DisasterTD: Disaster Toponym Disambiguation Using Multimodal LLMs and Cross-View Geolocalization

2026-07-25

AI Summary

This study addresses the challenge of ambiguous and imprecise georeferences in social media imagery during disaster scenarios, which hinders accurate location identification. The authors propose a novel geolocation framework that integrates multimodal large language models with multi-source, cross-view imagery—including social media posts, satellite remote sensing, and street-level views. The approach first leverages the semantic reasoning capabilities of large language models to disambiguate place names and generate candidate locations, followed by cross-view image verification to refine positional accuracy. To the best of the authors’ knowledge, this is the first work to apply multimodal large language models to toponym disambiguation in disaster contexts. Evaluated on the Hurricane Harvey dataset, the method achieves a 47.01% accuracy within 50 meters and reduces the median localization error to 0.68 kilometers, substantially outperforming existing approaches—particularly in highly ambiguous settings.

Abstract

Social media imagery (SMI) provides timely and fine-grained ground perspectives that are valuable for situational awareness and emergency response. Unlike satellite or aerial imagery, SMI can capture disaster impacts and ground-level conditions in a timely manner. However, geographic references in SMI are often vague or ambiguous, making accurate geolocalization challenging. To address this issue, we propose DisasterTD, a disaster toponym disambiguation framework that integrates multimodal large language model (MLLMs)-based semantic reasoning with cross-view geolocalization. First, MLLMs extract toponyms and generate candidate geolocations from noisy textual inputs. Then, cross-view matching between SMI, remote sensing imagery (RSI), and optionally street-view imagery (SVI) is used to verify and refine these candidate results. We evaluate DisasterTD on the Hurricane Harvey dataset, where SMI is augmented with collected RSI and SVI to construct a cross-view benchmark for disaster geolocalization. The dataset is divided into four categories based on toponym clarity and ambiguity, allowing a fine-grained performance analysis across scenarios. Results show that DisasterTD consistently outperforms MLLM-only and cross-view-only baselines without disambiguation, achieving geolocalization accuracies of 71.62% within 1000 m, 62.36% within 500 m, 57.99% within 250 m, 52.09% within 100 m, and 47.01% within 50 m, while reducing the mean and median errors to 11.33 km and 0.68 km, respectively. The largest improvements appear in ambiguous toponyms, where semantic reasoning with cross-view evidence reduces candidate dispersion and errors. These findings demonstrate the effectiveness of integrating MLLM-based candidate generation with cross-view verification for fine-grained disaster geolocalization.

Problem

• disaster toponym disambiguation
• geolocalization
• social media imagery
• ambiguous geographic references
• cross-view matching

Innovation

• multimodal LLMs
• cross-view geolocalization
• toponym disambiguation
• disaster situational awareness
• social media imagery