petemetefete

↗ HN profile [ 20.0 ms ] full profile

Karma: 0
Created: ()
Submissions: 0

Training large language models on narrow tasks can lead to broad misalignment (nature.com)

3 points 8mo ago 1