RobertLuo1/NeurIPS2023_SOC

Issues

• why not use query's padding_mask in vision language early-fusion

  #7 opened 9 months ago by tmliang
  2

• [MMF] Why implement t->v and v->t with video features from different layer?

  #6 opened 9 months ago by wjun0830
  2

• A2D dataset trian for scratch

  #5 opened 10 months ago by wangyihanwyh
  1

• An consultation on Deformable Transformer Block!!!

  #4 opened 10 months ago by Starboy-at-earth
  2

• batch size problem

  #2 opened a year ago by QiRenn
  1

• About class prediction

  #3 opened a year ago by Choi58
  2

• cannot get access to model checkpoints

  #1 opened a year ago by Pierre0089
  1