Tag: adversarial machine learning
Browse our exclusive articles!
Efficient Submodular Benchmark Selection for AI Models
Discover a novel submodular approach to select optimal AI benchmarks, reducing evaluation costs while maximizing insight into language model performance.
Clean-Label Backdoor Attacks on Vision Language Models
Discover how clean-label backdoor attacks exploit diffusion models to threaten Vision Language Models' security with stealthy, natural poisoned samples.
MEMAUDIT: Optimizing Budgeted Long-Term LLM Memory Writing
Discover MEMAUDIT, a precise evaluation protocol for efficient, budgeted long-term memory writing in large language models (LLMs).
Intervention Complexity: A New Measure of AI Intelligence
Discover intervention complexity, a novel AI intelligence measure refining reward functions for better evaluation of agent competence and learning efficien...
Model Spec Midtraining: Boosting Alignment Training Generalization
Discover how Model Spec Midtraining enhances alignment training to improve AI generalization and safety in language models.