Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training
Paper
•
2401.05566
•
Published
•
25
Reading List
Note TLDR: Anthropic tries to train a deliberately deceptive LLM.
-
On the Societal Impact of Open Foundation Models
Paper • 2403.07918 • Published • 16 -
JudgeLM: Fine-tuned Large Language Models are Scalable Judges
Paper • 2310.17631 • Published • 32 -
Instruction Tuning for Large Language Models: A Survey
Paper • 2308.10792 • Published • 1 -
An Empirical Study of LLM-as-a-Judge for LLM Evaluation: Fine-tuned Judge Models are Task-specific Classifiers
Paper • 2403.02839 • Published • 1 -
Holistic Safety and Responsibility Evaluations of Advanced AI Models
Paper • 2404.14068 • Published -
A Framework for Automated Measurement of Responsible AI Harms in Generative AI Applications
Paper • 2310.17750 • Published • 9