Uncertainty-Aware Alignment Network for Cross-Domain Video-Text Retrieval

Neural Information Processing Systems 

Adaptation Video-text Retrieval (UDA VR), assuming that training (source) data and testing (target) data are from different domains.