World Model Guided Multi-Modal Representation Learning for Disaster Response Using Remote Sensing Data
Keywords:
world model, multi-modal representation learning, remote sensing, disaster response, sensor fusion, structural trade-offs, fairness, policyAbstract
Disaster response operations increasingly depend on rapid, accurate analysis of heterogeneous remote sensing data, including optical imagery, synthetic aperture radar, hyperspectral cubes, and LiDAR point clouds. Traditional multi-modal fusion methods often treat each modality independently and lack a unified understanding of the underlying physical processes that govern disaster dynamics. This paper proposes a world model guided framework for multi-modal representation learning that explicitly incorporates a structured, generative model of the Earth system to align and integrate disparate sensor streams. The world model provides a common latent space constrained by physical laws, enabling the learning of invariant representations robust to sensor noise, temporal gaps, and missing modalities. We examine the architectural trade-offs between end-to-end learning and modular world model components, the governance challenges of deploying such systems across heterogeneous institutional boundaries, and the sustainability of continuous learning in dynamic disaster environments. The paper further addresses fairness concerns arising from biased training data and the policy implications of using foundation models for high-stakes humanitarian decisions. Case illustrations from hurricane damage assessment, wildfire progression monitoring, and flood mapping demonstrate the potential of the approach. The study concludes with forward-looking recommendations for building resilient, equitable, and interpretable disaster intelligence systems grounded in physical principles.
References
1. Schmitt, M., & Zhu, X. X. (2016). Data fusion and remote sensing: An ever-growing relationship. IEEE Geoscience and Remote Sensing Magazine, 4(3), 6–23.
2. Jean, N., Burke, M., Xie, M., Davis, W. M., Lobell, D. B., & Ermon, S. (2016). Combining satellite imagery and machine learning to predict poverty. Science, 353(6301), 790–794.
3. Ha, D., & Schmidhuber, J. (2018). World models. arXiv preprint arXiv:1803.10122.
4. Yuan, Q., Shen, H., Li, T., Li, Z., Li, S., Jiang, Y., Xu, H., Tan, W., Yang, Q., Wang, J., Gao, J., & Zhang, L. (2020). Deep learning in environmental remote sensing: Achievements and challenges. Remote Sensing of Environment, 241, 111716.
5. Arcomano, T., Szunyogh, I., Pathak, J., Wikner, A., Hunt, B. R., & Ott, E. (2020). A machine learning-based global atmospheric forecast model. Geophysical Research Letters, 47(18), e2020GL089022.
6. Arjovsky, M., Bottou, L., Gulrajani, I., & Lopez-Paz, D. (2019). Invariant risk minimization. arXiv preprint arXiv:1907.02893.
7. Kashinath, K., Mustafa, M., Albert, A., Wu, J. L., Jiang, C., Esmaeilzadeh, S., Azizzadenesheli, K., Wang, R., Chattopadhyay, A., Singh, A., Manepalli, A., Chirila, D., Yu, R., Walters, R., White, B., Xiao, H., Tchelepi, H. A., Marcus, P., Anandkumar, A., ..., & Mahoney, M. W. (2021). Physics-informed machine learning: Case studies for weather and climate modelling. Philosophical Transactions of the Royal Society A, 379(2194), 20200093.
8. Finn, C., Abbeel, P., & Levine, S. (2017). Model-agnostic meta-learning for fast adaptation of deep networks. International Conference on Machine Learning, 1126–1135.
9. Fraccaro, M., Kamronn, S., Paquet, U., & Winther, O. (2017). A disentangled recognition and nonlinear dynamics model for unsupervised learning. Advances in Neural Information Processing Systems, 30.
10. Chen, R. T. Q., Rubanova, Y., Bettencourt, J., & Duvenaud, D. (2018). Neural ordinary differential equations. Advances in Neural Information Processing Systems, 31.
11. French, R. M. (1999). Catastrophic forgetting in connectionist networks. Trends in Cognitive Sciences, 3(4), 128–135.
12. Satyanarayanan, M. (2017). The emergence of edge computing. Computer, 50(1), 30–39.
13. Goodfellow, I. J., Shlens, J., & Szegedy, C. (2015). Explaining and harnessing adversarial examples. International Conference on Learning Representations.
14. United Nations Satellite Centre. (2020). UNOSAT annual report 2020. United Nations Institute for Training and Research.
15. Strubell, E., Ganesh, A., & McCallum, A. (2019). Energy and policy considerations for deep learning in NLP. Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, 3645–3650.
16. Yang, J. X., Wang, J., Li, Z., Sui, C., Long, Z., & Zhou, J. (2025). HSLiNets: Evaluating Band Ordering Strategies in Hyperspectral and LiDAR Fusion. IEEE Geoscience and Remote Sensing Letters.
17. Xiong, Z., Song, Y., He, L., Xiong, W., Yuan, Y., Qiao, F., & Jacobs, N. (2026). PhysAlign: Physics-Coherent Image-to-Video Generation through Feature and 3D Representation Alignment. arXiv preprint arXiv:2603.13770.
18. Buolamwini, J., & Gebru, T. (2018). Gender shades: Intersectional accuracy disparities in commercial gender classification. Proceedings of the 1st Conference on Fairness, Accountability and Transparency, 77–91.
19. Hardt, M., Price, E., & Srebro, N. (2016). Equality of opportunity in supervised learning. Advances in Neural Information Processing Systems, 29.
20. United Nations. (2021). Guidelines for the long-term sustainability of outer space activities. Committee on the Peaceful Uses of Outer Space.
21. McMahan, B., Moore, E., Ramage, D., Hampson, S., & y Arcas, B. A. (2017). Communication-efficient learning of deep networks from decentralized data. Proceedings of the 20th International Conference on Artificial Intelligence and Statistics, 1273–1282.
22. United Nations Office for Outer Space Affairs. (2023). AI and space: A policy framework. UNOOSA.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Journal of Data Intelligence and AI Systems
This work is licensed under a Creative Commons Attribution 4.0 International License.
This article is published under the Creative Commons Attribution 4.0 International License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
