Will the model be attacked by the adversarial examples?

[zhanlaoban]

In Chinese text corpus, we can generate some adversarial examples by random insertion(RI), random deletion(RD) or synonym replacement(SR). I am wondering whether EDA method will cause the model such text classifier to be attacked by the adversarial examples generated by RI, RD or SR like EDA does.
Can you explain this? Because I did some experiments and they show a decrease in performance.
Thank you very much!

[jasonwei20]

I'm surprised to hear that you saw a decrease in performance, which repository did you use?
What synonym dictionary are you using?
What's the size of the training set?

You can see the tSNE figure in our paper to see how augmented sentences relate to original sentences. Generally, it seems like examples from EDA don't change the ground-truth label.