About co-attention problem

[ChengShiest]

Hi, thanks for your amazing work.

We notice that in your newly released version, you use co-attention from MCAN to achieve a much higher performance, I wonder that will this part of code be released?

Thanks!

[CurryYuan]

Hi @sH1cHEnG ,

You can refer to https://github.com/MILVLG/mcan-vqa for the implementaion. We may release the code in the future.

Best,

Zhihao

[longmalongma]

Hi @sH1cHEnG ,

You can refer to https://github.com/MILVLG/mcan-vqa for the implementaion. We may release the code in the future.

Best,

Zhihao

@CurryYuan 非常感谢您的伟大的工作，我想请教一下您用的MCAN里面的Co-Attention对三个打分进行加权求和，而且比将三个打分进行逐元素相加效果好，请问您是用的MCAN里面的哪一部分对得到的最后三个打分进行加权求和的呢？是用的Stacking,encoder-decoder，还是attention-reduce？期待您的回复，谢谢！