TyphoFormer: Language-Augmented Transformer for Accurate Typhoon Track Forecasting

TL;DR

TyphoFormer is a novel Transformer-based framework that enhances typhoon track forecasting by integrating natural language descriptions generated from meteorological data, leading to improved accuracy especially in complex scenarios.

Contribution

The paper introduces TyphoFormer, which incorporates language-augmented prompts into Transformer models to improve typhoon trajectory prediction accuracy.

Findings

Outperforms state-of-the-art baselines on HURDAT2 benchmark

Effective in scenarios with nonlinear path shifts

Handles limited historical data well

Abstract

Accurate typhoon track forecasting is crucial for early system warning and disaster response. While Transformer-based models have demonstrated strong performance in modeling the temporal dynamics of dense trajectories of humans and vehicles in smart cities, they usually lack access to broader contextual knowledge that enhances the forecasting reliability of sparse meteorological trajectories, such as typhoon tracks. To address this challenge, we propose TyphoFormer, a novel framework that incorporates natural language descriptions as auxiliary prompts to improve typhoon trajectory forecasting. For each time step, we use Large Language Model (LLM) to generate concise textual descriptions based on the numerical attributes recorded in the North Atlantic hurricane database. The language descriptions capture high-level meteorological semantics and are embedded as auxiliary special tokens prepended to the numerical time series input. By integrating both textual and sequential information within a unified Transformer encoder, TyphoFormer enables the model to leverage contextual cues that are otherwise inaccessible through numerical features alone. Extensive experiments are conducted on HURDAT2 benchmark, results show that TyphoFormer consistently outperforms other state-of-the-art baseline methods, particularly under challenging scenarios involving nonlinear path shifts and limited historical observations.