Edoardo Mosca, Shreyash Agarwal, Javier Rando-Ramirez, Georg Groh. "That Is a Suspicious Reaction!": Interpreting Logits Variation to Detect NLP Adversarial Attacks. In Smaranda Muresan, Preslav Nakov, Aline Villavicencio, editors, Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), ACL 2022, Dublin, Ireland, May 22-27, 2022. pages 7806-7816, Association for Computational Linguistics, 2022. [doi]
Abstract is missing.