CoDe applies blockwise best-of-N sampling during diffusion denoising, with Tweedie-based reward estimates, to align generated images to differentiable or non-differentiable rewards.
Adaptively-realistic image generation from stroke and sketch with diffusion model
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
CoDe: Blockwise Control for Denoising Diffusion Models
CoDe applies blockwise best-of-N sampling during diffusion denoising, with Tweedie-based reward estimates, to align generated images to differentiable or non-differentiable rewards.