Multimodal Guidance Network for Missing-Modality Inference in Content Moderation

Zhuokai Zhao, Harish Palani, Tianyi Liu +2

Multimodal deep learning, especially vision-language models, have gained significant traction in recent years, greatly improving performance on many downstream tasks, including con…