4 ms·
It's not even an anthropomorphization, the reward function in RLHF-like scenarios is usually quite literally "did the user think the output was good"
by Retr0id 8mo ago
It's not even an anthropomorphization, the reward function in RLHF-like scenarios is usually quite literally "did the user think the output was good"
- mjburgess 8mo ago> adversary that is trying to trick me