Communication Dans Un Congrès Année : 2025

CrisisTS: Coupling Social Media Textual Data and Meteorological Time Series for Urgency Classification

Résumé

This paper proposes CRISISTS, the first multimodal and multilingual dataset for urgency classification composed of benchmark crisis datasets that have been mapped with open source geocoded meteorological time series data. This mapping is based on a simple and effective strategy that allows for temporal and location alignment even in the absence of location mention in the text. A set of multimodal experiments have been conducted relying on transformers and LLMs to improve overall performances while ensuring model generalizability. Our results show that modality fusion outperforms text-only models.

Fichier principal
Vignette du fichier
2025.acl-long.783.pdf (682.66 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence

Dates et versions

hal-05246794 , version 1 (09-09-2025)

Licence

Identifiants

Citer

Romain Meunier, Farah Benamara, Véronique Moriceau, Zhongzheng Qiao, Savitha Ramasamy. CrisisTS: Coupling Social Media Textual Data and Meteorological Time Series for Urgency Classification. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Jul 2025, Vienna, France. pp.16082-16099, ⟨10.18653/v1/2025.acl-long.783⟩. ⟨hal-05246794⟩
2117 Consultations
191 Téléchargements

Altmetric

Partager

  • More