HoloGeo: Mitigating Landmark Bias in Geo-localization via Evidence-Driven Reasoning

2026-07-16

AI Summary

This work addresses the susceptibility of existing vision-language models (VLMs) to landmark bias in image geolocation, which leads to spurious correlations and localization inaccuracies. To mitigate this, the authors propose HoloGeo, an evidence-driven reasoning framework that guides models to equitably leverage diverse visual cues through structured multi-evidence reasoning chains for unbiased geolocation. The study introduces Bias Intensity and Bias Harmfulness—novel quantitative metrics for assessing landmark bias—and presents LandmarkBias-3K, a new benchmark for evaluation. Furthermore, a multidimensional reinforcement learning reward mechanism is designed, trained on the high-quality BF-30k dataset to foster joint reasoning. Experiments demonstrate that HoloGeo achieves state-of-the-art performance on IM2GPS3K and YFCC4k, and significantly outperforms existing open-source VLMs on LandmarkBias-3K, confirming its robustness and effectiveness.

Abstract

Recent advances in Vision-Language Models (VLMs) have significantly improved image geo-localization, yet existing models remain susceptible to landmark bias, causing them to overlook geographical cues or form spurious correlations, ultimately resulting in inaccurate localization. To systematically investigate this issue, we first design two quantitative metrics, Bias Intensity (BI) and Bias Harmfulness (BH), to characterize the impact of landmarks exerted on model reasoning, and establish a comprehensive benchmark, LandmarkBias-3K. To mitigate landmark bias, we further propose an evidence-driven reasoning framework, HoloGeo, to improve the reliability of geo-localization. HoloGeo is supported by a high-quality dataset, BF-30k, annotated with structured multi-evidence bias-free reasoning chains. By incorporating multi-dimensional rewards, HoloGeo explicitly encourages balanced attention over diverse visual cues and achieves evidence-driven joint reasoning. Extensive experiments demonstrate that HoloGeo not only maintains excellent performance on IM2GPS3K and YFCC4k but also significantly outperforms existing open-source VLMs on LandmarkBias-3K, validating its effectiveness for robust geospatial reasoning.

Problem

• landmark bias
• geo-localization
• Vision-Language Models
• spurious correlations
• geographical cues

Innovation

• landmark bias
• evidence-driven reasoning
• geo-localization
• Vision-Language Models
• bias mitigation