Multimodal Large Language Models have potential in Image Quality Assessment, with GPT-4V showing promise but room for improvement.
