mlsanityData to 7 Oct 2026

Studying Bias in GANs through the Lens of Race

arXiv 2209.02836 ↗First posted 6 Sep 2022cs.CV cs.LGcited by 31

Not on the map. Why?

Abstract (arXiv)

In this work, we study how the performance and evaluation of generative image models are impacted by the racial composition of their training datasets. By examining and controlling the racial distributions in various training datasets, we are able to observe the impacts of different training distributions on generated image quality and the racial distributions of the generated images. Our results show that the racial compositions of generated images successfully preserve that of the training data. However, we observe that truncation, a technique used to generate higher quality images during inference, exacerbates racial imbalances in the data. Lastly, when examining the relationship between image quality and race, we find that the highest perceived visual quality images of a given race come from a distribution where that race is well-represented, and that annotators consistently prefer generated images of white people over those of Black people.

Key facts. Cited by 31 papers in the map (data through 7 Oct 2026). Builds on 31 of its 53 references that are in the map. First posted on arXiv 6 Sep 2022.

Builds on

31 of 53 references.

References · 31

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  3. 2020Language Models are Few-Shot Learnerscited by 32,199
  4. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  5. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 5,986
  6. 2022Classifier-Free Diffusion Guidancecited by 5,717
  7. 2022Generative Adversarial Networkscited by 5,603
  8. 2014Going Deeper with Convolutionscited by 4,231
  9. 2013Auto-Encoding Variational Bayescited by 3,980
  10. 2019Analyzing and Improving the Image Quality of StyleGANcited by 3,580
  11. 2018Large Scale GAN Training for High Fidelity Natural Image Synthesiscited by 2,609
  12. 2017Improved Training of Wasserstein GANscited by 2,223
19 more
  1. 2019Generating Diverse High-Fidelity Images with VQ-VAE-2cited by 1,376
  2. 2020Training Generative Adversarial Networks with Limited Datacited by 1,184
  3. 2018Model Cards for Model Reportingcited by 1,068
  4. 2020GANSpace: Discovering Interpretable GAN Controlscited by 612
  5. 2020PULSE: Self-Supervised Photo Upsampling via Latent Space Exploration of Generative Modelscited by 358
  6. 2019Image2StyleGAN++: How to Edit the Embedded Images?cited by 356
  7. 2016NIPS 2016 Tutorial: Generative Adversarial Networkscited by 284
  8. 2019Towards Fairness in Visual Recognition: Effective Strategies for Bias Mitigationcited by 252
  9. 2018Pros and Cons of GAN Evaluation Measurescited by 224
  10. 2019Seeing What a GAN Cannot Generatecited by 148
  11. 2019Mode Seeking Generative Adversarial Networks for Diverse Image Synthesiscited by 138
  12. 2019Towards a Critical Race Methodology in Algorithmic Fairnesscited by 124
  13. 2018Understanding the Origins of Bias in Word Embeddingscited by 90
  14. 2017Megapixel Size Image Creation using Generative Adversarial Networkscited by 27
  15. 2020Imperfect ImaGANation: Implications of GANs Exacerbating Biases on Facial Data Augmentation and Snapchat Selfie Lensescited by 27
  16. 2017Do GANs actually learn the distribution? An empirical studycited by 23
  17. 2019Prescribed Generative Adversarial Networkscited by 23
  18. 2021Rethinking Common Assumptions to Mitigate Racial Bias in Face Recognition Datasetscited by 22
  19. 2018Imagining an Engineer: On GAN-Based Data Augmentation Perpetuating Biasescited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 13; 2024: 8; 2025: 2; 2026: 7.

31 papers.

Sort:

By area:

  1. 2023Art and the science of generative AI: A deeper divecited by 93
  2. 2023ITI-GEN: Inclusive Text-to-Image Generationcited by 64
  3. 2024Balancing Act: Distribution-Guided Debiasing in Diffusion Modelscited by 50
  4. 2023GANDiffFace: Controllable Generation of Synthetic Datasets for Face Recognition with Realistic Variationscited by 49
  5. 2023Analyzing Bias in Diffusion-based Face Generation Modelscited by 45
  6. 2023Mass-Producing Failures of Multimodal Systems with Language Modelscited by 36
  7. 2023MoSAR: Monocular Semi-Supervised Model for Avatar Reconstruction using Differentiable Shadingcited by 18
  8. 2024Fairness and Bias Mitigation in Computer Vision: A Surveycited by 12
  9. 2023Motion Matters: Neural Motion Transfer for Better Camera Physiological Measurementcited by 11
  10. 2022How to Boost Face Recognition with StyleGAN?cited by 10
  11. 2023Diverse Diffusion: Enhancing Image Diversity in Text-to-Image Generationcited by 8
  12. 2023On Measuring Fairness in Generative Modelscited by 8
19 more
  1. 2025Debiasing Diffusion Model: Enhancing Fairness through Latent Representation Learning in Stable Diffusion Modelcited by 5
  2. 2023Limitations of Face Image Generationcited by 4
  3. 2023Gaussian Harmony: Attaining Fairness in Diffusion-based Face Generation Modelscited by 4
  4. 2024Benchmarking the Fairness of Image Upsampling Methodscited by 3
  5. 2024Perceptual Fairness in Image Restorationcited by 2
  6. 2023What's color got to do with it? Face recognition in grayscalecited by 1
  7. 2024Examining Pathological Bias in a Generative Adversarial Network Discriminator: A Case Study on a StyleGAN3 Modelcited by 1
  8. 2024Disentangling Racial Phenotypes: Fine-Grained Control of Race-related Facial Phenotype Characteristicscited by 1
  9. 2025Accuracy and Fairness of Facial Recognition Technology in Low-Quality Police Images: An Experiment With Synthetic Facescited by 1
  10. 2026Discount Model Search for Quality Diversity Optimization in High-Dimensional Measure Spacescited by 1
  11. 2026BeautyGRPO: Aesthetic Alignment for Face Retouching via Dynamic Path Guidance and Fine-Grained Preference Modelingcited by 1
  12. 2023Fair GANs through model rebalancing for extremely imbalanced class distributionscited by 0
  13. 2024Sampling Strategies for Mitigating Bias in Face Synthesis Methodscited by 0
  14. 2024Mitigation of gender bias in automatic facial non-verbal behaviors generationcited by 0
  15. 2026GLEaN: A Text-to-image Bias Detection Approach for Public Comprehensioncited by 0
  16. 2026Phase-map synthesis from magnitude-only MR images using conditional score-based diffusion models with application in training of accelerated MRI reconstruction modelscited by 0
  17. 2026Detangled: A Framework for Creating, Editing, and Inferencing Feature Rich Hair Strandscited by 0
  18. 2026Position: Fairness Failure in Generative Models is an Evaluation Problemcited by 0
  19. 2026A Large-scale Evaluation of Text-guided Models for Facial Editingcited by 0