Large Language Models as Assessors: On the Impact of Relevance Scales
Traditionally, relevance judgments have relied on human annotators, but recent advances in Large Language Models (LLMs) have prompted growing interest in their use as a proxy for …
riccardo-zamolo