Glow (Generative Flow with Invertible 1x1 Convolutions):
Glow is a deep generative model for image synthesis that uses a sequence of invertible transformations, including invertible 1x1 convolutions, affine coupling layers, and squeezing operations, to generate high-quality images with a relatively small number of parameters. It is useful for tasks such as image editing, super-resolution, and inpainting.
Overall, Glow represents a significant advance in the field of generative modeling, and it is likely to be an important tool for researchers and practitioners in the field of computer vision and machine learning.