Adian Liusie

This framework improves text evaluation efficiency and accuracy using Large Language Models.

CrossCheckGPT provides a new way to evaluate model reliability and accuracy.

A new method enhances text evaluation by using soft probabilities for better accuracy.