Questionário vinculado ao guia · Difícil Nível

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.

Caminhos de guia relacionadosCalibrating LLM Judges Against Humans
Pergunta 1 de 8

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