CacheScope: Exact-Compute Allocation for Training-Free Diffusion Acceleration
arXiv:2610.12530v1 Announce Type: new Abstract: Training-free feature reuse accelerates diffusion transformer inference by replacing full denoising computation at selected steps with cached approximations. Existing methods mainly improve the reconstruction of skipped features, but we find that once…
Read the full story at arXiv cs.LG ↗
Timeline · 1 report
- 2026-10-12 04:00 · arXiv cs.LG
CacheScope: Exact-Compute Allocation for Training-Free Diffusion Acceleration