Zhiheng Ma

DLAM achieves superior temporal consistency and policy performance by modeling transitions as distributional latent actions, fundamentally changing how we approach action generation in VLA tasks.

Achieving state-of-the-art performance in mobile manipulation hinges on aligning temporal granularity and action space, revealing critical insights into effective world-action modeling.

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