Advertisement · 728 × 90

Posts by Stevie Bergman

This work couldn’t be more urgent. We need better measurement practices in AI evaluation — asap. Here, we aim to clarify and inform, and show what better looks like for accuracy metrics and confidence estimates, with bonuses such as deeper evaluation understanding. Excellent work, team!

2 months ago 1 0 0 0

I am *so proud* of this fantastic work!!

2 months ago 1 0 0 0

Come work with our team at CAISI! Applications due Feb 1

3 months ago 0 0 0 0