Large Language Models (LLMs) can be prompted using either Markdown or XML-style tags to delineate different sections of input. Markdown is generally preferred for its readability and ease of maintenance, especially for prompts that are primarily instructions or contain simple lists. XML-style tags are more effective when dealing with complex structures, nested content, or when explicit boundaries are needed to distinguish between instructions and provided data, such as in untrusted documents. AI
IMPACT Provides guidance on structuring LLM prompts for better performance and maintainability.
RANK_REASON The item discusses best practices for LLM prompting formats, drawing on documentation from AI labs, rather than announcing a new product or research.
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →