TY - UNPB
T1 - Can we predict the unpredictable? Leveraging disasterNet-LLM for multimodal disaster classification
AU - Kulahara, Manaswi
AU - Kashyap, Gautam Siddharth
AU - Joshi, Nipun
AU - Soni, Arpita
PY - 2025
Y1 - 2025
N2 - Effective disaster management requires timely and accurate insights, yet traditional methods struggle to integrate multimodal data such as images, weather records, and textual reports. To address this, we propose DisasterNet-LLM, a specialized Large Language Model (LLM) designed for comprehensive disaster analysis. By leveraging advanced pretraining, cross-modal attention mechanisms, and adaptive transformers, DisasterNet-LLM excels in disaster classification. Experimental results demonstrate its superiority over state-of-the-art models, achieving higher accuracy of 89.5%, an F1 score of 88.0%, AUC of 0.92%, and BERTScore of 0.88% in multimodal disaster classification tasks.
AB - Effective disaster management requires timely and accurate insights, yet traditional methods struggle to integrate multimodal data such as images, weather records, and textual reports. To address this, we propose DisasterNet-LLM, a specialized Large Language Model (LLM) designed for comprehensive disaster analysis. By leveraging advanced pretraining, cross-modal attention mechanisms, and adaptive transformers, DisasterNet-LLM excels in disaster classification. Experimental results demonstrate its superiority over state-of-the-art models, achieving higher accuracy of 89.5%, an F1 score of 88.0%, AUC of 0.92%, and BERTScore of 0.88% in multimodal disaster classification tasks.
UR - https://publons.com/wos-op/publon/78182441/
U2 - 10.48550/arXiv.2506.23462
DO - 10.48550/arXiv.2506.23462
M3 - Preprint
T3 - arXiv
BT - Can we predict the unpredictable? Leveraging disasterNet-LLM for multimodal disaster classification
PB - arXiv.org
ER -