A. Swenor, “Using Random Perturbations to Mitigate Adversarial Attacks on NLP Models”, AAAI, vol. 36, no. 11, pp. 13142-13143, Jun. 2022.