AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Beta-Tuned Timestep Diffusion Model

Tianyi Zheng, Peng-Tao Jiang, Ben Wan, Hao Zhang, Jinwei Chen, Jia Wang, Bo Li

ECCV 2024 · Image generation › Fast diffusion sampling · approach: Diffusion theory and learned samplers · 12 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
replace uniform timestep sampling with Beta(alpha<1, beta=1) samplingDDPM (ADM architecture) · FFHQ 128x128FID: 14.07 -> 10.28Improved
replace uniform timestep sampling with beta samplingDDPM · ImageNet 32x32FID: 3.65 -> 2.87Improved
add beta timestep samplingDDPM · ImageNet 32x32 training speedtraining speedup to equal FID: 4.71xImproved
add beta timestep samplingDDPM · AFHQ-D 256x256 training speedtraining speedup to equal FID: 3.33xImproved
use loss re-weighting methods (Debias, Min-SNR, P2-Weight) on low-resolution CIFARDDPM · CIFAR-10 32x32FID: 3.01 -> 3.17 / 3.23 / 3.19Hurt
add B-TTDM to P2-WeightP2-Weight · AFHQ-D 256x256FID: 16.18 -> 12.45Improved
add B-TTDM to Min-SNRMin-SNR · AFHQ-D 256x256FID: 17.03 -> 13.34Improved
add B-TTDM to DDPM-IPDDPM-IP · CIFAR-10 32x32FID: 2.72 -> 2.46Improved
use B-TTDM with fast samplers (DDIM, DPM-Solver) at 25 NFEDDPM · AFHQ-D 256x256FID: 24.22 -> 21.76 (DDIM, 25 NFE)Improved
use alpha=0.8, beta=1 instead of uniform alpha=1, beta=1 (ablation)B-TTDM · AFHQ-D 256x256FID: 17.21 -> 13.81Improved
lower alpha from 0.8 to 0.7 (ablation)B-TTDM · AFHQ-D 256x256FID: 13.81 -> 14.78Hurt
use beta sampling instead of exponential timestep sampling (ablation)B-TTDM · AFHQ 256x256FID: 15.12 -> 13.81Improved

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