Using Large Multimodal Models to Extract Knowledge Components for Knowledge Tracing from Multimedia Question Information

This paper proposes using large multimodal models (LMMs) to automatically extract knowledge components from educational content for knowledge tracing in intelligent tutoring systems, eliminating the need for manual expert labeling. The approach is validated across five educational domains by evaluating whether automatically extracted knowledge components can predict student performance as effectively as human-tagged labels.

Knowledge tracing models have enabled a range of intelligent tutoring systems to provide feedback to students. However, existing methods for knowledge tracing in learning sciences are predominantly reliant on statistical data and instructor-defined knowledge components, making it challenging to integrate AI-generated educational content with traditional established methods. We propose a method for automatically extracting knowledge components from educational content using instruction-tuned larg

Study Type

Source

Framework Categories

Tool Types

Tags