Jacob Pfau

PhD student at the NYU Alignment Research Group. Current research projects include:

• applying causal behavioral analysis to problems of introspective truthfulness in LMs
• studying scaling properties of LM performance as a function of empty space in prompts
• formalizing limitations of RL fine-tuning of LMs

I like to post about research on Twitter and Lesswrong. I also like to create prediction markets e.g. “Will an AI produce encyclopedia-worthy philosophy by 2026” on Manifold, and “Will transformer derived architectures accelerate progress in deep learning?” on Metaculus.