1 paper
Jin Lou, Zhiyuan Jing, Xupeng Wang +21
Large-scale vision--language--action (VLA) policies have advanced generalist robot control, yet most remain stimulus-to-action black boxes: actions are exposed, but their explanato…