SOTAVerified

Adversarial Text

Adversarial Text refers to a specialised text sequence that is designed specifically to influence the prediction of a language model. Generally, Adversarial Text attack are carried out on Large Language Models (LLMs). Research on understanding different adversarial approaches can help us build effective defense mechanisms to detect malicious text input and build robust language models.

Papers

Showing 1120 of 114 papers

TitleStatusHype
White-box Multimodal Jailbreaks Against Large Vision-Language ModelsCode1
Few-Shot Adversarial Prompt Learning on Vision-Language ModelsCode1
Boosting Transferability in Vision-Language Attacks via Diversification along the Intersection Region of Adversarial TrajectoryCode1
A Pilot Study of Query-Free Adversarial Attack against Stable DiffusionCode1
RIATIG: Reliable and Imperceptible Adversarial Text-to-Image Generation With Natural PromptsCode1
SemAttack: Natural Textual Attacks via Different Semantic SpacesCode1
"That Is a Suspicious Reaction!": Interpreting Logits Variation to Detect NLP Adversarial AttacksCode1
Semantic-Preserving Adversarial Text AttacksCode1
MATE-KD: Masked Adversarial TExt, a Companion to Knowledge DistillationCode1
Persistent Anti-Muslim Bias in Large Language ModelsCode1
Show:102550
← PrevPage 2 of 12Next →

No leaderboard results yet.