LESSWRONG
LW

2329
Approximating Human Preferences Using a Multi-Judge Learned System
by JoseFaustino