Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision
Collin Burns†, Pavel Izmailov†, Jan Hendrik Kirchner†, Bowen Baker†, Leo Gao†, Leopold Aschenbrenner†, Yining Chen†, Adrien Ecoffet†, Manas Joglekar†, Jan Leike, Ilya Sutskever, Jeff Wu†
ICML 2024 (Oral)
Discovering Latent Knowledge in Language Models Without Supervision
Collin Burns*, Haotian Ye*, Dan Klein, Jacob Steinhardt
ICLR 2023
Measuring Mathematical Problem Solving With the MATH Dataset
Dan Hendrycks, Collin Burns, Saurav Kadavath, Akul Arora, Steven Basart, Eric Tang, Dawn Song, Jacob Steinhardt
NeurIPS 2021 (Datasets and Benchmarks Track)
Measuring Massive Multitask Language Understanding
Dan Hendrycks, Collin Burns, Steven Basart, Andy Zou, Mantas Mazeika, Dawn Song, Jacob Steinhardt
ICLR 2021
(*: equal contribution, †: primary contributor)
Speedcubing. I used to be very involved in competitive Rubik's Cube solving ("speedcubing"). In 2015 I broke the official world record for a single 3x3 solve with a time of 5.25 seconds. I've also had a national championship title, a continental record, and four national records.