Deep Learning Detection Method for Large Language Models-Generated Scientific Content

Project Overview

The document explores the impact of generative AI, especially large language models (LLMs) such as ChatGPT, on the integrity of educational content by highlighting concerns over plagiarism and misinformation in academic research. It addresses the challenge posed by AI's ability to produce texts that closely resemble human-written work, which complicates the detection of academic dishonesty. To combat this issue, the authors introduce a new detection tool called AI-Catcher, which utilizes a combination of multilayer perceptron (MLP) and convolutional neural networks (CNN) to effectively identify AI-generated scientific texts. This innovative approach shows significant advancements in detection accuracy compared to existing methods, underscoring the critical need to uphold academic integrity within educational contexts. The findings indicate that while generative AI presents opportunities for enhancing learning and content creation, it also necessitates robust mechanisms to ensure that the authenticity and credibility of academic work are preserved.

Key Applications

Implementation Barriers

Project Team

Contact Information

For information about the paper, please contact the authors.

Authors: Bushra Alhijawi, Rawan Jarrar, Aseel AbuAlRub, Arwa Bader

Source Publication: View Original PaperLink opens in a new window