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

arXiv: https://arxiv.org/abs/2603.27959 · HTML: https://mlsanity.com/paper/2603.27959/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 37 references that are in the map.
- First posted on arXiv 30 Mar 2026.

## Builds on (most cited first, 20 of 20)

- [Scaling Rectified Flow Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2403.03206/) (2024)
- [MathVista: Evaluating Mathematical Reasoning of Foundation Models in Visual Contexts](https://mlsanity.com/paper/2310.02255/) (2023)
- [Scaling Autoregressive Models for Content-Rich Text-to-Image Generation](https://mlsanity.com/paper/2206.10789/) (2022)
- [Qwen-Image Technical Report](https://mlsanity.com/paper/2508.02324/) (2025)
- [Emerging Properties in Unified Multimodal Pretraining](https://mlsanity.com/paper/2505.14683/) (2025)
- [Gemini: A Family of Highly Capable Multimodal Models](https://mlsanity.com/paper/2312.11805/) (2023)
- [PixArt-α: Fast Training of Diffusion Transformer for Photorealistic Text-to-Image Synthesis](https://mlsanity.com/paper/2310.00426/) (2023)
- [Janus-Pro: Unified Multimodal Understanding and Generation with Data and Model Scaling](https://mlsanity.com/paper/2501.17811/) (2025)
- [Show-o: One Single Transformer to Unify Multimodal Understanding and Generation](https://mlsanity.com/paper/2408.12528/) (2024)
- [BLIP3-o: A Family of Fully Open Unified Multimodal Models-Architecture, Training and Dataset](https://mlsanity.com/paper/2505.09568/) (2025)
- [Infinity: Scaling Bitwise AutoRegressive Modeling for High-Resolution Image Synthesis](https://mlsanity.com/paper/2412.04431/) (2024)
- [OmniGen2: Towards Instruction-Aligned Multimodal Generation](https://mlsanity.com/paper/2506.18871/) (2025)
- [Seedream 4.0: Toward Next-generation Multimodal Image Generation](https://mlsanity.com/paper/2509.20427/) (2025)
- [Seedream 3.0 Technical Report](https://mlsanity.com/paper/2504.11346/) (2025)
- [HiDream-I1: A High-Efficient Image Generative Foundation Model with Sparse Diffusion Transformer](https://mlsanity.com/paper/2505.22705/) (2025)
- [GoT-R1: Unleashing Reasoning Capability of MLLM for Visual Generation with Reinforcement Learning](https://mlsanity.com/paper/2505.17022/) (2025)
- [Easier Painting Than Thinking: Can Text-to-Image Models Set the Stage, but Not Direct the Play?](https://mlsanity.com/paper/2509.03516/) (2025)
- [GenExam: A Multidisciplinary Text-to-Image Exam](https://mlsanity.com/paper/2509.14232/) (2025)
- [Demystifying Flux Architecture](https://mlsanity.com/paper/2507.09595/) (2025)
- [MathReal: We Keep It Real! A Real Scene Benchmark for Evaluating Math Reasoning in Multimodal Large Language Models](https://mlsanity.com/paper/2508.06009/) (2025)

## Built on by (most cited first, 1 of 1)

- [MathVis: Can AI Reason through Drawing? A Benchmark for Mathematical Figure Generation](https://mlsanity.com/paper/2610.12954/) (2026)

## Abstract (arXiv)

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 420 problems spanning seven core domains, including 350 Clean-Scene problems and 70 paired Open-Scene problems. Each problem is evaluated under a Script-as-a-Judge protocol with problem-specific verification criteria implemented through reusable executable scripts for deterministic and reproducible 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 53.7% overall accuracy, while open-source models achieve just 1--11%, often near 0% on structured tasks, particularly those requiring precise geometric and functional rendering. Overall, current T2I models remain far from reliable at even elementary mathematical visual generation.