Fetching the paper…

Backdoor Attack against NLP models with Robustness-Aware Perturbation defense · Around