MiniGPT-v2: large language model as a unified interface for vision-language multi-task learningJun Chen, Mohamed Elhoseiny, Raghuraman Krishnamoorthi et al.|arXiv (Cornell University)|2023Cited by 66