【ACL 2023获奖论文】主题论文奖:7.Weaker Than You Think: A Critical Look at Weakly Supervised Learning 写在最前面引言正文问题1:对WSL来说,clean data是否必要?问题2:WSL需要多少的clean data?问题3…
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation image-text contrastive learning(ITC)用在单模态,masked language modeling (MLM) and image-text matching (ITM) 用在多模态。 单模态编码器的表示上引入了中间图像…