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...

Fuld beskrivelse

Bibliografiske detaljer

Principais autores:	Xiangge Li, Hong Luo, Yan Sun
Format:	Artigo
Sprog:	Inglês
Udgivet:	MDPI AG 2024-04-01
Serier:	Applied Sciences
Fag:	natural language processing textual attack hard label adversarial samples model robustness
Online adgang:	https://www.mdpi.com/2076-3417/14/9/3831
Tags:	Tilføj Tag Ingen Tags, Vær først til at tagge denne postø!