此文章是给开发者使用的ChatGPT Prompt Engineering系列的第四篇,前三篇文章详见:指导原则、迭代获取提示词、应用举例。
这篇文章对课程ChatGPT Prompt Engineering for Developers做一个总结。
总结中保持课程中的英文表述。具体中文讲解可以参考之前的文章。
Introduction
- base LLMs
- Instruction tuned LLMs
guidelines
write clear and specific instruction
- use delimiters
- ask for structured output
- check whether conditions are satisfied
- few-shot prompting
give model time to think
- specify the steps to complete a task
- instruct the model to work out ites own solution before rushing to a conclusion
model Limitations
- hallucination
find relevant information then answer questions based on the relevant information
Iterative Prompt Development
- try something
- analyze where the result does not give what you want
- clarity instructions, give more time to think
- refine prompts with a batch of examples
Application
summarizing
infering
- sentiment
- identify types of emotions
- extract information
- inferring topics
transforming
- translation
- format conversion
- spell check / grammar check
expanding
- email reply
chatbot
- role
system user assistant - context
最后,给出一份整理好的思维导图,供大家参考。
课程思维导图
关注【算法工程笔记】公众号,回复提示词,获取课程对应的课件和jupyter notebook文件(仅限个人学习使用,请勿用于商业用途)。