jiangdaolime

Results 2 issues of jiangdaolime

When I used the model trained by the A2T attack to evaluate the robustness, the code used the other models to generate 1000 samples for evaluation. ![image](https://github.com/QData/TextAttack-A2T/assets/99568990/52bc6571-f997-4ca5-972d-f7c43d1bae07) Some of them,...

When I used the model trained by the A2T attack to evaluate the robustness, the code used the other models to generate 1000 samples for evaluation. ![1712895714705](https://github.com/QData/TextAttack/assets/99568990/a047ba0f-c547-4167-8b7a-32552a442d69) Some of them,...