Heavy-tailed noise in diffusion models leads to less favorable sampling-error bounds than light-tailed Gaussian noise by making the underlying statistical estimation problem harder.
We summarize their forward paths below
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Do Heavy Tails Help Diffusion? On the Subtle Trade-off Between Initialization and Training
Heavy-tailed noise in diffusion models leads to less favorable sampling-error bounds than light-tailed Gaussian noise by making the underlying statistical estimation problem harder.