ML researcher at Seoul National University: efficient GNN quantization, weak supervision, and applied computer vision
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 scored: the OpenAlex record lists h-index and i10-index of 0 with three 2026 arXiv preprints, so cumulative publication volume and venue quality cannot yet be measured reliably. This axis would track total works, venues, and authorship roles.
Impact—
Not yet scored: citation indicators in the OpenAlex record are currently zero and the three 2026 preprints are too recent to have accumulated citations. This axis would measure citation counts, h-index trajectory, and influence on follow-up work.
Focus—
Not yet scored: the five topic labels in the record span neural network applications, VLSI design, and image processing, but the sample is too small for a stable focus measure. This axis would quantify concentration across topics over time.
Collaboration53/100
The record shows 13 distinct coauthors across three recent works, more than four per paper, yielding a mid-range collaboration score of 53. Frequent coauthors include Dain Kwon, Kanghyun Choi, Jinho Lee, and Hyeyoon Lee, and all three catalog papers place this researcher in a middle author position, consistent with team-based projects rather than solo or corresponding authorship.
Activity40/100
All three works in the record are dated 2026, so publication activity is concentrated entirely in the current year, with three works in the last three years producing an activity score of 40. This is an early signal of an active preprint cadence on arXiv rather than a long-term record; future yearly counts will clarify the pace.
Reproducibility—
Not yet scored: none of the three catalog papers carries an Evolve verdict or reproducibility-readiness assessment, and no artifact links appear in the data. This axis would evaluate whether code, datasets, and experimental setups are released for others to replicate.
Research summary
Sunjong Park is an early-stage researcher at Seoul National University whose public record, resolved from OpenAlex bibliographic data, currently centers on applied machine learning. The catalog lists three 2026 arXiv preprints, each with a middle author position: SePArate (weakly supervised segmentation of patterns from defects in wafer manufacturing), TopGQ (fast post-training quantization for graph neural networks that leverages topology information), and STARLINC (satellite trail artifact removal using inter-frame correlation). Together they sketch a profile oriented toward efficient deep learning, quantized graph neural networks, and practical vision-and-signal problems drawn from semiconductor inspection and astronomical imaging. OpenAlex topic labels reinforce this spread: 'Advanced Neural Network Applications' (2 works) alongside low-power high-performance VLSI design, image and signal denoising, advanced image processing, and advanced graph neural networks. Bibliometric indicators are still thin: h-index and i10-index stand at 0, three works fall within the last three years, and the record shows 13 distinct coauthors, producing mid-range collaboration (53) and activity (40) scores. Frequent coauthors include Dain Kwon, Kanghyun Choi, Jinho Lee, and Hyeyoon Lee. This is an early-stage profile: citation-based output and impact, research-focus stability, and reproducibility cannot yet be scored, and the record should be re-evaluated as the preprints move through review and accumulate citations.
Collaboration pattern
Bibliometrics list 13 distinct coauthors across three 2026 works, with frequent collaborators including Dain Kwon, Kanghyun Choi, Jinho Lee, Hyeyoon Lee, Shingeon Kim, and Jaewon Jang. All three catalog papers place Sunjong Park in a middle author position without corresponding status, consistent with a contributing member of team-based projects spanning machine learning, computer vision, and hardware-aware computing at Seoul National University.
Research keywords
Highlights
SePArate: weakly supervised defect-pattern segmentation for wafer manufacturing
A 2026 arXiv preprint, co-authored in a middle position, that applies weak supervision to segment patterns from defects in wafer manufacturing, a labeling-efficiency approach for semiconductor inspection.
TopGQ: fast post-training quantization for graph neural networks
A 2026 arXiv preprint proposing a fast GNN post-training quantization method that leverages topology information, targeting efficient deployment of graph learning models.
STARLINC: satellite trail artifact removal via inter-frame correlation
A 2026 arXiv preprint describing a method that removes satellite trail artifacts from images using inter-frame correlation, an image-restoration contribution relevant to astronomical imaging and denoising.
Timeline
- 2026Co-authored SePArate (arXiv preprint), applying weak supervision to segment patterns from defects in wafer manufacturing.
- 2026Co-authored TopGQ (arXiv preprint), proposing fast GNN post-training quantization that leverages topology information.
- 2026Co-authored STARLINC (arXiv preprint), removing satellite trail artifacts from images using inter-frame correlation.
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 (h-index, i10-index, topics, coauthor counts) come from an OpenAlex identity record; citation coverage may be incomplete.
- All three catalog papers are 2026 arXiv preprints; no peer-reviewed venue, field label, Evolve verdict, or reproducibility assessment is available yet.
- Authorship is name-resolved from public bibliographic data; 'Sunjong Park' may overlap with other researchers of the same or similar names.
- The researcher holds middle author positions on all listed papers; no first- or corresponding-author work appears in the current data scope.