Repo for paper "CODIS: Benchmarking Context-Dependent Visual Comprehension for Multimodal Large Language Models".
Primary LanguageJavaScriptApache License 2.0Apache-2.0