6 citations · 11 across the 6 of their papers we have counts for
1 paper · 1 filter
Galen Pogoncheff, Michael Beyeler
Human-aligned deep learning models exhibit behaviors consistent with human values, such as robustness, fairness, and honesty. Transferring these behavioral properties to models tra…