StyleGAN-XL: Scaling StyleGAN to Large Diverse Datasets
Open accessThe DOI matches its Crossref record; 28 of the 30 references it cites were found in a registry or an index, and 2 are not checked yet.
Abstract
Computer graphics has experienced a recent surge of data-centric approaches for photorealistic and controllable content creation. StyleGAN in particular sets new standards for generative modeling regarding image quality and controllability. However, StyleGAN’s performance severely degrades on large unstructured datasets such as ImageNet. StyleGAN was designed for controllability; hence, prior works suspect its restrictive design to be unsuitable for diverse datasets. In contrast, we find the main limiting factor to be the current training strategy. Following the recently introduced Projected GAN paradigm, we leverage powerful neural network priors and a progressive growing strategy to successfully train the latest StyleGAN3 generator on ImageNet. Our final model, StyleGAN-XL, sets a new state-of-the-art on large-scale image synthesis and is the first to generate images at a resolution of 10242 at such a dataset scale. We demonstrate that this model can invert and edit images beyond the narrow domain of portraits or specific object classes. Code, models, and supplementary videos can be found at https://sites.google.com/view/stylegan-xl/ .
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“StyleGAN-XL overcomes the scalability issues of previous models, offering better sample diversity, generalization, and image fidelity.”
From Deepfake Media Forensics: Status and Future Challenges (2025) · cites this paper as [36]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Our hypothesis could explain the phenomenon that conditional generation is easier than unconditional (Mirza & Osindero, 2014;Liu et al., 2020;Sauer et al., 2022), as the data we condition on may have the same platonic structure as the data we are generating.”
From The Platonic Representation Hypothesis (2024) · cites this paper as Sauer et al., 2022
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“StyleGAN-XL [36] is an extension of the StyleGAN family, designed to generate high-quality images on diverse and unstructured datasets, particularly ImageNet.”
From Deepfake Media Forensics: Status and Future Challenges (2025) · cites this paper as [36]
The cited source says what the claim attributes to it.
In the citing paper
“The model achieves state-of-the-art performance in large-scale image synthesis, generating high-resolution images up to 1024 × 1024.”
From Deepfake Media Forensics: Status and Future Challenges (2025) · cites this paper as [36]
The cited source says what the claim attributes to it.
In the citing paper
“StyleGAN-XL incorporates Projected GAN training and a progressive growing strategy to enhance stability and scalability.”
From Deepfake Media Forensics: Status and Future Challenges (2025) · cites this paper as [36]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
ImageNet: A large-scale hierarchical image database
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009) · 2009 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2009.5206848 · Cited by 65,333 · cited 1×
“However, when trained on large and unstructured datasets like ImageNet [Deng et al. 2009], StyleGANs do not achieve satisfactory results yet.”
MethodGoing deeper with convolutions
Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., & Rabinovich, A. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298594 · Cited by 48,562
The unreasonable effectiveness of deep features as a perceptual metric
Zhang, R., Isola, P., Efros, A. A., Shechtman, E., & Wang, O. (2018) · arXiv · doi:10.48550/arxiv.1801.03924 · Cited by 24,812
The Unreasonable Effectiveness of Deep Features as a Perceptual Metric
Zhang, R., Isola, P., Efros, A. A., Shechtman, E., & Wang, O. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00068 · Cited by 14,623
Unsupervised representation learning with deep convolutional generative adversarial networks
Radford, A., Metz, L., & Chintala, S. (2015) · arXiv · doi:10.48550/arxiv.1511.06434 · Cited by 12,235 · cited 2×
“However, its architecture is more restrictive than a standard generator network [Karras et al. 2018; Radford et al. 2016] which seems to come at a price when training on complex”
BackgroundA Style-Based Generator Architecture for Generative Adversarial Networks
Karras, T., Laine, S., & Aila, T. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00453 · Cited by 11,714 · cited 7×
“Regularization improves results on uni-modal datasets like FFHQ [27] or LSUN [63], whereas it can be detrimental on multi-modal datasets [5, 19].”
BackgroundInfluentialProceedings of the 25th international conference on Machine learning - ICML '08
(2008) · doi:10.1145/1390156 · Cited by 6,680
Analyzing and Improving the Image Quality of StyleGAN
Karras, T., Laine, S., Aittala, M., Hellsten, J., Lehtinen, J., & Aila, T. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.00813 · Cited by 6,332
Analyzing and Improving the Image Quality of StyleGAN
Karras, T., Laine, S., Aittala, M., Hellsten, J., Lehtinen, J., & Aila, T. (2019) · arXiv · doi:10.48550/arxiv.1912.04958 · Cited by 6,332
SwinIR: Image Restoration Using Swin Transformer
Liang, J., Cao, J., Sun, G., Zhang, K., Van Gool, L., & Timofte, R. (2021) · 2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW) · doi:10.1109/iccvw54120.2021.00210 · Cited by 4,734 · cited 2×
“the data with a super-resolution network [Liang et al. 2021], see supplementary.”
Method
Research works printed but not linked: 46
- Yuval Alaluf, Or Patashnik, Zongze Wu, Asif Zamir, Eli Shechtman, Dani Lischinski, and Daniel Cohen-Or. 2022. Third Time's the Charm? Image and Video Editing with StyleGAN3. arXiv.org abs/2201.13433 (2022). https://arxiv.org/abs/2201.13433 · arXiv
- Yuval Alaluf, Or Patashnik, and Daniel Cohen-Or. 2021. ReStyle: A Residual-Based StyleGAN Encoder via Iterative Refinement. Proc. of the IEEE International Conf. on Computer Vision (ICCV) abs/2104.02699 (2021). https://arxiv.org/abs/2104.02699 · arXiv
- Andrew Brock, Jeff Donahue, and Karen Simonyan. 2019. Large Scale GAN Training for High Fidelity Natural Image Synthesis. In Proc. of the International Conf. on Learning Representations (ICLR) . OpenReview.net. https://openreview.net/forum? id=B1xsqj09Fm
- Arantxa Casanova, Marlène Careil, Jakob Verbeek, Michal Drozdzal, and Adriana Romero-Soriano. 2021. Instance-Conditioned GAN. In Advances in Neural Information Processing Systems (NeurIPS) .
- Lucy Chai, Jun-Yan Zhu, Eli Shechtman, Phillip Isola, and Richard Zhang. 2021. Ensembling With Deep Generative Views. In Proc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR) . Computer Vision Foundation / IEEE, 1499715007. https://openaccess.thecvf.com/content/CVPR2021/html/Chai\_Ensembling\_ With\_Deep\_Generative\_Views\_CVPR\_2021\_paper.html momentum contrastive learning. arxiv.org:2003.04297
- Ting-Yun Chang and Chi-Jen Lu. 2020. TinyGAN: Distilling BigGAN for Conditional Image Generation. In Proc. of the Asian Conf. on Computer Vision (ACCV) (Lecture Notes in Computer Science, Vol. 12625) , Hiroshi Ishikawa, Cheng-Lin Liu, Tomás Pajdla, and Jianbo Shi (Eds.). 509-525. https://doi.org/10.1007/978-3-030-69538-5\_31 Xinlei Chen, Haoqi Fan, and Ross Girshick Kaiming He. 2020. Improved baselines with · DOI
- Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, Jakob Uszkoreit, and Neil Houlsby. 2021. An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale. In Proc. of the International Conf. on Learning Representations (ICLR) . OpenReview.net. https://openreview.net/ forum?id=YicbFdNTTy
- Ian J. Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron C. Courville, and Yoshua Bengio. 2014. Generative Adversarial Nets. In Advances in Neural Information Processing Systems (NeurIPS) ,
- Tom George Grigg, Dan Busbridge, Jason Ramapuram, and Russ Webb. 2021. Do Self-Supervised and Supervised Methods Learn Similar Visual Representations? arxiv.org:2110.00528
- Timofey Grigoryev, Andrey Voynov, and Artem Babenko. 2022. When, Why, and Which Pretrained GANs Are Useful?. In Proc. of the International Conf. on Learning Representations (ICLR) . https://openreview.net/forum?id=4Ycr8oeCoIh
- Gwern. 2020. Making Anime Faces with StyleGAN . Retrieved January 16, 2022 from https://www.gwern.net/Faces#stylegan2-ext-modifications/
- Erik Härkönen, Aaron Hertzmann, Jaakko Lehtinen, and Sylvain Paris. 2020. GANSpace: Discovering Interpretable GAN Controls. In Advances in Neural Information Processing Systems (NeurIPS) . https://proceedings.neurips.cc/paper/2020/hash/ 6fe43269967adbb64ec6149852b5cc3e-Abstract.html
- Martin Heusel, Hubert Ramsauer, Thomas Unterthiner, Bernhard Nessler, and Sepp Hochreiter. 2017. GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium. In Advances in Neural Information Processing Systems (NeurIPS) . 6626-6637. https://proceedings.neurips.cc/paper/2017/hash/ 8a1d694707eb0fefe65871369074926d-Abstract.html
- Ali Jahanian, Lucy Chai, and Phillip Isola. 2020. On the "steerability" of generative adversarial networks. In Proc. of the International Conf. on Learning Representations (ICLR) . OpenReview.net. https://openreview.net/forum?id=HylsTT4FvB
- Tero Karras, Timo Aila, Samuli Laine, and Jaakko Lehtinen. 2018. Progressive Growing of GANs for Improved Quality, Stability, and Variation. In Proc. of the International Conf. on Learning Representations (ICLR) . OpenReview.net. https://openreview.net/ forum?id=Hk99zCeAb
- Tero Karras, Miika Aittala, Samuli Laine, Erik Härkönen, Janne Hellsten, Jaakko Lehtinen, and Timo Aila. 2021. Alias-Free Generative Adversarial Networks. In Advances in Neural Information Processing Systems (NeurIPS) .
- Umut Kocasari, Alara Dirik, Mert Tiftikci, and Pinar Yanardag. 2022. StyleMC: MultiChannel Based Fast Text-Guided Image Generation and Manipulation. Proc. of the IEEE Winter Conference on Applications of Computer Vision (WACV) (2022). https: //arxiv.org/abs/2112.08493
- Diederik P. Kingma and Jimmy Ba. 2015. Adam: A Method for Stochastic Optimization. In Proc. of the International Conf. on Learning Representations (ICLR) .
- Daiqing Li, Huan Ling, Seung Wook Kim, Karsten Kreis, Adela Barriuso, Sanja Fidler, and Antonio Torralba. 2022. BigDatasetGAN: Synthesizing ImageNet with Pixelwise Annotations. arXiv.org (2022). https://arxiv.org/abs/2201.04684 · arXiv
- Huan Ling, Karsten Kreis, Daiqing Li, Seung Wook Kim, Antonio Torralba, and Sanja Fidler. 2021. EditGAN: High-Precision Semantic Image Editing. arXiv.org abs/2111.03186 (2021). https://arxiv.org/abs/2111.03186 · arXiv
- Bingchen Liu, Yizhe Zhu, Kunpeng Song, and Ahmed Elgammal. 2021. Towards Faster and Stabilized GAN Training for High-fidelity Few-shot Image Synthesis. In Proc. of the International Conf. on Learning Representations (ICLR) . OpenReview.net. https: //openreview.net/forum?id=1Fqg133qRaI
- Lars M. Mescheder, Andreas Geiger, and Sebastian Nowozin. 2018. Which Training Methods for GANs do actually Converge?. In Proc. of the International Conf. on Machine learning (ICML) (Proceedings of Machine Learning Research, Vol. 80) . PMLR, 3478-3487. http://proceedings.mlr.press/v80/mescheder18a.html
- Takeru Miyato, Toshiki Kataoka, Masanori Koyama, and Yuichi Yoshida. 2018. Spectral Normalization for Generative Adversarial Networks. In Proc. of the International Conf. on Learning Representations (ICLR) . OpenReview.net. https://openreview.net/ forum?id=B1QRgziT-
- Alexander Quinn Nichol and Prafulla Dhariwal. 2021. Improved Denoising Diffusion Probabilistic Models. In Proc. of the International Conf. on Machine learning (ICML) (Proceedings of Machine Learning Research, Vol. 139) . PMLR, 8162-8171. http:// proceedings.mlr.press/v139/nichol21a.html
- Or Patashnik, Zongze Wu, Eli Shechtman, Daniel Cohen-Or, and Dani Lischinski. 2021. StyleCLIP: Text-Driven Manipulation of StyleGAN Imagery. In Proc. of the IEEE International Conf. on Computer Vision (ICCV) . 2085-2094.
- Etienne Perot, Pierre de Tournemire, Davide Nitti, Jonathan Masci, and Amos Sironi. 2020. Learning to Detect Objects with a 1 Megapixel Event Camera. In Advances in Neural Information Processing Systems (NeurIPS) . https://proceedings.neurips.cc/ paper/2020/hash/c213877427b46fa96cff6c39e837ccee-Abstract.html
- Phillip Pope, Chen Zhu, Ahmed Abdelkader, Micah Goldblum, and Tom Goldstein. 2021. The Intrinsic Dimension of Images and Its Impact on Learning. In Proc. of the International Conf. on Learning Representations (ICLR) . OpenReview.net. https: //openreview.net/forum?id=XJk19XzGq2J
- Maithra Raghu, Thomas Unterthiner, Simon Kornblith, Chiyuan Zhang, and Alexey Dosovitskiy. 2021. Do Vision Transformers See Like Convolutional Neural Networks?. In Advances in Neural Information Processing Systems (NeurIPS) .
- Daniel Roich, Ron Mokady, Amit H. Bermano, and Daniel Cohen-Or. 2021. Pivotal Tuning for Latent-based Editing of Real Images. arXiv.org (2021). https://arxiv.org/ abs/2106.05744 · arXiv
- Chitwan Saharia, Jonathan Ho, William Chan, Tim Salimans, David J. Fleet, and Mohammad Norouzi. 2021. Image Super-Resolution via Iterative Refinement. arxiv.org:2104.07636
- Tim Salimans, Ian J. Goodfellow, Wojciech Zaremba, Vicki Cheung, Alec Radford, and Xi Chen. 2016. Improved Techniques for Training GANs. In Advances in Neural Information Processing Systems (NeurIPS) . 2226-2234. https://proceedings.neurips. cc/paper/2016/hash/8a3363abe792db2d8761d6403605aeb7-Abstract.html
- Axel Sauer, Kashyap Chitta, Jens Müller, and Andreas Geiger. 2021. Projected GANs Converge Faster. In Advances in Neural Information Processing Systems (NeurIPS) .
- Axel Sauer and Andreas Geiger. 2021. Counterfactual Generative Networks. In Proc. of the International Conf. on Learning Representations (ICLR) . OpenReview.net. https: //openreview.net/forum?id=BXewfAYMmJw
- Yujun Shen and Bolei Zhou. 2021. Closed-Form Factorization of Latent Semantics in GANs. In Proc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR) . 1532-1540. https://openaccess.thecvf.com/content/CVPR2021/html/Shen\_Closed-Form\_Factorization\_of\_Latent\_Semantics\_in\_GANs\_CVPR\_2021\_paper.html
- Jiaming Song, Chenlin Meng, and Stefano Ermon. 2021. Denoising Diffusion Implicit Models. In Proc. of the International Conf. on Learning Representations (ICLR) . https: //openreview.net/forum?id=St1giarCHLP
- Nurit Spingarn, Ron Banner, and Tomer Michaeli. 2021. GAN "Steerability" without optimization. In Proc. of the International Conf. on Learning Representations (ICLR) . OpenReview.net. https://openreview.net/forum?id=zDy\_nQCXiIj
- Mingxing Tan and Quoc V. Le. 2019. EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks. In Proc. of the International Conf. on Machine learning (ICML) (Proceedings of Machine Learning Research, Vol. 97) , Kamalika Chaudhuri and
- Matthew Tancik, Pratul P. Srinivasan, Ben Mildenhall, Sara Fridovich-Keil, Nithin Raghavan, Utkarsh Singhal, Ravi Ramamoorthi, Jonathan T. Barron, and Ren Ng. 2020. Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domains. In Advances in Neural Information Processing Systems (NeurIPS) . https://proceedings.neurips.cc/paper/2020/hash/ 55053683268957697aa39fba6f231c68-Abstract.html
- Hugo Touvron, Matthieu Cord, Matthijs Douze, Francisco Massa, Alexandre Sablayrolles, and Hervé Jégou. 2021. Training data-efficient image transformers & distillation through attention. In Proc. of the International Conf. on Machine learning (ICML) (Proceedings of Machine Learning Research, Vol. 139) . PMLR, 10347-10357. http://proceedings.mlr.press/v139/touvron21a.html
- Aäron van den Oord, Oriol Vinyals, and Koray Kavukcuoglu. 2017. Neural Discrete Representation Learning. In Advances in Neural Information Processing Systems (NeurIPS) , Isabelle Guyon, Ulrike von Luxburg, Samy Bengio, Hanna M. Wallach, Rob Fergus, S. V. N. Vishwanathan, and Roman Garnett (Eds.). 6306-6315. https://proceedings. neurips.cc/paper/2017/hash/7a98af17e63a0ac09ce2e96d03992fbc-Abstract.html
- Daniel Watson, Jonathan Ho, Mohammad Norouzi, and William Chan. 2021. Learning to Efficiently Sample from Diffusion Probabilistic Models. CoRR abs/2106.03802 (2021). https://arxiv.org/abs/2106.03802 · arXiv
- Andrey Voynov and Artem Babenko. 2020. Unsupervised Discovery of Interpretable Directions in the GAN Latent Space. In Proc. of the International Conf. on Machine learning (ICML) . PMLR, 9786-9796. http://proceedings.mlr.press/v119/voynov20a. html
- Zongze Wu, Dani Lischinski, and Eli Shechtman. 2021. StyleSpace Analysis: Disentangled Controls for StyleGAN Image Generation. In Proc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR) . Computer Vision Foundation / IEEE, 12863-12872. https://openaccess.thecvf.com/content/CVPR2021/html/Wu\_ StyleSpace\_Analysis\_Disentangled\_Controls\_for\_StyleGAN\_Image\_Generation\_ CVPR\_2021\_paper.html
- Rui Xu, Xintao Wang, Kai Chen, Bolei Zhou, and Chen Change Loy. 2021. Positional Encoding As Spatial Inductive Bias in GANs. In Proc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR) . Computer Vision Foundation / IEEE, 1356913578. https://openaccess.thecvf.com/content/CVPR2021/html/Xu\_Positional\_ Encoding\_As\_Spatial\_Inductive\_Bias\_in\_GANs\_CVPR\_2021\_paper.html
- Fisher Yu, Yinda Zhang, Shuran Song, Ari Seff, and Jianxiong Xiao. 2015. LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loop. arxiv.org:1506.03365
- Shengyu Zhao, Zhijian Liu, Ji Lin, Jun-Yan Zhu, and Song Han. 2020. Differentiable Augmentation for Data-Efficient GAN Training. In Advances in Neural Information Processing Systems (NeurIPS) . https://proceedings.neurips.cc/paper/2020/hash/ 55479c55ebd1efd3ff125f1337100388-Abstract.html
Cited by
Items 401–500 of 793
P2D: Plug and Play Discriminator for accelerating GAN frameworks
Chong, M. J., Singh, K. K., Li, Y., Lu, J., & Forsyth, D. A. (2024) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv57701.2024.00534 · Cited by 6
“We investigate whether ProjectedGAN [24] and StyleGAN-XL [25] are plug-and-play.”
BackgroundMethodInfluentialYeo, S., Jang, Y., & Yoo, J. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-73223-2_7 · Cited by 6
Augmentation leak-prevention scheme using an auxiliary classifier in GAN-based image generation
Shim, J., Kim, H., Kim, E., & Hwang, E. (2023) · Journal of King Saud University - Computer and Information Sciences · doi:10.1016/j.jksuci.2023.101711 · Cited by 6
“…performance in various image generation tasks, such as image-to-image transformation (Emami et al., 2021; Mehmood et al., 2022), super-resolution (You et al., 2019), text-to-image generation (Zhu et al., 2019), medical image analysis (Ren et al., 2022) and image denoising (Sauer et al., 2022).”
Metrics to Quantify Global Consistency in Synthetic Medical Images
Scholz, D., Wiestler, B., Rueckert, D., & Menten, M. J. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-53767-7_3 · Cited by 6
Spider GAN: Leveraging Friendly Neighbors to Accelerate GAN Training
Asokan, S., & Seelamantula, C. S. (2023) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52729.2023.00378 · Cited by 6
“While StyleGAN-XL achieves marginally superior FID, it does so at the cost of a three-fold increase in network complexity [70].”
BackgroundImproving Few-shot Image Generation by Structural Discrimination and Textural Modulation
Yang, M., Wang, Z., Feng, W., Zhang, Q., & Xiao, T. (2023) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3581783.3611763 · Cited by 6
“…of Generative Adversarial Networks [14] such as impressive sample quality and flexible content controllability have significantly advanced visual applications including image [4, 25, 27, 42] and video generation [47, 51, 60], image editing [45, 46], image-to-image translation [34, 39, 41, 59], etc.”
BackgroundThe Challenges of Image Generation Models in Generating Multi-Component ImagesPreprint
Foong, T. Y., Kotyan, S., Mao, P. Y., & Vargas, D. V. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2311.13620 · Cited by 6
“…on either a) Variational Auto-Encoders (VAEs) like DCVAE [25], b) Generative Adversarial Networks (GANs) like, Attention GAN [38], Style GAN [16, 36], Big GAN [4] or c) the more recent Diffusion-based Models like Dall-E [7, 30, 31], Imagen [34], Stable Diffusion [32] to generate high-quality…”
BackgroundSMaRt: Improving GANs with Score Matching RegularityPreprint
Xia, M., Shen, Y., Yang, C., Yi, R., Wang, W., & Liu, Y. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2311.18208 · Cited by 7
Bidirectional Normalizing Flow: From Data to Noise and BackPreprint
Lu, Y., Sun, Q., Wang, X., Jiang, Z., Zhao, H., & He, K. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2512.10953 · Cited by 6
Continuous Visual Autoregressive Generation via Score MaximizationPreprint
Shao, C., Meng, F., & Zhou, J. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2505.07812 · Cited by 6
Hao, S., Liu, X., Qi, X., Zhao, S., Zi, B., Xiao, R., Han, K., & Wong, K.-Y. K. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2410.14672 · Cited by 6
FrequencyBooster: Full-Frequency Modeling for High-Fidelity Pixel DiffusionPreprint
Ma, L., Guo, Z., He, Y., Fu, X., Liu, L., Fu, J., Huang, J., & Li, Y. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2605.17759 · Cited by 6
Fast Certification of Vision-Language Models Using Incremental Randomized Smoothing
Nirala, A., Joshi, A., Sarkar, S., & Hegde, C. (2024) · IEEE Conference on Secure and Trustworthy Machine Learning (SaTML) · doi:10.1109/satml59370.2024.00019 · Cited by 5
“Deep learning systems have achieved state-of-the-art performance in various domains, including computer vision [23], speech [36], and more [13], [27], [41], occasionally surpassing human capabilities [18].”
BackgroundUnmasking Deepfakes: A Systematic Review of Generation Techniques and Detection Strategies
Eidan, S. S., & Eadan, S. (2025) · Iraqi Journal of Intelligent Computing and Informatics (IJICI) · doi:10.52940/ijici.v4i2.105 · Cited by 5
Mellado, D., Mayeta-Revilla, L., Sotelo, J. A., Querales, M., Godoy, E., Lever, S., Pardo, F., Chabert, S., & Salas, R. (2025) · Frontiers in Oncology · doi:10.3389/fonc.2025.1601929 · Cited by 5
Toward Accurate Image Generation via Dynamic Generative Image Transformer
Mao, Z., Huang, M., Lin, Y., Wang, Q., Zhang, L., & Zhang, Y. (2026) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2026.3653620 · Cited by 5
Liu, E., Chen, Q., Ning, X., Yan, S., Dai, G., Lin, Z., & Wang, Y. (2025) · neural information processing systems · doi:10.52202/085713-0193 · Cited by 5
OsteoGA: An Explainable AI Framework for Knee Osteoarthritis Severity Assessment
Trung, H. P., Thiet, S. N., Trung, T. N., Tan, L. L., Tran, M., & Quan, T. (2023) · International Symposium on Information and Communication Technology · doi:10.1145/3628797.3628933 · Cited by 5
A destructive active defense algorithm for deepfake face images
Yang, Y., Idris, N. B., Liu, C., Wu, H., & Yu, D. (2024) · PeerJ Computer Science · doi:10.7717/peerj-cs.2356 · Cited by 5
“…generative adversarial networks ( Wang et al., 2024 ) that introduce conditional variables, recurrent generative adversarial networks ( Li & Wang, 2021 ) that allow conversion between different domains, and StyleGAN that can finely control styles and attributes ( Sauer, Schwarz & Geiger, 2022 ).”
Spotting fully-synthetic facial images via local camera surface frames
Ciamarra, A., Caldelli, R., & Bimbo, A. D. (2024) · IEEE International Workshop on Information Forensics and Security · doi:10.1109/wifs61860.2024.10810698 · Cited by 5
“On the other hand, powerful GAN-based architectures, such as StyleGAN models [4]–[7], have also been introduced.”
Latent Patched Efficient Diffusion Model for High Resolution Image Synthesis
Jiang, W., Jangid, D. K., Lee, S.-J., & Sheikh, H. R. (2025) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw67362.2025.00633 · Cited by 5
“The rapid advancement of diffusion models [5, 15, 20, 21] has positioned them as the state-of-the-art approach for image synthesis, outperforming generative adversarial networks (GANs) [1, 6, 7, 19, 22] in generating high-fidelity images with superior diversity and realism.”
Phan, H. C., Tran, M. T., Oh, S., Kim, D.-K., & Lee, H. W. (2026) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2026.114543 · Cited by 5
Soft Curriculum for Learning Conditional GANs with Noisy-Labeled and Uncurated Unlabeled Data
Katsumata, K., Vo, D. M., Harada, T., & Nakayama, H. (2024) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv57701.2024.00524 · Cited by 5
“Significant breakthroughs [3, 6, 17, 23, 24, 26, 34, 35, 45] in class-conditional image generation (cGANs) yield images with high fidelity and diversity; yet they are all trained in a supervised fashion where the training data consist of carefully labeled samples.”
BackgroundLearning Images Across Scales Using Adversarial Training
Wolski, K., Djeacoumar, A., Javanmardi, A., Seidel, H., Theobalt, C., Cordonnier, G., Myszkowski, K., George Drettakis, B., Pan, X., & Leimkühler, T. (2024) · ACM Transactions on Graphics · doi:10.1145/3658190 · Cited by 5
CAGAN: Constrained neural architecture search for GANs
Yang, Y., Zhang, X., Zhu, Q., Chen, W., Wong, K.-C., & Lin, Q. (2024) · Knowledge-Based Systems · doi:10.1016/j.knosys.2024.112277 · Cited by 5
Manifold Constraint Regularization for Remote Sensing Image Generation
Su, X., Zheng, C., Qiang, W., Wu, F., Zhao, J., Sun, F., & Xiong, H. (2024) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2024.3441631 · Cited by 5
“Additionally, numerous efforts have been made to address this issue using regularization methods [24], [25], [26], [27] or modifying network architectures [15], [28], [29], [30], [31], [32], [33].”
MethodRecursive Reasoning in Minimax Games: A Level k Gradient Play Method
Liu, Z., & Pavel, L. (2022) · neural information processing systems · doi:10.52202/068431-1230 · Cited by 5
On Accelerating Diffusion-Based Sampling Process via Improved Integration ApproximationPreprint
Zhang, G., Niwa, K., & Kleijn, W. B. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2304.11328 · Cited by 5
“In comparison to generative adversarial networks (GANs) [8; 1; 9; 24], DPMs exhibit more stable training dynamics by avoiding adversarial learning.”
BackgroundAlteration-free and Model-agnostic Origin Attribution of Generated ImagesPreprint
Wang, Z., Chen, C., Zeng, Y., Lyu, L., & Ma, S. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2305.18439 · Cited by 5
“…N A ttributio N ) in PyTorch and evaluated it on three different types of generative models (i.e., unconditional model, class-to-image model, and text-to-image model) including various GANs [1, 10, 30–32], VAEs [6], and diffusion models such as latest Consistency Model [4] and Stable Diffusion [3].”
MethodInfluentialRezaei, P., Farnia, F., & Li, C. T. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2412.17622 · Cited by 5
“In the first experiment, we used five distinct generative models: LDM (Rombach et al., 2022), StyleGAN-XL (Sauer et al., 2022), Efficient-VDVAE (Hazami et al., 2022), InsGen (Yang et al., 2021), and StyleNAT (Walton et al., 2022), all trained on the FFHQ dataset (Karras et al., 2019).”
Next Visual Granularity Generation
Wang, Y., Wang, Z., Wu, Z., Tao, Q., Liao, K., & Loy, C. C. (2025) · International Conference on Learning Representations · doi:10.48550/arxiv.2508.12811 · Cited by 5
DRAG: Data Reconstruction Attack using Guided DiffusionPreprint
Lei, W.-K., Chen, J.-C., & Chen, S.-T. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2509.11724 · Cited by 5
“GLASS (Li et al., 2023) uses domain-specific priors: StyleGAN2-ADA (FFHQ) (Karras et al., 2020a) for facial images and StyleGAN-XL (ImageNet-1K) (Sauer et al., 2022) for others, assuming knowledge of the target distribution.”
Scalable GANs with TransformersPreprint
Hyun, S., Lee, M., & Heo, J. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2509.24935 · Cited by 5
“Generative Adversarial Networks (GANs) are trained via an adversarial game between a generator and a discriminator, with progress driven by architectural and objective-level innovations.”
InfluentialSingh, G., Banerjee, T., & Ghosh, N. (2025) · Preprints.org · doi:10.20944/preprints202511.0637.v1 · Cited by 5
Autoregressive Image Generation with Linear Complexity: A Spatial-Aware Decay PerspectivePreprint
Mao, Y., Qin, Z., Zhou, J., Deng, H., Shen, X., Fan, B., Zhang, J., Zhong, Y., & Dai, Y. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2507.01652 · Cited by 5
“Traditionally, image generation has been dominated by GAN-based models [8–10] or diffusion models [11], which produce high-quality images through adversarial training or iterative denoising processes.”
Improving GANs by leveraging the quantum noise from real hardwarePreprint
Jin, H., & Merz, K. M. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2507.01886 · Cited by 5
“The StyleGAN family[Karras et al., 2019, 2020, Sauer et al., 2022] adopts a novel architecture that implements style-based generation through adaptive instance normalization (AdaIN).”
Markovian Scale Prediction: A New Era of Visual Autoregressive GenerationPreprint
Zhang, Y., Liu, J., Shi, Y., Zhang, Q., Miao, D., Wang, C., & Cao, L. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2511.23334 · Cited by 5
“We conduct a board comparison on class-to-image generation on ImageNet 256 × 256 benchmark with more alternative-paradigm models, including GANs [3, 22, 44], diffusion models [10, 19, 37, 42], mask autoregressive models [4, 28, 29], and next-token prediction autoregressive models [13, 25, 50, 68].”
HyperNet-Adaptation for Diffusion-Based Test Case GenerationPreprint
Weißl, O., Riccio, V., Kacianka, S., & Stocco, A. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2601.15041 · Cited by 5
“However, existing generative test generation methods are typically limited to simple datasets [39], due to model architecture [19, 20, 41], or rely on curated training data [28], restricting their applicability to complex, high-dimensional vision tasks.”
Duality Models: An Embarrassingly Simple One-step Generation ParadigmPreprint
Sun, P., Shang, X., Lin, T., & Shen, Z. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2602.17682 · Cited by 5
SubFlow: Sub-mode Conditioned Flow Matching for Diverse One-Step GenerationPreprint
Lin, Y., Shi, J., Ye, S., Wang, W., Yao, Y., & Liu, T. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2604.12273 · Cited by 5
Wei, K., Qian, J., Jia, C.-K., Chen, J., Zhang, J., Wu, Y., Zhu, J., & Guo, Y. (2025) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2025.3593631 · Cited by 4
Friction on demand: A generative framework for the inverse design of metainterfaces
Mouton, V., & Mélot, A. (2026) · Tribology International · doi:10.1016/j.triboint.2026.111724 · Cited by 4
Data Bias Mitigation and Evaluation Framework for Diffusion-Based Generative Face Models
Yu, H., Betke, M., & Bargal, S. A. (2025) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw69036.2025.00037 · Cited by 4
“In particular, Generative Adversarial Networks (GANs) [9] and Diffusion models [14] are two most recent classes of methods, delivering state-of-the-art performance in generating realistic facial images [39, 42].”
Noiseless Diffusion-GAN: Scaling-based data augmentation for generative models
Koike, Y., Nakagawa, T., Waida, H., & Kanamori, T. (2025) · Neural Networks · doi:10.1016/j.neunet.2025.108458 · Cited by 4
A Survey of GAN Based Image Synthesis
Ni, J. (2022) · Journal of Information Hiding and Privacy Protection · doi:10.32604/jihpp.2022.039751 · Cited by 4
Zhu, B., Lv, Q., Yang, Y., Zhang, K., Sui, X., Tang, Y., & Tan, Z. (2023) · Remote Sensing · doi:10.3390/rs15112827 · Cited by 4
Structure-Guided Image Completion With Image-Level and Object-Level Semantic Discriminators
Zheng, H., Lin, Z., Lu, J., Cohen, S. D., Shechtman, E., Barnes, C., Zhang, J., Liu, Q., Amirghodsi, S., Zhou, Y., et al. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3393898 · Cited by 4
“However, generating unaligned objects in a complex natural scene is known to be challenging [72] for various tasks including inpainting [19, 73] and semantic image generation [73].”
BackgroundRethinking Image Skip Connections in StyleGAN2
Park, S., & Shin, Y.-G. (2024) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2024.3495633 · Cited by 4
“To more effectively control Style parameters, a variety of models such as StyleGAN2 [26], StyleGAN3 [27], and StyleGAN-XL [28] have been developed, where these technologies are capable of producing images with the high quality.”
BackgroundA lightweight diffusion model for image generation based on improved MobileViT
Liu, Y., & Sun, W. (2024) · International Conference on Image Processing, Machine Learning and Pattern Recognition (IPMLPR) · doi:10.1145/3700906.3700917 · Cited by 4
On the Generalisation Capability of Local Surface Frames in Detecting Diffusion-Based Facial Images
Ciamarra, A., Caldelli, R., & Bimbo, A. D. (2025) · IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW) · doi:10.1109/wacvw65960.2025.00154 · Cited by 4
“This class of architectures fall into two main categories: StyleGANs [12, 13, 28] and Diffusion Models [9, 27, 29].”
Exploring the Faithfulness of Synthetic Data by Generative Models
Marulli, F., Paganini, P., & Lancellotti, F. (2023) · International Conference on Machine Learning and Applications · doi:10.1109/icmla58977.2023.00334 · Cited by 4
“…to compare real and synthetic data and rate their visual or semantic similarity; ii) Improved Generative Ar-chitectures, such as StyleGAN [15]; iii) Fine-Tuning on target data and Conditional Generation, that enable the production of more faithful data tailored to specific applications…”
Understanding the Generalization of Pretrained Diffusion Models on Out-of-Distribution Data
Ramachandran, S. N., Mukhopadhyay, R., Agarwal, M., Jawahar, C. V., & Namboodiri, V. P. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i13.29395 · Cited by 5
Partial Camera Obstruction Detection Using Single Value Image Metrics and Data Augmentation
de Oliveira, F. A. C., Niemi, A., García-Ortiz, A., & Torres, F. S. (2022) · International Conference on System Reliability and Safety (ICSRS) · doi:10.1109/icsrs56243.2022.10067386 · Cited by 4
“Other strategies, that are based on ML, are texture and style transfer [19], and synthetic images created by Generative Adversarial Networks (GAN) [20], [21], [22].”
BackgroundMultimodal Controller for Generative Models
Diao, E., Ding, J., & Tarokh, V. (2023) · Lecture notes in networks and systems · doi:10.1007/978-981-19-7867-8_10 · Cited by 4
“Latest variants of StyleGAN show that auxiliary classifier improves the performance of class-conditional generation [11,30].”
BackgroundDeep Learning-based Visualization of Music Mood
Wang, H., & Sourin, A. (2023) · International Conference on Cyberworlds (CW) · doi:10.1109/cw58918.2023.00015 · Cited by 4
Interpreting Class Conditional GANs with Channel AwarenessPreprint
He, Y., Zhang, Z., Zhu, J., Shen, Y., & Chen, Q. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2203.11173 · Cited by 4
“, GAN [11] has received wide attention due to its impressive performance on both unconditional synthesis [22, 24–26] and conditional synthesis [6,36,49,64].”
BackgroundTarget-Aware Generative Augmentations for Single-Shot AdaptationPreprint
Thopalli, K., Subramanyam, R., Turaga, P. K., & Thiagarajan, J. J. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2305.13284 · Cited by 4
“For this experiment, we used the state-of-the-art StyleGAN-XL model (Sauer et al., 2022) trained on ImageNet (Russakovsky et al.”
MethodExact Diffusion Inversion via Bi-directional Integration ApproximationPreprint
Zhang, G., Lewis, J. P., & Kleijn, W. B. (2023) · doi:10.36227/techrxiv.23713515.v1 · Cited by 4
Structural Adversarial Objectives for Self-Supervised Representation LearningPreprint
Zhang, X., & Maire, M. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2310.00357 · Cited by 4
“…G ϕ and D θ are alternatively updated in an adversarial fashion, which can be formulated as a minimax problem: Much research on GANs has focused on improving the quality of generated data, yielding significant advances (Karras et al., 2017; 2019; 2020b; 2021; Sauer et al., 2022; Dai et al., 2022).”
BackgroundUnsupervised Model DiagnosisPreprint
Wang, Y. O., Li, E., Luo, J., Wang, Z., & de la Torre, F. N. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2410.06243 · Cited by 4
“Generative models, especially StyleGAN [17, 18, 39] and Diffusion Models [11, 38, 44], have semantic latent spaces that are differentiable, and can be used to edit image attributes [12, 40, 41].”
MethodCART: Compositional Auto-Regressive Transformer for Image GenerationPreprint
Roheda, S., Chowdhury, R., Bala, A., & Jaiswal, R. (2024) · AAAI Publications (The Association for the Advancement of Artificial Intelligence (AAAI)) · doi:10.48550/arxiv.2411.10180 · Cited by 4
Joint Learning of Depth and Appearance for Portrait Image AnimationPreprint
Ji, X., Zoss, G., Chandran, P., Yang, L., Cao, X., Solenthaler, B., & Bradley, D. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2501.08649 · Cited by 4
“What we learned from early models like StyleGAN [31, 55] is that image synthesis methods can produce stunning photo-realistic images of human faces, indistinguishable from reality.”
Qiu, Y., Liu, Y., Yu, H., Fang, H., Chen, B., Xia, S., & Xu, K. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2508.20613 · Cited by 4
“For the image reconstruction, we employ a pre-trained StyleGAN2 [37] as a strong generative prior, and we utilize StyleGAN-XL [49] in heterogeneous scenarios.”
Keïta, M., Hamidouche, W., Eutamene, H. B., Ahmed, A. T., & Hadid, A. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2508.03967 · Cited by 4
Neon: Negative Extrapolation From Self-Training Improves Image Generation
Alemohammad, S., Wang, Z., & Baraniuk, R. G. (2025) · International Conference on Learning Representations · doi:10.48550/arxiv.2510.03597 · Cited by 4
Marrying Autoregressive Transformer and Diffusion with Multi-Reference AutoregressionPreprint
Zhen, D., Qiao, Q., Xu, Z., Tan, Y., Wu, K., Zhang, Z., Liu, S., Yin, S., & Tao, M. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2506.09482 · Cited by 4
Hita: Holistic Tokenizer for Autoregressive Image GenerationPreprint
Zheng, A., Wang, H., Zhao, Y., Deng, W., Wang, T., Zhang, X., & Qi, X. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2507.02358 · Cited by 4
Exploiting Discriminative Codebook Prior for Autoregressive Image GenerationPreprint
Tang, L., Chu, R., Wang, X., Han, Y., Wu, P., He, C., Zhang, Y., Zhang, S., & Jia, J. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2508.10719 · Cited by 4
LSD-3D: Large-Scale 3D Driving Scene Generation with Geometry GroundingPreprint
Ost, J., Ramazzina, A., Joshi, A., Bömer, M., Bijelic, M., & Heide, F. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2508.19204 · Cited by 4
“These approaches – from generative adversarial networks (GANs) [27, 5, 40, 74], likelihood-based methods [45, 66, 89], to more prominently, diffusion models [14, 79, 15, 17, 33, 71, 90] – have been re-Method Table 1: Video and 3D Generation for Driving Scenes .”
FVAR: Visual Autoregressive Modeling via Next Focus PredictionPreprint
Li, X., Wu, C., Sun, Y., Zhou, J., Qu, D., Qu, Y., Bo, W., Yu, H., & Liang, D. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2511.18838 · Cited by 4
VIGIL: Part-Grounded Structured Reasoning for Generalizable Deepfake DetectionPreprint
Li, X., Xu, J., & Chen, J. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2603.21526 · Cited by 4
“Fake images are generated by unseen models within related paradigm families, including ADM [49], BigGAN [50], SDXL [51], and StyleGANXL [52].”
Frequency-Aware Flow Matching for High-Quality Image GenerationPreprint
Ren, S., Yu, Q., He, J., Shen, X., Yuille, A., & Chen, L.-C. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2604.15521 · Cited by 4
IncreFA: Breaking the Static Wall of Generative Model AttributionPreprint
Qin, H., Chang, D., Gao, Y., Tan, Y., Chen, L., & Ma, Z. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2604.17736 · Cited by 4
“The IABench dataset includes 4 categories of GANs [22, 23, 30, 46], 2 categories of autoregressive models [13, 37], and 22 categories of diffusion models [8, 11, 12, 20, 24, 26, 27, 32, 34, 36, 39, 44].”
Aligning Few-Step Generative Models by Amortizing Sample-based Variational InferencePreprint
Lee, J., Kang, H., Kim, D., Sim, K. S., Shin, W., Kim, M., Yun, T., Lee, J., Choi, S., Lee, T. E., et al. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2605.26552 · Cited by 4
“In image generator alignment, we fine-tune few-step generators at two scales: at ImageNet-256 [16], we align StyleGAN-XL [82], drifting model [17], inductive moment matching [113], and iMeanFlow [30] to optimize an aesthetic reward [83].”
Distilling Drifting Transformers with Representation AutoencodersPreprint
Zhang, J., Xia, M., Li, G., & Gu, Y. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2606.15553 · Cited by 4
One Step Diffusion via Shortcut Models
Frans, K., Hafner, D., Levine, S., & Abbeel, P. (2024) · arXiv · doi:10.48550/arxiv.2410.12557 · Cited by 3
“A A PPENDIX FID Sampling Steps Param Count Epochs Trained Adversarial approaches BigGAN-deep (Brock, 2018) 6.96 1 – – StyleGAN-XL (Sauer et al., 2022) 2.3 1 – – Denoising approaches DiT-XL (Peebles & Xie, 2023) 2.27 500 675M 640 ADM-G (Dhariwal & Nichol, 2021)”
InfluentialWildFake: A Large-scale Challenging Dataset for AI-Generated Images Detection
Hong, Y., & Zhang, J. (2024) · arXiv · doi:10.48550/arxiv.2402.11843 · Cited by 3
“Raging from Generative Adversarial Networks (GANs) [3, 8, 12, 19, 36, 38, 53, 63, 68, 69] to DMs (DMs) [59, 62], a multitude of users can now effortlessly produce images of high quality and diversity using varied models with personalized weights, and spread those generated images across various…”
BackgroundAttribute Diffusion: Diffusion Driven Diverse Attribute Editing
Parihar, R., Balaji, P., Magazine, R., Vora, S., Jampani, V., & Babu, R. V. (2025) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv61041.2025.00366 · Cited by 3
“Recent advancements in deep generative models [8, 24, 45, 47] have unlocked multiple image editing and synthesis applications.”
Jasim, R. N., Al-Qaraghuli, A. S. R., & Nasar, Z. (2025) · Civil Engineering and Architecture · doi:10.13189/cea.2025.130308 · Cited by 3
FLARE: Federated Learning Attack via Robust Expectation-Based Backdooring Using GAN
Faraoun, H., Bellafqira, R., Coatrieux, G., & Kallas, K. (2025) · International Conference on Emerging Technologies and Computing (IC_ETC) · doi:10.1109/ic_etc65981.2025.11141092 · Cited by 3
“We selected StyleGAN-XL [21] pretrained on CIFAR10, and a simpler DCGAN [22] model trained on MNIST, as the inference-based block to validate the first stage of our strategy.”
Soft Tail-Dropping for Adaptive Visual Tokenization
Chen, Z., Zhang, K., Tu, Z., & Xiong, Y. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37553-7_24 · Cited by 3
Ardebili, S., Kargar, M., & Hoseini, N. S. (2026) · Iran Journal of Computer Science · doi:10.1007/s42044-025-00369-y · Cited by 4
Efficient Dataset Distillation via Generative Pruning
Ma, Y., Li, M., Duan, G., Qin, K., Liang, S., & Zhang, D. (2026) · IEEE Transactions on Big Data · doi:10.1109/tbdata.2026.3668524 · Cited by 3
“Recent efforts have explored compressing large-scale GANs, particularly the highly successful StyleGAN [19], [20], [21], [22], to reduce computation costs while preserving generation quality.”
Enhancing Scene Transition Awareness in Video Generation via Post-Training
Shen, H., Lu, J., Cao, Y., & Yang, X. (2025) · International Joint Conference on Natural Language Processing and Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (IJCNLP-AACL) · doi:10.18653/v1/2025.findings-ijcnlp.41 · Cited by 3
Chen, X., Vaxman, Y., Baruch, E. B., Asulin, D., Moreshet, A., Sra, M., & Sen, P. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-72655-2_2 · Cited by 3
A Study in Dataset Distillation for Image Super-Resolution
Dietz, T., Moser, B. B., Nauen, T. C., Raue, F., Frolov, S., & Dengel, A. R. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-31335-5_5 · Cited by 3
Fan, W., Zhang, M., Li, H., Jiang, W., Chen, H., Yue, X., Michael Backes, A., & Zhang, X. (2025) · ACM SIGSAC Conference on Computer and Communications Security (CCS) · doi:10.1145/3719027.3744867 · Cited by 3
Demystifying Randomly Initialized Networks for Evaluating Generative Models
Lee, J., Kim, J., & Lee, J. (2023) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v37i7.26022 · Cited by 3
Controlling the Output of a Generative Model by Latent Feature Vector Shifting
Belanec, R., Lacko, P., & Malinovská, K. (2023) · World Symposium on Digital Intelligence for Systems and Machines (DISA) · doi:10.1109/disa59116.2023.10308936 · Cited by 3
“Following this line of research we have decided to continue using StyleGAN3 instead of the current state-of-the-art GAN model StyleGAN-XL [9].”
BackgroundMethodKatsumata, K., Vo, D. M., & Nakayama, H. (2024) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv57701.2024.00487 · Cited by 3
“The impressive success of conditional deep generative models [12,24,37,45,48,49,62] has been largely aided by a large amount of well-collected, balanced, and diverse data, typically consisting of not only a large number of images in total but also a certain number of images in each class.”
BackgroundPolynomial Implicit Neural Framework for Promoting Shape Awareness in Generative Models
Nath, U., Singh, R., Shukla, A., Kulkarni, K., & Turaga, P. K. (2024) · International Journal of Computer Vision · doi:10.1007/s11263-024-02270-w · Cited by 3
“We conduct our experiments on ImageNet-100 Tian et al. (2020) at 32 × 32, 64 × 64 and 128 × 128 resolution.”
InfluentialEnriching Operational High-Resolution Ensemble Forecasts with StyleGAN-2
Brochet, C., Moldovan, G. I., Rabault, J., Regan, C., & Raynaud, L. (2025) · Artificial Intelligence for the Earth Systems · doi:10.1175/aies-d-24-0058.1 · Cited by 3
FLODA: Harnessing Vision-Language Models for Deepfake Assessment
Park, S., Bae, Y.-H., Han, G.-H., & Olson, A. W. (2025) · International Conference on Consumer Electronics · doi:10.1109/icce63647.2025.10929816 · Cited by 3
“For testing, we employed 16 datasets, including 2 real (MS COCO [40], Flickr30k [42]) and 14 fake image datasets generated by various models (e.g., SD2 [21], SDXL [43], Deep-Floyd IF [44], DALLE-2 [45], SGXL [46]).”
Yu, H., Mallick, R., Betke, M., & Bargal, S. A. (2026) · IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW) · doi:10.1109/wacvw68408.2026.00100 · Cited by 3
“Recent advances in generative models ( e.g. , Generative Ad-versarial Networks [20, 55], Diffusion Models [24, 52]) have enabled the creation and use of synthetic face data for face-related computer-vision tasks such as face recognition, landmark localization, and face parsing [3, 6, 30, 36, 50,…”
REPRESENTATIVE GUIDANCE:DIFFUSION SAMPLING WITH COHERENCE
Dinh, A. D., Liu, D., & Xu, C. (2025) · UWA Profiles and Research Repository (UWA) · Cited by 3
QC-StyleGAN - Quality Controllable Image Generation and Manipulation
Nguyen, D. V. T., The, P. T., Dinh, T. M., Pham, C., & Tran, A. (2022) · neural information processing systems · doi:10.52202/068431-1369 · Cited by 3
Fundamentals and Challenges of Generative Adversarial Networks for Image-based Applications
de Souza, V. L. T., Marques, B. A. D., & Gois, J. P. (2022) · SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) · doi:10.1109/sibgrapi55357.2022.9991776 · Cited by 3
“Works such as Denoising Diffusion GAN [39] and StyleGAN-XL [40] aim to tackle all three requirements.”
BackgroundAssessing Neural Network Robustness via Adversarial Pivotal Tuning
Christensen, P. E., Snæbjarnarson, V., Dittadi, A., Belongie, S., & Benaim, S. (2024) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv57701.2024.00293 · Cited by 3
Zhao, G., Magoulianitis, V., You, S. O., & Kuo, C. ‐. J. (2024) · IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW) · doi:10.1109/wacvw60836.2024.00054 · Cited by 3
“Our experiments span across four diverse datasets to validate the robustness and generalizability of our method: CIFAR-10 [19] (32x32 resolution images), LSUN-Church, LSUN-Bedroom [46] (both with 256x256 resolution images), and LSUN-10 [8] (96x96 resolution images).”
MethodDive into Big Model TrainingPreprint
Liu, Q., & Jiang, Y. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2207.11912 · Cited by 3
“By scaling up the model size, big models benefit from significant performance improvements across a wide range of machine learning problems [11, 12, 3, 4].”
Background
Metadata
- Authors
- Sauer, A., Schwarz, K., & Geiger, A.
- Affiliations
- Max Planck Institute for Intelligent Systems (DE)
- Venue
- Special Interest Group on Computer Graphics and Interactive Techniques Conference Proceedings
- Publisher
- ACM
- Published
- 2022-08-07
- Pages
- 1-10
- OpenAlex
- W4285981784
- Cited by
- 353
- References
- 19
- Citations by year
- 2026: 42 · 2025: 84 · 2024: 122 · 2023: 86 · 2022: 19
- FWCI
- 28.20
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- ACM SIGGRAPH Conference (SIGGRAPH) (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Generative Adversarial Networks and Image Synthesis
- Physical Sciences › Computer Science › Computer Graphics and Computer-Aided Design › Computer Graphics and Visualization Techniques
- Physical Sciences › Computer Science › Computer Graphics and Computer-Aided Design › 3D Shape Modeling and Analysis
- Keywords
- StyleGAN3, StyleGAN, ImageNet, image inversion, image editing, high-resolution image generation, generative adversarial networks
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.