association for Computational Linguistics. · DOI
  22. Xiaolei Huang, Xing Linzi, Franck Dernoncourt, and Michael J. Paul. 2020a. Multilingual twitter corpus and baselines for evaluating demographic bias in hate speech recognition. In Proceedings of the Twelveth International Conference on Language Resources and Evaluation (LREC 2020), Marseille, France. Eu- ropean Language Resources Association (ELRA).
  23. Xiaolei Huang and Michael J. Paul. 2019. Neural user factor adaptation for text classification: Learning to generalize across author demographics. In Pro- ceedings of the Eighth Joint Conference on Lexical and Computational Semantics (*SEM 2019), pages 136-146, Minneapolis, Minnesota. Association for Computational Linguistics. · DOI
  24. Xiaolei Huang, Linzi Xing, Franck Dernoncourt, and Michael J. Paul. 2020b. Multilingual Twitter cor- pus and baselines for evaluating demographic bias in hate speech recognition. In Proceedings of the Twelfth Language Resources and Evaluation Confer- ence, pages 1440-1448, Marseille, France. European Language Resources Association.
  25. Masahiro Kaneko, Aizhan Imankulova, Danushka Bol- legala, and Naoaki Okazaki. 2022. Gender bias in masked language models for multiple languages. In Proceedings of the 2022 Conference of the North American Chapter of the Association for Computa- tional Linguistics: Human Language Technologies, pages 2740-2750, Seattle, United States. Association for Computational Linguistics. · DOI
  26. Sehoon Kim, Amir Gholami, Zhewei Yao, Michael W Mahoney, and Kurt Keutzer. 2021. I-bert: Integer- only bert quantization. In International conference on machine learning, pages 5506-5518. PMLR.
  27. Svetlana Kiritchenko and Saif Mohammad. 2018. Ex- amining gender and race bias in two hundred senti- ment analysis systems. In Proceedings of the Sev- enth Joint Conference on Lexical and Computational Semantics, pages 43-53, New Orleans, Louisiana. Association for Computational Linguistics. · DOI · Paper page
  28. Keita Kurita, Nidhi Vyas, Ayush Pareek, Alan W Black, and Yulia Tsvetkov. 2019. Measuring bias in con- textualized word representations. In Proceedings of the First Workshop on Gender Bias in Natural Lan- guage Processing, pages 166-172, Florence, Italy. Association for Computational Linguistics. · DOI · Paper page
  29. François Lagunas, Ella Charlaix, Victor Sanh, and Alexander M Rush. 2021. Block pruning for faster transformers. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Process- ing, pages 10619-10629.
  30. Anne Lauscher, Tobias Lueken, and Goran Glavaš. 2021. Sustainable modular debiasing of language models. In Findings of the Association for Computational Linguistics: EMNLP 2021, pages 4782-4797, Punta Cana, Dominican Republic. Association for Compu- tational 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 approach. CoRR, abs/1907.11692. · arXiv
  32. Nicholas Meade, Elinor Poole-Dayan, and Siva Reddy. 2022. An empirical survey of the effectiveness of debiasing techniques for pre-trained language models. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1878-1898, Dublin, Ireland. Association for Computational Linguistics. · DOI
  33. Alireza Mohammadshahi, Vassilina Nikoulina, Alexan- dre Berard, Caroline Brun, James Henderson, and Laurent Besacier. 2022. What do compressed multi- lingual machine translation models forget? · DOI
  34. Hadas Orgad and Yonatan Belinkov. 2022. Choose your lenses: Flaws in gender bias evaluation. In Proceed- ings of the 4th Workshop on Gender Bias in Natu- ral Language Processing (GeBNLP), pages 151-167, Seattle, Washington. Association for Computational Linguistics. · DOI
  35. Julian Salazar, Davis Liang, Toan Q. Nguyen, and Ka- trin Kirchhoff. 2020. Masked language model scor- ing. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 2699-2712, Online. Association for Computational Linguistics. · DOI
  36. Andrew Silva, Pradyumna Tambwekar, and Matthew Gombolay. 2021. Towards a comprehensive under- standing and accurate evaluation of societal biases in pre-trained transformers. In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 2383-2389, Online. Association for Computational Linguistics. · DOI · Paper page
  37. Richard Socher, Alex Perelygin, Jean Wu, Jason Chuang, Christopher D. Manning, Andrew Ng, and Christopher Potts. 2013. Recursive deep models for semantic compositionality over a sentiment treebank. In Proceedings of the 2013 Conference on Empiri- cal Methods in Natural Language Processing, pages 1631-1642, Seattle, Washington, USA. Association for Computational Linguistics.
  38. Yarden Tal, Inbal Magar, and Roy Schwartz. 2022. Fewer errors, but more stereotypes? the effect of model size on gender bias. In Proceedings of the 4th Workshop on Gender Bias in Natural Language Pro- cessing (GeBNLP), pages 112-120, Seattle, Wash- ington. Association for Computational Linguistics. · DOI
  39. Zeerak Talat, Aurélie Névéol, Stella Biderman, Miruna Clinciu, Manan Dey, Shayne Longpre, Sasha Luc- cioni, Maraim Masoud, Margaret Mitchell, Dragomir Radev, Shanya Sharma, Arjun Subramonian, Jaesung Tae, Samson Tan, Deepak Tunuguntla, and Oskar Van Der Wal. 2022. You reap what you sow: On the chal- lenges of bias evaluation under multilingual settings. In Proceedings of BigScience Episode #5 -Workshop on Challenges & Perspectives in Creating Large Lan- guage Models, pages 26-41, virtual+Dublin. Associ- ation for Computational Linguistics. · DOI
  40. Ziheng Wang, Jeremy Wohlwend, and Tao Lei. 2020b. Structured pruning of large language models. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 6151-6162.
  41. Kellie Webster, Xuezhi Wang, Ian Tenney, Alex Beutel, Emily Pitler, Ellie Pavlick, Jilin Chen, Ed Chi, and Slav Petrov. 2020. Measuring and reducing gendered correlations in pre-trained models. · DOI · Paper page
  42. Thomas Wolf, Lysandre Debut, Victor Sanh, Julien Chaumond, Clement Delangue, Anthony Moi, Pier- ric Cistac, Tim Rault, Rémi Louf, Morgan Funtowicz, Joe Davison, Sam Shleifer, Patrick von Platen, Clara Ma, Yacine Jernite, Julien Plu, Canwen Xu, Teven Le Scao, Sylvain Gugger, Mariama Drame, Quentin Lhoest, and Alexander M. Rush. 2020. Transform- ers: State-of-the-art natural language processing. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: System Demonstrations, pages 38-45, Online. Association for Computational Linguistics.
  43. Guangxuan Xu and Qingyuan Hu. 2022. Can model compression improve nlp fairness. ArXiv, abs/2201.08542. · arXiv
  44. Ofir Zafrir, Guy Boudoukh, Peter Izsak, and Moshe Wasserblat. 2019a. Q8bert: Quantized 8bit bert. In 2019 Fifth Workshop on Energy Efficient Machine Learning and Cognitive Computing-NeurIPS Edition (EMC2-NIPS), pages 36-39. IEEE.
  45. Ofir Zafrir, Guy Boudoukh, Peter Izsak, and Moshe Wasserblat. 2019b. Q8BERT: quantized 8bit BERT. CoRR, abs/1910.06188. · arXiv
  46. Ofir Zafrir, Ariel Larey, Guy Boudoukh, Haihao Shen, and Moshe Wasserblat. 2021. Prune once for all: Sparse pre-trained language models. arXiv preprint arXiv:2111.05754. · arXiv
  47. 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 Computational Lin- guistics, pages 2896-2907, Online. Association for Computational Linguistics. · DOI

Cited by

Sorted by citations

Authors
Ramesh, K., Chavan, A., Pandit, S., & Sitaram, S.
Affiliations
Microsoft (United States) (US)
ORCID
Venue
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
Publisher
Association for Computational Linguistics
Published
2023
Pages
15762-15782
ISSN
0736-587X (ISSN-L)
DOI
10.18653/v1/2023.acl-long.878
OpenAlex
W4385572485
Cited by
15
References
13
Citations by year
2026: 2 · 2025: 7 · 2024: 3 · 2023: 3
FWCI
3.20
Type
conference paper
Language
en
Open access
Licence
cc-by
Locations
Proceedings of the conference - Association for Computational Linguistics. Meeting (publishedVersion, cc-by)
Topics
  • Social Sciences › Safety Research › Ethics and Social Impacts of AI
  • Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
Keywords
model compression, algorithmic fairness, language models, pruning, knowledge distillation, quantization, multilingual models, fairness metrics

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