Search papers, labs, and topics across Lattice.
5
0
6
0
Evaluator quality for robotic policies hinges more on long-horizon consistency than on short-term visual fidelity, reshaping our approach to world model design.
R2RDreamer achieves spatial generalization improvements in manipulation tasks by leveraging 3D-aware data augmentation without the pitfalls of complex scene setups or sim-to-real gaps.
Treating raw visual images as action representations revolutionizes embodied control, outperforming traditional methods in accuracy and generalization.
VLN agents can now navigate more effectively by reasoning about their own state and task progress, closing the gap between end-to-end VLMs and explicit scene mapping.
Legged robots can now recover from sensor noise and crazy user commands with 10x greater reliability, thanks to a new method that respects the robot's competence boundaries.
