EDUMATH: Generating Standards-aligned Educational Math Word Problems

This paper develops EDUMATH, a system for generating math word problems (MWPs) aligned with K-12 math standards and customized to student interests. The work includes a teacher-annotated dataset of over 11,000 generated MWPs evaluated on four criteria (solvability, accuracy, educational appropriateness, standards alignment), trained models for generation, and a classroom study with grade school students showing similar performance but higher preference for customized problems.

Math word problems (MWPs) are critical K-12 educational tools, and customizing them to students'interests and ability levels can increase learning outcomes. However, teachers struggle to find time to customize MWPs for each student given large class sizes and increasing burnout. We propose that LLMs can support math education by generating MWPs customized to student interests and math education standards. To this end, we use a joint human expert-LLM judge approach to evaluate over 11,000 MWPs ge

Study Type

Source

Framework Categories

Tool Types

Tags