Vision language models (VLMs) exhibit vast knowledge of the physical world, including intuition of physical and spatial properties, affordances, and motion. With fine-tuning, VLMs can also natively ...
Results that may be inaccessible to you are currently showing.
Hide inaccessible results