Tales from a Graph: a Pipeline for Mathematical Problem Generation

Published: 17 Oct 2025, Last Modified: 21 Nov 2025MATH-AI 2025 PosterEveryoneRevisionsBibTeXCC BY 4.0
Keywords: mathematical reasoning, LLM
Abstract: We present a framework for generating pairs of mathematical and word problems with controllable complexity, where the shared underlying mathematical steps are guaranteed to be correct. Our approach enables controlled studies revealing systematic performance gaps between mathematical and word problem variants, while providing verifiable training signals for reasoning models as each problem pair is based on an underlying graph where each step is solved symbolically.
Submission Number: 3
Loading