Layerwise universal adversarial attack on NLP models.

Olga Tsymboi, Danil Malaev, Andrei Petrovskii, Ivan V. Oseledets

Browse the full ACL paper archive.