A deep dive into spectral analysis of diffusion models of images, revealing how they implicitly perform a form of autoregression in the frequency domain.| Sander Dieleman
The noise schedule is a key design parameter for diffusion models. Unfortunately it is a superfluous abstraction that entangles several different model aspects. Do we really need it?| Sander Dieleman
Thoughts on the tension between iterative refinement as the thing that makes diffusion models work, and our continual attempts to make it _less_ iterative.| Sander Dieleman
A summary of my current thoughts on typicality, and its relevance to likelihood-based generative models.| Sander Dieleman
By Danielle Belgrave, Angela Fan, Ulrich Paquet, Jakub Tomczak, Cheng Zhang| NeurIPS Blog
Imagen 3 is our highest quality text-to-image model, capable of generating images with even better detail, richer lighting and fewer distracting artifacts than our previous models.| Google DeepMind
Delivering high-quality, photorealistic outputs that are closely aligned and consistent with the user’s prompt.| Google DeepMind