Guide-yakabatana mibvunzo · Zvakaoma Nhanho

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.

Nzira dzekutungamira dzinoenderanaCalibrating LLM Judges Against Humans
Mubvunzo 1 ye 8

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