Abstract: Visual language models (VLMs) have made significant advances in accuracy in recent years. However, their efficiency has received much less attention. This paper introduces NVILA, a family of ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results