Tom Henighan
5 papers · Latest:
Discovering Language Model Behaviors with Model-Written Evaluations
This paper introduces a method to automatically generate high-quality evaluations using language models themselves, revealing new and unexpected behaviors as models scale.
Constitutional AI: Harmlessness from AI Feedback
Constitutional AI trains harmless AI assistants using AI-generated feedback guided by a set of human-defined principles, minimizing the need for human-labeled data.
Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback
This paper demonstrates that reinforcement learning from human feedback (RLHF) can effectively fine-tune language models to be both helpful and harmless, improving performance across NLP tasks while maintaining specialized skills.
Language Models are Few-Shot Learners
GPT-3, a 175 billion parameter language model, demonstrates strong few-shot learning abilities across diverse NLP tasks without task-specific fine-tuning.
Scaling Laws for Neural Language Models
This paper identifies power-law scaling relationships between language model performance and factors like model size, dataset size, and compute, enabling optimal training strategies under fixed compute budgets.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.