Adversarial Training for Pixel Diffusion

Adversarial learning can serve as an effective post-training correction for pixel diffusion models, which generate RGB images directly without an autoencoder but systematically underrepresent fine-scale natural-image statistics, according to new research. The work applies adversarial training to correct this deficiency in pixel diffusion outputs.

Pixel diffusion models generate RGB images directly, avoiding the bottleneck of an autoencoder, yet their outputs still systematically underrepresent fine-scale natural-image statistics. We show that adversarial learning provides an effective post-training correction for this deficiency. Starting fr