Can Many-Shot In-Context Learning Help LLMs as Evaluators? A Preliminary Empirical Study

TL;DR

This study explores how many-shot in-context learning prompts can improve the reliability of large language models when used as evaluators, showing that specific prompt designs enhance evaluation consistency and accuracy.

Contribution

The paper introduces two novel many-shot ICL prompt templates, MSwR and MSoR, to mitigate biases in LLM evaluators and demonstrates their effectiveness with GPT-4o.

Findings

GPT-4o performs better in many-shot regimes.

MSwR prompts outperform MSoR in evaluation tasks.

Increasing in-context examples improves evaluation quality.

Abstract

Utilizing Large Language Models (LLMs) as evaluators to assess the performance of LLMs has garnered attention. However, this kind of evaluation approach is affected by potential biases within LLMs, raising concerns about the accuracy and reliability of the evaluation results of LLMs. To address this problem, we propose and study two many-shot In-Context Learning (ICL) prompt templates to help LLM evaluators mitigate potential biases: Many-Shot with Reference (MSwR) and Many-Shot without Reference (MSoR). Specifically, the former utilizes in-context examples with model-generated evaluation rationales as references, while the latter does not include these references. Using these prompt designs, we investigate the impact of increasing the number of in-context examples on the consistency and quality of the evaluation results. Experimental results show that advanced LLMs, such as GPT-4o, perform better in the many-shot regime than in the zero-shot and few-shot regimes. Furthermore, when using GPT-4o as an evaluator in the many-shot regime, adopting MSwR as the prompt template performs better than MSoR.