mlsanityData to 14 Oct 2026

Feature Unlearning for Pre-trained GANs and VAEs

arXiv 2303.05699 ↗First posted 10 Mar 2023cs.CV cs.LGcited by 16

On the map: LLM safety and security → Vision-language and text-to-image safety

Abstract (arXiv)

We tackle the problem of feature unlearning from a pre-trained image generative model: GANs and VAEs. Unlike a common unlearning task where an unlearning target is a subset of the training set, we aim to unlearn a specific feature, such as hairstyle from facial images, from the pre-trained generative models. As the target feature is only presented in a local region of an image, unlearning the entire image from the pre-trained model may result in losing other details in the remaining region of the image. To specify which features to unlearn, we collect randomly generated images that contain the target features. We then identify a latent representation corresponding to the target feature and then use the representation to fine-tune the pre-trained model. Through experiments on MNIST, CelebA, and FFHQ datasets, we show that target features are successfully removed while keeping the fidelity of the original models. Further experiments with an adversarial attack show that the unlearned model is more robust under the presence of malicious parties.

Key facts. Cited by 16 papers in the map (data through 14 Oct 2026). Placed in Vision-language and text-to-image safety, one of 12 directions in LLM safety and security. Builds on 35 of its 52 references that are in the map. First posted on arXiv 10 Mar 2023.

Builds on

35 of 52 references.

References · 35

  1. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,585
  2. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  3. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 6,013
  4. 2022Generative Adversarial Networkscited by 5,627
  5. 2016Image-to-Image Translation with Conditional Adversarial Networkscited by 4,527
  6. 2017MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applicationscited by 3,845
  7. 2017Progressive Growing of GANs for Improved Quality, Stability, and Variationcited by 2,988
  8. 2016Improved Techniques for Training GANscited by 2,970
  9. 2016Perceptual Losses for Real-Time Style Transfer and Super-Resolutioncited by 2,694
  10. 2015Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networkscited by 2,054
  11. 2016Context Encoders: Feature Learning by Inpaintingcited by 1,428
  12. 2016Colorful Image Colorizationcited by 1,114
23 more
  1. 2018Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representationscited by 941
  2. 2019Interpreting the Latent Space of GANs for Semantic Face Editingcited by 695
  3. 2019Eternal Sunshine of the Spotless Net: Selective Forgetting in Deep Networkscited by 618
  4. 2020GANSpace: Discovering Interpretable GAN Controlscited by 614
  5. 2019Making AI Forget You: Data Deletion in Machine Learningcited by 465
  6. 2020Closed-Form Factorization of Latent Semantics in GANscited by 408
  7. 2017Adversarial Examples: Attacks and Defenses for Deep Learningcited by 351
  8. 2022A Survey of Machine Unlearningcited by 312
  9. 2020Unsupervised Discovery of Interpretable Directions in the GAN Latent Spacecited by 289
  10. 2020Very Deep VAEs Generalize Autoregressive Models and Can Outperform Them on Imagescited by 264
  11. 2021Fast Yet Effective Machine Unlearningcited by 217
  12. 2019GANalyze: Toward Visual Definitions of Cognitive Image Propertiescited by 193
  13. 2021Adaptive Machine Unlearningcited by 176
  14. 2022Zero-Shot Machine Unlearningcited by 133
  15. 2020Variational Bayesian Unlearningcited by 117
  16. 2020Machine Unlearning: Linear Filtration for Logit-based Classifierscited by 95
  17. 2018Morpho-MNIST: Quantitative Assessment and Diagnostics for Representation Learningcited by 56
  18. 2021WarpedGANSpace: Finding non-linear RBF paths in GAN latent spacecited by 48
  19. 2022StudioGAN: A Taxonomy and Benchmark of GANs for Image Synthesiscited by 38
  20. 2021The Geometry of Deep Generative Image Models and its Applicationscited by 35
  21. 2022Few-Shot Unlearning by Model Inversioncited by 19
  22. 2022Data Redaction from Pre-trained GANscited by 18
  23. 2022Efficient Attribute Unlearning: Towards Selective Removal of Input Attributes from Feature Representationscited by 17

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 3; 2024: 6; 2025: 5; 2026: 2.

16 papers.

Sort:

By area:

  1. 2023Exploring the Landscape of Machine Unlearning: A Comprehensive Survey and Taxonomycited by 51
  2. 2024Machine Unlearning in Generative AI: A Surveycited by 49
  3. 2024Unified Parameter-Efficient Unlearning for LLMscited by 25
  4. 2024A Survey on Machine Unlearning: Techniques and New Emerged Privacy Riskscited by 15
  5. 2023Adapt then Unlearn: Exploring Parameter Space Semantics for Unlearning in Generative Adversarial Networkscited by 11
  6. 2024Ferrari: Federated Feature Unlearning via Optimizing Feature Sensitivitycited by 11
  7. 2023FAST: Feature Aware Similarity Thresholding for Weak Unlearning in Black-Box Generative Modelscited by 5
  8. 2024Moderating the Generalization of Score-based Generative Modelcited by 4
  9. 2025Generative Model Unlearning: A Survey through Target Events, Unlearning Operators, and Evaluation Protocolscited by 4
  10. 2025Efficient Knowledge Graph Unlearning with Zeroth-order Informationcited by 3
  11. 2025Backdoor Defense in Diffusion Models via Spatial Attention Unlearningcited by 2
  12. 2025Prompting Forgetting: Unlearning in GANs via Textual Guidancecited by 1
4 more
  1. 2025Unlearning in Diffusion models under Data Constraints: A Variational Inference Approachcited by 1
  2. 2024Targeted Therapy in Data Removal: Object Unlearning Based on Scene Graphscited by 0
  3. 2026IREU: Identity-Related Encoder-Only Unlearning for Customized Portrait Generationcited by 0
  4. 2026DECAF: De-Clustering for Adaptive Representational Unlearningcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025UNO: Unlearning via Orthogonalization in Generative models
  2. 2024Unlearn and Burn: Adversarial Machine Unlearning Requests Destroy Model Accuracy
  3. 2024Defensive Unlearning with Adversarial Training for Robust Concept Erasure in Diffusion Models
  4. 2024Machine Unlearning for Image-to-Image Generative Models
  5. 2025AMUN: Adversarial Machine UNlearning
  6. 2023Generative Adversarial Networks Unlearning
  7. 2025Toward Reliable Machine Unlearning: Theory, Algorithms, and Evaluation
  8. 2024Attack and Reset for Unlearning: Exploiting Adversarial Noise toward Machine Unlearning through Parameter Re-initialization