First author on reachability-based GNN verification and NNV3 neural-network verification tool, Vanderbilt University
2
papers here
2
reviewed by Evolve
2
first-author
1
corresponding
—
works · OpenAlex
—
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-02) 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—
Output is not yet scored: this snapshot contains no publication-count or productivity data for the researcher beyond the two cataloged 2026 preprints, so the axis — which would measure annual publication volume and output trajectory — cannot be assessed.
Impact—
Impact is not yet scored: no citation counts, h-index, or uptake metrics are present in the input data. This axis would measure citations and influence of the work across the verification and machine-learning communities.
Focus0/100
The focus score is 0 because the field distribution splits evenly (0.5/0.5) between two labels — formal verification of machine learning and GNN verification for power systems — rather than concentrating in one. In substance both labels describe a single tightly-connected theme, reachability-based neural-network verification, so the low score reflects how the fields are labeled in the data more than genuine topical scatter.
Collaboration55/100
Collaboration scores 55, with 15 distinct co-authors recorded in the snapshot. The pattern is group-centered rather than solitary: both cataloged papers are first-authored, with frequent co-authors including Taylor T. Johnson, Diego Manzanas Lopez, Ben Wooding, Samuel Sasaki, and Tyler Derr, alongside a wider circle (Waseem Abbas, Ipek Oguz, Meiyi Ma, Hongchao Zhang, Navid Hashemi, Muhammad Usama Zubair, Zhenxuan Shao) spanning verification and applied ML.
Activity—
Activity is not yet scored: the snapshot provides no per-year publication cadence or recent-output timeline beyond two 2026 preprints. It would measure research tempo — papers per year and sustained engagement over time.
Reproducibility100/100
Reproducibility scores 100: the mean across 2 papers with reproducibility assessments (reproMean 100). Both cataloged papers — the GraphStar reachability paper and NNV3 — are rated high reproducibility readiness, consistent with tool artifacts: NNV3 packages its verification methods in a MATLAB tool, and the GNN work compares against the CORA reachability toolbox.
Research summary
Anne M. Tumlin is a researcher at Vanderbilt University working on the formal verification of machine learning, with a particular focus on graph neural networks (GNNs) and reachability analysis. This profile is name-resolved from public bibliographic data linked to ORCID 0009-0000-1635-8793. The Evolve catalog contains two 2026 arXiv preprints, both first-authored. The first, on reachability-based formal verification of GNNs with node and edge features, claims the first edge-aware formal verification of GNNs via GraphStar sets; Evolve's review finds the method sound with a strong comparison against CORA, while flagging issues in PF feature design and selection. The second, NNV3, unifies weight-perturbation, video, GNN, probabilistic, and fairness verification within a single MATLAB tool; the integration is judged credible but the evaluations remain small. Both papers are rated high for reproducibility readiness, giving a reproducibility mean of 100 across 2 assessed papers. Her field distribution splits evenly (0.5/0.5) between two adjacent labels — formal verification of machine learning and GNN verification for power systems — and the snapshot records 15 distinct co-authors, including Taylor T. Johnson and Diego Manzanas Lopez. Citation counts, publication volume, and per-year cadence are not available, so output, impact, and activity are unscored in this early-stage profile.
Collaboration pattern
The snapshot records 15 distinct co-authors and a collaboration score of 55, consistent with work inside a Vanderbilt-centered verification group. Frequent co-authors span the NNV3/reachability cluster — Ben Wooding, Diego Manzanas Lopez, Taylor T. Johnson, Samuel Sasaki, Muhammad Usama Zubair, Navid Hashemi — and a broader network including Hongchao Zhang, Waseem Abbas, Ipek Oguz, Meiyi Ma, Zhenxuan Shao, and Tyler Derr. Both cataloged papers are first-authored; the GNN verification paper is also corresponding.
Research keywords
Highlights
First edge-aware formal verification of GNNs via GraphStar sets
NNV3: unifying neural-network verification in one MATLAB tool
Timeline
- 2026First-authored (corresponding) arXiv preprint on reachability-based formal verification of graph neural networks with node and edge features.
- 2026First-authored arXiv preprint introducing NNV3, expanding neural-network verification to new architectures and domains.
- 2026Both cataloged papers receive high reproducibility-readiness ratings in Evolve's review (reproMean 100 across 2 papers).
Papers in the Evolve catalog
Frequent co-authors
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.
- Bibliometric scores (collaboration, focus, reproducibility) are computed from a limited snapshot covering only the 2 reviewed 2026 arXiv preprints in the Evolve catalog.
- Output, impact, and activity are unscored: the input data contains no citation counts, h-index, or per-year publication volume.
- Both cataloged items are arXiv preprints; no peer-reviewed venue, award, or career-stage information is included in the data.
- Country is not provided, and the profile is name-resolved from public bibliographic data via ORCID 0009-0000-1635-8793, so common-name collisions cannot be fully excluded.