Tag: machine learning security
Browse our exclusive articles!
Jailbreaking Vision-Language Models via Visual Attacks
Discover how visual modality exploits vulnerabilities in vision-language models and learn key strategies to enhance AI safety and alignment.
Detecting Multi-Turn Attacks in LLMs via Activation Probing
Discover how adaptive probing of LLM activations detects multi-turn adversarial attacks, boosting AI security with 93.8% accuracy.
AdaBFL: Adaptive Multi-Layer Defense for Robust FL
Discover AdaBFL, a multi-layer adaptive aggregation framework enhancing Byzantine-robust federated learning with dynamic defenses against attacks.
Robust Federated Learning Against Adversarial Attacks
Enhance federated learning security with loss-based client clustering to resist Byzantine attacks and improve model robustness effectively.
DSIPA: Detect LLM-Generated Texts via Sentiment Analysis
DSIPA detects AI-generated texts using sentiment-invariant patterns, boosting accuracy without training or model access. Effective across multiple domains.