FROM EQUATION TO EXPLANATION: LEVERAGING LARGE LANGUAGE MODELS FOR INTERACTIVE MATH EDUCATION
Universitat Politècnica de València (SPAIN)
About this paper:
Conference name: 19th International Technology, Education and Development Conference
Dates: 3-5 March, 2025
Location: Valencia, Spain
Abstract:
This paper introduces a groundbreaking educational tool that leverages Artificial Intelligence (AI) to revolutionize mathematics learning. Unlike traditional methods, such as textbooks, static online resources, and even recent advancements like ChatGPT-style models for problem-solving assistance, our system offers a fully immersive learning experience. It transforms static mathematical equations into dynamic, step-by-step explanatory videos, employing AI and Large Language Models (LLMs) to interpret mathematical content from equation images and generate visually engaging animations with synchronized audio narration. This novel approach has the potential to significantly enhance student engagement and understanding across various difficulty levels, bridging the divide between abstract mathematical concepts and interactive, personalized learning experiences.
The paper outlines the tool's architecture, detailing the workflow from equation image input to final educational video production. We highlight the significant potential of this AI-driven approach to address common barriers in mathematics education, such as the need for immediate, personalized explanations for complex problems. The system's narrative and visual components create a more immersive and engaging content experience compared to text-based AI assistants. This potential to address common barriers in mathematics education is a reason for hope about the tool's effectiveness. Additionally, we explore the potential integration of our tool with Intelligent Tutoring Systems (ITS), enabling these systems to incorporate specific videos when deemed pedagogically appropriate. While large-scale evaluation is pending, we present our methodology for future assessment of the tool's effectiveness in real educational settings. This research contributes to the field of educational technology by demonstrating a novel application of AI in creating adaptive, multi-modal learning experiences. Future work will focus on comprehensive evaluation, system refinement, and exploration of broader applications across various mathematical topics, including potential synergies with existing AI-powered educational tools.Keywords:
Artificial Intelligence, Large Language Models, Interactive Learning, Student Engagement.