← Back to Benchmarks
simmediumvision-robotmetric · varies

Unsupervised Decomposition and Recombination with Discriminator-Driven Diffusion Models

Description

Decomposing complex data into factorized representations can reveal reusable components and enable synthesizing new samples via component recombination. We investigate this in the context of diffusion-based models that learn factorized latent spaces without factor-level supervision. In images, factors can capture background, illumination, and object attributes; in robotic videos, they can capture reusable motion components. To improve both latent factor discovery and quality of compositional gen

Source

http://arxiv.org/abs/2601.22057v2