Tong Zhao

The paper proposes VERA, a decoupled policy that uses an action-free video world model combined with an embodiment-specific Inverse Dynamics Model (IDM) to achieve generalizable, zero-shot robot control across different hardware.

The paper introduces extsc{Ptah}, a multi-agent harness designed to improve verifiable multimodal deep research by orchestrating the entire report generation process, ensuring factual grounding and visual consistency.

This paper proposes a novel mode-selectable pinching antenna-assisted integrated sensing and communication framework to optimize waveguide mode selection, transmit beamforming, and transmit/receive PA positions for maximum sensing signal-to-noise ratio and multi-user quality-of-service.