Saved in:
Bibliographic Details
Main Authors: Tang, Kexian, Gao, Junyao, Zeng, Yanhong, Duan, Haodong, Sun, Yanan, Xing, Zhening, Liu, Wenran, Lyu, Kaifeng, Chen, Kai
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2503.19990
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911014705954816
author Tang, Kexian
Gao, Junyao
Zeng, Yanhong
Duan, Haodong
Sun, Yanan
Xing, Zhening
Liu, Wenran
Lyu, Kaifeng
Chen, Kai
author_facet Tang, Kexian
Gao, Junyao
Zeng, Yanhong
Duan, Haodong
Sun, Yanan
Xing, Zhening
Liu, Wenran
Lyu, Kaifeng
Chen, Kai
contents Multi-step spatial reasoning entails understanding and reasoning about spatial relationships across multiple sequential steps, which is crucial for tackling complex real-world applications, such as robotic manipulation, autonomous navigation, and automated assembly. To assess how well current Multimodal Large Language Models (MLLMs) have acquired this fundamental capability, we introduce LEGO-Puzzles, a scalable benchmark designed to evaluate both spatial understanding and sequential reasoning in MLLMs through LEGO-based tasks. LEGO-Puzzles consists of 1,100 carefully curated visual question-answering (VQA) samples spanning 11 distinct tasks, ranging from basic spatial understanding to complex multi-step reasoning. Based on LEGO-Puzzles, we conduct a comprehensive evaluation of 20 state-of-the-art MLLMs and uncover significant limitations in their spatial reasoning capabilities: even the most powerful MLLMs can answer only about half of the test cases, whereas human participants achieve over 90% accuracy. Furthermore, based on LEGO-Puzzles, we design generation tasks to investigate whether MLLMs can transfer their spatial understanding and reasoning abilities to image generation. Our experiments show that only GPT-4o and Gemini-2.0-Flash exhibit a limited ability to follow these instructions, while other MLLMs either replicate the input image or generate completely irrelevant outputs. Overall, LEGO-Puzzles exposes critical deficiencies in existing MLLMs' spatial understanding and sequential reasoning capabilities, and underscores the need for further advancements in multimodal spatial reasoning.
format Preprint
id arxiv_https___arxiv_org_abs_2503_19990
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle LEGO-Puzzles: How Good Are MLLMs at Multi-Step Spatial Reasoning?
Tang, Kexian
Gao, Junyao
Zeng, Yanhong
Duan, Haodong
Sun, Yanan
Xing, Zhening
Liu, Wenran
Lyu, Kaifeng
Chen, Kai
Artificial Intelligence
Multi-step spatial reasoning entails understanding and reasoning about spatial relationships across multiple sequential steps, which is crucial for tackling complex real-world applications, such as robotic manipulation, autonomous navigation, and automated assembly. To assess how well current Multimodal Large Language Models (MLLMs) have acquired this fundamental capability, we introduce LEGO-Puzzles, a scalable benchmark designed to evaluate both spatial understanding and sequential reasoning in MLLMs through LEGO-based tasks. LEGO-Puzzles consists of 1,100 carefully curated visual question-answering (VQA) samples spanning 11 distinct tasks, ranging from basic spatial understanding to complex multi-step reasoning. Based on LEGO-Puzzles, we conduct a comprehensive evaluation of 20 state-of-the-art MLLMs and uncover significant limitations in their spatial reasoning capabilities: even the most powerful MLLMs can answer only about half of the test cases, whereas human participants achieve over 90% accuracy. Furthermore, based on LEGO-Puzzles, we design generation tasks to investigate whether MLLMs can transfer their spatial understanding and reasoning abilities to image generation. Our experiments show that only GPT-4o and Gemini-2.0-Flash exhibit a limited ability to follow these instructions, while other MLLMs either replicate the input image or generate completely irrelevant outputs. Overall, LEGO-Puzzles exposes critical deficiencies in existing MLLMs' spatial understanding and sequential reasoning capabilities, and underscores the need for further advancements in multimodal spatial reasoning.
title LEGO-Puzzles: How Good Are MLLMs at Multi-Step Spatial Reasoning?
topic Artificial Intelligence
url https://arxiv.org/abs/2503.19990