Last author on RankGround (GUI grounding) and HiLNO (neural operators), spanning vision and scientific ML
2
papers here
2
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
—
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-09-19) 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—
Insufficient data to score: only two Evolve-reviewed 2026 preprints are on record, with no fuller publication count or venue history. This axis would measure paper volume, venue tier, and preprint-to-publication conversion over time.
Impact—
No citation or influence metrics are available for this profile, so impact cannot be scored. This axis would measure citations, uptake of RankGround and HiLNO by later work, and adoption of the proposed methods in GUI-agent and neural-operator research.
Focus0/100
Focus scores 0, driven directly by the raw field distribution: output is split exactly 50/50 between computer vision (GUI agents and vision-language models) and scientific machine learning (PDE surrogate modeling / numerical analysis). The two 2026 papers sit in different fields — RankGround in GUI grounding, HiLNO in neural operators — so the profile currently reads as dual-track rather than one concentrated line; the even split, not paper quality, produces the zero.
Collaboration41/100
Collaboration scores 41, computed from 7 distinct coauthors across the two papers: Liyang Fan, Xinping Bi, Yitai Li, Shuaimin Li, Hui Li, Zhicheng Hu, and Jiacheng Li. Min Yang sits in the last author position on both preprints without the corresponding-author flag, a pattern consistent with a senior or supervisory role, though the small coauthor pool keeps the score moderate.
Activity—
Activity cannot be scored: the record covers only two 2026 preprints, with no multi-year publication series or submission cadence. This axis would measure publishing tempo — outputs per year, recency, and sustained presence in arXiv and peer-reviewed venues.
Reproducibility55/100
Reproducibility scores 55 — the mean readiness rating across two reviewed papers, both rated medium. Evolve's verdicts explain the middling placement: RankGround's results are strong and honestly analyzed but its +5.5% headline gain needs scrutiny, while HiLNO is well ablated yet rests on single-run results and transcribed baselines, tempering confidence in exact replication.
Research summary
Min Yang is profiled here from public bibliographic data and two Evolve-reviewed preprints, both from 2026 on arXiv, on which Min Yang appears in the last author position without the corresponding-author flag. The record splits evenly — 50/50 in the field distribution — between two quite different areas. The first is computer vision for GUI agents and vision-language models: RankGround (2026) proposes efficient high-resolution GUI grounding through a lightweight reranker that guides crop selection within a single vision-language-model call; Evolve's verdict calls the results strong and honestly analyzed, while flagging that the +5.5% headline gain needs scrutiny. The second is scientific machine learning: HiLNO (2026) is a hierarchical latent neural operator with multi-scale supervision for PDEs on general geometries, reported to reach near-state-of-the-art accuracy with far fewer parameters than LinearNO; Evolve notes thorough ablations but single-run results and transcribed baselines temper confidence. Radar scores computed from this data place collaboration at 41 (7 distinct coauthors), reproducibility at 55 (the mean across 2 papers, both rated medium readiness), and focus at 0, reflecting the even two-field split. The output, impact, and activity axes lack sufficient data to score. No institution or ORCID is attached, so identity resolution carries the usual common-name caveat.
Collaboration pattern
Across the two reviewed preprints Min Yang works with a compact pool of 7 distinct coauthors — Liyang Fan, Xinping Bi, Yitai Li, Shuaimin Li, Hui Li, Zhicheng Hu, and Jiacheng Li — and consistently occupies the last author position without the corresponding-author flag. The radar collaboration score of 41 reflects this small, stable team. The two papers sit in different fields, suggesting separate working groups rather than one continuous collaboration network; beyond this, no cross-institution data is on record.
Research keywords
Highlights
Efficient high-resolution GUI grounding with a single VLM call
Near-SOTA PDE accuracy with far fewer parameters
Timeline
- 2026Last-author arXiv preprint RankGround introduces lightweight reranker-guided crop selection for efficient high-resolution GUI grounding.
- 2026Last-author arXiv preprint HiLNO presents a hierarchical latent neural operator with multi-scale supervision for PDEs on general geometries.
- 2026Both preprints enter the Evolve catalog with medium reproducibility readiness; radar scores of collaboration 41, reproducibility 55, and focus 0 are recorded from this two-paper base.
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.
- Name is common; this profile is name-resolved from public bibliographic data, and no institution, country, or ORCID is attached — identity should be verified before citing.
- Bibliometrics come from a public bibliometric snapshot plus the two Evolve-reviewed papers only; output, impact, and activity axes are unscored for lack of data.
- Radar scores (collaboration 41, focus 0, reproducibility 55) are computed from just two 2026 preprints, a very small base.
- Both papers are recent arXiv preprints; peer-review status and code availability are not recorded here.