Scholarlink

Disaster relief operations can be chaotic and complex, often characterised by fragmented communication, conflicting information and duplicated effort across agencies. Relief workers may have to sift through large volumes of information from multiple sources to identify relevant and reliable details under time pressure, which can be a daunting task. This poster introduces Vyron, a secure Artificial Intelligence (AI) assistant designed to streamline disaster relief coordination. Combining small language models (SLMs) for mobile devices with large language models (LLMs) and external data sources, Vyron offers a unified platform for real-time information management and decision support. Through illustrative scenarios, we demonstrate Vyron’s capabilities to prioritise conflicting information, recommend safe routes based on evolving weather and road conditions and optimise the distribution of critical resources like food and medical supplies. The poster also addresses trust and transparency in AI systems, highlighting design principles that empowered users to question and verify information. This work advances the field of Human-Computer Interaction (HCI) by illustrating how AI technologies can support human teams in high-stress, real-world environments. By showcasing Vyron at HCI2025, we hope to spark discussions about the role of AI in disaster response and encourage collaboration to develop innovative systems for future crisis management.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Leveraging Large Language Models to Address Communication and Misinformation Challenges to Enhance Information Coordination in Disaster Relief Operations

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Daria Shcherbak,
  • Helen Muncie

摘要

Disaster relief operations can be chaotic and complex, often characterised by fragmented communication, conflicting information and duplicated effort across agencies. Relief workers may have to sift through large volumes of information from multiple sources to identify relevant and reliable details under time pressure, which can be a daunting task. This poster introduces Vyron, a secure Artificial Intelligence (AI) assistant designed to streamline disaster relief coordination. Combining small language models (SLMs) for mobile devices with large language models (LLMs) and external data sources, Vyron offers a unified platform for real-time information management and decision support. Through illustrative scenarios, we demonstrate Vyron’s capabilities to prioritise conflicting information, recommend safe routes based on evolving weather and road conditions and optimise the distribution of critical resources like food and medical supplies. The poster also addresses trust and transparency in AI systems, highlighting design principles that empowered users to question and verify information. This work advances the field of Human-Computer Interaction (HCI) by illustrating how AI technologies can support human teams in high-stress, real-world environments. By showcasing Vyron at HCI2025, we hope to spark discussions about the role of AI in disaster response and encourage collaboration to develop innovative systems for future crisis management.