Тест, свързан с ръководство · трудно Ниво

Calibrating LLM Judges Against Humans Quiz

Calibrate LLM judges with shared rubrics, independent human labels, confusion matrices, suitable agreement statistics, and repeated held-out checks.

Свързани ръководни пътекиCalibrating LLM Judges Against Humans
Въпрос 1 на 8

Why is Cohen's kappa preferred over raw percent agreement when calibrating an LLM judge?