Learn Where Outcomes Diverge: Efficient VLA RL via Probabilistic Chunk Masking
返回详情
VLA / Vision-Language-Action 每日论文卡