MathGen: Revealing the Illusion of Mathematical Competence through Text-to-Image Generation

TL;DR

MathGen is a benchmark revealing that current text-to-image models struggle significantly with generating accurate mathematical visuals, highlighting a major gap in their capabilities.

Contribution

The paper introduces MathGen, a comprehensive benchmark with an evaluation protocol to assess the mathematical visual generation ability of T2I models.

Findings

Best closed-source model achieves 42.0% accuracy.

Open-source models achieve 1-11% accuracy.

Models perform near 0% on structured tasks.

Abstract

Modern generative models have demonstrated the ability to solve challenging mathematical problems. In many real-world settings, however, mathematical solutions must be expressed visually through diagrams, plots, geometric constructions, and structured symbolic layouts, where correctness depends on precise visual composition. This naturally raises the question of whether generative models can still do so when the answer must be rendered visually rather than written in text? To study this problem, we introduce MathGen, a rigorous benchmark of 900 problems spanning seven core domains, each paired with an executable verifier under a Script-as-a-Judge protocol for deterministic and objective evaluation. Experiments on representative open-source and proprietary text-to-image models show that mathematical fidelity remains a major bottleneck: even the best closed-source model reaches only 42.0% overall accuracy, while open-source models achieve just ~ 1-11%, often near 0% on structured tasks. Overall, current T2I models remain far from competent at even elementary mathematical visual generation.