Unifying Cross-Lingual and Cross-Modal Modeling Towards Weakly Supervised Multilingual Vision-Language Pre-training

38535

Knowledge Triples

Source Entity	Relationship	Target Entity
Multilingual Vision Language Pre training Task	addresses	Weakly Supervised Multilingual VLP Method
Multilingual Vision Language Pre training Task	learns	Universal Multilingual Multimodal Representations OtherEntity
Unified Cross Lingual Cross Modal Learning Method	guides	Multi Modal Tasks Task
Weakly Supervised Multilingual VLP Method	supports	CrossLingualTransfer Method
Weakly Supervised Multilingual VLP Method	uses	MultilingualTextCorpora Material