Breaking the Regional Barrier: Inductive Semantic Topology Learning for Worldwide Air Quality Forecasting

2026-01-29

AI Summary

Global air quality forecasting is hindered by strong spatial heterogeneity and poor generalization of models to unseen regions. To address this, this work proposes OmniAir, a novel framework that introduces an inductive semantic topology learning mechanism tailored for global station-level prediction. OmniAir generates generalizable station identities by encoding invariant physical-environmental attributes and dynamically constructs an adaptive sparse graph to capture non-Euclidean long-range dependencies and physical dispersion patterns. Evaluated on the WorldAir dataset encompassing over 7,800 monitoring stations, OmniAir significantly outperforms 18 baseline models while achieving nearly a tenfold speedup in inference, effectively mitigating monitoring blind spots in data-sparse regions.

Technology Category

• Planning, Routing, and Scheduling: Planning with Language Models
• Machine Learning: Graph-based Machine Learning
• Computer Vision: Remote Sensing / Geospatial AI

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Global air quality forecasting grapples with extreme spatial heterogeneity and the poor generalization of existing transductive models to unseen regions. To tackle this, we propose OmniAir, a semantic topology learning framework tailored for global station-level prediction. By encoding invariant physical environmental attributes into generalizable station identities and dynamically constructing adaptive sparse topologies, our approach effectively captures long-range non-Euclidean correlations and physical diffusion patterns across unevenly distributed global networks. We further curate WorldAir, a massive dataset covering over 7,800 stations worldwide. Extensive experiments show that OmniAir achieves state-of-the-art performance against 18 baselines, maintaining high efficiency and scalability with speeds nearly 10 times faster than existing models, while effectively bridging the monitoring gap in data-sparse regions.

Problem

• spatial heterogeneity
• generalization
• air quality forecasting
• unseen regions
• global prediction

Innovation

• inductive learning
• semantic topology
• global air quality forecasting
• adaptive sparse graph
• spatial heterogeneity