mlsanityData to 14 Oct 2026

Breaking the Spurious Causality of Conditional Generation via Fairness Intervention with Corrective Sampling

arXiv 2212.02090 ↗First posted 5 Dec 2022cs.CV cs.AI cs.LGcited by 5

Not on the map. Why?

Abstract (arXiv)

To capture the relationship between samples and labels, conditional generative models often inherit spurious correlations from the training dataset. This can result in label-conditional distributions that are imbalanced with respect to another latent attribute. To mitigate this issue, which we call spurious causality of conditional generation, we propose a general two-step strategy. (a) Fairness Intervention (FI): emphasize the minority samples that are hard to generate due to the spurious correlation in the training dataset. (b) Corrective Sampling (CS): explicitly filter the generated samples and ensure that they follow the desired latent attribute distribution. We have designed the fairness intervention to work for various degrees of supervision on the spurious attribute, including unsupervised, weakly-supervised, and semi-supervised scenarios. Our experimental results demonstrate that FICS can effectively resolve spurious causality of conditional generation across various datasets.

Key facts. Cited by 5 papers in the map (data through 14 Oct 2026). Builds on 49 of its 61 references that are in the map. First posted on arXiv 5 Dec 2022.

Builds on

49 of 61 references.

References · 49

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2020Language Models are Few-Shot Learnerscited by 32,419
  3. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,255
  4. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,585
  5. 2022Generative Adversarial Networkscited by 5,627
  6. 2014Deep Learning Face Attributes in the Wildcited by 3,693
  7. 2020Taming Transformers for High-Resolution Image Synthesiscited by 3,253
  8. 2018Large Scale GAN Training for High Fidelity Natural Image Synthesiscited by 2,626
  9. 2017Improved Training of Wasserstein GANscited by 2,235
  10. 2014Conditional Generative Adversarial Netscited by 2,162
  11. 2020FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidencecited by 2,085
  12. 2020Shortcut Learning in Deep Neural Networkscited by 1,969
37 more
  1. 2019Invariant Risk Minimizationcited by 1,854
  2. 2018Spectral Normalization for Generative Adversarial Networkscited by 1,636
  3. 2018ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustnesscited by 1,432
  4. 2019Distributionally Robust Neural Networks for Group Shifts: On the Importance of Regularization for Worst-Case Generalizationcited by 1,247
  5. 2021Alias-Free Generative Adversarial Networkscited by 1,184
  6. 2020HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesiscited by 1,154
  7. 2017Counterfactual Fairnesscited by 804
  8. 2020NVAE: A Deep Hierarchical Variational Autoencodercited by 637
  9. 2021Just Train Twice: Improving Group Robustness without Training Group Informationcited by 523
  10. 2021Tackling the Generative Learning Trilemma with Denoising Diffusion GANscited by 516
  11. 2020Differentiable Augmentation for Data-Efficient GAN Trainingcited by 363
  12. 2020Noise or Signal: The Role of Image Backgrounds in Object Recognitioncited by 343
  13. 2022CogView2: Faster and Better Text-to-Image Generation via Hierarchical Transformerscited by 307
  14. 2020Environment Inference for Invariant Learningcited by 306
  15. 2019Learning De-biased Representations with Biased Representationscited by 199
  16. 2022Generating Videos with Dynamics-aware Implicit Generative Adversarial Networkscited by 186
  17. 2018FairGAN: Fairness-aware Generative Adversarial Networkscited by 164
  18. 2020Freeze the Discriminator: a Simple Baseline for Fine-Tuning GANscited by 132
  19. 2020DP-CGAN: Differentially Private Synthetic Data and Label Generationcited by 118
  20. 2019Fair Generative Modeling via Weak Supervisioncited by 114
  21. 2021DECAF: Generating Fair Synthetic Data Using Causally-Aware Generative Networkscited by 109
  22. 2018BAGAN: Data Augmentation with Balancing GANcited by 90
  23. 2022Spread Spurious Attribute: Improving Worst-group Accuracy with Spurious Attribute Estimationcited by 88
  24. 2018Bias and Generalization in Deep Generative Models: An Empirical Studycited by 83
  25. 2019Bias Correction of Learned Generative Models using Likelihood-Free Importance Weightingcited by 70
  26. 2018Discriminator Rejection Samplingcited by 49
  27. 2020Improving the Fairness of Deep Generative Models without Retrainingcited by 46
  28. 2020Inclusive GAN: Improving Data and Minority Coverage in Generative Modelscited by 44
  29. 2018InstaGAN: Instance-aware Image-to-Image Translationcited by 37
  30. 2021Object-aware Contrastive Learning for Debiased Scene Representationcited by 35
  31. 2021Fairness for Image Generation with Uncertain Sensitive Attributescited by 31
  32. 2020Imperfect ImaGANation: Implications of GANs Exacerbating Biases on Facial Data Augmentation and Snapchat Selfie Lensescited by 27
  33. 2020MASKER: Masked Keyword Regularization for Reliable Text Classificationcited by 25
  34. 2021BARACK: Partially Supervised Group Robustness With Guaranteescited by 25
  35. 2021MaGNET: Uniform Sampling from Deep Generative Network Manifolds Without Retrainingcited by 23
  36. 2021Self-Diagnosing GAN: Diagnosing Underrepresented Samples in Generative Adversarial Networkscited by 13
  37. 2019Mining GOLD Samples for Conditional GANscited by 11

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 1; 2024: 3; 2025: 0; 2026: 1.

5 papers.

  1. 2023Discovering and Mitigating Visual Biases through Keyword Explanationcited by 58
  2. 2024Balancing Act: Distribution-Guided Debiasing in Diffusion Modelscited by 52
  3. 2024DiffusionPID: Interpreting Diffusion via Partial Information Decompositioncited by 16
  4. 2024FairGen: Enhancing Fairness in Text-to-Image Diffusion Models via Self-Discovering Latent Directionscited by 8
  5. 2026Stay Fair! Ensuring Group Fairness in Diffusion Models Across Guidance Scalescited by 2