Κουίζ που συνδέεται με οδηγό · Σκληρό Επίπεδο

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.

Σχετικές διαδρομές οδηγώνCalibrating LLM Judges Against Humans
Ερώτηση 1 του 8

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