用大模型模拟地震影响，提前预测灾情，助力防灾准备。
LLMs as World Models: Data-Driven and Human-Centered Pre-Event Simulation for Disaster Impact Assessment
- 基于多模态数据和大模型生成震感强度预测
- 在两个地震案例中相关性达0.88，误差仅0.77
- 视觉数据显著提升预测精度，适合应急规划者使用
高效模拟对突发性灾害（如地震）的主动应对至关重要。大型语言模型作为世界模型在复杂场景模拟中展现出潜力。本研究评估多种LLM，用于提前估算地震感知影响。利用包括地理空间、社会经济、建筑及街景图像在内的多模态数据，框架在邮编和县级别生成修正麦卡利震感强度（MMI）预测。基于2014年纳帕和2019年里奇克雷斯特地震的美国地质调查局“你感觉到了吗？（DYFI）”报告评估显示，预测结果与真实报告高度一致，邮编级别相关系数达0.88，均方根误差仅为0.77。检索增强生成（RAG）和提示学习（ICL）可提升性能，视觉输入相比纯结构化数据显著提高准确性。结果表明，大模型在灾害影响模拟中具有应用前景，有助于加强灾前规划。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。