AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

ERTACache: Error Rectification and Timesteps Adjustment for Efficient Diffusion

Xurui Peng, Chenqian Yan, Hong Liu, Rui Ma, Fangmin Chen, XING WANG, Zhihua Wu, Songwei Liu, Mingbao Lin

ICLR 2026 · Image generation › Fast diffusion sampling · approach: Training-free diffusion acceleration · 10 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
use offline-calibrated cache policy instead of uniform cache (ablation)ERTACache · Wan2.1-1.3BPSNR: 13.7640 -> 22.0920Improved
use offline-calibrated cache policy instead of uniform cache (ablation)ERTACache · Flux-dev 1.0PSNR: 15.9428 -> 19.4069Improved
add trajectory-aware timestep adjustment (ablation)ERTACache offline policy · Wan2.1-1.3BPSNR: 22.0920 -> 22.9413Improved
add trajectory-aware timestep adjustment (ablation)ERTACache offline policy · Flux-dev 1.0PSNR: 19.4069 -> 20.3586Improved
add explicit error rectification (ablation)ERTACache with time adjustment · Wan2.1-1.3BPSNR: 22.9413 -> 23.7680Improved
add explicit error rectification (ablation)ERTACache with time adjustment · Wan2.1-1.3BVBench (%): 80.89% -> 80.73%Hurt
add explicit error rectification (ablation)ERTACache with time adjustment · Flux-dev 1.0PSNR: 20.3586 -> 20.5146Improved
ERTACache instead of TeaCache at similar speedupCogVideoX fast · CogVideoXPSNR: 20.97 -> 26.44Improved
ERTACache instead of TeaCacheWan2.1-1.3B · Wan2.1-1.3BVBench (%): 76.04% -> 80.73%Improved
ERTACache-slow instead of full-step Open-Sora 1.2Open-Sora 1.2 · Open-Sora 1.2VBench (%): 79.22% -> 79.36%No clear effect
ERTACache instead of FasterCache at same speedupCogVideoX slow · CogVideoXVBench (%): 79.83% -> 79.30%Hurt

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