AI for Education and Scientific Learning
We develop multimodal AI systems that reason about student thinking, instructional practice, and the evidence educators use to make decisions.
Our education research is not general-purpose tutoring or chatbots. It focuses on the multimodal evidence of learning — classroom video, student scientific drawings, and teacher decision-making — and on AI that does not merely recognize classroom activity but reasons about student understanding, instructional quality, and the evidence behind educational decisions.
What we work on
- Multimodal understanding of classroom instruction
- AI-supported teacher feedback and decision-making
- Student drawing and scientific model assessment
- Diagnostic reasoning about student understanding
- Generative simulation of student work
- Video-based analysis of teaching and learning
- Evaluation of multimodal foundation models in education
- Human–AI collaboration for educators
- Personalized and evidence-grounded learning support
Selected work
-
Can Multimodal LLMs See Science Instruction? Benchmarking Pedagogical Reasoning in K-12 Classroom Videos
AIED 2026 -
DrawSim-PD: Simulating Student Science Drawings to Support NGSS-Aligned Teacher Diagnostic Reasoning
AIED 2026 -
Automatically Assess Elementary Students' Hand-Drawn Scientific Models Using Deep Learning
ICLS 2023