CMMaTH: A Chinese Multi-modal Math Skill Evaluation Benchmark for Foundation Models
This provides a domain-specific evaluation tool for researchers and educators assessing multimodal large language models in Chinese educational contexts, though it is incremental as it builds on existing benchmarks like MathVista.
The authors tackled the lack of fine-grained evaluation tools for multimodal mathematical capabilities in Chinese K12 education by creating CMMaTH, a benchmark with 23k multimodal math questions, which is the largest such dataset to date.
Due to the rapid advancements in multimodal large language models, evaluating their multimodal mathematical capabilities continues to receive wide attention. Despite the datasets like MathVista proposed benchmarks for assessing mathematical capabilities in multimodal scenarios, there is still a lack of corresponding evaluation tools and datasets for fine-grained assessment in the context of K12 education in Chinese language. To systematically evaluate the capability of multimodal large models in solving Chinese multimodal mathematical problems, we propose a Chinese Multi-modal Math Skill Evaluation Benchmark, named CMMaTH, contraining 23k multimodal K12 math related questions, forming the largest Chinese multimodal mathematical problem benchmark to date. CMMaTH questions from elementary to high school levels, provide increased diversity in problem types, solution objectives, visual elements, detailed knowledge points, and standard solution annotations. We have constructed an open-source tool GradeGPT integrated with the CMMaTH dataset, facilitating stable, rapid, and cost-free model evaluation. Our data and code are available.