multimodal models

"The article's context provides the title, which directly indicates the core research question: the performance of LMMs in visual species recognition."

"MM-CoT is a benchmark for probing visual chain-of-thought reasoning in Multimodal Models."