AI RESEARCH
NEWTON: Agentic Planning for Physically Grounded Video Generation
arXiv CS.CV
•
ArXi:2605.18396v1 Announce Type: new Video generation models produce visually compelling results but systematically violate physical commonsense -- on VideoPhy-2, the best model achieves only 32.6% joint accuracy. We identify a specification bottleneck: text prompts are lossy compression of the physical world, omitting the parameters that fully determine dynamics, and no amount of model scaling can recover what was never specified.