Efficient and Accurate Prompt Optimization: the Benefit of Memory in Exemplar-Guided Reflection

TL;DR

This paper introduces a memory-augmented exemplar-guided reflection method for prompt optimization in large language models, significantly improving performance and efficiency by utilizing historical feedback and better exemplar retrieval.

Contribution

It proposes a novel memory-based mechanism for prompt optimization that leverages historical feedback and improves exemplar selection, outperforming previous methods.

Findings

Improves F1 score by 10.1 on LIAR dataset

Reduces optimization steps by half on ProTeGi

Outperforms state-of-the-art prompt optimization methods

Abstract

Automatic prompt engineering aims to enhance the generation quality of large language models (LLMs). Recent works utilize feedbacks generated from erroneous cases to guide the prompt optimization. During inference, they may further retrieve several semantically-related exemplars and concatenate them to the optimized prompts to improve the performance. However, those works only utilize the feedback at the current step, ignoring historical and unseleccted feedbacks which are potentially beneficial. Moreover, the selection of exemplars only considers the general semantic relationship and may not be optimal in terms of task performance and matching with the optimized prompt. In this work, we propose an Exemplar-Guided Reflection with Memory mechanism (ERM) to realize more efficient and accurate prompt optimization. Specifically, we design an exemplar-guided reflection mechanism where the feedback generation is additionally guided by the generated exemplars. We further build two kinds of memory to fully utilize the historical feedback information and support more effective exemplar retrieval. Empirical evaluations show our method surpasses previous state-of-the-arts with less optimization steps, i.e., improving F1 score by 10.1 on LIAR dataset, and reducing half of the optimization steps on ProTeGi.