Groma: Localized Visual Tokenization for Grounding Multimodal Large Language Models Paper • 2404.13013 • Published Apr 19, 2024 • 30
