Streamline Your Study Routine: AI Tips for Summarizing Class Readings

Students, have you started using ChatGPT or another LLM to summarize your assigned readings?

If you haven’t, it might be time to consider how these tools can support your academic journey. Think of ChatGPT or your preferred Large Language Model as a partner—a collaborator in your practical work projects. Let me share a quick story to illustrate the importance of context when working with LLMs.

Imagine this scenario:

You’re handed a document by a random stranger. They provide absolutely no context, simply shove it in your hands, and aggressively say, “Summarize this!” You’re not allowed to ask any questions. You might scan the document first, looking for clues like letterheads, signatories, or branding, before diving in.

Now, picture a different situation. A friend comes into your office, asking for help summarizing an email strand for your boss. They explain that the boss needs the summary for a report to the board. You’re already familiar with the organization, the audience, and the desired outcomes. Plus, you know the unspoken nuances of office politics and culture, so you know exactly what your friend needs.

Even if you were given the same document in both scenarios, the context matters. It’s not the document itself, but the context you bring to the table that leads to the quality of the summary you need.

Here’s how you can apply this approach to your readings:

When using ChatGPT for summarization, provide it with as much context as possible to get the most accurate and useful results. Here’s an example prompt you can use:

Prompt Challenge:

"I would like to provide you with the article that I am reading. The class I am reading for is IS XXX, Course Title. Many of the articles I will be providing are seminal articles in their field. I need to understand 1) the overall concepts being discussed in each article. 2) please pull out quotes that might be connected to my research interests [INSERT RESEARCH INTERESTS]. 3) In the course of this class, I am going to write a paper on either information behavior. I want to make sure that I keep notes on each source for that purpose.

Keeping these things in mind, can you help me with summaries?"

Important Reminders:

• Check with Your Instructor: Before using these summaries, make sure your instructor allows it. Also, avoid relying on these summaries for quizzes on readings. They are best used as a supplement to your own analysis and understanding.
• Copy-Pasting Text Over Uploading Documents: I’ve found that summaries are often more accurate when you copy-paste the text directly into the model rather than uploading a document. This ensures that the model focuses on the content without being distracted by formatting issues. Don't worry about losing formatting; the focus should be on the text itself.
• Be Aware of Biases: LLMs like ChatGPT are trained on vast amounts of data, which includes biases. For example, if you ask for a summary of a Wikipedia article on “scientists,” the model might exclude women or people of color from its summary, reflecting the biases in the training data. The model determines what’s important based on these patterns, so always critically evaluate the summaries you receive.

By using these strategies, you can make your reading more efficient and focused, allowing you to engage more deeply with the material that matters most to your academic work.

To learn more about summarization: https://christy-moss.ck.page/posts/ai-insights-from-bland-to-brilliant-elevate-your-ai-game-with-this-simple-hack