超越提示內容：通過內容格式集成優化提示來增強LLM性能

Beyond Prompt Content: Enhancing LLM Performance via Content-Format Integrated Prompt Optimization

February 6, 2025

作者: Yuanye Liu, Jiahang Xu, Li Lyna Zhang, Qi Chen, Xuan Feng, Yang Chen, Zhongxin Guo, Yuqing Yang, Cheng Peng

cs.AI

摘要

大型語言模型（LLMs）展示了在各種任務中的顯著能力，其在現實世界中的效果通常取決於提示設計。儘管最近的研究集中於優化提示內容，但提示格式的作用，作為一個關鍵但常常被忽視的維度，卻受到了有限的系統性調查。在本文中，我們介紹了內容-格式一體化提示優化（CFPO），這是一種創新的方法論，通過迭代的細化過程共同優化提示內容和格式。CFPO利用自然語言變異來探索內容變化，並採用動態格式探索策略，系統性地評估各種格式選項。我們在多個任務和開源LLMs上進行了廣泛的評估，結果顯示CFPO相較於僅優化內容的方法，表現出可衡量的性能改進。這突顯了整合內容-格式優化的重要性，並提供了一種實用的、與模型無關的方法來增強LLM的性能。代碼將在https://github.com/HenryLau7/CFPO 上提供。

English

Large Language Models (LLMs) have shown significant capability across various tasks, with their real-world effectiveness often driven by prompt design. While recent research has focused on optimizing prompt content, the role of prompt formatting, a critical but often overlooked dimension, has received limited systematic investigation. In this paper, we introduce Content-Format Integrated Prompt Optimization (CFPO), an innovative methodology that jointly optimizes both prompt content and formatting through an iterative refinement process. CFPO leverages natural language mutations to explore content variations and employs a dynamic format exploration strategy that systematically evaluates diverse format options. Our extensive evaluations across multiple tasks and open-source LLMs demonstrate that CFPO demonstrates measurable performance improvements compared to content-only optimization methods. This highlights the importance of integrated content-format optimization and offers a practical, model-agnostic approach to enhancing LLM performance. Code will be available at https://github.com/HenryLau7/CFPO.