Hangjun Ye

Discrete-WAM enables compositional causal reasoning in autonomous driving, outperforming traditional methods that struggle with complex state-action dynamics.

OneVLA unifies navigation and manipulation tasks into a single framework, enabling robots to seamlessly interpret commands and interact with their environments like never before.

Ditch pixel-perfect reconstruction: LVDrive shows that learning future scene representations in a high-level latent space dramatically improves autonomous driving performance.

Robots can now navigate complex outdoor environments using only high-level human instructions and readily available GPS/map data, bypassing the need for expensive HD maps or limited short-horizon policies.

Latent reasoning can beat explicit Chain-of-Thought – but only if you force it to learn causal dynamics via a visual world model, not just language.

Endowing VLMs with intrinsic 3D geometric awareness and physical interaction cues via XEmbodied substantially boosts performance on spatial reasoning and embodied tasks, surpassing existing 2D image-text pretrained models.

Autonomous driving models can now achieve remarkable zero-shot generalization by leveraging the power of large-scale video generation models to jointly predict future actions and visuals.