LLMs as Academic Reading Companions: Extending HCI Through Synthetic Personae
Project Overview
The document examines the role of generative AI, particularly large language models (LLMs) like Claude.ai, in enhancing education by serving as academic reading companions that facilitate students' understanding of complex texts. An exploratory study presented in the text illustrates that students utilizing these AI tools demonstrate significant improvements in reading comprehension and engagement compared to those engaged in traditional independent study methods. Despite the promising findings, the document raises important concerns regarding the potential for over-reliance on AI, ethical implications surrounding its use, and the necessity for responsible integration of such technologies into educational practices. Overall, while generative AI shows great promise for enriching the learning experience, careful consideration of its implementation is essential to mitigate risks and maximize benefits in educational settings.
Key Applications
Claude.ai, an LLM-based interactive assistant
Context: Graduate-level courses at the University of Maryland
Implementation: Participants were randomly assigned to a control group or an experimental group using Claude.ai over a semester.
Outcomes: Tangible improvements in reading comprehension and engagement for the experimental group using Claude.ai.
Challenges: Potential for overreliance on AI for learning, ethical considerations regarding AI's role in education.
Implementation Barriers
Ethical concerns
The risk of students becoming overly reliant on AI tools, which may impair their critical thinking and metacognitive skills.
Proposed Solutions: Implementation of structured scaffolding that fades over time, ensuring students retain their skills.
Technical limitations
LLMs can produce biased or inaccurate outputs, which may mislead students.
Proposed Solutions: Establish ethical safeguards and provide informed consent and moderation during testing.
Project Team
Celia Chen
Researcher
Alex Leitch
Researcher
Contact Information
For information about the paper, please contact the authors.
Authors: Celia Chen, Alex Leitch
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