The rapid progress in the field of large multimodal foundation models, especially vision-language...