AI RESEARCH

SHIFT: Motion Alignment in Video Diffusion Models with Adversarial Hybrid Fine-Tuning

arXiv CS.CV

ArXi:2603.17426v1 Announce Type: new Image-conditioned Video diffusion models achieve impressive visual realism but often suffer from weakened motion fidelity, e.g., reduced motion dynamics or degraded long-term temporal coherence, especially after fine-tuning. We study the problem of motion alignment in video diffusion models post-