mlsanityData to 7 Oct 2026

Gender Bias in Contextualized Word Embeddings

arXiv 1904.03310 ↗First posted 5 Apr 2019cs.CLcited by 203

Not on the map. Why?

Abstract (arXiv)

In this paper, we quantify, analyze and mitigate gender bias exhibited in ELMo's contextualized word vectors. First, we conduct several intrinsic analyses and find that (1) training data for ELMo contains significantly more male than female entities, (2) the trained ELMo embeddings systematically encode gender information and (3) ELMo unequally encodes gender information about male and female entities. Then, we show that a state-of-the-art coreference system that depends on ELMo inherits its bias and demonstrates significant bias on the WinoBias probing corpus. Finally, we explore two methods to mitigate such gender bias and show that the bias demonstrated on WinoBias can be eliminated.

Key facts. Cited by 203 papers in the map (data through 7 Oct 2026). First posted on arXiv 5 Apr 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 36; 2022: 55; 2023: 37; 2024: 41; 2025: 21; 2026: 13.

203 papers.

Sort:

By area:

  1. 2022LaMDA: Language Models for Dialog Applicationscited by 1,218
  2. 2024The FineWeb Datasets: Decanting the Web for the Finest Text Data at Scalecited by 1,104
  3. 2021Ethical and social risks of harm from Language Modelscited by 957
  4. 2023Bias and Fairness in Large Language Models: A Surveycited by 626
  5. 2023Toxicity in ChatGPT: Analyzing Persona-assigned Language Modelscited by 356
  6. 2022Challenges and Strategies in Cross-Cultural NLPcited by 216
  7. 2023Transformer-Patcher: One Mistake worth One Neuroncited by 196
  8. 2023Gender bias and stereotypes in Large Language Modelscited by 185
  9. 2022"I'm sorry to hear that": Finding New Biases in Language Models with a Holistic Descriptor Datasetcited by 160
  10. 2021RedditBias: A Real-World Resource for Bias Evaluation and Debiasing of Conversational Language Modelscited by 149
  11. 2023Debiasing Vision-Language Models via Biased Promptscited by 147
  12. 2021Bias Out-of-the-Box: An Empirical Analysis of Intersectional Occupational Biases in Popular Generative Language Modelscited by 141
38 more
  1. 2021Harms of Gender Exclusivity and Challenges in Non-Binary Representation in Language Technologiescited by 128
  2. 2021Debiasing Pre-trained Contextualised Embeddingscited by 107
  3. 2022A Prompt Array Keeps the Bias Away: Debiasing Vision-Language Models with Adversarial Learningcited by 104
  4. 2021Sustainable Modular Debiasing of Language Modelscited by 98
  5. 2021Trustworthy AI: A Computational Perspectivecited by 93
  6. 2023Evaluating the Social Impact of Generative AI Systems in Systems and Societycited by 91
  7. 2024A Survey on Multilingual Large Language Models: Corpora, Alignment, and Biascited by 82
  8. 2021On Measures of Biases and Harms in NLPcited by 81
  9. 2021What's in the Box? A Preliminary Analysis of Undesirable Content in the Common Crawl Corpuscited by 79
  10. 2024Bias in Large Language Models: Origin, Evaluation, and Mitigationcited by 77
  11. 2022On the Intrinsic and Extrinsic Fairness Evaluation Metrics for Contextualized Language Representationscited by 75
  12. 2022Exploring the Limits of Domain-Adaptive Training for Detoxifying Large-Scale Language Modelscited by 67
  13. 2024Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systemscited by 63
  14. 2022Theories of "Gender" in NLP Bias Researchcited by 62
  15. 2021Unmasking the Mask -- Evaluating Social Biases in Masked Language Modelscited by 60
  16. 2021Stereotype and Skew: Quantifying Gender Bias in Pre-trained and Fine-tuned Language Modelscited by 57
  17. 2024Fairness in Large Language Models: A Taxonomic Surveycited by 57
  18. 2021Worst of Both Worlds: Biases Compound in Pre-trained Vision-and-Language Modelscited by 52
  19. 2022Perturbation Augmentation for Fairer NLPcited by 51
  20. 2023ChatGPT and a New Academic Reality: Artificial Intelligence-Written Research Papers and the Ethics of the Large Language Models in Scholarly Publishingcited by 51
  21. 2021Automated Self-Supervised Learning for Graphscited by 50
  22. 2024Towards Implicit Bias Detection and Mitigation in Multi-Agent LLM Interactionscited by 50
  23. 2022BERTScore is Unfair: On Social Bias in Language Model-Based Metrics for Text Generationcited by 48
  24. 2022On the Explainability of Natural Language Processing Deep Modelscited by 37
  25. 2025FairSteer: Inference Time Debiasing for LLMs with Dynamic Activation Steeringcited by 37
  26. 2022Towards Identifying Social Bias in Dialog Systems: Frame, Datasets, and Benchmarkscited by 36
  27. 2021Improving Gender Fairness of Pre-Trained Language Models without Catastrophic Forgettingcited by 34
  28. 2021Learning Bias-Invariant Representation by Cross-Sample Mutual Information Minimizationcited by 32
  29. 2023A Trip Towards Fairness: Bias and De-Biasing in Large Language Modelscited by 32
  30. 2022MABEL: Attenuating Gender Bias using Textual Entailment Datacited by 31
  31. 2024Whispering Experts: Neural Interventions for Toxicity Mitigation in Language Modelscited by 31
  32. 2022Choose Your Lenses: Flaws in Gender Bias Evaluationcited by 29
  33. 2022Counterfactually Measuring and Eliminating Social Bias in Vision-Language Pre-training Modelscited by 28
  34. 2022Causally Testing Gender Bias in LLMs: A Case Study on Occupational Biascited by 28
  35. 2023Language Models Get a Gender Makeover: Mitigating Gender Bias with Few-Shot Data Interventionscited by 28
  36. 2023Sociodemographic Bias in Language Models: A Survey and Forward Pathcited by 27
  37. 2022Toxicity in Multilingual Machine Translation at Scalecited by 26
  38. 2023Survey of Social Bias in Vision-Language Modelscited by 26

50 most cited of 203 All 203 (JSON)