Deep learning researcher at Seoul National University — wafer defect analysis, GNN quantization, image restoration
3
papers here
0
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
0
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-17) 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—
Not yet scorable: only three 2026 arXiv preprints are recorded, with no venue or cumulative-count history beyond OpenAlex's h-index 0 and i10-index 0. This axis would measure total publication volume and its distribution over years and venues.
Impact—
Insufficient data for a score. OpenAlex currently lists h-index 0 and i10-index 0, and the three preprints are too recent for citation accumulation. This axis would measure citations, citing works, and influence on follow-up research.
Focus—
No thematic-concentration score is computed. OpenAlex assigns five topics — Advanced Neural Network Applications (2) plus VLSI design, denoising, image processing, and graph neural networks (1 each) — suggesting breadth, but the axis would quantify topical concentration over time.
Collaboration53/100
Radar scores collaboration at 53. OpenAlex records 13 distinct coauthors against just 3 works in the last three years — a comparatively wide partner set for the output volume. All three 2026 preprints place Lee in the last author position, the conventional senior slot, and the recurring collaborator list includes Dain Kwon, Kanghyun Choi, Sunjong Park, and Hyeyoon Lee. The pattern suggests small-team, Lee-led projects spanning manufacturing, graph learning, and astronomy.
Activity40/100
Radar scores activity at 40. The available countsByYear series contains a single point — 3 works in 2026 — and recent3yWorks is likewise 3, matching the three arXiv preprints in the Evolve catalog. All recorded output is concentrated in one year, so a long-term publishing cadence cannot yet be characterized. Within 2026 the record shows a burst of near-simultaneous submissions across three distinct application areas rather than a steady multi-year stream.
Reproducibility—
No score: reproReadiness is null for all three catalog papers and no code or dataset release was reviewed. This axis would assess whether artifacts — code, data, trained models — are publicly available and sufficient to reproduce each paper's results.
Research summary
Jinho Lee is a researcher at Seoul National University (South Korea) whose public bibliographic profile, resolved through OpenAlex (ORCID 0000-0003-4010-6611), centers on applied deep learning. The Evolve catalog records three 2026 arXiv preprints — SePArate (weak supervision for wafer-defect pattern segmentation), TopGQ (fast post-training quantization for graph neural networks leveraging topology information), and STARLINC (satellite trail artifact removal via inter-frame correlation) — on each of which Lee holds the last author position, the slot conventionally occupied by the senior or supervising researcher. OpenAlex topic assignments reinforce this applied orientation: Advanced Neural Network Applications (2 works) alongside Low-power high-performance VLSI design, Image and Signal Denoising Methods, Advanced Image Processing Techniques, and Advanced Graph Neural Networks (1 each). Reported indicators are early-stage: h-index 0, i10-index 0, 3 works in the last three years (all dated 2026), and 13 distinct coauthors. Radar scores are available only for collaboration (53) and activity (40); output, impact, focus, and reproducibility lack sufficient underlying data. Overall, this is an early-stage, senior-author profile of an active applied machine-learning researcher whose recent output spans semiconductor manufacturing inspection, efficient graph learning, and astronomical image restoration.
Collaboration pattern
OpenAlex records 13 distinct coauthors across only 3 recent works, with Lee in the last author position on all three 2026 preprints — a senior-supervision pattern on small teams. Recurring names include Dain Kwon, Kanghyun Choi, Sunjong Park, Hyeyoon Lee, and Shingeon Kim; the wider list (Mi-Ryang Kim, Jeong-Eun Lee, Jaewon Jang, others) is consistent with parallel projects in semiconductor inspection, graph learning, and astronomy. Radar's collaboration score is 53.
Research keywords
Highlights
SePArate: weakly supervised segmentation of wafer-defect patterns (last author)
A 2026 arXiv preprint proposing SePArate, which uses weak supervision to segment patterns from defects in wafer manufacturing. Lee is listed in the last author position and is not marked corresponding. No venue, Evolve verdict, keywords, or reproducibility assessment has been recorded yet.
TopGQ: fast GNN post-training quantization leveraging topology
A 2026 arXiv preprint describing TopGQ, a fast post-training quantization method for graph neural networks that leverages topology information. Lee appears as the last-listed author. The paper is unreviewed in the catalog, with no verdict or repro-readiness data recorded.
STARLINC: satellite trail artifact removal via inter-frame correlation
A 2026 arXiv preprint presenting STARLINC, which removes satellite trail artifacts from images using inter-frame correlation, aligning with the profile's image-denoising and image-processing topics. Lee is the last-listed author; no Evolve verdict or reproducibility data is available.
Timeline
- 2026Released the SePArate arXiv preprint on weakly supervised segmentation of wafer-defect patterns, in the last author position.
- 2026Released the TopGQ arXiv preprint on fast GNN post-training quantization leveraging topology information, in the last author position.
- 2026Released the STARLINC arXiv preprint on satellite trail artifact removal using inter-frame correlation, in the last author position.
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 indicators come from OpenAlex only and may lag or under-count recent arXiv preprints; h-index 0 and i10-index 0 reflect this early indexing state rather than a confirmed absence of influence.
- Jinho Lee is a common Korean name; this profile is name-resolved from public bibliographic data (OpenAlex identity, ORCID 0000-0003-4010-6611) and could mix records if the underlying index is ambiguous.
- All three catalog papers are 2026 arXiv preprints with no venue, peer-review status, Evolve verdict, keywords, or reproducibility assessment recorded.
- Radar scores exist only for collaboration (53) and activity (40); output, impact, focus, and reproducibility are not scored because the underlying data is insufficient.