Guidelänkat frågesport · Hårt Nivå
Reinforcement Fine-Tuning with Graders Quiz
Tests how graders score model outputs, which tasks suit reinforcement fine-tuning, and how reward hacking and weak training signals arise.
Fråga 1 av 8
Guidelänkat frågesport · Hårt Nivå
Tests how graders score model outputs, which tasks suit reinforcement fine-tuning, and how reward hacking and weak training signals arise.