Fetching the paper…

WenLan: Bridging Vision and Language by Large-Scale Multi-Modal Pre-Training · Around