From Noise to Signal to Selbstzweck: Reframing Human Label Variation in the Era of Post-training in NLP
"人类价值观" 相关文章
From Noise to Signal to Selbstzweck: Reframing Human Label Variation in the Era of Post-training in NLP
VAL-Bench: Measuring Value Alignment in Language Models
Messy on Purpose: Part 2 of A Conservative Vision for the Future
The Morality of Probability: How Implicit Moral Biases in LLMs May Shape the Future of Human-AI Symbiosis
Interpretability as Alignment: Making Internal Understanding a Design Principle
Former Intel CEO launches a benchmark to measure AI alignment
用科幻建立AI行为准则？DeepMind提出首个此类基准并构建了机器人宪法
Starting Thoughts on RLHF
Building AI safety benchmark environments on themes of universal human values
人类自身都对不齐，怎么对齐AI？新研究全面审视偏好在AI对齐中的作用
Values Are Real Like Harry Potter
We Don't Know Our Own Values, but Reward Bridges The Is-Ought Gap
Comment on Counterarguments to the basic AI x-risk case by Jonathan
Against Explosive Growth
Ten counter-arguments that AI is (not) an existential risk (for now)
Can This AI Save Teenage Spy Alex Rider From A Terrible Fate?
AI Alignment: Why Solving It Is Impossible
