More on Prompting

As we have seen in the preceding pages, prompts can take various formats and levels of complexity. They may comprise contextual information, guidelines, multiple question-answer examples, and even additional prompts. Yes, you read that right!

To illustrate, here's a sample prompt that incorporates context, instructions, and multiple examples:

Twitter is a social media platform where users can share brief messages known as 'tweets.' These tweets can express positive or negative sentiments, and our goal is to classify them accordingly. Below are examples of positive and negative tweets. Please ensure that you correctly classify the last tweet. Q: Tweet: "What a beautiful day!" Is this tweet positive or negative? A: positive Q: Tweet: "I hate this class" Is this tweet positive or negative? A: negative Q: Tweet: "I love pockets on jeans" A:

Supplementing AI models with additional context and examples can significantly enhance their performance on diverse tasks. The following chapter delves into slightly more advanced prompting techniques to help you achieve even better results.

Notes

Please note that in the upcoming chapters, the terms 'AI', 'model', and 'LLM' may be used interchangeably. For more information, please refer to the vocabulary reference.

Additionally, the next few sections will cover self-augmented prompts, which are prompts that contain sub-prompts within them.