Mono-InternVL: Pushing the Boundaries of Monolithic Multimodal Large Language Models with Endogenous Visual Pre-training Paper • 2410.08202 • Published 28 days ago • 1