analiza aiqbench
This study presents a multimodal generative AI pipeline that synthesizes residential building datasets from public records, effectively overcoming data scarcity and privacy constraints in urban energy modeling. By demonstrating superior visual focus over standard GPT-based alternatives and achieving over 95% overlap with national reference data for key variables, the framework lowers barriers to scalable ML-driven retrofit analysis and simulation. The work offers a practical solution for researchers needing high-fidelity synthetic data without relying on expensive or restricted proprietary sources.
Synthetic Homes: A Multimodal Generative AI Pipeline for Residential Building Data Generation under Data Scarcity
arXiv:2509.09794v5 Announce Type: replace-cross Abstract: Computational models have emerged as powerful tools for multi-scale energy modeling research at the building and urban scale, supporting data-driven analysis across building and urban energy systems. However, these models require large amounts of building parameter data that is often inaccessible, expensive to collect, or subject to privacy constraints. We introduce a modular, multimodal generative Artificial Intelligence (AI) framework that integrates image, tabular, and simulation-based components and produces synthetic residential building datasets from publicly available county records and images, and present an end-to-end pipeline instantiating this framework. To reduce typical Large Language Model (LLM) challenges, we evaluate our model's components using occlusion-based visual focus analysis. Our analysis demonstrates that our selected vision-language model achieves greater visual focus than a GPT-based alternative for building image processing. We also assess realism of our results against a national reference dataset, finding that our synthetic data overlaps more than 95% for three of the four selected variables. This work reduces dependence on costly or restricted data sources, lowering barriers to building-scale energy research and Machine Learning (ML)-driven urban energy modeling, and therefore enabling scalable downstream tasks such as energy modeling, retrofit analysis, and urban-scale simulation under data scarcity.
Abstract:Computational models have emerged as powerful tools for multi-scale energy modeling research at the building and urban scale, supporting data-driven analysis across building and urban energy systems. However, these models require large amounts of building parameter data that is often inaccessible, expensive to collect, or subject to privacy constraints. We introduce a modular, multimodal generative Artificial Intelligence (AI) framework that integrates image, tabular, and simulation-based components and produces synthetic residential building datasets from publicly available county records and images, and present an end-to-end pipeline instantiating this framework. To reduce typical Large Language Model (LLM) challenges, we evaluate our model's components using occlusion-based visual focus analysis. Our analysis demonstrates that our selected vision-language model achieves greater visual focus than a GPT-based alternative for building image processing. We also assess realism of our results against a national reference dataset, finding that our synthetic data overlaps more than 95% for three of the four selected variables. This work reduces dependence on costly or restricted data sources, lowering barriers to building-scale energy research and Machine Learning (ML)-driven urban energy modeling, and therefore enabling scalable downstream tasks such as energy modeling, retrofit analysis, and urban-scale simulation under data scarcity.
37 pages; 2 appendices; 6 figures; 2 tables. Code available at this https URL
Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
arXiv:2509.09794 [cs.AI]
(or arXiv:2509.09794v5 [cs.AI] for this version)
Text integral sub licența deschisă a publicației · 222 cuvinte
Subiecte conexe
Energy Vision--Language--Action: A Controlled Multimodal Benchmark for Intent-Conditioned Residential Energy Management
arXiv · 30 sept. 2026
Multimodal Cultural Heritage Architectural Style Classification for Residential Buildings in the UAE Based on CLIP Embeddings and SVM
arXiv · 16 sept. 2026
Harness Robotic OS: A Unified Embodied-Agent Runtime for Closed-Loop Quadruped Inspection
arXiv · 11 sept. 2026
Multimodal Large Language Models Predict Urban Safety Perception but Encode Non-Neutral Demographic Priors
arXiv · 1 sept. 2026