Jinhua Zhu

In my experiment, these arguments are the same as Fairseq. Can you paste your log?

sorry, I did not save these scripts for these two language pairs. But you refer to [this file](https://github.com/teslacool/preprocess_iwslt/blob/master/preprocess.sh) for details.

yes, I used "share-all-embedding" and this vocabulary size is 10000 (you can see my paper for details). I also notice that you have changed your learning rate. I use the...

Removing it directly doesn't seem quite right. I recommend keeping it and applying the flip operation to both 'cur_img' and 'img_pos_list'.