AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Leveraging Early-Stage Robustness in Diffusion Models for Efficient and High-Quality Image Synthesis

Yulhwa Kim, Dongwon Jo, Hyesung Jeon, Taesu Kim, Daehyun Ahn, Hyungjun Kim, jae-joon kim

NeurIPS 2023 · Image generation › Fast diffusion sampling · approach: Training-free diffusion acceleration · 4 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
apply 4-bit activation quantization only in early steps 1.0T:0.8T instead of late steps 0.2T:0.0T (ablation)LDM-8 W4 activation quantization · LSUN-Churches 256x256 unconditional (5k images)FID: 34.30 -> 6.10Improved
use step-wise activation bits (4b early to 8b late, effective 6) instead of uniform 6-bit (ablation)RAQ LDM-8 · LSUN-Churches 256x256 unconditionalFID: 5.31 -> 4.64Improved
use step-wise activation bits (effective 6) instead of uniform 6-bit (ablation)RAQ LDM-4 · LSUN-Bedrooms 256x256 unconditionalFID: 5.58 -> 3.99Improved
replace Q-diffusion W4/A8 with RAQ at effective 6-bit activationsLDM-8 · LSUN-Churches 256x256 unconditionalTBOPs: 148.6 -> 108.8Improved
replace Q-diffusion W4/A8 with RAQ at effective 6-bit activationsLDM-4 · LSUN-Bedrooms 256x256 unconditionalTBOPs: 681.2 -> 504.8Improved
replace Q-diffusion W4/A8 with RAQ at effective 6-bit activationsLDM-8 · LSUN-Churches 256x256 unconditionalFID: 4.45 -> 4.64No clear effect
replace Q-diffusion W4/A8 with RAQ at effective 6-bit activationsLDM-4 · LSUN-Bedrooms 256x256 unconditionalFID: 4.17 -> 3.99Improved
lower effective activation bits further to 5.60 / 5.35 (relaxed FID) (ablation)RAQ LDM-8 · LSUN-Churches 256x256 unconditionalFID: 4.64 -> 5.12 -> 5.51Hurt
lower effective activation bits further to 5.50 / 5.45 (ablation)RAQ LDM-4 · LSUN-Bedrooms 256x256 unconditionalFID: 3.99 -> 4.94 -> 5.32Hurt
lower effective activation bits further to 5.60 / 5.35 (ablation)RAQ LDM-8 · LSUN-Churches 256x256 unconditionalTBOPs: 108.8 -> 101.3 -> 96.4Improved
reduce weight bits down to 4 (weight-only quantization) (ablation)LDM-8 BRECQ · LSUN-Churches 256x256 unconditionalFIDNo clear effect
reduce activation bits below 8 (ablation)LDM-8 · LSUN-Churches 256x256 unconditionalFIDHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy