Search papers, labs, and topics across Lattice.
Corresponding author: muyao@sjtu.edu.cn
2
0
6
3
Forget painstakingly creating 3D assets for robot training - ManiTwin automates the process, turning single images into simulation-ready objects at scale.
A 7B model trained with RL can outperform 72B-scale general MLLMs in robotic manipulation process supervision by explicitly reasoning about progress toward the final task goal.
