safety
Safety in the context of AI introduces measures and protocols designed to ensure that AI systems operate without causing harm to users or their environment. This includes establishing guidelines for ethical behavior, data management, and system security when deploying AI technologies.
- Double Descent Meets Out-of-Distribution Detection: Theoretical Insights and Empirical Analysis on the Role of Model Complexity
- Enhancing Safety in Reinforcement Learning with Human Feedback via Rectified Policy Optimization
- Inference-Time Reward Hacking in Large Language Models
- Integrating Drug Substructures and Longitudinal Electronic Health Records for Personalized Drug Recommendation
- MJ-Bench: Is Your Multimodal Reward Model Really a Good Judge for Text-to-Image Generation?
- Measuring what Matters: Construct Validity in Large Language Model Benchmarks
- On the creation of narrow AI: hierarchy and nonlocality of neural network skills
- RespoDiff: Dual-Module Bottleneck Transformation for Responsible & Faithful T2I Generation
- Safe and Stable Control via Lyapunov-Guided Diffusion Models
- Statistically Valid Post-Deployment Monitoring Should Be Standard for AI-Based Digital Health