Mu Xu

For the first time, a scaling law for quadruped motion tracking reveals that performance consistently improves with larger training datasets, unlocking new capabilities in robotic locomotion.

Agents can now explore environments more efficiently by thinking like humans, prioritizing key landmarks and semantic information during online memory construction.

Achieve stable, controllable, and semantically consistent long-form video generation by decoupling local dynamics from global semantic anchors.

Ditch discrete waypoints: VLA models can now generate smooth, physically plausible robot trajectories by directly regressing continuous action functions.

Autonomous driving gets a human-like reasoning boost: MindDriver uses progressive multimodal reasoning to bridge the gap between semantic understanding and physical trajectory planning.

Forget task-specific architectures: a single Vision-Language-Action foundation model, ABot-N0, now dominates embodied navigation across five distinct tasks.

By learning to project actions onto a low-dimensional manifold, ABot-M0 achieves faster and more stable robotic control policies compared to directly predicting actions in the full high-dimensional space.