amazon-science/QA-ViT

Issues

• Issues with the reproduction of results

  #8 opened 4 months ago by lky-violet
  1

• qa-vit training steps

  #5 opened 5 months ago by Zyf139
  1

• Question embedding module for Llava-1.5

  #7 opened 5 months ago by xl1990
  1

• Gate_value is all 0 in the ViT-T5-base repleased model

  #6 opened 5 months ago by Zhiyuan-R
  1

• When the training code will be released?

  #4 opened 6 months ago by Zhiyuan-R
  2

• Question about instruct_embeds, image_features, input_embeds

  #3 opened 7 months ago by xiaokj37
  7

• Question about "Employing GradCAM highlights the focus areas"

  #1 opened 7 months ago by xiaokj37
  4

• Can QAVit be applied to multi-turn conversations?

  #2 opened 8 months ago by echo840
  2