Andrew White 🐦⬛
@andrewwhite01
Automating science. Cofounder @EdisonSci. Cofounder @FutureHouseSF. Former prof of chem eng.
Joined April 2009
1.9K Following 33.7K Followers
Bixbench3: Can models reproduce the entire analysis from papers? We had agents repro papers end-to-end, with over 1B tokens spent on some. Frontier agents still score below 50%, and due to real problems like environment misconfig, quitting, or faking data 1/6
Show more
