State of Adversarial AI

3 papers · avg viability 5.7

Download CSV View topic page

Top papers

Learning to Inject: Automated Prompt Injection via Reinforcement Learning(6.0)
Not-in-Perspective: Towards Shielding Google's Perspective API Against Adversarial Negation Attacks(6.0)
LLM-Based Adversarial Persuasion Attacks on Fact-Checking Systems(5.0)