AIED

A Multi-Agent Approach to Validate and Refine LLM-Generated Personalized Math Problems

Wed Jul 1, 10:45 AM–11:10 AM · North 209

Part of Multi-Agent Systems for Mathematics Education

AI Agents & Agentic Systems Mathematics Education Generative AI & Large Language Models Adaptive & Personalized Learning

★ Notable speakers

Neil Heffernan ★★ — ASSISTments intelligent tutoring platform; educational data mining at scale

Candace Walkington ★ — Personalization of mathematics instruction by connecting math to students' interests; intelligent tutoring, generative AI, AR/VR in math learning

Andrew Lan ★ — Machine learning and NLP for personalized learning, knowledge tracing, automated assessment

A multi-agent LLM pipeline that validates and refines automatically generated personalized math problems to improve their quality.

Authors

Fareya Ikram, Nischal Ashok Kumar, Junyang Lu, Hunter McNichols, Candace Walkington, Neil Heffernan, Andrew Lan