Prompt optimization (PO) offers a practical alternative to fine-tuning large language models (LLMs), enabling performance improvements without altering model weights. Existing methods typically rely on advanced, large-scale LLMs like GPT-4 to generate optimized prompts. However, due to limited downward compatibility, verbose, instruction-heavy prompts from advanced LLMs can overwhelm lightweight inference models and degrade response quality. In this work, we rethink prompt optimization through the lens of interpretable design. We first identify a set of model-agnostic prompt quality merits and empirically validate their effectiveness in enhancing prompt and response quality. We then introduce MePO, a merit-guided, lightweight, and locally deployable prompt optimizer trained on our preference dataset built from merit-aligned prompts generated by a lightweight LLM. Unlike prior work, MePO avoids online optimization reliance, reduces cost and privacy concerns, and, by learning clear, interpretable merits, generalizes effectively to both large-scale and lightweight inference models. Experiments demonstrate that MePO achieves better results across diverse tasks and model types, offering a scalable and robust solution for real-world deployment. Our model and dataset are available at:this https URL
View on arXiv@article{zhu2025_2505.09930,
title={ Rethinking Prompt Optimizers: From Prompt Merits to Optimization },
author={ Zixiao Zhu and Hanzhang Zhou and Zijian Feng and Tianjiao Li and Chua Jia Jim Deryl and Mak Lee Onn and Gee Wah Ng and Kezhi Mao },
journal={arXiv preprint arXiv:2505.09930},
year={ 2025 }
}