SurveyAlign

New benchmark aims to align LLM survey evaluators with human reviewers

Researchers have introduced SurveyReview, a new benchmark and dataset designed to evaluate large language models (LLMs) when they are used as survey evaluators. This benchmark addresses the lack of systematic alignment …