Search papers, labs, and topics across Lattice.
Affiliation:
6
0
6
6
Facet-0 achieves a staggering 82% success rate in contact-rich robotic assembly, showcasing a leap in precision and robustness over existing methods.
EMERGE-Policy reveals that a robot's intelligence can thrive through a collaborative network of specialized agents rather than a monolithic policy.
Where you measure entropy can drastically change the learned policy geometry, with implications for how we design continuous control systems.
SCORE achieves a remarkable 53.23% Top-1 accuracy in EEG-to-image retrieval, outperforming existing methods by over 17 percentage points, even without target labels.
Dynamic affordance trajectories can significantly improve robotic manipulation success rates, achieving 70% on challenging benchmarks.
DockAnywhere lets you train mobile manipulation policies that generalize to new viewpoints from a single demonstration, sidestepping the costly data collection typically needed for robust visuomotor control.
