LLMs as World Models: Data-Driven and Human-Centered Pre-Event Simulation for Disaster Impact Assessment
By: Lingyao Li , Dawei Li , Zhenhui Ou and more
Potential Business Impact:
Predicts earthquake shaking to help people prepare.
Efficient simulation is essential for enhancing proactive preparedness for sudden-onset disasters such as earthquakes. Recent advancements in large language models (LLMs) as world models show promise in simulating complex scenarios. This study examines multiple LLMs to proactively estimate perceived earthquake impacts. Leveraging multimodal datasets including geospatial, socioeconomic, building, and street-level imagery data, our framework generates Modified Mercalli Intensity (MMI) predictions at zip code and county scales. Evaluations on the 2014 Napa and 2019 Ridgecrest earthquakes using USGS ''Did You Feel It? (DYFI)'' reports demonstrate significant alignment, as evidenced by a high correlation of 0.88 and a low RMSE of 0.77 as compared to real reports at the zip code level. Techniques such as RAG and ICL can improve simulation performance, while visual inputs notably enhance accuracy compared to structured numerical data alone. These findings show the promise of LLMs in simulating disaster impacts that can help strengthen pre-event planning.
Similar Papers
A Multimodal, Multilingual, and Multidimensional Pipeline for Fine-grained Crowdsourcing Earthquake Damage Evaluation
Computation and Language
Helps find disaster damage faster using pictures and words.
Subnational Geocoding of Global Disasters Using Large Language Models
Artificial Intelligence
Maps disaster locations automatically and accurately.
Mechanics-Informed Machine Learning for Geospatial Modeling of Soil Liquefaction: Global and National Surrogate Models for Simulation and Near-Real-Time Response
Computational Engineering, Finance, and Science
Predicts earthquake damage to the ground.