Neural Approximate Mirror Maps for Constrained Diffusion Models
Diffusion models excel at creating visually-convincing images, but they often struggle to
meet subtle constraints inherent in the training data. Such constraints could be physics …
meet subtle constraints inherent in the training data. Such constraints could be physics …
A Generative Model of Symmetry Transformations
Correctly capturing the symmetry transformations of data can lead to efficient models with
strong generalization capabilities, though methods incorporating symmetries often require …
strong generalization capabilities, though methods incorporating symmetries often require …