mlsanityData to 7 Oct 2026

Target-Agnostic Gender-Aware Contrastive Learning for Mitigating Bias in Multilingual Machine Translation

arXiv 2305.14016 ↗First posted 23 May 2023cs.CLcited by 10

Not on the map. Why?

Abstract (arXiv)

Gender bias is a significant issue in machine translation, leading to ongoing research efforts in developing bias mitigation techniques. However, most works focus on debiasing bilingual models without much consideration for multilingual systems. In this paper, we specifically target the gender bias issue of multilingual machine translation models for unambiguous cases where there is a single correct translation, and propose a bias mitigation method based on a novel approach. Specifically, we propose Gender-Aware Contrastive Learning, GACL, which encodes contextual gender information into the representations of non-explicit gender words. Our method is target language-agnostic and is applicable to pre-trained multilingual machine translation models via fine-tuning. Through multilingual evaluation, we show that our approach improves gender accuracy by a wide margin without hampering translation performance. We also observe that incorporated gender information transfers and benefits other target languages regarding gender accuracy. Finally, we demonstrate that our method is applicable and beneficial to models of various sizes.

Key facts. Cited by 10 papers in the map (data through 7 Oct 2026). Builds on 37 of its 49 references that are in the map. First posted on arXiv 23 May 2023.

Builds on

37 of 49 references.

References · 37

  1. 2022Scaling Instruction-Finetuned Language Modelscited by 2,997
  2. 2020Supervised Contrastive Learningcited by 2,803
  3. 2021SimCSE: Simple Contrastive Learning of Sentence Embeddingscited by 2,168
  4. 2022No Language Left Behind: Scaling Human-Centered Machine Translationcited by 1,290
  5. 2018Gender Bias in Coreference Resolution: Evaluation and Debiasing Methodscited by 657
  6. 2019How multilingual is Multilingual BERT?cited by 642
  7. 2020Making Monolingual Sentence Embeddings Multilingual using Knowledge Distillationcited by 591
  8. 2020Beyond English-Centric Multilingual Machine Translationcited by 559
  9. 2020Null It Out: Guarding Protected Attributes by Iterative Nullspace Projectioncited by 442
  10. 2019Lipstick on a Pig: Debiasing Methods Cover up Systematic Gender Biases in Word Embeddings But do not Remove Themcited by 268
  11. 2019Evaluating Gender Bias in Machine Translationcited by 247
  12. 2019Massively Multilingual Neural Machine Translationcited by 246
25 more
  1. 2020Measuring and Reducing Gendered Correlations in Pre-trained Modelscited by 222
  2. 2023Towards Making the Most of ChatGPT for Machine Translationcited by 138
  3. 2018Assessing Gender Bias in Machine Translation -- A Case Study with Google Translatecited by 121
  4. 2021Gender Bias in Machine Translationcited by 116
  5. 2021Sustainable Modular Debiasing of Language Modelscited by 98
  6. 2020Reducing Gender Bias in Neural Machine Translation as a Domain Adaptation Problemcited by 86
  7. 2021Larger-Scale Transformers for Multilingual Masked Language Modelingcited by 86
  8. 2021FairFil: Contrastive Neural Debiasing Method for Pretrained Text Encoderscited by 75
  9. 2021Facebook AI WMT21 News Translation Task Submissioncited by 57
  10. 2022Entropy-based Attention Regularization Frees Unintended Bias Mitigation from Listscited by 54
  11. 2021Text Detoxification using Large Pre-trained Neural Modelscited by 52
  12. 2020Neural Machine Translation Doesn't Translate Gender Coreference Right Unless You Make Itcited by 40
  13. 2022Probing Classifiers are Unreliable for Concept Removal and Detectioncited by 40
  14. 2022MT-GenEval: A Counterfactual and Contextual Dataset for Evaluating Gender Accuracy in Machine Translationcited by 38
  15. 2022MABEL: Attenuating Gender Bias using Textual Entailment Datacited by 31
  16. 2022Choose Your Lenses: Flaws in Gender Bias Evaluationcited by 29
  17. 2021Evaluating Gender Bias in Natural Language Inferencecited by 27
  18. 2022Under the Morphosyntactic Lens: A Multifaceted Evaluation of Gender Bias in Speech Translationcited by 23
  19. 2020Gender Coreference and Bias Evaluation at WMT 2020cited by 21
  20. 2020Nurse is Closer to Woman than Surgeon? Mitigating Gender-Biased Proximities in Word Embeddingscited by 20
  21. 2021Investigating Failures of Automatic Translation in the Case of Unambiguous Gendercited by 20
  22. 2022Overcoming Catastrophic Forgetting beyond Continual Learning: Balanced Training for Neural Machine Translationcited by 18
  23. 2022SMaLL-100: Introducing Shallow Multilingual Machine Translation Model for Low-Resource Languagescited by 16
  24. 2021Breaking Down Multilingual Machine Translationcited by 8
  25. 2022Mitigating Gender Bias in Machine Translation through Adversarial Learningcited by 8

Built on by

Cited by year

Citing papers per year: 2023: 1; 2024: 9.

10 papers.

  1. 2024Multilingual Large Language Model: A Survey of Resources, Taxonomy and Frontierscited by 49
  2. 2024VLind-Bench: Measuring Language Priors in Large Vision-Language Modelscited by 44
  3. 2023Sociodemographic Bias in Language Models: A Survey and Forward Pathcited by 27
  4. 2024Can LLMs Recognize Toxicity? A Structured Investigation Framework and Toxicity Metriccited by 23
  5. 2024What the Harm? Quantifying the Tangible Impact of Gender Bias in Machine Translation with a Human-centered Studycited by 15
  6. 2024Watching the Watchers: Exposing Gender Disparities in Machine Translation Quality Estimationcited by 10
  7. 2024A Contrastive Learning Approach to Mitigate Bias in Speech Modelscited by 9
  8. 2024Whose wife is it anyway? Assessing bias against same-gender relationships in machine translationcited by 6
  9. 2024Current State-of-the-Art of Bias Detection and Mitigation in Machine Translation for African and European Languages: a Reviewcited by 1
  10. 2024Fine-grained Gender Control in Machine Translation with Large Language Modelscited by 0