Generating Effective Student Handouts with Iterative Prompting

In this lesson, you will learn how to use LLMs to quickly create student handouts from your lesson materials. This is a common task for educators, but it can be time-consuming to organize key points, define important terms, and present information in a clear way. LLMs can help you draft a handout in just a few steps, saving you time and giving you a strong starting point to edit and personalize for your students.

You have already learned how to use LLMs to generate different versions of assessments. Now, we will focus on a new use case: turning a lesson into a student-friendly handout. This lesson will show you how to break this task into smaller steps and use LLMs effectively.

Before we start, let’s talk about a key concept: iterative prompting. This means you don’t have to ask the LLM to do everything at once. Instead, you guide the model step by step, using the output from one prompt as the input for the next. This approach helps you get more accurate and useful results.

For example, if you want to create a handout from a lesson, you might first ask the LLM to summarize the lesson. Then, you use that summary to ask for a handout with specific features, like key terms and tables.

This step-by-step process is called iterative prompting because you are iterating — moving from one step to the next, refining the output as you go.

Let’s start with the first step: getting a summary of your lesson. Instead of asking the LLM to create a handout right away, it’s better to begin with a summary. This helps the model focus on the main ideas and makes the next step easier.

Suppose you have a lesson saved as a PDF file. You want the LLM to read it and give you a short summary. Here’s how you might write your prompt:

I have a lesson provided to you in the appended PDF file. Read it and provide me with a short summary of this lesson.

This prompt is clear and direct. You tell the LLM where the lesson is (in the PDF) and what you want (a short summary). By starting with a summary, you help the model focus on the most important points, which will be useful for the next step.