DeepMath-Creative: A Benchmark for Evaluating Mathematical Creativity of Large Language Models

TL;DR

This paper introduces DeepMath-Creative, a benchmark for assessing the mathematical creativity of large language models, revealing current models' limited creative problem-solving abilities across various mathematical domains.

Contribution

It presents a new benchmark dataset for evaluating mathematical creativity in LLMs and systematically assesses mainstream models' performance on constructive problems.

Findings

Models achieve up to 70% accuracy on basic tasks

Performance drops significantly on complex problems

Current models rely on pattern recombination rather than true creativity

Abstract

To advance the mathematical proficiency of large language models (LLMs), the DeepMath team has launched an open-source initiative aimed at developing an open mathematical LLM and systematically evaluating its mathematical creativity. This paper represents the initial contribution of this initiative. While recent developments in mathematical LLMs have predominantly emphasized reasoning skills, as evidenced by benchmarks on elementary to undergraduate-level mathematical tasks, the creative capabilities of these models have received comparatively little attention, and evaluation datasets remain scarce. To address this gap, we propose an evaluation criteria for mathematical creativity and introduce DeepMath-Creative, a novel, high-quality benchmark comprising constructive problems across algebra, geometry, analysis, and other domains. We conduct a systematic evaluation of mainstream LLMs' creative problem-solving abilities using this dataset. Experimental results show that even under lenient scoring criteria -- emphasizing core solution components and disregarding minor inaccuracies, such as small logical gaps, incomplete justifications, or redundant explanations -- the best-performing model, O3 Mini, achieves merely 70% accuracy, primarily on basic undergraduate-level constructive tasks. Performance declines sharply on more complex problems, with models failing to provide substantive strategies for open problems. These findings suggest that, although current LLMs display a degree of constructive proficiency on familiar and lower-difficulty problems, such performance is likely attributable to the recombination of memorized patterns rather than authentic creative insight or novel synthesis.