Yuhan Zhang

—

Unclaimed · AI-structured profileOpenAlex

Computer vision researcher in cross-view geo-localization and adversarial attack; last author of 2026 arXiv papers

papers here

reviewed by Evolve

first-author

corresponding

—

works · OpenAlex

—

h-index · OpenAlex

Bibliometric figures come from OpenAlex (2026-09-21) and cover the author's entire public record, not just the papers listed here. Reproducibility signal comes from Evolve's own reviews.

Research profile

Output—

Unscored: this snapshot provides no publication-volume data. Once a complete, disambiguated bibliographic record exists, this axis would measure cumulative output — total papers, annual cadence, and venue types — rather than the two cataloged preprints alone.

Impact—

Unscored: no citation or influence metrics are available for this profile. This axis would measure citation counts, h-index-style aggregates, and influential citations once the OpenAlex record is confidently linked to this specific researcher.

Focus100/100

Focus scores 100 because the field distribution for this snapshot contains a single labeled field: computer vision (cross-view geo-localization / remote sensing), assigned to one cataloged paper. The second paper, on attacking visible-infrared object detectors, is unlabeled but thematically adjacent. With only one field-labeled paper, the perfect concentration should be read as narrow early evidence rather than a long-term specialization record.

Collaboration61/100

Collaboration scores 61, grounded in 20 distinct coauthors across the snapshot. The frequent-coauthor list is topped by Qi Ming, who appears twice, alongside Xuyu Fan, Zhu Han, Liuqian Wang, Siyuan Cao, Xiaohan Zhang, Xudong Zhao, Mingjing Zhao, Yueqi Zhu, Guo Cheng, and Yongkang Zhang. Holding last-author position on both cataloged papers suggests a recurring senior or supervisory slot within a stable team.

Activity—

Unscored: the snapshot lacks a full year-by-year publication history. This axis would measure recent publication cadence and continuity — papers per recent year — beyond the two 2026 arXiv preprints currently in the catalog.

Reproducibility55/100

Reproducibility scores 55, the mean over one assessed paper: the MVLGeo cross-view geo-localization work, which Evolve rates at medium repro-readiness. Its verdict credits a unified MoE model with VLM reranking and SAM ellipse priors and strong UAV gains, while flagging partial street-view state-of-the-art and unaccounted VLM pipeline costs. With a single assessed paper, this score is an early, low-confidence signal.

Research summary

Yuhan Zhang is a computer vision researcher whose cataloged work centers on cross-view geo-localization and adversarial robustness in visual sensing. Evolve's catalog lists two 2026 arXiv preprints, both with Zhang in the last author position. The first, 'Multi-View Mixture-of-Experts with Vision-Language Reranking for Cross-View Object Geo-Localization,' is reviewed by Evolve as MVLGeo: a single mixture-of-experts model unifying UAV and street-view cross-view object geo-localization, with vision-language reranking and SAM-based ellipse priors, strong UAV gains, only partial street-view state-of-the-art, and unaccounted VLM pipeline costs; its repro-readiness is medium. The second, 'Stealthy in Semantics, Antagonistic in Space,' attacks visible-infrared object detectors via object-level misalignment and awaits full review. The radar shows focus at 100, derived from a field distribution containing exactly one field — computer vision (cross-view geo-localization / remote sensing); collaboration at 61 across 20 distinct coauthors; and reproducibility at 55 from one assessed paper. Output, impact, and activity are unscored for missing data. Repeated co-authorship with Qi Ming and a stable circle suggests team-based work. As 'Yuhan Zhang' is a common name, this profile is name-resolved from public bibliographic data; no institution or country is recorded.

Collaboration pattern

Across the snapshot, Yuhan Zhang has 20 distinct coauthors. The frequent list is led by Qi Ming, who appears twice, followed by Xuyu Fan, Zhu Han, Liuqian Wang, Siyuan Cao, Xiaohan Zhang, Xudong Zhao, Mingjing Zhao, Yueqi Zhu, Guo Cheng, and Yongkang Zhang — a stable core team. Zhang sits in the last author position on both cataloged papers, a placement typically associated with senior or supervisory roles. No institutional affiliation data is available.

Research keywords

cross-view object geo-localizationmixture-of-expertsvision-language rerankingSAM (Segment Anything)anisotropic Gaussian priorUAV / drone navigationremote sensingvisible-infrared object detectionadversarial attackmulti-view learning

Highlights

One model for UAV and street-view object geo-localization
The MVLGeo paper unifies UAV and street-view cross-view object geo-localization in a single mixture-of-experts model, adding vision-language reranking and SAM-based anisotropic Gaussian ellipse priors.
Strong UAV gains, with open caveats
Evolve's verdict credits strong UAV improvements but notes street-view state-of-the-art is only partial and the cost of the vision-language reranking pipeline is not accounted for; repro-readiness is medium.
Object-level misalignment attacks on visible-infrared detectors
A second 2026 arXiv paper proposes attacks that are stealthy in semantics and antagonistic in space against visible-infrared object detectors; it has not yet received a full Evolve review.

Timeline

2026Posted 'Multi-View Mixture-of-Experts with Vision-Language Reranking for Cross-View Object Geo-Localization' to arXiv as last author; Evolve later reviewed it, rating repro-readiness medium.
2026Co-authored 'Stealthy in Semantics, Antagonistic in Space: Attacking Visible-Infrared Object Detectors via Object-Level Misalignment' on arXiv, again in the last author position.
2026Both cataloged outputs land in 2026, concentrating current activity on cross-view geo-localization and adversarial attack in visual sensing; institution and venue details remain unavailable.

Papers in the Evolve catalog

Frequent co-authors

Qi Ming×2 Xuyu Fan×1 Zhu Han×1 Liuqian Wang×1 Siyuan Cao×1 Xiaohan Zhang×1 Xudong Zhao×1 Mingjing Zhao×1 Yueqi Zhu×1 Qi Ming×1 Guo Cheng×1 Yongkang Zhang×1

About this profile

This profile is AI-structured by Evolve.Pub from public bibliographic data (OpenAlex) and Evolve's own paper reviews. It is not written or endorsed by the researcher unless the profile is claimed. Radar dimensions are computed from real bibliometrics with fixed formulas — raw figures are shown alongside; missing dimensions mean insufficient data, never zero. Errors? Claim the profile to correct or hide it.

The name 'Yuhan Zhang' is common; this profile is name-resolved from public bibliographic data and may conflate different researchers.
Bibliometric scores rest on a thin snapshot: one field-labeled paper (focus 100), 20 distinct coauthors (collaboration 61), and a single reproducibility assessment (reproMean 55, reproPapers 1).
No institution, country, or ORCID is available; output, impact, and activity dimensions are unscored due to insufficient data.
Both cataloged papers are 2026 arXiv preprints; the second lacks field, keyword, and review metadata, and no citation counts are quoted anywhere in this profile.