Search papers, labs, and topics across Lattice.
University of California, Los Angeles, 2 University of California, San Diego
5
0
3
15
Tactile feedback integration boosts robot manipulation success rates by over 40% in contact-rich tasks, redefining the limits of vision-language-action models.
Robot actions can serve as powerful geometric supervision, enabling sparse 3D representations that are both reusable and effective across diverse manipulation tasks.
GRAIL achieves an impressive 84% success rate in real-world object pick-up tasks using only synthetic data, revolutionizing humanoid robot training.
Humanoid robots can now perform complex loco-manipulation tasks with more natural and stable movements by decomposing control into VLM-orchestrated expert policies trained with human motion priors.
Origami, the "Hello, World!" of physical intelligence, is now tractable: Learn2Fold uses LLMs and graph-structured world models to generate valid folding sequences from text.
