arXiv:2606. 06217v1 Announce Type: cross Abstract: When a disaster unfolds, responders must answer not only what is happening, but also why it is happening, what will happen next, and what to do now, often from noisy low-altitude UAV views and under tight on-site compute constraints.
By Tan Zhang, Quanyou Li, Lu Zhang, Jun Liu, Xiaofeng Zhu, Ping Hu
arXiv:2607. 03591v1 Announce Type: cross Abstract: Recent studies on multimodal traffic accident understanding have mainly relied on infrastructure-camera footage, satellite imagery, or structured crash records.
By Ryosei Tamura, Andrew Shin
arXiv:2608. 09230v1 Announce Type: new Abstract: Industrial-safety understanding requires more than detecting workers, equipment, and personal protective equipment.
By Yuanchi Zhu, Kang An, Tengyue Wang, Zhongyu Yang, Chenxu Du, Xinqi Yang, Hebao Zhu, Bokai Zhao, Tianyu Liang, Ziliang Wang, Faqiang Qian, Yunli Yang, Weiyang Shi, Qibing Ren
The paper introduces GRDisaster, a multi-task geospatial reasoning framework that leverages vision‑language models to interpret, geolocalize, and assess damage in crowdsourced disaster imagery. It builds on a new benchmark dataset of 26,340 images from PhotoMappers, linking volunteer geographic information, street‑view imagery, and remote sensing data across multiple disaster events from 2018 to 2024. GRDisaster combines deterministic and probabilistic cross‑view geolocalization with multi‑view fusion, and introduces spatial reasoning indicators to validate cross‑view matches and quantify disaster severity using expert‑verified annotations.
By Wenping Yin, Fabian Desuer, Ziqi Liu, Naixia Mou, Weijia Li, Pedram Ghamisi, Xiao Xiang Zhu, Hao Li
The paper introduces MuPHI, a diagnostic benchmark that isolates multimodal harm arising solely from image‑text composition, providing benign counterfactuals and annotated harm rationales. Evaluation with MuPHI shows that vision‑language models can ground individual modalities but struggle to bind evidence into a coherent harm mechanism, and that existing harm detection methods transfer poorly to new harm distributions. To address this, the authors propose Compositional Harm Interpretation Learning (CHIL), which rewards evidence identification, cross‑modal binding, and verdict consistency, leading to significant improvements in out‑of‑distribution transfer and counterfactual robustness.
By Anisha Saha, Varsha Suresh, Teodora Kamova, Sophia Wiedmann, Timothy Hospedales, Vera Demberg
DisasterRoute introduces a dual-system routing framework for visual disaster question answering that balances fast direct decisions with selective step-by-step reasoning. The system first uses Disaster-S1, a vision‑language model trained for option discrimination, to produce a quick answer. If the model’s confidence is low or evidence suggests uncertainty, the framework routes the query to DisasterVL, a more comprehensive reasoning model, thereby improving accuracy while maintaining low latency.
By Tan Zhang, Xing Deng, Lu Zhang, Xiaofeng Zhu, Ping Hu