1 paper · 1 filter
Shivin Dass, Alaa Khaddaj, Logan Engstrom +3
Recently, the robotics community has amassed ever larger and more diverse datasets to train generalist policies. However, while these policies achieve strong mean performance acros…