1 paper · 1 filter
Zhuoling Li, Liangliang Ren, Jinrong Yang +5
The effectiveness of scaling up training data in robotic manipulation is still limited. A primary challenge in manipulation is the tasks are diverse, and the trained policy would b…