mlsanityData to 14 Oct 2026

Towards Fairness in Visual Recognition: Effective Strategies for Bias Mitigation

arXiv 1911.11834 ↗First posted 26 Nov 2019cs.CVcited by 252

Not on the map. Why?

Abstract (arXiv)

Computer vision models learn to perform a task by capturing relevant statistics from training data. It has been shown that models learn spurious age, gender, and race correlations when trained for seemingly unrelated tasks like activity recognition or image captioning. Various mitigation techniques have been presented to prevent models from utilizing or learning such biases. However, there has been little systematic comparison between these techniques. We design a simple but surprisingly effective visual recognition benchmark for studying bias mitigation. Using this benchmark, we provide a thorough analysis of a wide range of techniques. We highlight the shortcomings of popular adversarial training approaches for bias mitigation, propose a simple but similarly effective alternative to the inference-time Reducing Bias Amplification method of Zhao et al., and design a domain-independent training technique that outperforms all other methods. Finally, we validate our findings on the attribute classification task in the CelebA dataset, where attribute presence is known to be correlated with the gender of people in the image, and demonstrate that the proposed technique is effective at mitigating real-world gender bias.

Key facts. Cited by 252 papers in the map (data through 14 Oct 2026). First posted on arXiv 26 Nov 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 41; 2022: 55; 2023: 61; 2024: 46; 2025: 37; 2026: 12.

252 papers.

Sort:

By area:

  1. 2023Scaling Vision Transformers to 22 Billion Parameterscited by 694
  2. 2022A Survey of Machine Unlearningcited by 312
  3. 2023PaLI-X: On Scaling up a Multilingual Vision and Language Modelcited by 255
  4. 2021A Unified Survey on Anomaly, Novelty, Open-Set, and Out-of-Distribution Detection: Solutions and Future Challengescited by 154
  5. 2023Debiasing Vision-Language Models via Biased Promptscited by 149
  6. 2021Counterfactual Generative Networkscited by 111
  7. 2021EnD: Entangling and Disentangling deep representations for bias correctioncited by 105
  8. 2021Trustworthy AI: A Computational Perspectivecited by 93
  9. 2021OoD-Bench: Quantifying and Understanding Two Dimensions of Out-of-Distribution Generalizationcited by 86
  10. 2022A Whac-A-Mole Dilemma: Shortcuts Come in Multiples Where Mitigating One Amplifies Otherscited by 85
  11. 2022Fair Contrastive Learning for Facial Attribute Classificationcited by 71
  12. 2024FairCLIP: Harnessing Fairness in Vision-Language Learningcited by 69
38 more
  1. 2022MEDFAIR: Benchmarking Fairness for Medical Imagingcited by 68
  2. 2023ITI-GEN: Inclusive Text-to-Image Generationcited by 64
  3. 2022Handling Bias in Toxic Speech Detection: A Surveycited by 61
  4. 2024OpenBias: Open-set Bias Detection in Text-to-Image Generative Modelscited by 61
  5. 2021Fair Feature Distillation for Visual Recognitioncited by 59
  6. 2022FairPrune: Achieving Fairness Through Pruning for Dermatological Disease Diagnosiscited by 59
  7. 2021Fairness in Cardiac MR Image Analysis: An Investigation of Bias Due to Data Imbalance in Deep Learning Based Segmentationcited by 57
  8. 2022Leveling Down in Computer Vision: Pareto Inefficiencies in Fair Deep Classifierscited by 53
  9. 2023SocialCounterfactuals: Probing and Mitigating Intersectional Social Biases in Vision-Language Models with Counterfactual Examplescited by 53
  10. 2023TIBET: Identifying and Evaluating Biases in Text-to-Image Generative Modelscited by 52
  11. 2021Understanding and Mitigating Annotation Bias in Facial Expression Recognitioncited by 50
  12. 2022Discover and Mitigate Unknown Biases with Debiasing Alternate Networkscited by 47
  13. 2024A Decade's Battle on Dataset Bias: Are We There Yet?cited by 47
  14. 2024FairMedFM: Fairness Benchmarking for Medical Imaging Foundation Modelscited by 47
  15. 2021Discover the Unknown Biased Attribute of an Image Classifiercited by 46
  16. 2022A Causal Lens for Controllable Text Generationcited by 45
  17. 2022Diagnosing failures of fairness transfer across distribution shift in real-world medical settingscited by 45
  18. 2021NeuronFair: Interpretable White-Box Fairness Testing through Biased Neuron Identificationcited by 42
  19. 2022Unbiased Supervised Contrastive Learningcited by 40
  20. 2022Gender Artifacts in Visual Datasetscited by 38
  21. 2022Quantifying Societal Bias Amplification in Image Captioningcited by 37
  22. 2022Studying Bias in GANs through the Lens of Racecited by 31
  23. 2022FairGRAPE: Fairness-aware GRAdient Pruning mEthod for Face Attribute Classificationcited by 30
  24. 2022Bias Mimicking: A Simple Sampling Approach for Bias Mitigationcited by 30
  25. 2022Rethinking Bias Mitigation: Fairer Architectures Make for Fairer Face Recognitioncited by 26
  26. 2023Survey of Social Bias in Vision-Language Modelscited by 26
  27. 2024Depth Prompting for Sensor-Agnostic Depth Estimationcited by 26
  28. 2022Bayes-Optimal Classifiers under Group Fairnesscited by 24
  29. 2021UDIS: Unsupervised Discovery of Bias in Deep Visual Recognition Modelscited by 23
  30. 2023FairTune: Optimizing Parameter Efficient Fine Tuning for Fairness in Medical Image Analysiscited by 23
  31. 2021Responsible AI: Gender bias assessment in emotion recognitioncited by 22
  32. 2021Rethinking Common Assumptions to Mitigate Racial Bias in Face Recognition Datasetscited by 22
  33. 2022FairDisCo: Fairer AI in Dermatology via Disentanglement Contrastive Learningcited by 22
  34. 2023Consensus and Subjectivity of Skin Tone Annotation for ML Fairnesscited by 22
  35. 2022Men Also Do Laundry: Multi-Attribute Bias Amplificationcited by 21
  36. 2021Technical Challenges for Training Fair Neural Networkscited by 20
  37. 2024Fairness and Bias in Multimodal AI: A Surveycited by 20
  38. 2021Domain-Incremental Continual Learning for Mitigating Bias in Facial Expression and Action Unit Recognitioncited by 19

50 most cited of 252 All 252 (JSON)