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 2130 of 114 papers

TitleStatusHype
Few-Shot Adversarial Prompt Learning on Vision-Language ModelsCode1
End-to-End Adversarial Text-to-SpeechCode1
Adversarial Decoding: Generating Readable Documents for Adversarial ObjectivesCode1
T3: Tree-Autoencoder Constrained Adversarial Text Generation for Targeted AttackCode1
Adversarial Text Rewriting for Text-aware Recommender SystemsCode1
Persistent Anti-Muslim Bias in Large Language ModelsCode1
RIATIG: Reliable and Imperceptible Adversarial Text-to-Image Generation With Natural PromptsCode1
Breaking BERT: Gradient Attack on Twitter Sentiment Analysis for Targeted MisclassificationCode0
Less is More: Removing Text-regions Improves CLIP Training Efficiency and RobustnessCode0
BinarySelect to Improve Accessibility of Black-Box Attack ResearchCode0
Show:102550
← PrevPage 3 of 12Next →

No leaderboard results yet.