Kyle Cox

Post-hoc reasoning in chain of thought

26 Dec 2024

Active typo correction outperforms passive typo correction in GPT-3.5

03 Jan 2024

Run your simulations hot

16 Nov 2023

Notes on Risks from Learned Optimization in Advanced Machine Learning Systems

21 Sep 2023

Notes on The Alignment Problem from a Deep Learning Perspective

20 Sep 2023

Augmenting decision trees with LLMs

12 Sep 2023

Basketball power ratings as stochastic gradient descent

15 Aug 2023

Three approaches to selecting examples for in-context learning

02 Aug 2023