Diffusion model generalization is explained by the covariance of the noisy score-matching target, which adds state-dependent noise to sampling and fills gaps in the training distribution.
Align your latents: High-resolution video synthesis with latent diffusion models
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Generalization through variance: how noise shapes inductive biases in diffusion models
Diffusion model generalization is explained by the covariance of the noisy score-matching target, which adds state-dependent noise to sampling and fills gaps in the training distribution.