Samuel R. Bowman

Language Models Learn to Mislead Humans via RLHF

Future Events as Backdoor Triggers: Investigating Temporal Vulnerability in LLMs

GPQA: A Graduate-Level Google-Proof Q&A Benchmark

LLM Evaluators Recognize and Favor Their Own Generations

Towards Understanding Sycophancy in Language Models

Many-shot Jailbreaking

Let’s Think Dot by Dot: Hidden computation in transformer language models