Rethinking What to Cache in Few-Step Diffusion Transformers: Solver-Aware Target Selection

Paper recorded by Signals 4 on 2026-10-02 in cs.AI. Abstract reproduced from arXiv; link to the original below.

Published 2026-10-02 on arXiv · recorded by Signals 4 on 2026-10-05

Category: cs.AI · 人工智能 · first seen 2026-10-05

Abstract

Diffusion Transformers (DiTs) can generate high-quality images and videos, but generating each sample requires multiple costly DiT forward passes. Two common ways to accelerate DiT sampling are step distillation, which reduces the number of sampling steps, and caching, which skips some DiT evaluations by reusing a tensor computed at an earlier step. Most caching methods decide in advance which ten

Read on arXiv →