多模态预训练vision-language pre-training (VLP)

🍉 vision-language pre-training (VLP)

🍉🍉 Paired vision-language pre-training 对齐的视觉-文本预训练

🍉🍉 Unpaired vision-language pre-training 非对齐的视觉-文本预训练

NAACL_2021_Unsupervised vision-and-language pretraining without parallel images and captions

ICML_2022_VLMixer: Unpaired Vision-Language Pre-training via Cross-Modal CutMix