研究利用 ChatGPT 多模态视觉能力按贫困程度排序卫星图像
Leveraging ChatGPT's Multimodal Vision Capabilities to Rank Satellite Images by Poverty Level: Advancing Tools for Social Science Research
AI 导读
一项 arXiv 研究用具备视觉能力的 ChatGPT 对卫星图像进行村庄级贫困预测。采用成对比较方法后，ChatGPT 排序卫星图像贫困水平的准确性与基于人类专家标注特征训练的模型相当，并优于此前报告的专家直接财富分类结果。研究同时提出 DHS 等公开数据集经坐标加噪匿名化后的财富指数可靠性问题，代码和数据已公开。
正文 · 原文
Abstract:This paper investigates the novel application of Large Language Models (LLMs) with vision capabilities to analyze satellite imagery for village-level poverty prediction. Although LLMs were originally designed for natural language understanding, their adaptability to perform multimodal tasks, including geospatial analysis, has opened new frontiers in data-driven research. By leveraging advancements in vision-enabled LLMs, we assess their ability to provide interpretable, scalable, and reliable insights into human poverty from satellite images. Using a pairwise comparison approach, we demonstrate that ChatGPT can rank satellite images based on poverty levels with an accuracy comparable to a model trained on human-expert-identified image features and better than direct expert wealth classifications reported previously. These findings highlight both the promise and the limitations of LLMs in socioeconomic research, providing a foundation for their integration into poverty assessment workflows. This study contributes to the ongoing exploration of unconventional data sources for welfare analysis and opens pathways for cost-effective, large-scale poverty monitoring. Our results also put forward the question of how reliable public datasets, such as DHS, are for retrieving wealth indices that are anonymized by adding noise to the location coordinates. The code and data used for the analyses in this paper are publicly available.
| Comments: | Extra experiments are added to the paper and the repository analyzing the quality of comparisons |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2501.14546 [cs.CV] |
| (or arXiv:2501.14546v4 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2501.14546 arXiv-issued DOI via DataCite |
Submission history
From: Hamid Sarmadi [view email]
[v1]
Fri, 24 Jan 2025 14:49:00 UTC (11,865 KB)
[v2]
Wed, 22 Jul 2026 15:44:08 UTC (11,827 KB)
[v3]
Tue, 28 Jul 2026 14:57:28 UTC (11,827 KB)
[v4]
Thu, 5 Nov 2026 15:47:58 UTC (11,834 KB)
来源：arXiv cs.AI · arxiv.org