Tag
This paper proposes a cost-aware Bayesian optimization framework with level-set estimation to guide UAVs for rapid post-disaster damage assessment, reducing uncertainty while minimizing operational costs.
Introduces Obshazard-bench, a real-time, observation-driven benchmark for evaluating multimodal foundation models on disaster intelligence from raw Earth observation streams, spanning 8 disaster categories across 60+ countries.
NASA's weekly update highlights celebrations of America's aerospace legacy with Freedom250 flyovers, preparations for new ISS crew, Artemis III assembly milestones, and global disaster response support.
New satellite images from Copernicus and NASA reveal the destruction caused by the June 24 earthquakes in Venezuela, aiding rescue efforts and damage assessment.
NASA satellites are providing critical support for earthquake response in Venezuela, capturing data to help assess impacts and guide efforts.
The article discusses the collapse of traditional home insurance in disaster-prone areas due to climate change and the rise of AI-assisted parametric insurance as a potential alternative, though it notes potential pitfalls.
A study evaluating the Prithvi-EO-2.0 foundation model for satellite-based flood mapping across 19 diverse global flood events, finding that detection accuracy is jointly governed by land cover and flood type.
This paper presents an empirical evaluation of LLM-guided semi-supervised learning for classifying social media crisis data. It demonstrates that LG-CoTrain outperforms classical baselines in low-resource settings and highlights the potential of transferring knowledge from LLMs to smaller, deployable models for disaster response.
OpenAI, the Gates Foundation, and ADPC hosted an inaugural AI Jam in Bangkok bringing together 50 disaster management leaders from 13 Asian countries to develop practical AI applications for emergency response. The initiative aims to help governments and nonprofits use AI to improve coordination, data management, and decision-making in disaster situations.
The Universities Space Research Association and Meta are collaborating to apply the Segment Anything Model to support USGS water observing systems for flood emergency response.