N-Tree Diffusion for Long-Horizon Wildfire Risk Forecasting

2026-03-07

AI Summary

This work addresses the challenges of probabilistic spatial field modeling and multi-horizon computational efficiency in long-term wildfire risk prediction under sparse event supervision. We propose a hierarchical diffusion model based on an N-Tree architecture that enables efficient joint modeling across multiple forecast horizons by sharing early denoising stages and branching into refined pathways at later stages. This design preserves prediction accuracy for each time step while substantially reducing redundant sampling and computational overhead. Integrating probabilistic modeling grounded in wildfire risk maps with a hierarchical diffusion mechanism, our approach achieves higher predictive accuracy and lower inference cost compared to baseline methods on a newly curated real-world wildfire dataset.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Reasoning under Uncertainty: Relational Probabilistic Models
• Machine Learning: Probabilistic Circuits and Graphical Models

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Long-horizon wildfire risk forecasting requires generating probabilistic spatial fields under sparse event supervision while maintaining computational efficiency across multiple prediction horizons. Extending diffusion models to multi-step forecasting typically repeats the denoising process independently for each horizon, leading to redundant computation. We introduce N-Tree Diffusion (NT-Diffusion), a hierarchical diffusion model designed for long-horizon wildfire risk forecasting. Fire occurrences are represented as continuous Fire Risk Maps (FRMs), which provide a smoothed spatial risk field suitable for probabilistic modeling. Instead of running separate diffusion trajectories for each predicted timestamp, NT-Diffusion shares early denoising stages and branches at later levels, allowing horizon-specific refinement while reducing redundant sampling. We evaluate the proposed framework on a newly collected real-world wildfire dataset constructed for long-horizon probabilistic prediction. Results indicate that NT-Diffusion achieves consistent accuracy improvements and reduced inference cost compared to baseline forecasting approaches.

Problem

• long-horizon forecasting
• wildfire risk
• probabilistic spatial fields
• sparse event supervision
• computational efficiency

Innovation

• N-Tree Diffusion
• long-horizon forecasting
• wildfire risk prediction
• hierarchical diffusion model
• Fire Risk Maps