Evaluating MLLMs with Multimodal Multi-image Reasoning Benchmark

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Cheng, Ziming, Xu, Binrui, Gong, Lisheng, Song, Zuhe, Zhou, Tianshuo, Zhong, Shiqi, Ren, Siyu, Chen, Mingxiang, Meng, Xiangchao, Zhang, Yuxin, Li, Yanlin, Ren, Lei, Chen, Wei, Huang, Zhiyuan, Zhan, Mingjie, Wang, Xiaojie, Feng, Fangxiang
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866916780350373888
author Cheng, Ziming
Xu, Binrui
Gong, Lisheng
Song, Zuhe
Zhou, Tianshuo
Zhong, Shiqi
Ren, Siyu
Chen, Mingxiang
Meng, Xiangchao
Zhang, Yuxin
Li, Yanlin
Ren, Lei
Chen, Wei
Huang, Zhiyuan
Zhan, Mingjie
Wang, Xiaojie
Feng, Fangxiang
author_facet Cheng, Ziming
Xu, Binrui
Gong, Lisheng
Song, Zuhe
Zhou, Tianshuo
Zhong, Shiqi
Ren, Siyu
Chen, Mingxiang
Meng, Xiangchao
Zhang, Yuxin
Li, Yanlin
Ren, Lei
Chen, Wei
Huang, Zhiyuan
Zhan, Mingjie
Wang, Xiaojie
Feng, Fangxiang
contents With enhanced capabilities and widespread applications, Multimodal Large Language Models (MLLMs) are increasingly required to process and reason over multiple images simultaneously. However, existing MLLM benchmarks focus either on single-image visual reasoning or on multi-image understanding tasks with only final-answer evaluation, leaving the reasoning capabilities of MLLMs over multi-image inputs largely underexplored. To address this gap, we introduce the $\textbf{Multimodal Multi-image Reasoning Benchmark (MMRB)}$, the first benchmark designed to evaluate structured visual reasoning across multiple images. MMRB comprises $\textbf{92 sub-tasks}$ covering spatial, temporal, and semantic reasoning, with multi-solution, CoT-style annotations generated by GPT-4o and refined by human experts. A derivative subset is designed to evaluate multimodal reward models in multi-image scenarios. To support fast and scalable evaluation, we propose a sentence-level matching framework using open-source LLMs. Extensive baseline experiments on $\textbf{40 MLLMs}$, including 9 reasoning-specific models and 8 reward models, demonstrate that open-source MLLMs still lag significantly behind commercial MLLMs in multi-image reasoning tasks. Furthermore, current multimodal reward models are nearly incapable of handling multi-image reward ranking tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2506_04280
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Evaluating MLLMs with Multimodal Multi-image Reasoning Benchmark
Cheng, Ziming
Xu, Binrui
Gong, Lisheng
Song, Zuhe
Zhou, Tianshuo
Zhong, Shiqi
Ren, Siyu
Chen, Mingxiang
Meng, Xiangchao
Zhang, Yuxin
Li, Yanlin
Ren, Lei
Chen, Wei
Huang, Zhiyuan
Zhan, Mingjie
Wang, Xiaojie
Feng, Fangxiang
Computer Vision and Pattern Recognition
Artificial Intelligence
68T50
I.2.7
With enhanced capabilities and widespread applications, Multimodal Large Language Models (MLLMs) are increasingly required to process and reason over multiple images simultaneously. However, existing MLLM benchmarks focus either on single-image visual reasoning or on multi-image understanding tasks with only final-answer evaluation, leaving the reasoning capabilities of MLLMs over multi-image inputs largely underexplored. To address this gap, we introduce the $\textbf{Multimodal Multi-image Reasoning Benchmark (MMRB)}$, the first benchmark designed to evaluate structured visual reasoning across multiple images. MMRB comprises $\textbf{92 sub-tasks}$ covering spatial, temporal, and semantic reasoning, with multi-solution, CoT-style annotations generated by GPT-4o and refined by human experts. A derivative subset is designed to evaluate multimodal reward models in multi-image scenarios. To support fast and scalable evaluation, we propose a sentence-level matching framework using open-source LLMs. Extensive baseline experiments on $\textbf{40 MLLMs}$, including 9 reasoning-specific models and 8 reward models, demonstrate that open-source MLLMs still lag significantly behind commercial MLLMs in multi-image reasoning tasks. Furthermore, current multimodal reward models are nearly incapable of handling multi-image reward ranking tasks.
title Evaluating MLLMs with Multimodal Multi-image Reasoning Benchmark
topic Computer Vision and Pattern Recognition
Artificial Intelligence
68T50
I.2.7
url https://arxiv.org/abs/2506.04280