Yifan Xie

Mobile manipulation policies do not need joint action spaces to coordinate effectively: decoupling arm and base streams via masked joint attention and training-time world imagination drives state-of-the-art coordination across 17 distinct embodiments.

Action-conditioned verification can boost success rates in mobile manipulation tasks by over 8% while enhancing timely recall by nearly 30%.

Adaptive routing of perception priors allows PerceptDrive to generate optimal driving trajectories in real-time without complex post-processing.

Achieving a 57.6% success rate on RoboCasa365, Xiaomi-Robotics-1 sets a new standard for vision-language-action models in real-world robotic manipulation.

Achieving over 555 FPS in tactile simulations, TaCauchy delivers unprecedented accuracy in mechanical stress computation for robotics applications.