INSIGHT: Bridging the Student-Teacher Gap in Times of Large Language Models
Project Overview
The document explores the use of generative AI, particularly Large Language Models (LLMs), in education through a system known as INSIGHT, which seeks to improve student-teacher interactions by delivering personalized learning experiences while maintaining data privacy. INSIGHT enables educators to derive valuable insights from student interactions, facilitating tailored support and adaptive learning strategies that cater to individual needs. However, the implementation of such technology raises important concerns regarding the potential reduction in human interaction, increased reliance on technological solutions, and privacy implications for students. The document emphasizes the necessity for future development to focus on validating the effectiveness of INSIGHT in actual educational environments to ensure that it enhances, rather than detracts from, the educational experience. Overall, while generative AI offers promising applications in personalizing education and improving engagement, careful consideration of its impacts on human relationships and privacy is crucial.
Key Applications
INSIGHT - Intelligent Student and Instructor Guidance through Human-Centered Technology
Context: Higher education courses, targeting both students and teaching staff
Implementation: Modular design allowing integration into various courses; uses LLMs for student interactions and FAQ generation.
Outcomes: Improved personalized support for students, enhanced student-teacher interaction, and identification of knowledge gaps.
Challenges: Potential over-reliance on AI, degradation of human interaction, and privacy concerns regarding data collection.
Implementation Barriers
Technical
Integration of AI tools may lead to over-reliance and degradation of face-to-face interactions.
Proposed Solutions: INSIGHT emphasizes the complementary role of AI tools alongside teaching staff, ensuring human interaction remains key.
Privacy
AI tools collect vast amounts of data which may raise privacy concerns among students.
Proposed Solutions: INSIGHT offers an anonymous mode and uses local LLMs to ensure sensitive queries remain private.
Project Team
Jarne Thys
Researcher
Sebe Vanbrabant
Researcher
Davy Vanacken
Researcher
Gustavo Rovelo Ruiz
Researcher
Contact Information
For information about the paper, please contact the authors.
Authors: Jarne Thys, Sebe Vanbrabant, Davy Vanacken, Gustavo Rovelo Ruiz
Source Publication: View Original PaperLink opens in a new window
Project Contact: Dr. Jianhua Yang
LLM Model Version: gpt-4o-mini-2024-07-18
Analysis Provider: Openai