LATIDIA · Investigación

MathVis: ¿Puede la IA razonar a través del dibujo? Un punto de referencia para la generación de figuras matemáticas

arXiv: 2610.12954v1Announce Type: new Resumen: Este artículo presenta MathVis, un punto de referencia diseñado para evaluar la capacidad de los LLM para visualizar problemas matemáticos de frontera y realizar razonamiento asistido por diagrama. El benchm

La noticia

arXiv:2610.12954v1 Announce Type: new Abstract: This paper introduces MathVis, a benchmark designed to evaluate LLMs' ability to visualize frontier mathematical problems and perform diagram-assisted reasoning. The benchmark consists of 512 tasks, each comprising a mathematical statement, a reference diagram, executable TikZ or SVG code, a Gold Intermediate Representation (IR), atomic rubrics, and error codes. Through rigorous evaluation, GPT-6-Astra, Claude Fable 5, and Gemini 3.1 Pro Preview achieve pass@1 scores of 31.25%, 26.56%, and 22.85%, respectively. We found that although the tasks require visualization, LLMs produce outputs dominated by formal languages, symbols, and graphics programs, rather than explicit diagram-mediated reasoning. Furthermore, while the solutions generated by LLMs may contain all the