Does Visual Grounding Enhance the Understanding of Embodied Knowledge in Large Language Models?

Open in new window