AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

SenCache: Accelerating Diffusion Model Inference via Sensitivity-Aware Caching

Yasaman Haghighi, Alexandre Alahi

CVPR 2026 · Image generation › Fast diffusion sampling · approach: Training-free diffusion acceleration · 7 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 sensitivity-based criterion instead of MagCache residual-magnitude criterion (aggressive regime)caching for Wan 2.1 · Wan 2.1 (VBench, fast)LPIPS: 0.0603 -> 0.0540 (same NFE 21)Improved
use sensitivity-based criterion instead of MagCache (conservative regime)caching for Wan 2.1 · Wan 2.1 (VBench, slow)LPIPS: 0.0390 -> 0.0390No clear effect
use sensitivity criterion instead of TeaCachecaching for Wan 2.1 slow · Wan 2.1 (VBench, slow)NFE: 33 -> 25Improved
use sensitivity criterion instead of MagCachecaching for CogVideoX · CogVideoX (VBench)LPIPS: 0.1952 -> 0.1901Improved
use sensitivity criterion instead of MagCachecaching for LTX-Video · LTX-Video (VBench)LPIPS: 0.1795 -> 0.1625Improved
increase max consecutive cache steps n from 1 to 4 (ablation)SenCache on Wan 2.1 · Wan 2.1 (ablation n)NFE: 32 -> 23Improved
increase n beyond 4 (4 to 7) (ablation)SenCache on Wan 2.1 · Wan 2.1 (ablation n)NFE: 23 -> 23No clear effect
increase n from 4 to 7 (ablation)SenCache on Wan 2.1 · Wan 2.1 (ablation n)LPIPS: 0.0558 -> 0.0760Hurt
increase tolerance epsilon from 0.04 to 0.13 (ablation)SenCache on Wan 2.1 · Wan 2.1 (ablation eps)NFE: 25 -> 21Improved
increase tolerance epsilon from 0.04 to 0.13 (ablation)SenCache on Wan 2.1 · Wan 2.1 (ablation eps)LPIPS: 0.0455 -> 0.0513Hurt
calibration set of 8 videos instead of 4096 (ablation)SenCache sensitivity estimation · sensitivity profileprofile matchNo clear effect

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