Search papers, labs, and topics across Lattice.
Affiliation:
6
0
7
Robotic manipulation and scene prediction are unified in a single 3D-native world action model that directly reasons about metric space, achieving SOTA performance across diverse tasks.
Local contact reasoning, not global scene modeling, unlocks 30%+ accuracy gains in rigid-body physics simulation.
A unified benchmark that combines synthetic data with real-world testing could revolutionize how we train robots for complex manipulation tasks.
Vid2WAM achieves superior task generalization and data efficiency by leveraging video diffusion priors, even with minimal expert demonstrations.
A new 3D representation and BEV alignment framework drastically enhance the generalization of robotic manipulation policies across varied environments and robot types.
