Prompt Orchestration Markup Language

TL;DR

POML is a new markup language designed to improve the organization, data integration, and presentation of prompts for large language models, enhancing development and application complexity management.

Contribution

It introduces a component-based markup system with styling, templating, and developer tools to systematically address prompt structuring challenges.

Findings

Enhanced prompt organization in case studies

Improved accuracy in TableQA task

Positive user feedback on development efficiency

Abstract

Large Language Models (LLMs) require sophisticated prompting, yet current practices face challenges in structure, data integration, format sensitivity, and tooling. Existing methods lack comprehensive solutions for organizing complex prompts involving diverse data types (documents, tables, images) or managing presentation variations systematically. To address these gaps, we introduce POML (Prompt Orchestration Markup Language). POML employs component-based markup for logical structure (roles, tasks, examples), specialized tags for seamless data integration, and a CSS-like styling system to decouple content from presentation, reducing formatting sensitivity. It includes templating for dynamic prompts and a comprehensive developer toolkit (IDE support, SDKs) to improve version control and collaboration. We validate POML through two case studies demonstrating its impact on complex application integration (PomLink) and accuracy performance (TableQA), as well as a user study assessing its effectiveness in real-world development scenarios.