Saved in:
Bibliographic Details
Main Authors: Wang, Yuandong, Cui, Yao, Zhao, Yuxin, Yang, Zhen, Zhu, Yangfu, Shao, Zhenzhou
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2511.23112
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908680012693504
author Wang, Yuandong
Cui, Yao
Zhao, Yuxin
Yang, Zhen
Zhu, Yangfu
Shao, Zhenzhou
author_facet Wang, Yuandong
Cui, Yao
Zhao, Yuxin
Yang, Zhen
Zhu, Yangfu
Shao, Zhenzhou
contents Recent advances in Vision-Language Models (VLMs) have achieved impressive progress in multimodal mathematical reasoning. Yet, how much visual information truly contributes to reasoning remains unclear. Existing benchmarks report strong overall performance but seldom isolate the role of the image modality, leaving open whether VLMs genuinely leverage visual understanding or merely depend on linguistic priors. To address this, we present MathSight, a university-level multimodal mathematical reasoning benchmark designed to disentangle and quantify the effect of visual input. Each problem includes multiple visual variants -- original, hand-drawn, photo-captured -- and a text-only condition for controlled comparison. Experiments on state-of-the-art VLMs reveal a consistent trend: the contribution of visual information diminishes with increasing problem difficulty. Remarkably, Qwen3-VL without any image input surpasses both its multimodal variants and GPT-5, underscoring the need for benchmarks like MathSight to advance genuine vision-grounded reasoning in future models.
format Preprint
id arxiv_https___arxiv_org_abs_2511_23112
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MathSight: A Benchmark Exploring Have Vision-Language Models Really Seen in University-Level Mathematical Reasoning?
Wang, Yuandong
Cui, Yao
Zhao, Yuxin
Yang, Zhen
Zhu, Yangfu
Shao, Zhenzhou
Computer Vision and Pattern Recognition
Machine Learning
Recent advances in Vision-Language Models (VLMs) have achieved impressive progress in multimodal mathematical reasoning. Yet, how much visual information truly contributes to reasoning remains unclear. Existing benchmarks report strong overall performance but seldom isolate the role of the image modality, leaving open whether VLMs genuinely leverage visual understanding or merely depend on linguistic priors. To address this, we present MathSight, a university-level multimodal mathematical reasoning benchmark designed to disentangle and quantify the effect of visual input. Each problem includes multiple visual variants -- original, hand-drawn, photo-captured -- and a text-only condition for controlled comparison. Experiments on state-of-the-art VLMs reveal a consistent trend: the contribution of visual information diminishes with increasing problem difficulty. Remarkably, Qwen3-VL without any image input surpasses both its multimodal variants and GPT-5, underscoring the need for benchmarks like MathSight to advance genuine vision-grounded reasoning in future models.
title MathSight: A Benchmark Exploring Have Vision-Language Models Really Seen in University-Level Mathematical Reasoning?
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2511.23112