There are are a number of prompt frameworks circulating right now. At a minimum, I suggest specifying the Role, Task, Audience, and Desired Output.
Role:
Description: Clearly define the role or context for which you need content. Is it for a writer, marketer, student, or developer?
Example: “As a travel blogger, create an engaging introduction for an article about exploring hidden gems in Kyoto.”
Task:
Description: Specify the task you want the AI to perform. Is it writing, summarizing, generating code, or creating visual content?
Example: “Generate a poem inspired by the theme of solitude.”
Audience:
Description: Identify the intended audience. Is it for children, professionals, enthusiasts, or a specific demographic?
Example: “Write a humorous dialogue suitable for a teenage audience.”
Desired Output:
Description: Be clear about the format or type of content you expect. Is it a blog post, tweet, code snippet, or image?
Example: “Provide a Python code snippet that calculates the Fibonacci sequence.”
Remember that thoughtful prompts lead to better results. By incorporating these guidelines, users can tailor their requests effectively and obtain more relevant and accurate outputs from generative AI systems. 🌟
Be specific in your descrption of the task or intent
Ask the AI to assume a role or persona
Provide examples of the desired task or output
Ask the AI to go step-by-step
Use positive rather than negative language
Provide encouragement or incentives for accurate replies
Delineate parts of the prompt with hash marks, e.g. ##instructions##
Evaluate the initial output before using or proceeding with your prompt
Refine through iteration
1. Write Clear Instructions:
- Be specific: Clarity in instructions leads to more relevant outcomes.
- Define the desired output length and complexity.
- Demonstrate preferred formats.
- Minimize ambiguity to enhance model accuracy.
2. Provide Reference Text:
- Counteract potential fabrications with concrete reference materials.
- Reference texts guide the model towards accurate and reliable answers.
3. Split Complex Tasks into Simpler Subtasks:
- Break down tasks to reduce errors and improve manageability.
- Consider tasks as workflows of simpler, interconnected steps.
4. Give the Model Time to "Think":
- Allow the model to process and reason, similar to a human solving a complex problem.
- Encourage a "chain of thought" approach for more accurate reasoning.
5. Use External Tools:
- Supplement the model's capabilities with specialized tools for specific tasks.
- Leverage resources like text retrieval systems or code execution engines.
6. Test Changes Systematically:
- Measure improvements with a comprehensive testing approach.
- Ensure that modifications lead to overall performance enhancements.
Gewirtz, D. (2024, January 24). How to write better ChatGPT prompts in 5 steps. ZDNET. https://www.zdnet.com/article/how-to-write-better-chatgpt-prompts-in-5-steps/
Lee, L., & Syam, A. (2024, January 9). ️ Prompt Engineering for Education. AI x Education. https://aixeducation.substack.com/p/prompt-engineering-for-education?utm_source=post-email-title&publication_id=1689008&post_id=140465989&utm_campaign=email-post-title&isFreemail=true&r=1ukey9
This work is licensed by Emily Rush under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.