Codebook-based additive quantization compresses an ImageNet diffusion model to under 2 bits per weight with better reported FID and sFID than the full-precision model.
fvcore: Collection of common code that’s shared among different research projects in fair computer vision team.https://github.com/facebookresearch/ fvcore, 2019
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
other 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
other 1polarities
unclear 1representative citing papers
citing papers explorer
-
AQUATIC-Diff: Additive Quantization for Truly Tiny Compressed Diffusion Models
Codebook-based additive quantization compresses an ImageNet diffusion model to under 2 bits per weight with better reported FID and sFID than the full-precision model.