Saved in:
Bibliographic Details
Main Authors: Ma, Zihui, Li, Lingyao, Li, Juan, Hua, Wenyue, Liu, Jingxiao, Feng, Qingyuan, Miura, Yuki
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2506.03360
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916779077402624
author Ma, Zihui
Li, Lingyao
Li, Juan
Hua, Wenyue
Liu, Jingxiao
Feng, Qingyuan
Miura, Yuki
author_facet Ma, Zihui
Li, Lingyao
Li, Juan
Hua, Wenyue
Liu, Jingxiao
Feng, Qingyuan
Miura, Yuki
contents Rapid, fine-grained disaster damage assessment is essential for effective emergency response, yet remains challenging due to limited ground sensors and delays in official reporting. Social media provides a rich, real-time source of human-centric observations, but its multimodal and unstructured nature presents challenges for traditional analytical methods. In this study, we propose a structured Multimodal, Multilingual, and Multidimensional (3M) pipeline that leverages multimodal large language models (MLLMs) to assess disaster impacts. We evaluate three foundation models across two major earthquake events using both macro- and micro-level analyses. Results show that MLLMs effectively integrate image-text signals and demonstrate a strong correlation with ground-truth seismic data. However, performance varies with language, epicentral distance, and input modality. This work highlights the potential of MLLMs for disaster assessment and provides a foundation for future research in applying MLLMs to real-time crisis contexts. The code and data are released at: https://github.com/missa7481/EMNLP25_earthquake
format Preprint
id arxiv_https___arxiv_org_abs_2506_03360
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Multimodal, Multilingual, and Multidimensional Pipeline for Fine-grained Crowdsourcing Earthquake Damage Evaluation
Ma, Zihui
Li, Lingyao
Li, Juan
Hua, Wenyue
Liu, Jingxiao
Feng, Qingyuan
Miura, Yuki
Computation and Language
Computers and Society
Social and Information Networks
Rapid, fine-grained disaster damage assessment is essential for effective emergency response, yet remains challenging due to limited ground sensors and delays in official reporting. Social media provides a rich, real-time source of human-centric observations, but its multimodal and unstructured nature presents challenges for traditional analytical methods. In this study, we propose a structured Multimodal, Multilingual, and Multidimensional (3M) pipeline that leverages multimodal large language models (MLLMs) to assess disaster impacts. We evaluate three foundation models across two major earthquake events using both macro- and micro-level analyses. Results show that MLLMs effectively integrate image-text signals and demonstrate a strong correlation with ground-truth seismic data. However, performance varies with language, epicentral distance, and input modality. This work highlights the potential of MLLMs for disaster assessment and provides a foundation for future research in applying MLLMs to real-time crisis contexts. The code and data are released at: https://github.com/missa7481/EMNLP25_earthquake
title A Multimodal, Multilingual, and Multidimensional Pipeline for Fine-grained Crowdsourcing Earthquake Damage Evaluation
topic Computation and Language
Computers and Society
Social and Information Networks
url https://arxiv.org/abs/2506.03360