differential-privacy
New JAX-based library enables differentially private training of machine learning models, addressing critical privacy concerns in AI development and synthetic media generation.
agentic-ai
ArXiv paper examines critical reliability issues facing autonomous AI agents, from unpredictable behavior to safety concerns. Researchers outline technical challenges and propose frameworks for building more dependable agentic systems.
AI-safety
New research reveals that AI image generators can regenerate supposedly 'unlearned' harmful content through adversarial prompts, posing challenges for deepfake prevention.
ai-research
New research reveals that current model editing techniques rely on fragile shortcuts rather than true semantic understanding, calling into question the foundation of AI knowledge updates.