Yue Zhao

University of Minnesota · United States

Unclaimed · AI-structured profileOpenAlex

Early-career University of Minnesota researcher spanning graph anomaly detection and functional data analysis

papers here

reviewed by Evolve

first-author

corresponding

—

works · OpenAlex

h-index · OpenAlex

Bibliometric figures come from OpenAlex (2026-10-15) 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—

No output score is available yet: the record holds only two 2026 arXiv preprints. This axis would measure publication volume and authorship role over time — too few works to compute.

Impact—

No impact score is available yet. This axis would measure citation-based influence — citations, h-index and i10-index relative to field and career stage. OpenAlex currently reports h-index 0 and i10-index 0, consistent with a preprint-only record.

Focus0/100

Focus scores 0, reflecting an evenly split record: OpenAlex assigns 50% to machine learning (graph learning / anomaly detection) and 50% to statistics (functional data analysis), and the three detected topics — Statistical Methods and Inference, Advanced Statistical Methods and Models, Statistical Methods and Applications — each contain a single paper. At this stage the two-field portfolio reads as exploratory breadth rather than an established specialization.

Collaboration44/100

Collaboration scores 44, based on 8 distinct coauthors. The frequent-coauthor pool — Suprim Nakarmi, Chahana Dahal, Junggab Son, Zuobin Xiong, Hossein Haghbin, Rebecca Sanders and Mehdi Maadooliat — spans both of the researcher's fields, though the data does not map individual names to specific papers. Both cataloged works are multi-authored, with one first-author and one middle-author position and no corresponding-author role recorded.

Activity20/100

Activity scores 20, a low reading driven by a thin recent record: recent3yWorks is 1 and countsByYear contains a single entry, 2026, with 1 work. Both cataloged preprints are dated 2026, so visible output concentrates entirely in one year, with no longer publication history to establish a sustained cadence. The score is best read as early-career sparsity rather than inactivity.

Reproducibility78/100

Reproducibility scores 78, a mean across the two reviewed papers (reproPapers: 2). The first-author functional SVD paper rates high readiness while FoundAna rates medium, so 78 places this small body of work in a comparatively strong position. Evolve's notes temper the number: the statistics paper lacks a convergence proof and its simulations favor its own assumptions, and FoundAna's evidence against graph foundation model baselines is weak.

Research summary

Yue Zhao, affiliated with the University of Minnesota (US), is an early-stage researcher whose OpenAlex-resolved record splits evenly between two methodological territories: machine learning for graph learning and anomaly detection (50% of the field distribution) and statistics, specifically functional data analysis (50%). The cataloged output comprises two 2026 arXiv preprints. The first-author paper, 'A Functional SVD Framework for Regularized Multivariate Functional PCA with Dual Penalization,' adds score-sparsity to regularized multivariate functional PCA via a functional SVD formulation; Evolve judges the theory solid but notes no convergence proof and simulations that favor the method's own assumptions, with high reproducibility readiness. FoundAna, a middle-author contribution, pairs a GNN with a transformer and four positional encodings for cross-graph anomaly detection; its few-shot averages lead reported baselines, though Evolve finds the evidence against graph foundation model baselines weak, with medium reproducibility readiness. Aggregate signals remain thin: OpenAlex reports h-index 0 and i10-index 0, one work in countsByYear (2026), and one recent-3-year work. Among reviewed dimensions, reproducibility scores 78 (mean across two papers), collaboration 44 (8 distinct coauthors), and activity 20. This is an early-stage profile built from limited evidence.

Collaboration pattern

Yue Zhao's collaboration footprint is compact: 8 distinct coauthors and a collaboration score of 44. The frequent-coauthor list — Suprim Nakarmi, Chahana Dahal, Junggab Son, Zuobin Xiong, Hossein Haghbin, Rebecca Sanders, Mehdi Maadooliat — spans both fields (graph learning and functional data analysis), though the data does not link names to specific papers. Both cataloged works are co-authored, with one first-author slot and no corresponding-author role recorded.

Research keywords

graph anomaly detectiongraph foundation modelgraph transformerpositional encodingfew-shot learningzero-shot transferfunctional data analysismultivariate functional PCAfunctional SVDroughness penaltysparsityPC scores

Highlights

Functional SVD brings score-sparsity to multivariate functional PCA
As first author, Yue Zhao proposes a functional SVD framework that adds score-sparsity to regularized multivariate functional PCA under dual penalization. Evolve rates the theory solid but flags the missing convergence proof and simulations that favor the method's own assumptions; reproducibility readiness is high.
FoundAna: GNN-assisted foundation model for graph anomaly detection
In a middle-author contribution, FoundAna combines a GNN with a transformer architecture and four positional encodings to target cross-graph anomaly detection. Few-shot averages lead reported baselines, though Evolve judges the evidence against graph foundation model baselines weak; reproducibility readiness is medium.
Weak evidence against graph foundation model baselines
Evolve's review of FoundAna notes that while few-shot averages lead, comparisons against graph foundation model baselines provide weak evidence — the key open question for a paper claiming foundation-model status in graph anomaly detection.

Timeline

2026First-author arXiv preprint proposing a functional SVD framework for regularized multivariate functional PCA with dual penalization (statistics / functional data analysis).
2026Middle-author arXiv preprint FoundAna: a GNN-assisted foundation model combining a graph transformer with four positional encodings for graph anomaly detection.
2026OpenAlex countsByYear records 1 work in 2026 and recent-3-year works total 1, marking the visible start of this research record.

Papers in the Evolve catalog

2026-09-17
FoundAna: A GNN-assisted Foundation Model for Graph Anomaly Detection
2026-09-13first author
A Functional SVD Framework for Regularized Multivariate Functional PCA with Dual Penalization

Frequent co-authors

Suprim Nakarmi×1 Chahana Dahal×1 Yue Zhao×1 Junggab Son×1 Zuobin Xiong×1 Hossein Haghbin×1 Rebecca Sanders×1 Mehdi Maadooliat×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.

Bibliometrics are computed from OpenAlex only; h-index (0) and i10-index (0) reflect a preprint-only record.
'Yue Zhao' is a common name; this profile is name-resolved from public bibliographic data, and no ORCID is available to confirm identity.
Both cataloged papers are 2026 arXiv preprints; no peer-reviewed venue information is present in the data.
The frequent-coauthor list contains a duplicate entry of the researcher's own name ('Yue Zhao'), a likely data artifact.