Skip to content

"That Is a Suspicious Reaction!": Interpreting Logits Variation to Detect NLP Adversarial Attacks.

Edoardo Mosca, Shreyash Agarwal, Javier Rando-Ramirez, Georg Groh

VenueA*ACL
Year2022
ProceedingsACL (1)

Browse the full ACL paper archive.