Abstract: Vision-Language Models (VLMs) bring powerful understanding and reasoning capabilities to multimodal tasks. Meanwhile, the great need for capable aritificial intelligence on mobile devices ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback