Pretrained Vision-Language-Action Models are Surprisingly Resistant to Forgetting in Continual Learning

返回详情 VLA / Vision-Language-Action 每日论文卡