Human-Guided Harm Recovery for Safer AI Agents
Explore human-guided harm recovery methods that help AI agents safely recover from harmful actions, ensuring responsible computer system use.
Quantum Qutrit Neural Networks for Real-Time Finance
Explore how quantum qutrit neural networks outperform traditional models in real-time financial forecasting with superior accuracy and efficiency.
AI Scientists Fail to Reason Scientifically in Research
Study reveals AI scientists produce results without proper scientific reasoning, highlighting gaps in evidence use and belief revision.
ARES: Enhancing AI Safety with Adaptive Red-Teaming
Discover how ARES improves AI safety by adaptive red-teaming and repairing policy-reward systems in large language models.
Visualizing Language Model Output Distributions with GROVE
Discover GROVE, a tool for visualizing and comparing language model output distributions, enhancing understanding of generation diversity and structure.