Large language model (LLM)
Leveraging linguistic features to improve machine learning models for detecting ChatGPT usage on exams
Abstract
This work presents a case study, linguistic analyses, and the results of experiments on using linguistic features to improve the detection mechanism of the use of large language models (LLM) to generate solutions for exams that require domain-specific knowledge. The study involves analyzing the responses of three groups of students: a group who verbatim copied outputs of ChatGPT to plagiarize solutions, another group who referred to external non-LLM resources (e.g., web search) to plagiarize solutions, a control group who did not plagiarize. Linguistic analyses show that ...
Get Artificial Intelligence now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
