LLMs as World Models: Data-Driven and Human-Centered Pre-Event Simulation for Disaster Impact Assessment

• URL: http://arxiv.org/abs/2506.06355v1
• Date: Mon, 02 Jun 2025 22:07:53 GMT
• Title: LLMs as World Models: Data-Driven and Human-Centered Pre-Event Simulation for Disaster Impact Assessment
• Authors: Lingyao Li, Dawei Li, Zhenhui Ou, Xiaoran Xu, Jingxiao Liu, Zihui Ma, Runlong Yu, Min Deng,
• Abstract summary: This study examines multiple large language models (LLMs) to estimate perceived earthquake impacts.<n>Our framework generates Modified Mercalli Intensity (MMI) predictions at zip code and county scales.<n> Evaluations on the 2014 Napa and 2019 Ridgecrest earthquakes using USGS ''Did You Feel It? (DYFI)'' reports demonstrate significant alignment.
• Score: 6.787695140978638
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 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.

Related papers

• OKG-LLM: Aligning Ocean Knowledge Graph with Observation Data via LLMs for Global Sea Surface Temperature Prediction [70.48962924608033]
  This work presents the first systematic effort to construct an Ocean Knowledge Graph (OKG) specifically designed to represent diverse ocean knowledge for SST prediction.<n>We develop a graph embedding network to learn the comprehensive semantic and structural knowledge within the OKG, capturing both the unique characteristics of individual sea regions and the complex correlations between them. Finally, we align the learned knowledge with fine-grained numerical SST data and leverage a pre-trained LLM to model SST patterns for accurate prediction.
  arXiv  Detail & Related papers  (2025-07-31T02:06:03Z)
• Mitigating Forgetting in LLM Fine-Tuning via Low-Perplexity Token Learning [61.99353167168545]
  We show that fine-tuning with LLM-generated data improves target task performance and reduces non-target task degradation.<n>This is the first work to provide an empirical explanation based on token perplexity reduction to mitigate catastrophic forgetting in LLMs after fine-tuning.
  arXiv  Detail & Related papers  (2025-01-24T08:18:56Z)
• Learning Traffic Crashes as Language: Datasets, Benchmarks, and What-if Causal Analyses [76.59021017301127]
  We propose a large-scale traffic crash language dataset, named CrashEvent, summarizing 19,340 real-world crash reports. We further formulate the crash event feature learning as a novel text reasoning problem and further fine-tune various large language models (LLMs) to predict detailed accident outcomes. Our experiments results show that our LLM-based approach not only predicts the severity of accidents but also classifies different types of accidents and predicts injury outcomes.
  arXiv  Detail & Related papers  (2024-06-16T03:10:16Z)
• QuakeBERT: Accurate Classification of Social Media Texts for Rapid Earthquake Impact Assessment [7.777478408048141]
  Social media aids disaster response but suffers from noise, hindering accurate impact assessment and decision making for resilient cities. This study proposes the first domain-specific large language model (LLM) and an integrated method for rapid earthquake impact assessment. Results show that the proposed approach can effectively enhance the impact assessment process by accurate detection of noisy microblogs.
  arXiv  Detail & Related papers  (2024-05-06T10:52:21Z)
• Storm Surge Modeling in the AI ERA: Using LSTM-based Machine Learning for Enhancing Forecasting Accuracy [0.7149367973754319]
  We propose and analyze the use of an LSTM-based deep learning network machine learning architecture. The overall goal of this work is to predict the systemic error of the physics model and use it to improve the accuracy of the simulation results post factum.
  arXiv  Detail & Related papers  (2024-03-07T13:19:38Z)
• Data-Driven Prediction of Seismic Intensity Distributions Featuring Hybrid Classification-Regression Models [21.327960186900885]
  This study develops linear regression models capable of predicting seismic intensity distributions based on earthquake parameters. The dataset comprises seismic intensity data from earthquakes that occurred in the vicinity of Japan between 1997 and 2020. The proposed model can predict even abnormal seismic intensity distributions, a task at conventional GMPEs often struggle.
  arXiv  Detail & Related papers  (2024-02-03T13:39:22Z)
• Simulation-Enhanced Data Augmentation for Machine Learning Pathloss Prediction [9.664420734674088]
  This paper introduces a novel simulation-enhanced data augmentation method for machine learning pathloss prediction. Our method integrates synthetic data generated from a cellular coverage simulator and independently collected real-world datasets. The integration of synthetic data significantly improves the generalizability of the model in different environments.
  arXiv  Detail & Related papers  (2024-02-03T00:38:08Z)
• To Repeat or Not To Repeat: Insights from Scaling LLM under Token-Crisis [50.31589712761807]
  Large language models (LLMs) are notoriously token-hungry during pre-training, and high-quality text data on the web is approaching its scaling limit for LLMs. We investigate the consequences of repeating pre-training data, revealing that the model is susceptible to overfitting. Second, we examine the key factors contributing to multi-epoch degradation, finding that significant factors include dataset size, model parameters, and training objectives.
  arXiv  Detail & Related papers  (2023-05-22T17:02:15Z)
• CAFE: Learning to Condense Dataset by Aligning Features [72.99394941348757]
  We propose a novel scheme to Condense dataset by Aligning FEatures (CAFE) At the heart of our approach is an effective strategy to align features from the real and synthetic data across various scales. We validate the proposed CAFE across various datasets, and demonstrate that it generally outperforms the state of the art.
  arXiv  Detail & Related papers  (2022-03-03T05:58:49Z)
• A CNN-BiLSTM Model with Attention Mechanism for Earthquake Prediction [0.0]
  This paper proposes a novel prediction method based on attention mechanism (AM), convolution neural network (CNN), and bi-directional long short-term memory (BiLSTM) models. It can predict the number and maximum magnitude of earthquakes in each area of mainland China-based on the earthquake catalog of the region.
  arXiv  Detail & Related papers  (2021-12-26T20:16:20Z)
• Deep Learning Based Cloud Cover Parameterization for ICON [55.49957005291674]
  We train NN based cloud cover parameterizations with coarse-grained data based on realistic regional and global ICON simulations. Globally trained NNs can reproduce sub-grid scale cloud cover of the regional simulation. We identify an overemphasis on specific humidity and cloud ice as the reason why our column-based NN cannot perfectly generalize from the global to the regional coarse-grained data.
  arXiv  Detail & Related papers  (2021-12-21T16:10:45Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.