WordBlitz: An Efficient Hard-Label Textual Adversarial Attack Method Jointly Leveraging Adversarial Transferability and Word Importance

Existing textual attacks mostly perturb keywords in sentences to generate adversarial examples by relying on the prediction confidence of victim models. In practice, attackers can only access the prediction label, meaning that the victim model can easily defend against such hard-label attacks by den...

Descripció completa

Dades bibliogràfiques

Autors principals:	Xiangge Li, Hong Luo, Yan Sun
Format:	Artigo
Idioma:	Inglês
Publicat:	MDPI AG 2024-04-01
Col·lecció:	Applied Sciences
Matèries:	natural language processing textual attack hard label adversarial samples model robustness
Accés en línia:	https://www.mdpi.com/2076-3417/14/9/3831
Etiquetes:	Afegir etiqueta Sense etiquetes, Sigues el primer a etiquetar aquest registre!