The Prompt Report: A Systematic Survey of Prompt Engineering Techniques

TL;DR

This paper tackles the lack of standardized terminology and a coherent framework in prompt engineering by conducting a large-scale, PRISMA-guided survey. It introduces a taxonomy of 58 text-based prompting techniques, a 33-term vocabulary, and 40 non-text prompting techniques, covering multilingual and multimodal extensions and agent-based prompts. The authors also present best-practice guidelines, meta-analytic insights into prefix-prompting, and two benchmarking case studies to illustrate practical workflows and evaluation strategies. Overall, the work provides the most comprehensive catalog to date for researchers and practitioners and aims to standardize language, methods, and evaluation in prompt engineering. The study advances the field by clarifying terminology, outlining robust evaluation frameworks, and offering actionable guidance for deploying prompting techniques across domains and modalities.

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.

Figures (35)

• Figure 1: Categories within the field of prompting are interconnected. We discuss 7 core categories that are well described by the papers within our scope.
• Figure 2: Prompts and prompt templates are distinct concepts; a prompt template becomes a prompt when input is inserted into it.
• Figure 3: A Terminology of prompting. Terms with links to the appendix are not sufficiently critical to describe in the main paper, but are important to the field of prompting. Prompting techniques are shown in Figure \ref{['fig:taxonomy']}
• Figure 4: The Prompt Engineering Process consists of three repeated steps 1) performing inference on a dataset 2) evaluating performance and 3) modifying the prompt template. Note that the extractor is used to extract a final response from the LLM output (e.g. "This phrase is positive" $\rightarrow$ "positive"). See more information on extractors in Section \ref{['sec:answer-engineering']}.
• Figure 5: The PRISMA systematic literature review process. We accumulate 4,247 unique records from which we extract 1,565 relevant records.