Jul 2026
VIPER: Visual In-Context Physics Reasoning for Physically Plausible Video Generation
Experiments on an unseen validation set show that VIPER achieves stronger reference-video physical similarity and higher human preference than representative video generation and video-as-prompt baselines, while maintaining competitive general video quality.
Tianxi Chen, Han-Mo Chen, Hua-Jin Chen et al.
· arXiv.org · 0 citations