The Prompt Report: A Systematic Survey of Prompt Engineering Techniques

TL;DR

This paper provides a comprehensive systematic survey of prompt engineering techniques, establishing a structured taxonomy, vocabulary, and best practices for effectively interacting with generative AI systems like ChatGPT.

Contribution

It introduces the most detailed taxonomy and vocabulary for prompt engineering, consolidates best practices, and offers a meta-analysis of natural language prefix-prompting research.

Findings

Developed a taxonomy of 58 prompting techniques

Created a vocabulary of 33 terms for prompt engineering

Provided guidelines for prompting state-of-the-art LLMs

Abstract

Generative Artificial Intelligence (GenAI) systems are increasingly being deployed across diverse industries and research domains. Developers and end-users interact with these systems through the use of prompting and prompt engineering. Although prompt engineering is a widely adopted and extensively researched area, it suffers from conflicting terminology and a fragmented ontological understanding of what constitutes an effective prompt due to its relatively recent emergence. We establish a structured understanding of prompt engineering by assembling a taxonomy of prompting techniques and analyzing their applications. We present a detailed vocabulary of 33 vocabulary terms, a taxonomy of 58 LLM prompting techniques, and 40 techniques for other modalities. Additionally, we provide best practices and guidelines for prompt engineering, including advice for prompting state-of-the-art (SOTA) LLMs such as ChatGPT. We further present a meta-analysis of the entire literature on natural language prefix-prompting. As a culmination of these efforts, this paper presents the most comprehensive survey on prompt engineering to date.