Search papers, labs, and topics across Lattice.
3
2
6
2
Explicitly modeling depth in world-action models significantly boosts planning robustness and future prediction quality for autonomous driving.
Achieve ViT efficiency gains without the optimization headaches: ToaSt's decoupled pruning framework delivers better accuracy-FLOPs trade-offs than existing methods by strategically targeting different ViT components.
DrivingGen reveals that current generative driving world models either look good but break physics, or capture motion realistically but lack visual fidelity, exposing a critical trade-off for autonomous driving applications.
