Rhetorical Analysis of Predictive LLMs

This assignment, by Alan Knowles, is from the TextGenEd collection in the WAC Clearinghouse Repository.

The Abstract from the site explains:

This assignment asks students to train a large language model (LLM) to generate Twitter posts in the style of specific accounts via a process known as few-shot learning, which trains the LLM on a small number of sample posts. Students use the trained LLM to generate tweets, then they rhetorically analyze the generated tweets. The assignment was originally developed for an entry-level Professional and Technical writing (PTW) course, but can be easily adapted to other disciplines and course levels.

Key Features of This Assignment

Few-Shot Learning

Students are tasked with training a large language model (LLM) using a small set of sample tweets to generate content in the style of specific Twitter accounts. This process, known as few-shot learning, introduces students to the intricacies of language model training and generation.

Rhetorical Analysis

After generating tweets with the trained LLM, students conduct a rhetorical analysis of the generated content. This analysis involves examining the language, style, and rhetorical strategies used by the LLM to mimic the target Twitter account.

Adaptability and Application

The assignment is designed for an entry-level Professional and Technical Writing (PTW) course but is easily adaptable to other disciplines and educational levels. It provides a practical application of LLMs in a real-world context, enhancing students’ understanding of both AI and rhetorical principles.

Find the full version of this assignment at the WAC Clearinghouse.