Prompting a Weighting Mechanism into LLM-as-a-Judge in Two-Step: A Case Study

TL;DR

This paper introduces a prompt design mechanism that enables Large Language Models to better evaluate natural language generation by effectively weighting topic importance, leading to more accurate assessments.

Contribution

It proposes a novel prompt engineering approach that incorporates explicit importance weighting, improving LLMs' judgment accuracy in NLG evaluation tasks.

Findings

Achieved an average 6% improvement in Human Alignment Rate (HAR).

Enhanced LLMs' ability to prioritize relevant information.

Demonstrated effectiveness through a case study.

Abstract

While Large Language Models (LLMs) have emerged as promising tools for evaluating Natural Language Generation (NLG) tasks, their effectiveness is limited by their inability to appropriately weigh the importance of different topics, often overemphasizing minor details while undervaluing critical information, leading to misleading assessments. Our work proposes an efficient prompt design mechanism to address this specific limitation and provide a case study. Through strategic prompt engineering that incorporates explicit importance weighting mechanisms, we enhance using LLM-as-a-Judge ability to prioritize relevant information effectively, as demonstrated by an average improvement of 6% in the Human Alignment Rate (HAR) metric.