Search papers, labs, and topics across Lattice.
5
0
6
9
Achieving 86.4% grasp stability and 83.3% real-world success, SynManDex bridges the gap between human dexterity and robotic manipulation.
The hardest AI tasks remain largely unsolved, with current models achieving only a 2.6% success rate on economically valuable workflows.
Generating realistic human interactions with objects and scenes is now possible even with limited data, thanks to a clever training strategy that combines synthetic and real data.
Forget painstakingly creating 3D assets for robot training - ManiTwin automates the process, turning single images into simulation-ready objects at scale.
Finally, a robot can reliably pick out that specific shirt from a messy pile, thanks to a new vision-language pipeline that reasons about garment affordances.
