SynopticMind: An Instruction Tuning MLLM For Weather Forecasting Report Generation

Zinan Zheng, Yang Liu, Nuo Chen, Juepeng Zheng, Hong Cheng, Jia Li

Keywords: MLLM, Weather Report

Abstract: Accurate weather forecast reporting enables individuals and communities to better plan daily activities, agricultural operations, and transportation. However, the current reporting process primarily relies on manual analysis of multi-source data, which often leads to information overload and reduced efficiency. With the rapid advancement of multimodal large language models (MLLMs), leveraging data-driven models to analyze and generate reports in the weather forecasting domain remains largely underexplored. In this work, we propose the Weather Forecasting Report (WFR) task and construct the first instruction-tuning dataset for this task, named WFInstruct. Based on this corpus, we develop the first model, SynopticMind, specialized in generating weather forecast reports. Experiments on our dataset show that SynopticMind surpasses leading GPT-5. In addition, we analyze the generalization ability of the model, examine the influence of different visual inputs, and evaluate the contribution of individual categories of meteorological variables. SynopticMind offers valuable insight for developing MLLMs specialized in weather report generation.

Primary Area: applications to computer vision, audio, language, and other modalities

Submission Number: 1248