Ian Sefferman on Mythryl: “Interesting to think about the difficulty in creating *good* evals.”

Interesting to think about the difficulty in creating *good* evals.

Techmeme@techmeme.com

A look at the more challenging AI evaluations emerging in response to the rapid progress of models, including FrontierMath, Humanity's Last Exam, and RE-Bench (Tharin Pillay/Time) Main Link | Techmeme Permalink

25/12/24, 11:27 pm0 reposts1 like

Post