Underwater sonar AI and LLM interpretability researcher at NSWC Panama City, with an early but accelerating record
2
papers here
2
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
2
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-14) 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—
The output axis is unscored: this OpenAlex snapshot lacks a computed productivity dimension. It would measure cumulative publication volume, authorship roles and per-year output trends; only fragmentary signals (h-index 2, i10-index 0, per-year counts) exist, so no score is given.
Impact—
The impact axis is unscored in this snapshot. It would capture total citations, per-work citation rates and field-normalized influence over time; beyond h-index (2) and i10-index (0), no citation-level data was available, so no estimate is offered.
Focus0/100
Focus scores 0 on the 0–100 scale, reflecting an evenly split fieldDist: 0.5 for computer vision for underwater sonar sensing (SAS automatic target recognition) and 0.5 for machine learning (LLM interpretability and safety). Topic counts confirm the two-track pattern — Advanced SAR Imaging Techniques (9) and Underwater Acoustics Research (9) sit alongside Advanced Neural Network Applications (4) and Image and Signal Denoising Methods (2).
Collaboration41/100
Collaboration scores 41 on the 0–100 scale, built from 7 distinct co-authors across 9 works in the recent three-year window. Both Evolve-reviewed papers are team outputs: Vetaw is a middle author on Deep Noir and last author on the LoRA SAS paper, neither as corresponding author. Frequent co-authors include Frank E. Bobe III, Dan Zimmerman, Amelia L. McCormack, Darshan W. Bryner, Jose L. Salas-Vernis and Matthew (G.) Cook.
Activity67/100
Activity scores 67 on the 0–100 scale. countsByYear shows 1 work in 2021, 2 in 2023, 1 in 2024, then 4 in 2025 and 4 in 2026 — 9 works inside the recent three-year window. After thin and intermittent output through 2024, publication cadence has roughly quadrupled across 2025–2026, spread across the sonar ATR and transformer interpretability tracks.
Reproducibility38/100
Reproducibility scores 38 (reproMean over 2 reviewed papers). The LoRA SAS paper earns credit for rigorous matched-control nulls but relies on private sonar data, limiting replication (readiness: low); Deep Noir delivers broad ablations yet omits key automated baselines and code (readiness: medium). Neither paper currently pairs complete artifacts with public release, keeping the axis below mid-scale.
Research summary
Gregory D. Vetaw is a researcher at NSWC Panama City Division whose OpenAlex profile shows an early-stage but accelerating publication record: h-index 2, i10-index 0, and 9 works in the most recent three-year window, with output concentrated in 2025 (4 works) and 2026 (4 works). His portfolio splits evenly (0.5/0.5) between two lines: computer vision for underwater sonar sensing — synthetic aperture sonar automatic target recognition — and machine learning for LLM interpretability and safety. Topic counts mirror this: Advanced SAR Imaging Techniques (9) and Underwater Acoustics Research (9) lead, with Advanced Neural Network Applications (4) behind. Two 2026 arXiv preprints anchor the Evolve catalog. The SAS paper reports LoRA alone lifting ATR AUPRC from 0.300 to 0.679 on Vision Transformers, with contrastive mining and SupCon adding nothing; the review credits rigorous matched-control nulls while noting private data limits replication. Deep Noir automates steering-direction discovery via architectural chronometry, logit lens and head attribution; results and ablations are broad, but key automated baselines and code are missing. Reproducibility averages 38 across the two reviews, and collaboration is modest (7 distinct co-authors), with Vetaw typically in middle or last rather than corresponding-author positions. Identity is resolved from public bibliographic records (OpenAlex; ORCID 0000-0003-4680-8171), and citation-level impact data is not yet available.
Collaboration pattern
Collaboration scores 41: 7 distinct co-authors across 9 works in the recent three-year window. Both catalog papers are team efforts — Vetaw sits middle author on Deep Noir and last author on the LoRA SAS paper, neither as corresponding author. Recurring names include Frank E. Bobe III, Dan Zimmerman, Amelia L. McCormack, Darshan W. Bryner and Jose L. Salas-Vernis; 'Matthew Cook' and 'Matthew G. Cook' appear as separate entries, likely variants of one collaborator.
Research keywords
Highlights
LoRA alone lifts SAS ATR AUPRC from 0.300 to 0.679
Automated discovery of steering directions in transformers
Timeline
- 20211 work recorded — the earliest activity in the OpenAlex snapshot.
- 20232 works recorded.
- 20241 work recorded — a thin year before the ramp-up.
- 2025Output accelerates: 4 works recorded in a single year.
- 2026Most active year in the snapshot: 4 works recorded, including both Evolve-reviewed arXiv preprints — the SAS LoRA paper and Deep Noir.
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 figures come from an OpenAlex identity record (identity: openalex) and may undercount output or citations.
- Output and impact dimension scores were not computed; only h-index (2), i10-index (0) and per-year work counts are available.
- Both reviewed papers are arXiv preprints; peer-review status and formal publication venues are not reflected in the snapshot.
- The frequent-coauthor list mixes apparent name variants ('Matthew Cook' / 'Matthew G. Cook'), and country data is absent.