اختبار مرتبط بالدليل · صعب المستوى
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.
سؤال 1 من 8
اختبار مرتبط بالدليل · صعب المستوى
Tests how graders score model outputs, which tasks suit reinforcement fine-tuning, and how reward hacking and weak training signals arise.