Paper
Can Large Multimodal Models Inspect Buildings? A Hierarchical Benchmark for Structural Pathology Reasoning
Authors
Hui Zhong, Yichun Gao, Luyan Liu, Hai Yang, Wang Wang, Haowei Zhang, Xinhu Zheng
Abstract
Automated building facade inspection is a critical component of urban resilience and smart city maintenance. Traditionally, this field has relied on specialized discriminative models (e.g., YOLO, Mask R-CNN) that excel at pixel-level localization but are constrained to passive perception and worse generization without the visual understandng to interpret structural topology. Large Multimodal Models (LMMs) promise a paradigm shift toward active reasoning, yet their application in such high-stakes engineering domains lacks rigorous evaluation standards. To bridge this gap, we introduce a human-in-the-loop semi-automated annotation framework, leveraging expert-proposal verification to unify 12 fragmented datasets into a standardized, hierarchical ontology. Building on this foundation, we present \textit{DefectBench}, the first multi-dimensional benchmark designed to interrogate LMMs beyond basic semantic recognition. \textit{DefectBench} evaluates 18 state-of-the-art (SOTA) LMMs across three escalating cognitive dimensions: Semantic Perception, Spatial Localization, and Generative Geometry Segmentation. Extensive experiments reveal that while current LMMs demonstrate exceptional topological awareness and semantic understanding (effectively diagnosing "what" and "how"), they exhibit significant deficiencies in metric localization precision ("where"). Crucially, however, we validate the viability of zero-shot generative segmentation, showing that general-purpose foundation models can rival specialized supervised networks without domain-specific training. This work provides both a rigorous benchmarking standard and a high-quality open-source database, establishing a new baseline for the advancement of autonomous AI agents in civil engineering.
Metadata
Related papers
Vibe Coding XR: Accelerating AI + XR Prototyping with XR Blocks and Gemini
Ruofei Du, Benjamin Hersh, David Li, Nels Numan, Xun Qian, Yanhe Chen, Zhongy... • 2026-03-25
Comparing Developer and LLM Biases in Code Evaluation
Aditya Mittal, Ryan Shar, Zichu Wu, Shyam Agarwal, Tongshuang Wu, Chris Donah... • 2026-03-25
The Stochastic Gap: A Markovian Framework for Pre-Deployment Reliability and Oversight-Cost Auditing in Agentic Artificial Intelligence
Biplab Pal, Santanu Bhattacharya • 2026-03-25
Retrieval Improvements Do Not Guarantee Better Answers: A Study of RAG for AI Policy QA
Saahil Mathur, Ryan David Rittner, Vedant Ajit Thakur, Daniel Stuart Schiff, ... • 2026-03-25
MARCH: Multi-Agent Reinforced Self-Check for LLM Hallucination
Zhuo Li, Yupeng Zhang, Pengyu Cheng, Jiajun Song, Mengyu Zhou, Hao Li, Shujie... • 2026-03-25
Raw Data (Debug)
{
"raw_xml": "<entry>\n <id>http://arxiv.org/abs/2603.20148v1</id>\n <title>Can Large Multimodal Models Inspect Buildings? A Hierarchical Benchmark for Structural Pathology Reasoning</title>\n <updated>2026-03-20T17:24:46Z</updated>\n <link href='https://arxiv.org/abs/2603.20148v1' rel='alternate' type='text/html'/>\n <link href='https://arxiv.org/pdf/2603.20148v1' rel='related' title='pdf' type='application/pdf'/>\n <summary>Automated building facade inspection is a critical component of urban resilience and smart city maintenance. Traditionally, this field has relied on specialized discriminative models (e.g., YOLO, Mask R-CNN) that excel at pixel-level localization but are constrained to passive perception and worse generization without the visual understandng to interpret structural topology. Large Multimodal Models (LMMs) promise a paradigm shift toward active reasoning, yet their application in such high-stakes engineering domains lacks rigorous evaluation standards. To bridge this gap, we introduce a human-in-the-loop semi-automated annotation framework, leveraging expert-proposal verification to unify 12 fragmented datasets into a standardized, hierarchical ontology. Building on this foundation, we present \\textit{DefectBench}, the first multi-dimensional benchmark designed to interrogate LMMs beyond basic semantic recognition. \\textit{DefectBench} evaluates 18 state-of-the-art (SOTA) LMMs across three escalating cognitive dimensions: Semantic Perception, Spatial Localization, and Generative Geometry Segmentation. Extensive experiments reveal that while current LMMs demonstrate exceptional topological awareness and semantic understanding (effectively diagnosing \"what\" and \"how\"), they exhibit significant deficiencies in metric localization precision (\"where\"). Crucially, however, we validate the viability of zero-shot generative segmentation, showing that general-purpose foundation models can rival specialized supervised networks without domain-specific training. This work provides both a rigorous benchmarking standard and a high-quality open-source database, establishing a new baseline for the advancement of autonomous AI agents in civil engineering.</summary>\n <category scheme='http://arxiv.org/schemas/atom' term='cs.CV'/>\n <published>2026-03-20T17:24:46Z</published>\n <arxiv:primary_category term='cs.CV'/>\n <author>\n <name>Hui Zhong</name>\n </author>\n <author>\n <name>Yichun Gao</name>\n </author>\n <author>\n <name>Luyan Liu</name>\n </author>\n <author>\n <name>Hai Yang</name>\n </author>\n <author>\n <name>Wang Wang</name>\n </author>\n <author>\n <name>Haowei Zhang</name>\n </author>\n <author>\n <name>Xinhu Zheng</name>\n </author>\n </entry>"
}