1 paper
Youguang Xing, Xu Luo, Junlin Xie +3
Generalist robot policies trained on large-scale datasets such as Open X-Embodiment (OXE) demonstrate strong performance across a wide range of tasks. However, they often struggle…