A RAG-Based Multi-Agent LLM System for Natural Hazard Resilience and Adaptation

2025-04-24

AI Summary

To address the weak factual grounding and poor contextual adaptability of large language models (LLMs) in natural disaster scenarios, this work proposes WildfireGPT—a user-centered, resilience-oriented multi-agent RAG framework. Methodologically, it integrates meteorological observations, hazard forecasts, and scientific literature via spatiotemporally aware retrieval-augmented generation, domain-adaptive prompt engineering, and role-specialized multi-agent collaborative reasoning to enable cross-source knowledge fusion and interpretable risk analysis. Its key contributions include: (i) the first context-aware response mechanism supporting diverse stakeholders, and (ii) an expert-validated evaluation protocol. Evaluated on 10 real-world wildfire cases, WildfireGPT achieves a 37% improvement in risk recommendation accuracy over baseline LLMs; domain experts unanimously endorse its interpretability and decision-support utility.

Technology Category

• Natural Language Processing: Language Grounding & Multi-modal NLP
• Machine Learning: Large Multimodal Models (LMMs)
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Large language models (LLMs) are a transformational capability at the frontier of artificial intelligence and machine learning that can support decision-makers in addressing pressing societal challenges such as extreme natural hazard events. As generalized models, LLMs often struggle to provide context-specific information, particularly in areas requiring specialized knowledge. In this work we propose a retrieval-augmented generation (RAG)-based multi-agent LLM system to support analysis and decision-making in the context of natural hazards and extreme weather events. As a proof of concept, we present WildfireGPT, a specialized system focused on wildfire hazards. The architecture employs a user-centered, multi-agent design to deliver tailored risk insights across diverse stakeholder groups. By integrating natural hazard and extreme weather projection data, observational datasets, and scientific literature through an RAG framework, the system ensures both the accuracy and contextual relevance of the information it provides. Evaluation across ten expert-led case studies demonstrates that WildfireGPT significantly outperforms existing LLM-based solutions for decision support.

Problem

• Enhancing decision-making for natural hazard resilience using LLMs
• Providing context-specific information for extreme weather events
• Improving accuracy and relevance of hazard data through RAG

Innovation

• RAG-based multi-agent LLM system
• Integrates hazard data and literature
• User-centered design for stakeholders