返回所有成员
SJTU Scalelab
内容创作者
"Slow Science"
Robotic Learning & Reinforcement Learning (RL^2)
RL is all you need when expert data run out
World Model is easier to Scaling than VLA
