AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

WorldCache: Accelerating World Models for Free via Heterogeneous Token Caching

Weilun Feng, Guoxin Fan, Haotong Qin, Mingqiang Wu, Yuqi Li, Xiangqi Li, Zhulin An, Libo Huang, Dingrui Wang, Longlong Liao, Michele Magno, Yongjun Xu and 1 more

ICML 2026 · Image generation › Fast diffusion sampling · approach: Training-free diffusion acceleration · 0 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 caching with heterogeneous token caching (WorldCache)diffusion world models · Voyager-13B image-to-worldPSNR vs no-cache: 21.76 (EasyCache) -> 23.49Improved
apply WorldCache to Voyager-13BHunyuanVoyager-13B · WorldScoreWorldScore: 46.40 -> 45.43No clear effect
apply WorldCache to Voyager-13BHunyuanVoyager-13B · end-to-end latencyspeedup: 1054s -> 289s, 3.65xImproved
use layer-wise caching baselines (ToCa, DuCa, TaylorSeer, HiCache)Voyager-13B · memorymemory overhead: > 100GBHurt
replace uniform reuse, linear or damped prediction by curvature-based grouping (CHTP) (ablation)WorldCache · Voyager/AetherPSNR, SSIM, LPIPSImproved
use random token grouping instead of curvature grouping (ablation)WorldCache predictor · token prediction ablationfidelityHurt
vary grouping percentiles over nine {ps,pc} pairs (ablation)WorldCache · VoyagerPSNR: 22.77 to 23.52No clear effect
use curvature-only triggering for skipping (ablation)adaptive skipping · skipping strategy ablationfidelityHurt
use CAS instead of global difference or norm guided or fixed-interval skipping (ablation)WorldCache · skipping strategy ablationPSNR, SSIM, LPIPSImproved
increase CAS threshold (ablation)CAS on Aether · adaptive skipping thresholdfidelity vs latencyMixed
use CAS instead of TeaCache skipping (ablation)WorldCache · Aetherquality and speedImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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