mlsanityData to 14 Oct 2026

Fair Attribute Classification through Latent Space De-biasing

arXiv 2012.01469 ↗First posted 2 Dec 2020cs.CVcited by 125

Not on the map. Why?

Abstract (arXiv)

Fairness in visual recognition is becoming a prominent and critical topic of discussion as recognition systems are deployed at scale in the real world. Models trained from data in which target labels are correlated with protected attributes (e.g., gender, race) are known to learn and exploit those correlations. In this work, we introduce a method for training accurate target classifiers while mitigating biases that stem from these correlations. We use GANs to generate realistic-looking images, and perturb these images in the underlying latent space to generate training data that is balanced for each protected attribute. We augment the original dataset with this perturbed generated data, and empirically demonstrate that target classifiers trained on the augmented dataset exhibit a number of both quantitative and qualitative benefits. We conduct a thorough evaluation across multiple target labels and protected attributes in the CelebA dataset, and provide an in-depth analysis and comparison to existing literature in the space.

Key facts. Cited by 125 papers in the map (data through 14 Oct 2026). First posted on arXiv 2 Dec 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 8; 2022: 35; 2023: 33; 2024: 28; 2025: 14; 2026: 7.

125 papers.

Sort:

By area:

  1. 2022A Survey of Machine Unlearningcited by 312
  2. 2023Debiasing Vision-Language Models via Biased Promptscited by 149
  3. 2021Understanding and Evaluating Racial Biases in Image Captioningcited by 108
  4. 2021Generative Models as a Data Source for Multiview Representation Learningcited by 104
  5. 2022Fair Contrastive Learning for Facial Attribute Classificationcited by 71
  6. 2024FairCLIP: Harnessing Fairness in Vision-Language Learningcited by 69
  7. 2023ITI-GEN: Inclusive Text-to-Image Generationcited by 64
  8. 2022Towards Intersectionality in Machine Learning: Including More Identities, Handling Underrepresentation, and Performing Evaluationcited by 59
  9. 2023Diagnosing and Rectifying Vision Models using Languagecited by 58
  10. 2022Leveling Down in Computer Vision: Pareto Inefficiencies in Fair Deep Classifierscited by 53
  11. 2024Balancing Act: Distribution-Guided Debiasing in Diffusion Modelscited by 52
  12. 2022Fairness-aware Adversarial Perturbation Towards Bias Mitigation for Deployed Deep Modelscited by 47
38 more
  1. 2022Overlooked factors in concept-based explanations: Dataset choice, concept learnability, and human capabilitycited by 46
  2. 2023Inspecting the Geographical Representativeness of Images from Text-to-Image Modelscited by 44
  3. 2021NeuronFair: Interpretable White-Box Fairness Testing through Biased Neuron Identificationcited by 42
  4. 2022On-Demand Sampling: Learning Optimally from Multiple Distributionscited by 39
  5. 2022Gender Artifacts in Visual Datasetscited by 38
  6. 2023Human-Centric Multimodal Machine Learning: Recent Advances and Testbed on AI-based Recruitmentcited by 33
  7. 2023Overwriting Pretrained Bias with Finetuning Datacited by 32
  8. 2023Harvard Glaucoma Fairness: A Retinal Nerve Disease Dataset for Fairness Learning and Fair Identity Normalizationcited by 31
  9. 2022Evidence for Hypodescent in Visual Semantic AIcited by 30
  10. 2022FairGRAPE: Fairness-aware GRAdient Pruning mEthod for Face Attribute Classificationcited by 30
  11. 2022Bias Mimicking: A Simple Sampling Approach for Bias Mitigationcited by 30
  12. 2024Causal Concept Graph Models: Beyond Causal Opacity in Deep Learningcited by 29
  13. 2022Generative Visual Prompt: Unifying Distributional Control of Pre-Trained Generative Modelscited by 27
  14. 2023Balancing the Picture: Debiasing Vision-Language Datasets with Synthetic Contrast Setscited by 27
  15. 2023Survey of Social Bias in Vision-Language Modelscited by 26
  16. 2023Benchmarking Algorithmic Bias in Face Recognition: An Experimental Approach Using Synthetic Faces and Human Evaluationcited by 25
  17. 2023Zero-shot Model Diagnosiscited by 23
  18. 2022Mitigating Dataset Bias by Using Per-sample Gradientcited by 22
  19. 2022Debiasing Methods for Fairer Neural Models in Vision and Language Research: A Surveycited by 22
  20. 2022Open Long-Tailed Recognition in a Dynamic Worldcited by 21
  21. 2022Men Also Do Laundry: Multi-Attribute Bias Amplificationcited by 21
  22. 2022FairStyle: Debiasing StyleGAN2 with Style Channel Manipulationscited by 20
  23. 2024FairDomain: Achieving Fairness in Cross-Domain Medical Image Segmentation and Classificationcited by 19
  24. 2024SANER: Annotation-free Societal Attribute Neutralizer for Debiasing CLIPcited by 19
  25. 2021Latent Space Smoothing for Individually Fair Representationscited by 18
  26. 2021Anatomizing Bias in Facial Analysiscited by 18
  27. 2024Decompose-and-Compose: A Compositional Approach to Mitigating Spurious Correlationcited by 18
  28. 2021LARGE: Latent-Based Regression through GAN Semanticscited by 17
  29. 2022Efficient Attribute Unlearning: Towards Selective Removal of Input Attributes from Feature Representationscited by 17
  30. 2023Beyond Skin Tone: A Multidimensional Measure of Apparent Skin Colorcited by 17
  31. 2021Fair Representation: Guaranteeing Approximate Multiple Group Fairness for Unknown Taskscited by 16
  32. 2022Fair Bayes-Optimal Classifiers Under Predictive Paritycited by 16
  33. 2023Image retrieval outperforms diffusion models on data augmentationcited by 14
  34. 2023Post-hoc Bias Scoring Is Optimal For Fair Classificationcited by 14
  35. 2023FairVision: Equitable Deep Learning for Eye Disease Screening via Fair Identity Scalingcited by 13
  36. 2024Fairness and Bias Mitigation in Computer Vision: A Surveycited by 13
  37. 2025Beyond Visual Perception: Insights from Smartphone Interaction of Visually Impaired Users with Large Multimodal Modelscited by 13
  38. 2023From Fake to Real: Pretraining on Balanced Synthetic Images to Prevent Spurious Correlations in Image Recognitioncited by 12

50 most cited of 125 All 125 (JSON)