Closed pritamqu closed 3 months ago
can you please confirm the model sizes/variants of the reported results in Tables 2 and 3?
The sizes of LLaVA and LLaVA 1.5 are 13B. InstructBLIP is 7B. mplug-owl is 7B. Minigpt-4 is 7b. Multimodal-GPT is 7b.
can you please confirm the model sizes/variants of the reported results in Tables 2 and 3?