Search papers, labs, and topics across Lattice.
Peking University
2
0
4
Current action-conditioned world models may ignore commanded actions, but WorldSync significantly enhances their reliability for off-expert trajectories.
Achieving a mean visual order consistency of 0.9872, Robo-Dopamine 2.0 significantly outperforms traditional reward models, demonstrating its superior ability to handle OOD scenarios in robotic manipulation.
