7 ms·
Considering most SOTA LLMs are also multimodal/vision models, could they get better results if the LLM gets visual feedback with it?
by NullCascade 1y ago
Considering most SOTA LLMs are also multimodal/vision models, could they get better results if the LLM gets visual feedback with it?