Ziliang Wang

—

Unclaimed · AI-structured profileOpenAlex

Computer-vision researcher advancing medical image segmentation, lead author of failure-aware ultrasound repair (FAPR).

papers here

reviewed by Evolve

first-author

corresponding

—

works · OpenAlex

—

h-index · OpenAlex

Bibliometric figures come from OpenAlex (2026-09-21) 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 available: the snapshot lacks the publication-volume and productivity counts this axis measures, such as total papers and output over time. It will be scored only once catalog and bibliographic coverage of this name expands.

Impact—

Impact is unscored because the snapshot contains no citation counts, citation-year curves, or h-index-type measures. This axis would aggregate how often the published work is cited and taken up by others; it cannot be estimated from current data.

Focus100/100

Focus scores 100 because every classified paper in the snapshot sits in one field: medical image segmentation (computer vision), the sole entry in the field distribution. The reviewed first-author work on ultrasound lesion segmentation reinforces that concentration. The signal is thin, however — only one paper carries a classified field, and the robotics co-authorship suggests possible breadth the snapshot does not yet capture.

Collaboration62/100

Collaboration scores 62 on the strength of 22 distinct coauthors recorded in the snapshot, with a recurring core of 12 named collaborators. The pattern points to steady team-based research: a first-author medical-imaging paper alongside a multi-author robotics paper. The mid-range score suggests a moderately sized, sustained network rather than lone-author work or a very large rotating consortium.

Activity—

Activity is unscored: the underlying data — recent publication cadence and year-by-year output — is missing. With only two 2026 papers recorded and no multi-year coverage, the researcher's current tempo cannot be measured yet.

Reproducibility20/100

Reproducibility scores 20, computed from a single evaluated paper (reproPapers = 1, mean 20). Evolve's review of the first-author ultrasound segmentation work rates replication readiness as low: reported hard-case gains lack significance tests, very-hard evaluation subsets are tiny, and code is not publicly available. One paper is a narrow basis, but the open gaps are concrete and named.

Research summary

Ziliang Wang is a computer-vision researcher whose reviewed work centers on medical image segmentation; this profile is name-resolved from public bibliographic data rather than a verified institutional record. The Evolve catalog contains two 2026 arXiv papers. The first, 'Evolving Error States: Failure-Aware Progressive Repair for Ultrasound Lesion Segmentation,' is a first-author contribution that recasts post-hoc correction of ultrasound segmentation as ordered, state-transitioning, failure-specific repairs; Evolve's review credits it with impressive hard-case gains while flagging tiny very-hard subsets, absent significance tests, and unavailable code. The second, 'Acting in Meters: Learning Metric Interactions for Precise Robotic Manipulation,' is a middle-author paper extending the footprint toward robot learning, though it has not yet been reviewed and carries no field classification in this snapshot. Radar dimensions are partial: focus scores a perfect 100 on the single classified field, collaboration scores 62 against 22 distinct coauthors with a recurring core of 12, and reproducibility scores 20 from one evaluated paper rated low for replication readiness. Output, impact, and activity remain unscored for lack of volume and citation data. Overall this is an early-stage full profile: a focused, team-oriented researcher with one substantive first-author contribution whose claims still await independent verification through released code and stronger statistical evidence.

Collaboration pattern

Collaboration scores 62 with 22 distinct coauthors recorded. A recurring core of 12 collaborators anchors the network, including Lijie Wang, Zheng Lu, Yiming Wang, Heyang Yu, Kenghou Hoi, Bowen Hu, Di Cui, Tianyu Xin, Haoran Liao, Wanqi Zhong, Xingjie Fan, and Yizhao Xu. The pattern suggests stable team-based research spanning a first-author medical-imaging paper and a multi-author robotics paper, rather than lone-author work or a very large rotating consortium.

Research keywords

medical image segmentationultrasound lesion segmentationpost-hoc correctionfailure-aware routingiterative error feedbackprogressive mask refinementfalse-positive suppressioncomputer visionrobotic manipulationmetric-scale interaction learning

Highlights

First-author FAPR: failure-aware progressive repair for ultrasound lesion segmentation
The 2026 first-author paper recasts post-hoc correction of ultrasound segmentation as ordered, state-transitioning, failure-specific repairs built on iterative error feedback and progressive mask refinement. Evolve credits impressive hard-case gains but flags tiny very-hard subsets, absent significance tests, and unavailable code, rating replication readiness low.
Co-author on metric-scale interaction learning for robotic manipulation
'Acting in Meters: Learning Metric Interactions for Precise Robotic Manipulation' (2026, arXiv) lists the researcher as a middle author, extending the publication footprint from medical imaging toward robot learning. It has not yet been reviewed by Evolve and carries no field, keyword, or reproducibility data in this snapshot.
A concentrated profile with explicit verification gaps
Radar shows perfect focus (100) on the single classified field and mid-range collaboration (62, with 22 distinct coauthors), while reproducibility sits at 20 on one evaluated paper. The clearest near-term lever for this profile is closing FAPR's evidence gaps: released code, significance testing, and larger very-hard subsets.

Timeline

2026Posted the first-author paper 'Evolving Error States' to arXiv, introducing failure-aware progressive repair (FAPR) for ultrasound lesion segmentation.
2026Co-authored 'Acting in Meters: Learning Metric Interactions for Precise Robotic Manipulation' on arXiv, adding a robotics-learning collaboration.
2026Entered the Evolve catalog with a full review of the first-author paper, scored 20/100 for reproducibility and rated low for replication readiness.

Papers in the Evolve catalog

2026-09-16first author
Evolving Error States: Failure-Aware Progressive Repair for Ultrasound Lesion Segmentation
2026-09-16
Acting in Meters: Learning Metric Interactions for Precise Robotic Manipulation

Frequent co-authors

Lijie Wang×1 Zheng Lu×1 Yiming Wang×1 Heyang Yu×1 Kenghou Hoi×1 Bowen Hu×1 Di Cui×1 Tianyu Xin×1 Haoran Liao×1 Wanqi Zhong×1 Xingjie Fan×1 Yizhao Xu×1

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.

Identity is name-resolved from public bibliographic data; 'Ziliang Wang' is a common name and no institutional affiliation, country, or ORCID is on record.
Bibliometric scores and co-author counts come from a limited snapshot, and only one paper carries a field classification.
Output, impact, and activity axes are unscored because publication-volume and citation data are unavailable in this snapshot.
Only the first-author ultrasound paper has been reviewed; the robotics paper lacks a verdict, keywords, and reproducibility data.