CSCL Poster
#579: Hierarchical Multi-Agent System for Instructional Design in Music Knowledge Building
Wed Jun 17, 4:15 PM–5:45 PM · Outdoors
AI in Education Knowledge Building & Knowledge Creation Generative AI & Large Language Models Arts-Based & Creative Learning Teacher Learning & Professional Development
This study presents a hierarchical multi-agent system (MAS) to support music teachers in designing Knowledge Building (KB)-oriented lessons. The system integrates four expert agents (KB, musician, music pedagogy and music psychology) to generate holistic instructional design that contain embodied and creative music activities. A preliminary qualitative interview with five primary school teachers shows positive feedback of creativity and applicability. Further work will validate the lesson plan quality, compare with single-agent systems, and examine classroom impact through longitudinal studies.
Speakers
- Haiyang Xin — CocoRobo Ltd
Authors
Menglian Zhang, Mingyang Zhu, Haiyang Xin, Yibing Zhang