Diffusion models have emerged as a powerful new family of deep generative models with
record-breaking performance in many applications, including image synthesis, video …

Deep generative models are a class of techniques that train deep neural networks to model
the distribution of training samples. Research has fragmented into various interconnected …

We present high quality image synthesis results using diffusion probabilistic models, a class
of latent variable models inspired by considerations from nonequilibrium thermodynamics …

Energy-Based Models (EBMs), also known as non-normalized probabilistic models, specify
probability density or mass functions up to an unknown normalizing constant. Unlike most …

Deep learning approaches have produced substantial breakthroughs in fields such as
image classification and natural language processing and are making rapid inroads in the …

Humans are able to rapidly understand scenes by utilizing concepts extracted from prior
experience. Such concepts are diverse, and include global scene descriptors, such as the …

Contrastive divergence is a popular method of training energy-based models, but is known
to have difficulties with training stability. We propose an adaptation to improve contrastive …

The visual world around us can be described as a structured set of objects and their
associated relations. An image of a room may be conjured given only the description of the …

Energy-based models (EBMs) have recently been successful in representing complex
distributions of small images. However, sampling from them requires expensive Markov …

We propose an energy-based model (EBM) in the latent space of a generator model, so that
the EBM serves as a prior model that stands on the top-down network of the generator …