mlsanityData to 14 Oct 2026

UCD: Unconditional Discriminator Promotes Nash Equilibrium in GANs

arXiv 2510.00624 ↗First posted 1 Oct 2025cs.CVcited by 0

On the map: Diffusion models → Diffusion foundations and sampling

Abstract (arXiv)

Adversarial training turns out to be the key to one-step generation, especially for Generative Adversarial Network (GAN) and diffusion model distillation. Yet in practice, GAN training hardly converges properly and struggles in mode collapse. In this work, we quantitatively analyze the extent of Nash equilibrium in GAN training, and conclude that redundant shortcuts by inputting condition in D disables meaningful knowledge extraction. We thereby propose to employ an unconditional discriminator (UCD), in which D is enforced to extract more comprehensive and robust features with no condition injection. In this way, D is able to leverage better knowledge to supervise G, which promotes Nash equilibrium in GAN literature. Theoretical guarantee on compatibility with vanilla GAN theory indicates that UCD can be implemented in a plug-in manner. Extensive experiments confirm the significant performance improvements with high efficiency. For instance, we achieved 1.47 FID on the ImageNet-64 dataset, surpassing StyleGAN-XL and several state-of-the-art one-step diffusion models. The code will be made publicly available.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models. Builds on 52 of its 59 references that are in the map. First posted on arXiv 1 Oct 2025.

Builds on

52 of 59 references.

Closest ancestors · 18

Same direction, newest first

  1. 2024Improved Distribution Matching Distillation for Fast Image Synthesis23 May
  2. 2024PaGoDA: Progressive Growing of a One-Step Generator from a Low-Resolution Diffusion Teacher23 May
  3. 2024Score identity Distillation: Exponentially Fast Distillation of Pretrained Diffusion Models for One-Step Generation5 Apr
  4. 2024Scaling Rectified Flow Transformers for High-Resolution Image Synthesis5 Mar
  5. 2024SDXL-Lightning: Progressive Adversarial Diffusion Distillation21 Feb
  6. 2023Analyzing and Improving the Training Dynamics of Diffusion Models5 Dec
  7. 2023One-step Diffusion with Distribution Matching Distillation30 Nov
  8. 2023SMaRt: Improving GANs with Score Matching Regularity30 Nov
  9. 2023Adversarial Diffusion Distillation28 Nov
  10. 2023Improved Techniques for Training Consistency Models22 Oct
  11. 2023Consistency Trajectory Models: Learning Probability Flow ODE Trajectory of Diffusion1 Oct
  12. 2023Consistency Models2 Mar
6 more
  1. 2022Elucidating the Design Space of Diffusion-Based Generative Models1 Jun
  2. 2022Fast Sampling of Diffusion Models with Exponential Integrator29 Apr
  3. 2021Diffusion Models Beat GANs on Image Synthesis11 May
  4. 2020Score-Based Generative Modeling through Stochastic Differential Equations26 Nov
  5. 2020Denoising Diffusion Implicit Models6 Oct
  6. 2020Denoising Diffusion Probabilistic Models19 Jun

Other references · 34

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,968
  3. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 6,013
  4. 2016Image-to-Image Translation with Conditional Adversarial Networkscited by 4,527
  5. 2019Analyzing and Improving the Image Quality of StyleGANcited by 3,588
  6. 2014Conditional Generative Adversarial Netscited by 2,162
  7. 2019Semantic Image Synthesis with Spatially-Adaptive Normalizationcited by 1,299
  8. 2020Training Generative Adversarial Networks with Limited Datacited by 1,186
  9. 2021Alias-Free Generative Adversarial Networkscited by 1,184
  10. 2024Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Predictioncited by 1,023
  11. 2016Least Squares Generative Adversarial Networkscited by 1,001
  12. 2019Improved Precision and Recall Metric for Assessing Generative Modelscited by 944
22 more
  1. 2019Interpreting the Latent Space of GANs for Semantic Face Editingcited by 695
  2. 2020Contrastive Learning for Unpaired Image-to-Image Translationcited by 639
  3. 2016Conditional Image Synthesis With Auxiliary Classifier GANscited by 565
  4. 2022StyleGAN-XL: Scaling StyleGAN to Large Diverse Datasetscited by 492
  5. 2023Scaling up GANs for Text-to-Image Synthesiscited by 492
  6. 2017Towards Principled Methods for Training Generative Adversarial Networkscited by 464
  7. 2020Closed-Form Factorization of Latent Semantics in GANscited by 408
  8. 2020Differentiable Augmentation for Data-Efficient GAN Trainingcited by 363
  9. 2023StyleGAN-T: Unlocking the Power of GANs for Fast Large-Scale Text-to-Image Synthesiscited by 212
  10. 2022Diffusion-GAN: Training GANs with Diffusioncited by 205
  11. 2021Instance-Conditioned GANcited by 121
  12. 2020TSIT: A Simple and Versatile Framework for Image-to-Image Translationcited by 64
  13. 2025The GAN is dead; long live the GAN! A Modern GAN Baselinecited by 59
  14. 2021Data-Efficient Instance Generation from Instance Discriminationcited by 58
  15. 2021Deceive D: Adaptive Pseudo Augmentation for GAN Training with Limited Datacited by 53
  16. 2021Training GANs with Stronger Augmentations via Contrastive Discriminatorcited by 45
  17. 2021Improving GAN Equilibrium by Raising Spatial Awarenesscited by 22
  18. 2022Region-Based Semantic Factorization in GANscited by 22
  19. 2022Improving GANs with A Dynamic Discriminatorcited by 20
  20. 2023LinkGAN: Linking GAN Latents to Pixels for Controllable Image Synthesiscited by 15
  21. 2023Exploring Sparse MoE in GANs for Text-conditioned Image Synthesiscited by 13
  22. 2024MCGAN: Enhancing GAN Training with Regression-Based Generator Losscited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025SDAKD: Student Discriminator Assisted Knowledge Distillation for Super-Resolution Generative Adversarial Networks
  2. 2022Discriminator-Cooperated Feature Map Distillation for GAN Compression
  3. 2023KD-DLGAN: Data Limited Image Generation via Knowledge Distillation
  4. 2021Training Generative Adversarial Networks in One Stage
  5. 2026Improving Generative Adversarial Networks with Self-Distillation
  6. 2018Variational Discriminator Bottleneck: Improving Imitation Learning, Inverse RL, and GANs by Constraining Information Flow
  7. 2021Content-Aware GAN Compression
  8. 2025Generative Adversarial Networks Bridging Art and Machine Intelligence