Euclid: Supercharging Multimodal LLMs with Synthetic High-Fidelity Visual Descriptions Paper • 2412.08737 • Published 14 days ago • 51
TLDR: Token-Level Detective Reward Model for Large Vision Language Models Paper • 2410.04734 • Published Oct 7 • 16