Can Multimodal Foundation Models Understand Schematic Diagrams? An Empirical Study on Information-Seeking QA over Scientific Papers

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhao, Yilun, Wang, Chengye, Li, Chuhan, Cohan, Arman
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911055735685120
author Zhao, Yilun
Wang, Chengye
Li, Chuhan
Cohan, Arman
author_facet Zhao, Yilun
Wang, Chengye
Li, Chuhan
Cohan, Arman
contents This paper introduces MISS-QA, the first benchmark specifically designed to evaluate the ability of models to interpret schematic diagrams within scientific literature. MISS-QA comprises 1,500 expert-annotated examples over 465 scientific papers. In this benchmark, models are tasked with interpreting schematic diagrams that illustrate research overviews and answering corresponding information-seeking questions based on the broader context of the paper. We assess the performance of 18 frontier multimodal foundation models, including o4-mini, Gemini-2.5-Flash, and Qwen2.5-VL. We reveal a significant performance gap between these models and human experts on MISS-QA. Our analysis of model performance on unanswerable questions and our detailed error analysis further highlight the strengths and limitations of current models, offering key insights to enhance models in comprehending multimodal scientific literature.
format Preprint
id arxiv_https___arxiv_org_abs_2507_10787
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Can Multimodal Foundation Models Understand Schematic Diagrams? An Empirical Study on Information-Seeking QA over Scientific Papers
Zhao, Yilun
Wang, Chengye
Li, Chuhan
Cohan, Arman
Computation and Language
Computer Vision and Pattern Recognition
This paper introduces MISS-QA, the first benchmark specifically designed to evaluate the ability of models to interpret schematic diagrams within scientific literature. MISS-QA comprises 1,500 expert-annotated examples over 465 scientific papers. In this benchmark, models are tasked with interpreting schematic diagrams that illustrate research overviews and answering corresponding information-seeking questions based on the broader context of the paper. We assess the performance of 18 frontier multimodal foundation models, including o4-mini, Gemini-2.5-Flash, and Qwen2.5-VL. We reveal a significant performance gap between these models and human experts on MISS-QA. Our analysis of model performance on unanswerable questions and our detailed error analysis further highlight the strengths and limitations of current models, offering key insights to enhance models in comprehending multimodal scientific literature.
title Can Multimodal Foundation Models Understand Schematic Diagrams? An Empirical Study on Information-Seeking QA over Scientific Papers
topic Computation and Language
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2507.10787