Skip to main content
  1. PaperReading/
  2. MICCAI/

Masked Vision and Language Pre-training with Unimodal and Multimodal Contrastive Losses for Medical Visual Question Answering

MICCAI 2023

架构

img

  • 在开放问题和Y/N问题上的效果

img

  • 数据集

    • pretrain
      ROCO 80,000
      MedICaT 217,000
      ImageCLEF2022 90,000
      test
      VQA-RAD 315+3064
      SLAKE 14028 7:1.5:1.5
      PathVQA 32799 5:3:2

img

  • ITM&MLM不可或缺

img