Prompting a Weighting Mechanism into LLM-as-a-Judge in Two-Step: A Case Study
This work addresses a specific limitation in LLM-based evaluation for NLG tasks, offering an incremental improvement for researchers and practitioners.
The paper tackled the problem of LLMs overemphasizing minor details and undervaluing critical information when used as judges for NLG tasks, resulting in an average 6% improvement in Human Alignment Rate through a prompt design mechanism.
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.