3 ms·SimVLM: Simple Visual Language Model Pre-Training with Weak Supervision2 points by theafh 5y ago