Early-career robotics researcher uniting LLM-agent process planning and metrology-gated VLA inspection
2
papers here
2
reviewed by Evolve
1
first-author
0
corresponding
—
works · OpenAlex
1
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 computed: this record's indexed counts are thin and skewed toward arXiv preprints, so a reliable total-output measure is not yet available. This axis would measure total works, first/corresponding-author share, and the output trajectory once indexing matures.
Impact—
Impact is not scored yet: the available signals — h-index 1 and i10-index 0 on OpenAlex — come from an early-stage, preprint-heavy record. This axis would measure citations, citing works, and scholarly uptake once coverage matures.
Focus0/100
The focus score is 0, reflecting an evenly split field distribution: 50% robotics and manufacturing automation versus 50% industrial inspection and robot learning, with no dominant line. OpenAlex topic tagging is likewise diffuse — five distinct topics (neural network applications, video surveillance and tracking, robotic mechanisms, robot manipulation and learning) each carry a count of one. The two 2026 papers share an agentic-AI-plus-robotics thread, but the record is too short for focus to consolidate.
Collaboration41/100
Collaboration scores 41. The OpenAlex record lists 7 distinct coauthors, and Evolve's frequent-coauthor set adds Ruimin Chen, Farhad Imani, Azadeh Haghighi, Jiong Tang, Zhiling Chen, Matthew P. Castanier, and David Gorsich. Authorship roles differ across the two 2026 papers: first author on the agentic process-planning paper, middle author on the VLA inspection paper, with no corresponding-author entry in either. The network is small but already spans two subfields.
Activity47/100
Activity scores 47 on a very recent but thin record. OpenAlex counts 4 works in the last three years (recent3yWorks: 4), and countsByYear shows 1 work in 2023 versus 4 in 2026, making 2026 the most productive indexed year. Both Evolve-reviewed papers are 2026 arXiv preprints, so output is clustered almost entirely in the current year rather than accumulated steadily across time.
Reproducibility20/100
Reproducibility scores 20 (reproMean: 20) across the 2 papers Evolve has reviewed, and both are individually rated low for reproduction readiness. The agentic additive-manufacturing planner reports gains against the worst candidate of its own grid rather than external baselines, and the inspection paper's end-to-end results are small-scale and figure-only, so an independent reproduction would currently lack external baselines and published end-to-end numbers.
Research summary
Jingzhan Ge is an early-stage robotics researcher at the University of Connecticut whose profile is name-resolved from public OpenAlex bibliographic data (ORCID 0009-0006-5335-8171). The record is young and preprint-heavy: OpenAlex indexes 4 works in the last three years, with 1 in 2023 and 4 in 2026, and citation indicators remain minimal (h-index 1, i10-index 0). Two 2026 arXiv papers define the visible agenda. The first, first-authored, proposes a kinematics-grounded agentic AI system for robotic additive manufacturing process planning, combining LLM triage, tool-grounded planning, and an inverse-kinematics and trajectory pipeline; Evolve's review judges the kinematics pipeline credible while noting that reported gains are measured against the worst candidate of the method's own grid rather than external baselines. The second, a middle-authorship contribution, presents FRAME, which steers vision-language-action manipulation with dimensional-metrology checks and deterministic evidence gating for task-specified inspection; strong ablations accompany small-scale, figure-only end-to-end results. Field attribution splits evenly between robotics and manufacturing automation and industrial inspection with robot learning, and OpenAlex topic tags spread across five single-count areas, yielding a focus score of 0. Both reviewed papers carry low reproduction readiness, and the reproducibility score is 20 across 2 papers. With 7 distinct coauthors, this is an active but early profile whose citation impact is not yet scorable.
Collaboration pattern
Collaboration scores 41 on a small network: 7 distinct coauthors in OpenAlex, with Ruimin Chen, Farhad Imani, Azadeh Haghighi, Jiong Tang, Zhiling Chen, Matthew P. Castanier, and David Gorsich recurring in Evolve's coauthor set. Roles differ across the two 2026 papers — first author on the agentic planning paper, middle author on the FRAME inspection paper — and neither lists Ge as corresponding author.
Research keywords
Highlights
Kinematics-grounded agentic planner for robotic additive manufacturing
FRAME: metrology-gated VLA manipulation for inspection
A converging agentic-AI plus metrology research line
Across both 2026 papers, OpenAlex field attribution splits 50/50 between robotics and manufacturing automation and industrial inspection with robot learning, and five topics each carry a count of one. This reads as one emerging thread — grounded planning plus measured verification — rather than two separate agendas, with the focus score at 0 pending consolidation.
Timeline
- 2023OpenAlex indexes the first work on this record (countsByYear: 1 in 2023), predating the current agentic-robotics agenda.
- 2026Output concentrates in 2026: countsByYear records 4 works, matching the 4 recent-3-year works on the record.
- 2026First-author arXiv paper on kinematics-grounded agentic AI for robotic additive manufacturing process planning enters the Evolve catalog.
- 2026Middle-authorship arXiv paper on measurement-steered VLA manipulation with deterministic evidence gating (FRAME) is reviewed by Evolve.
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.
- This profile is name-resolved from public bibliographic data (OpenAlex / ORCID); 'Jingzhan Ge' is a common name, so identity confidence rests on the University of Connecticut affiliation link.
- Bibliometrics come from OpenAlex only; the recent outputs are arXiv preprints, so citation and venue coverage is likely incomplete.
- Output and impact radar scores are withheld: with h-index 1, i10-index 0, and a preprint-heavy record, the underlying data is insufficient to score.
- Both reviewed papers carry low reproduction readiness (reproMean 20 across 2 papers) and unconfirmed peer-review status, so reported gains should be read cautiously.