MathVerse Benchmark

What is MathVerse?

MathVerse is intended for evaluating MLLMs' visual math problem-solving, containing 2,612 high-quality, multi-subject math problems with diagrams.

Released

2024-03-21

Modality

multimodal

Evaluates

多模态, Multimodal, 推理, Reasoning, 数学, Math, 多模态模型, VLM, 逻辑推理, 数理能力, 开源收录, Open-Source, 不支持, Unsupported

Openness

open

Importer

OpenCompass Hub

Reported scores

0

Source provenance

• Original evidence https://hub.opencompass.org.cn/dataset-detail/MathVerse

MathVerse paper, code and dataset

• Paper https://arxiv.org/abs/2403.14624
• Code https://github.com/ZrrSkywalker/MathVerse
• Dataset https://huggingface.co/datasets/AI4Math/MathVerse

Which sources cite MathVerse?

• MathVerse

Related 多模态, Multimodal, 推理, Reasoning, 数学, Math, 多模态模型, VLM, 逻辑推理, 数理能力, 开源收录, Open-Source, 不支持, Unsupported benchmarks

• 3DCodeBench
• A-Bench
• A-OKVQA
• AbsPyramid
• ActionAtlas
• ActiveFly-Bench
• Ada-LEval
• AdvancedMathBench