| Projected GANs Converge Faster | ✓ Link | 1.59 | | | | | Projected GAN | 2021-11-01 |
| Ensembling Off-the-shelf Models for GAN Training | ✓ Link | 1.72 | 1.72 ± 0.01 | | | | Vision-aided GAN | 2021-12-16 |
| Diffusion-GAN: Training GANs with Diffusion | ✓ Link | 1.85 | | | | | Diffusion ProjectedGAN | 2022-06-05 |
| []() | | 2.66 | | | | | Patch Diffusion | |
| Image Generators with Conditionally-Independent Pixel Synthesis | ✓ Link | 2.92 | | | | | CIPS | 2020-11-27 |
| StyleSwin: Transformer-based GAN for High-resolution Image Generation | ✓ Link | 2.95 | | | | | StyleSwin | 2021-12-20 |
| Diffusion-GAN: Training GANs with Diffusion | ✓ Link | 3.17 | | | | | Diffusion StyleGAN2 | 2022-06-05 |
| StyleNAT: Giving Each Head a New Perspective | ✓ Link | 3.4 | | | | | StyleNAT | 2022-11-10 |
| CLR-GAN: Improving GANs Stability and Quality via Consistent Latent Representation and Reconstruction | ✓ Link | 3.43 | | | 0.48 | | CLR-GAN | 2024-09-30 |
| Analyzing and Improving the Image Quality of StyleGAN | ✓ Link | 3.86 | 4.28 ± 0.03 | | | | StyleGAN2 | 2019-12-03 |
| Polarity Sampling: Quality and Diversity Control of Pre-Trained Generative Networks via Singular Values | ✓ Link | 3.92 | | | | | Polarity-StyleGAN2 | 2022-03-03 |
| Truncated Diffusion Probabilistic Models and Diffusion-based Adversarial Auto-Encoders | ✓ Link | 3.98 | | 100 | | | TDPM+ (TTrunc=99) | 2022-02-19 |
| Adversarial Generation of Continuous Images | ✓ Link | 4.04 | | | | | INR-GAN-bil | 2020-11-24 |
| Unleashing Transformers: Parallel Token Prediction with Discrete Absorbing Diffusion for Fast High-Resolution Image Generation from Vector-Quantized Codes | ✓ Link | 4.07 | | | | | Unleashing Transformers | 2021-11-24 |
| A Style-Based Generator Architecture for Generative Adversarial Networks | ✓ Link | 4.21 | 4.75 ± 0.01 | | | | StyleGAN | 2018-12-12 |
| SWAGAN: A Style-based Wavelet-driven Generative Model | ✓ Link | 4.97 | | | | | SWAGAN-Bi | 2021-02-11 |
| Wavelet Diffusion Models are fast and scalable Image Generators | ✓ Link | 5.06 | | 4 | 0.40 | | WaveDiff | 2022-11-29 |
| MSG-GAN: Multi-Scale Gradients for Generative Adversarial Networks | ✓ Link | 5.2 | 5.38 ± 0.03 | | | | MSG-StyleGAN | 2019-03-14 |
| Tackling the Generative Learning Trilemma with Denoising Diffusion GANs | ✓ Link | 5.25 | | | | | DDGAN | 2021-12-15 |
| Flow Matching in Latent Space | ✓ Link | 5.54 | | | | | LFM | 2023-07-17 |
| Progressive Growing of GANs for Improved Quality, Stability, and Variation | ✓ Link | 6.42 | 6.43 ± 0.05 | | | | PGGAN | 2017-10-27 |
| Denoising Diffusion Probabilistic Models | ✓ Link | 7.89 | | | | | Denoising Diffusion Probabilistic Model | 2020-06-19 |
| Pseudo Numerical Methods for Diffusion Models on Manifolds | ✓ Link | 8.69 | | | | | PNDM | 2022-02-20 |
| TransGAN: Two Pure Transformers Can Make One Strong GAN, and That Can Scale Up | ✓ Link | 8.94 | | | | | TransGAN | 2021-02-14 |
| Gotta Go Fast When Generating Data with Score-Based Models | ✓ Link | 26.46 | | | | | VE (erel=0.01) | 2021-05-28 |
| Gotta Go Fast When Generating Data with Score-Based Models | ✓ Link | 26.46 | | | | | VE (erel=0.02) | 2021-05-28 |
| Bellman Optimal Stepsize Straightening of Flow-Matching Models | ✓ Link | | | | | 13.21 | BOSS | 2023-12-27 |