VisioMath: Benchmarking Figure-based Mathematical Reasoning in LMMs

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Li, Can, Liu, Ying, Zhang, Ting, Wang, Mei, Huang, Hua
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912946664243200
author Li, Can
Liu, Ying
Zhang, Ting
Wang, Mei
Huang, Hua
author_facet Li, Can
Liu, Ying
Zhang, Ting
Wang, Mei
Huang, Hua
contents Large Multimodal Models have achieved remarkable progress in integrating vision and language, enabling strong performance across perception, reasoning, and domain-specific tasks. However, their capacity to reason over multiple, visually similar inputs remains insufficiently explored. Such fine-grained comparative reasoning is central to real-world tasks, especially in mathematics and education, where learners must often distinguish between nearly identical diagrams to identify correct solutions. To address this gap, we present VisioMath, a curated benchmark of 1,800 high-quality K-12 mathematics problems in which all candidate answers are diagrams with subtle visual similarities. A comprehensive evaluation of state-of-the-art LMMs, covering both leading closed-source systems and widely adopted open-source models, reveals a consistent decline in accuracy as inter-image similarity increases. Analysis indicates that the dominant failure mode stems from image-text misalignment: rather than grounding reasoning in textual cues, models often resort to shallow positional heuristics, resulting in systematic errors. We further explore three alignment-oriented strategies, spanning training-free approaches and finetuning, and achieve substantial accuracy gains. We hope that VisioMath will serve as a rigorous benchmark and catalyst for developing LMMs toward deeper diagram understanding, precise comparative reasoning, and grounded multi-image-text integration.
format Preprint
id arxiv_https___arxiv_org_abs_2506_06727
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle VisioMath: Benchmarking Figure-based Mathematical Reasoning in LMMs
Li, Can
Liu, Ying
Zhang, Ting
Wang, Mei
Huang, Hua
Artificial Intelligence
Computer Vision and Pattern Recognition
Large Multimodal Models have achieved remarkable progress in integrating vision and language, enabling strong performance across perception, reasoning, and domain-specific tasks. However, their capacity to reason over multiple, visually similar inputs remains insufficiently explored. Such fine-grained comparative reasoning is central to real-world tasks, especially in mathematics and education, where learners must often distinguish between nearly identical diagrams to identify correct solutions. To address this gap, we present VisioMath, a curated benchmark of 1,800 high-quality K-12 mathematics problems in which all candidate answers are diagrams with subtle visual similarities. A comprehensive evaluation of state-of-the-art LMMs, covering both leading closed-source systems and widely adopted open-source models, reveals a consistent decline in accuracy as inter-image similarity increases. Analysis indicates that the dominant failure mode stems from image-text misalignment: rather than grounding reasoning in textual cues, models often resort to shallow positional heuristics, resulting in systematic errors. We further explore three alignment-oriented strategies, spanning training-free approaches and finetuning, and achieve substantial accuracy gains. We hope that VisioMath will serve as a rigorous benchmark and catalyst for developing LMMs toward deeper diagram understanding, precise comparative reasoning, and grounded multi-image-text integration.
title VisioMath: Benchmarking Figure-based Mathematical Reasoning in LMMs
topic Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.06727