AI RESEARCH
VDE: Training-Free Accelerating Rectified Flow Model via Velocity Decomposition and Estimation
arXiv CS.CV
•
ArXi:2605.23381v1 Announce Type: new Though rectified flow models have achieved remarkable performance in image, video, and 3D generation, their practical deployments are challenged by slow inference speeds. Prior acceleration methods reuse cached features from previous steps, which neglects the growing mismatch between static caches and the evolving input, leading to reduced output fidelity. This work proposes Velocity Decomposition and Estimation (VDE), a