给开发者使用的ChatGPT Prompt Engineering之课程总结

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

1. base LLMs
2. Instruction tuned LLMs

guidelines

write clear and specific instruction

1. use delimiters
2. ask for structured output
3. check whether conditions are satisfied
4. few-shot prompting

give model time to think

1. specify the steps to complete a task
2. instruct the model to work out ites own solution before rushing to a conclusion

model Limitations

1. hallucination
   find relevant information then answer questions based on the relevant information

Iterative Prompt Development

1. try something
2. analyze where the result does not give what you want
3. clarity instructions, give more time to think
4. refine prompts with a batch of examples

Application

summarizing

infering

1. sentiment
2. identify types of emotions
3. extract information
4. inferring topics

transforming

1. translation
2. format conversion
3. spell check / grammar check

expanding

• email reply

chatbot

1. role
   system user assistant
2. 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）。