1 paper · 1 filter
Ting Huang, Dongjian Li, Rui Yang +3
Grounding natural-language instructions into continuous control for quadruped robots remains a fundamental challenge in vision language action. Existing methods struggle to bridge…