Search papers, labs, and topics across Lattice.
Shanghai AI Lab
3
0
6
Forget generic video memory caches – Q-GeoMem selectively remembers only the geometrically relevant bits, boosting spatial reasoning performance.
Embodied agents can now collaboratively reason about space and manipulate objects in the real world, thanks to a new reinforcement learning approach that fuses their egocentric viewpoints into a world-centric understanding.
Skip the costly robot teleoperation data: ZeroWBC learns surprisingly natural humanoid control policies directly from human egocentric videos.
