The paper introduces a simulation‑grounded vision‑language model (VLM) framework for wildfire monitoring that converts 2D wildfire simulations into labeled video episodes using a fixed Blender mapping to create low‑detail 3D proxies. These proxies, along with controllable video generation, provide a multimodal memory that a training‑free multi‑agent VLM system uses to retrieve reference episodes, reconcile visual and memory‑based predictions, and generate structured wildfire reports. The system achieves 77.3% accuracy on six simulator‑derived report fields, outperforming direct VLM querying and text‑only memory baselines.
By Duowen Chen, Yuchen Sun, Zhiqi Li, Yuxuan Liao, Sinan Wang, Bart van Bloemen Waanders, Bo Zhu
WildfireSpreadBench evaluates machine‑learning models for predicting next‑day wildfire spread, comparing five discriminative and one generative architecture on the WildfireSpreadTS dataset. The study shows that model rankings differ markedly when using Average Precision versus threshold‑dependent metrics such as F1 and IoU, revealing three distinct prediction profiles—over‑predicting, balanced, and under‑predicting—that AP alone cannot distinguish. Expanding input channels modestly affected AP, underscoring that AP may favor models with predictions poorly suited for operational use.
By Arin Gopakumar, Marco Pannozzo
SAFIRE is a large-scale benchmark for fire and smoke understanding in multimodal large language models (MLLMs), featuring 83,000 captioned images across 20 scenarios and 193,000 multiple-choice VQA questions derived from a 9.7K-image subset. The benchmark evaluates 10 dimensions of performance, from basic perception to higher-order reasoning, and employs a GPT‑5.4-assisted verification pipeline to ensure annotation quality. Experiments on ten open-source MLLMs (8B–38B) reveal an average accuracy of 61.9%, highlighting significant gaps in safety-critical reasoning, while fine-tuning vision encoders on just 7% of SAFIRE data boosts fire-scene classification accuracy from 20.1% to 64.5%. All resources are publicly available at https://risys-lab.github.io/SAFIRE/.
By Pengfei Li, Naufal Suryanto, Sicheng Zhang, Mohammad Alsharid, Muzammal Naseer
The paper introduces an AI surrogate modelling framework that accelerates wildfire hazard mapping across Canada. By training a multi‑task deep learning model on national‑scale data, the surrogate reproduces key outputs of the computationally intensive BurnP3+ simulator—burn probability, mean fire intensity, and mean rate of spread—in seconds instead of days or weeks. The framework includes standardized training, fine‑tuning, and evaluation protocols, domain‑specific metrics, and counterfactual evaluation to ensure robust out‑of‑distribution performance, achieving up to 806‑fold CPU speed‑up and 13% memory usage compared to the original simulator.
By Hager Radi Abdelwahed, Toluwaleke Olutayo, Chirag Goel, Jana Osea, Aldo Zaimi, Tom Swystun, Sandy Erni, Francisco Emiliano Lopez Saavedra, Yan Boulanger, Ga\'etan Marceau Caron
arXiv:2607. 07951v1 Announce Type: new Abstract: Wildfire smoke events produce extreme PM$_{2.
By Yongcan Huang, Li Jiang, Ze Yu Liu
Annual Earth‑observation embeddings, specifically Tessera, can encode wildfire disturbance signals well enough to map burned areas without needing curated fire‑specific imagery or dense time‑series analysis. In tests, linear models using a single Tessera embedding matched or outperformed paired pre‑ and post‑fire HLS imagery and post‑fire imagery alone, achieving high F1 scores for burn‑scar delineation and regional mapping. The approach successfully mapped all same‑year fires in benchmark scenes, recovered 97% of California burned area without California training data, transferred to European fires with high accuracy, and even estimated ignition timing within a 13‑day error margin.
"whyItMatters":"The study demonstrates that pre‑trained annual embeddings can simplify and scale burned‑area mapping, reducing reliance on dense time‑series data and enabling more efficient wildfire monitoring."
By Jovana Knezevic, Clement Atzberger, Zhengpeng Feng, Adam F. A. Pellegrini, Srinivasan Keshav, David Coomes
arXiv:2609.05782v1 Announce Type: new
Abstract: Vision-language models (VLMs) offer a promising alternative to conventional fire detection systems by reasoning about the semantic context of a scene a...
By Mohammad Kazzazi, Zixuan Liu, Siavash Khajavi
The paper presents modular deep learning augmentations for next‑day wildfire spread prediction, including wind‑ and slope‑conditioned attention biases, physics‑feature retrieval‑augmented output correction, and fire‑conditioned dual‑stream gating. These modules are evaluated on five backbone models using the Next Day Wildfire Spread benchmark, with staged ablations, directional audits, retrieval perturbations, calibration measures, and computational comparisons. The best augmented SwinUNETR model achieves an F1 score of 0.4216 and an AUC‑PR of 0.3673, while a mixed ensemble reaches 0.4292 and 0.3790, demonstrating that predictive performance, operational trustworthiness, and computational practicality can be simultaneously improved.
By Miguel Esparza, Aydin Ayanzadeh Ahmad Mousavi, Ali Mostafavi
DisasterInsight is a building‑centric benchmark designed to evaluate vision‑language models (VLMs) for disaster response. Built on the xBD satellite dataset, it adds OpenStreetMap‑derived functional labels to 134,108 building instances and offers 15 task types, including instance assessment, scene counting, multi‑instance reasoning, and structured report generation. Experiments show that VLMs excel at visible damage detection but struggle with building function, multi‑instance reasoning, counting, and grounded reporting, and instruction tuning only partially mitigates these gaps.
By Sara Tehrani, Yonghao Xu, Leif Haglund, Amanda Berg, Gulnaz Zhambulova, Michael Felsberg
arXiv:2609.16199v2 Announce Type: replace-cross
Abstract: This article describes an open image dataset for developing and evaluating active-fire segmentation methods in satellite imagery. The dataset...
By Shreyan Mitra, Mohammadreza Narimani, Parastoo Farajpoor
arXiv:2601.18493v2 Announce Type: replace
Abstract: Vision--language models (VLMs) show promise for disaster-response remote sensing, but existing benchmarks mainly emphasize scene-level or damage-ce...
By Sara Tehrani, Yonghao Xu, Leif Haglund, Amanda Berg, Gulnaz Zhambulova, Michael Felsberg
The article introduces an open image dataset for active‑fire segmentation in satellite imagery, comprising 2,148 image‑mask pairs from 25 California wildfires captured between July 2020 and August 2026. Each 512×512 pixel, three‑channel image is a Sentinel‑2 Level‑2A composite of bands B12, B11, and B8A, with a fixed linear rendering applied uniformly. Masks distinguish background, short‑wave‑infrared rule‑based active fire, and invalid observations, and the dataset includes chip‑level metadata, an incident‑disjoint split, and a mask‑blind analyst review of 233 test chips.
By Shreyan Mitra, Mohammadreza Narimani, Parastoo Farajpoor