arXiv cs.AI· Jiyue Jiang, Zhiyu Huo, Yuhan Chen, Yanruisheng Shao, Gong Zhang, Lei Deng·· 3 天前AI 评分24
AlignCache：面向视频扩散模型加速的校准与纠错缓存框架
Heuristics are Not Enough: Calibrated and Corrected Caching for Video Diffusion Acceleration
AI 导读
AlignCache 是一个闭环缓存框架，通过阶段自适应预算路由、闭环阈值校准和动量增强草稿合成三个组件，将视频扩散模型的缓存调度从静态风险累积转变为反馈驱动的自校正控制器。该方法针对现有自适应缓存依赖启发式差异度量、单一全局阈值且从不校验预算估计的问题，在多个 SOTA 视频生成模型上取得更优的加速-质量权衡，代码已开源。
正文 · 原文
Abstract:Feature caching has become a mainstream training-free strategy for accelerating video diffusion models by reusing intermediate outputs across denoising steps. However, existing adaptive methods rely on heuristic difference measures misaligned with actual generation quality, employ a single global threshold that ignores heterogeneous phase dynamics, and never validate their budget estimates against true approximation error, allowing systematic bias to propagate uncorrected. We propose \textbf{AlignCache}, a closed-loop caching framework that directly aligns scheduling decisions with their true influence on the final generated video through three tightly coupled components: (1) phase-adaptive budget routing that partitions the denoising trajectory into structurally distinct phases, each governed by an independent threshold and a staleness-aware guard; (2) closed-loop threshold calibration that exploits counterfactual error measurement at the classifier-free guided output level to adjust decision boundaries online via an EMA feedback loop; and (3) momentum-augmented draft synthesis combining direction-magnitude gated residual extrapolation with anchor-guided unconditional correction. By transforming scheduling from a static risk accumulator into a feedback-driven, self-correcting controller, AlignCache consistently achieves superior acceleration-quality trade-offs across state-of-the-art video generation models. Our code is available at this https URL.
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.16545 [cs.CV] |
| (or arXiv:2610.16545v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2610.16545 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Lei Deng [view email]
[v1]
Mon, 12 Oct 2026 10:28:47 UTC (2,147 KB)
来源：arXiv cs.AI · arxiv.org