arXiv 2511.140102025-11-19cs.CLcs.AI

面向侦察报告多灾害理解的知识驱动智能体大语言模型

Knowledge-Grounded Agentic Large Language Models for Multi-Hazard Understanding from Reconnaissance Reports

organization= Department One , addressline= Address One , city= City One , postcode= 00000 , state= State One , country= Country One；organization= Department Two , addressline= Address Two , city= City Two , postcode= 22222 , state= State Two , country= Country Two；organization= Zachry Department of Civil \& Environmental Engineering, Texas A\&M University , addressline= 3136 TAMU , city= College Station , postcode= 77843 , state= TX , country= USA；organization= Department of Engineering Technology；Industrial Distribution, Texas A\&M University , city= College Station , postcode= 77843 , state= TX , country= USA；organization= Department of Civil；Environmental Engineering, University of Illinois Urbana-Champaign , city= Urbana , postcode= 61801 , state= IL , country= USA

Chenchen Kuai, Zihao Li, Braden Rosen, Stephanie Paal, Navid Jafari, Jean-Louis Briaud, Yunlong Zhang, Youssef M. A. Hashash, Yang Zhou

备注：: 17 pages, 5 figures

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对灾后侦察报告非结构化导致知识转移难、大语言模型分析易幻觉的问题，提出融合混合检索、智能体分块与验证循环的MoRA-RAG框架，构建HazardRecQA数据集，实现更高准确率与更少幻觉，让开源模型媲美专有模型。

摘要：

灾后侦察报告包含理解多灾害相互作用的关键证据，但其非结构化的叙述形式使得系统性知识转移难以实现。大语言模型（LLM）为分析这些报告提供了新的潜力，但在缺乏领域知识支撑时，往往会生成不可靠或存在幻觉的输出。本研究提出了混合检索智能体检索增强生成（Mixture-of-Retrieval Agentic RAG, MoRA-RAG），这是一个知识驱动的大语言模型框架，可将侦察报告转化为多灾害推理的结构化基础。该框架整合了混合检索机制，能在不同灾害专属数据库之间动态路由查询，同时利用智能体分块策略在检索过程中保持上下文连贯性。它还包含一个验证循环，用于评估证据充分性、优化查询，并在信息不完整时启动针对性搜索。我们通过从GEER侦察报告中提取问答对构建了HazardRecQA数据集，该数据集记录了涵盖7种主要灾害类型的90起全球事件。MoRA-RAG的准确率最高可达94.5%，比零样本大语言模型高出30%，比最先进的RAG系统高出10%，同时在多种大语言模型架构下均减少了幻觉现象。MoRA-RAG还能让开源大语言模型达到与专有模型相当的性能。它为将灾后文档转化为可用于灾害韧性建设的可执行、可信情报建立了新范式。

英文摘要：

Post-disaster reconnaissance reports contain critical evidence for understanding multi-hazard interactions, yet their unstructured narratives make systematic knowledge transfer difficult. Large language models (LLMs) offer new potential for analyzing these reports, but often generate unreliable or hallucinated outputs when domain grounding is absent. This study introduces the Mixture-of-Retrieval Agentic RAG (MoRA-RAG), a knowledge-grounded LLM framework that transforms reconnaissance reports into a structured foundation for multi-hazard reasoning. The framework integrates a Mixture-of-Retrieval mechanism that dynamically routes queries across hazard-specific databases while using agentic chunking to preserve contextual coherence during retrieval. It also includes a verification loop that assesses evidence sufficiency, refines queries, and initiates targeted searches when information remains incomplete. We construct HazardRecQA by deriving question-answer pairs from GEER reconnaissance reports, which document 90 global events across seven major hazard types. MoRA-RAG achieves up to 94.5 percent accuracy, outperforming zero-shot LLMs by 30 percent and state-of-the-art RAG systems by 10 percent, while reducing hallucinations across diverse LLM architectures. MoRA-RAG also enables open-weight LLMs to achieve performance comparable to proprietary models. It establishes a new paradigm for transforming post-disaster documentation into actionable, trustworthy intelligence for hazard resilience.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。