AI Paper Atlas

Research map / Adversarial robustness and security / Attacks on text and graphs

GeoShield: Safeguarding Geolocation Privacy from Vision-Language Models via Adversarial Perturbations

Xinwei Liu, Xiaojun Jia, Yuan Xun, Simeng Qin, Xiaochun Cao

AAAI 2026 · Adversarial robustness and security › Attacks on text and graphs · approach: Vision-language model robustness · 6 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add GeoShield perturbations (eps 8/255) to street-view imagesGPT-4o · Google Street View acc@1kmaccuracy (%): 7.3 -> 1.1Improved
use GeoShield instead of M-AttackGPT-4o · Google Street View acc@2500kmaccuracy (%): 71.1 -> 33.8Improved
use GeoShield instead of M-AttackGemini-2.5 · Google Street View acc@2500kmaccuracy (%): 86.3 -> 80.9Improved
add feature disentanglement module (GNFD) (ablation)GeoShield on GPT-4o · avg distancekm: 4481 -> 7564Improved
add exposure element identification (Geo-EE) (ablation)GeoShield on o1 · avg distancekm: 5840 -> 6780Improved
add scale-adaptive enhancement (PSAE) (ablation)GeoShield on GPT-4o · avg distancekm: 4932 -> 7564Improved
add GNFD, Geo-EE or PSAE (ablation)GeoShield · semantic consistency BERT-SBERT-S: 0.8937-0.8954 -> 0.8986No clear effect
use GeoShield instead of M-Attack for semantic preservationGPT-4o · semantic similarity eps 8/255BLEU: 0.09 -> 0.11Improved
use GeoShield instead of AdvDiffVLMGPT-4o · semantic similarity eps 8/255BLEU: 0.14 -> 0.11Hurt
use targeted baselines (AdvDiffVLM, AnyAttack)GPT-4o · Google Street View acc@1kmaccuracy (%): 7.3 -> 5.5No clear effect

Later papers that compared with this method

1 later accepted paper reports GeoShield as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy