Fetching the paper…

Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts · Around