CLLMate: A Multimodal Benchmark for Weather and Climate Events Forecasting

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Haobo, Wang, Zhaowei, Wang, Jiachen, Wang, Yueya, Lau, Alexis Kai Hon, Qu, Huamin
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912232667873280
author Li, Haobo
Wang, Zhaowei
Wang, Jiachen
Wang, Yueya
Lau, Alexis Kai Hon
Qu, Huamin
author_facet Li, Haobo
Wang, Zhaowei
Wang, Jiachen
Wang, Yueya
Lau, Alexis Kai Hon
Qu, Huamin
contents Forecasting weather and climate events is crucial for making appropriate measures to mitigate environmental hazards and minimize losses. However, existing environmental forecasting research focuses narrowly on predicting numerical meteorological variables (e.g., temperature), neglecting the translation of these variables into actionable textual narratives of events and their consequences. To bridge this gap, we proposed Weather and Climate Event Forecasting (WCEF), a new task that leverages numerical meteorological raster data and textual event data to predict weather and climate events. This task is challenging to accomplish due to difficulties in aligning multimodal data and the lack of supervised datasets. To address these challenges, we present CLLMate, the first multimodal dataset for WCEF, using 26,156 environmental news articles aligned with ERA5 reanalysis data. We systematically benchmark 23 existing MLLMs on CLLMate, including closed-source, open-source, and our fine-tuned models. Our experiments reveal the advantages and limitations of existing MLLMs and the value of CLLMate for the training and benchmarking of the WCEF task.
format Preprint
id arxiv_https___arxiv_org_abs_2409_19058
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle CLLMate: A Multimodal Benchmark for Weather and Climate Events Forecasting
Li, Haobo
Wang, Zhaowei
Wang, Jiachen
Wang, Yueya
Lau, Alexis Kai Hon
Qu, Huamin
Machine Learning
Artificial Intelligence
Computation and Language
Atmospheric and Oceanic Physics
Forecasting weather and climate events is crucial for making appropriate measures to mitigate environmental hazards and minimize losses. However, existing environmental forecasting research focuses narrowly on predicting numerical meteorological variables (e.g., temperature), neglecting the translation of these variables into actionable textual narratives of events and their consequences. To bridge this gap, we proposed Weather and Climate Event Forecasting (WCEF), a new task that leverages numerical meteorological raster data and textual event data to predict weather and climate events. This task is challenging to accomplish due to difficulties in aligning multimodal data and the lack of supervised datasets. To address these challenges, we present CLLMate, the first multimodal dataset for WCEF, using 26,156 environmental news articles aligned with ERA5 reanalysis data. We systematically benchmark 23 existing MLLMs on CLLMate, including closed-source, open-source, and our fine-tuned models. Our experiments reveal the advantages and limitations of existing MLLMs and the value of CLLMate for the training and benchmarking of the WCEF task.
title CLLMate: A Multimodal Benchmark for Weather and Climate Events Forecasting
topic Machine Learning
Artificial Intelligence
Computation and Language
Atmospheric and Oceanic Physics
url https://arxiv.org/abs/2409.19058