Similar Links:

• Diffusion Models Beat GANs on Image Synthesis

• Improved Denoising Diffusion Probabilistic Models

• Analyzing and Improving the Training Dynamics of Diffusion Models

• Exposing flaws of generative model evaluation metrics and their unfair treatment of diffusion models

• Diffusion-GAN: Training GANs with Diffusion

• On Convergence and Stability of GANs

• Do GANs learn the distribution? Some Theory and Empirics

• Seeing What a GAN Cannot Generate

• Large Scale GAN Training for High Fidelity Natural Image Synthesis

• Large Scale Adversarial Representation Learning

• Making Anime With BigGAN

• Deep Double Descent: We show that the double descent phenomenon occurs in CNNs, ResNets, and transformers: performance first improves, then gets worse, and then improves again with increasing model size, data size, or training time

• How AI Training Scales

• not-so-BigGAN: Generating High-Fidelity Images on Small Compute with Wavelet-based Super-Resolution

• The GAN is dead; long live the GAN! R3GAN: A Modern GAN Baseline

• Are GANs Created Equal? A Large-Scale Study

• Stretching Each Dollar: Diffusion Training from Scratch on a Micro-Budget

• Search: GS; Google; site