ericmichael
@ericmichael
UTRGV School of Medicine / UTHealth RGV. Data + AI.
ID: 15463728
17-07-2008 03:44:21
127 Tweet
787 Followers
635 Following
iāve seen teams try to apply evals and get the same outcome and gain reluctance to eval. this is because they used off-the-shelf evals (by evals platforms) like āfaithfulnessā, ācohesionā, āhelpfulnessā š generic evals arenāt useful. your evals must be aligned with your user