Salvato in:
Dettagli Bibliografici
Autori principali: Li, Yun, Zhang, Yiming, Lin, Tao, Liu, Xiangrui, Cai, Wenxiao, Liu, Zheng, Zhao, Bo
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2503.23765
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913945394085888
author Li, Yun
Zhang, Yiming
Lin, Tao
Liu, Xiangrui
Cai, Wenxiao
Liu, Zheng
Zhao, Bo
author_facet Li, Yun
Zhang, Yiming
Lin, Tao
Liu, Xiangrui
Cai, Wenxiao
Liu, Zheng
Zhao, Bo
contents The use of Multimodal Large Language Models (MLLMs) as an end-to-end solution for Embodied AI and Autonomous Driving has become a prevailing trend. While MLLMs have been extensively studied for visual semantic understanding tasks, their ability to perform precise and quantitative spatial-temporal understanding in real-world applications remains largely unexamined, leading to uncertain prospects. To evaluate models' Spatial-Temporal Intelligence, we introduce STI-Bench, a benchmark designed to evaluate MLLMs' spatial-temporal understanding through challenging tasks such as estimating and predicting the appearance, pose, displacement, and motion of objects. Our benchmark encompasses a wide range of robot and vehicle operations across desktop, indoor, and outdoor scenarios. The extensive experiments reveals that the state-of-the-art MLLMs still struggle in real-world spatial-temporal understanding, especially in tasks requiring precise distance estimation and motion analysis.
format Preprint
id arxiv_https___arxiv_org_abs_2503_23765
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle STI-Bench: Are MLLMs Ready for Precise Spatial-Temporal World Understanding?
Li, Yun
Zhang, Yiming
Lin, Tao
Liu, Xiangrui
Cai, Wenxiao
Liu, Zheng
Zhao, Bo
Computer Vision and Pattern Recognition
The use of Multimodal Large Language Models (MLLMs) as an end-to-end solution for Embodied AI and Autonomous Driving has become a prevailing trend. While MLLMs have been extensively studied for visual semantic understanding tasks, their ability to perform precise and quantitative spatial-temporal understanding in real-world applications remains largely unexamined, leading to uncertain prospects. To evaluate models' Spatial-Temporal Intelligence, we introduce STI-Bench, a benchmark designed to evaluate MLLMs' spatial-temporal understanding through challenging tasks such as estimating and predicting the appearance, pose, displacement, and motion of objects. Our benchmark encompasses a wide range of robot and vehicle operations across desktop, indoor, and outdoor scenarios. The extensive experiments reveals that the state-of-the-art MLLMs still struggle in real-world spatial-temporal understanding, especially in tasks requiring precise distance estimation and motion analysis.
title STI-Bench: Are MLLMs Ready for Precise Spatial-Temporal World Understanding?
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2503.23765