These two new AI benchmarks could help make models less biased
"情境意识" 相关文章
“Alignment Faking” frame is somewhat fake
Catastrophic sabotage as a major threat model for human-level AI systems
Owain Evans on Situational Awareness and Out-of-Context Reasoning in LLMs
AI Safety at the Frontier: Paper Highlights, July '24
Me, Myself, and AI: the Situational Awareness Dataset (SAD) for LLMs
