on: May 27, 2026
in: arXiv

VERA: Turning Video Models into Generalist Robot Policies

Sizhe Lester Li *
Evan Kim *
Xingjian Bai *
Tong Zhao
Tao Pang
Max Simchowitz
Vincent Sitzmann

*: shared first author

@article{li2026turningvideomodelsgeneralist,
    title = { VERA: Turning Video Models into Generalist Robot Policies },
    author = { Li, Sizhe Lester and 
               Kim, Evan and 
               Bai, Xingjian and 
               Zhao, Tong and 
               Pang, Tao and 
               Simchowitz, Max and 
               Sitzmann, Vincent },
    year = { 2026 },
    booktitle = { arXiv },
}

Copy to Clipboard

We pair a 14B video generative model with an embodiment-specific inverse dynamics model to solve a wide range of robotics challenges, from zero-shot pick-and-place on a real-world Panda arm to contact-rich re-orientation of a cube with a 16-DoF multi-fingered hand. We leave the video planner untouched and instead train an embodiment-specific inverse dynamics model (IDM) built on the robot’s Jacobian. The resulting closed-loop policy — VERA — translates dreamed futures into low-level actions faithfully, achieving zero-shot manipulation on a Panda arm and 16-DoF dexterous in-hand reorientation with the same planner. Decoupled video planning plus faithful translation is a viable path to generalizable robot control.