A Survey of Automatic Prompt Optimization with Instruction-focused Heuristic-based Search Algorithm

TL;DR

This survey reviews heuristic-based search algorithms for automatic prompt optimization in large language models, categorizing methods, datasets, tools, and discussing future challenges for enhancing prompt refinement.

Contribution

It provides a comprehensive taxonomy of automatic prompt optimization methods, categorizing them by optimization points, criteria, operators, and algorithms, and highlights supporting datasets and tools.

Findings

Categorizes prompt optimization methods systematically

Identifies key datasets and tools for prompt refinement

Discusses open challenges and future directions

Abstract

Recent advances in Large Language Models have led to remarkable achievements across a variety of Natural Language Processing tasks, making prompt engineering increasingly central to guiding model outputs. While manual methods can be effective, they typically rely on intuition and do not automatically refine prompts over time. In contrast, automatic prompt optimization employing heuristic-based search algorithms can systematically explore and improve prompts with minimal human oversight. This survey proposes a comprehensive taxonomy of these methods, categorizing them by where optimization occurs, what is optimized, what criteria drive the optimization, which operators generate new prompts, and which iterative search algorithms are applied. We further highlight specialized datasets and tools that support and accelerate automated prompt refinement. We conclude by discussing key open challenges pointing toward future opportunities for more robust and versatile LLM applications.