Evolve.Pub
Sign inScan your paper
TZ

Tian Zhou

—

Unclaimed · AI-structured profileOpenAlex

Early-career ML researcher: tabular foundation models, in-context representation refinement, glucose health AI

3

papers here

0

reviewed by Evolve

3

first-author

0

corresponding

—

works · OpenAlex

0

h-index · OpenAlex

Bibliometric figures come from OpenAlex (2026-09-28) 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 dataOutputImpact: insufficient dataImpactFocus: insufficient dataFocusCollaboration: 48/100CollaborationActivity: 40/100ActivityReproducibility: insufficient dataReproducibility

Output—

Output scoring is not yet available: the source record shows an h-index and i10-index of 0 with only three recent works, so publication volume, venue quality, and authorship roles cannot be meaningfully quantified yet.

Impact—

Impact data is insufficient for scoring: no citation counts or uptake signals are recorded beyond zero-valued indices, so this axis, which would measure scholarly influence and downstream use of the work, is left unscored.

Focus—

Focus cannot be scored yet: the five OpenAlex topic labels each attach to a single work, spanning neural networks, retrieval, and diabetes management, so topical concentration has no stable signal to measure.

Collaboration48/100

Scoring 48/100, collaboration is moderate for an early-career profile. OpenAlex records 10 distinct coauthors in the recent window; the recurring set includes Bingqing Peng, Wenwei Wang, Mengni Ye, Beverly Jin, Jinjie Gu, Liang Sun, Linxiao Yang, Xue Wang, and Zuyi Zhu. All three cataloged papers are first-authored without a corresponding-author flag, suggesting Zhou leads the writing while technical contributions are distributed across a small, stable team.

Activity40/100

Activity scores 40/100. The recent three-year window contains 3 works, and OpenAlex counts-by-year attributes all 3 to 2026, indicating a concentrated burst of first-authored arXiv output rather than sustained multi-year publication volume. As the profile matures, this axis will show whether the 2026 pace holds or the earlier record remains sparse.

Reproducibility—

Reproducibility is unscored: none of the three cataloged preprints carries an Evolve verdict or repro-readiness assessment, and code or artifact availability is not recorded, so this axis, which would measure released code, data, and replication evidence, has no basis yet.

Research summary

Tian Zhou is an early-career machine-learning researcher whose 2026 output centers on tabular foundation models and representation learning for sequential health data. The Evolve catalog lists three first-authored arXiv preprints, all dated 2026: a support-compiled feature-folding method that targets more evidence at lower memory across tabular foundation models; a study of what tabular foundation models compute in context, framed as in-situ representation refinement through attention-gated updates; and a transferable evidence-reconstruction approach for longitudinal glucose representations. OpenAlex topic labels reinforce this picture, spanning advanced neural network applications, graph neural networks, domain adaptation and few-shot learning, information retrieval and search behavior, and diabetes management — each attached to a single work, indicating breadth that has not yet consolidated into a settled identity. Radar scores place collaboration at 48/100, with 10 distinct coauthors in the recent window and recurring partners such as Bingqing Peng, Wenwei Wang, and Linxiao Yang; activity sits at 40/100 on a recent-3-year count of 3 works. Citation-based output and impact metrics read as zero in the source data and are treated as insufficient rather than scored, so this profile should be read as an early-stage record.

Collaboration pattern

Zhou's recorded collaboration footprint is compact: 10 distinct coauthors in the recent OpenAlex window, with Bingqing Peng, Wenwei Wang, Mengni Ye, Beverly Jin, Jinjie Gu, Liang Sun, Linxiao Yang, Xue Wang, and Zuyi Zhu recurring most often. All three cataloged papers are first-authored with no corresponding-author flag, consistent with a lead-author role inside a small stable group rather than a large distributed network. The collaboration score of 48/100 reflects this early, team-centered pattern.

Research keywords

Tabular foundation modelsIn-context learningFeature foldingMemory-efficient inferenceRepresentation learningAttention mechanismsTransfer learningDomain adaptationFew-shot learningGraph neural networksLongitudinal glucose modelingDiabetes management

Highlights

  • Support-compiled feature folding: more evidence at lower memory

    First-authored 2026 arXiv preprint proposing a support-compiled feature-folding scheme that targets more evidence at lower memory across tabular foundation models, positioning memory efficiency as a first-class constraint for tabular inference.

  • What tabular foundation models compute in context

    First-authored 2026 arXiv study asks what tabular foundation models compute in context, characterizing in-situ representation refinement driven by attention-gated updates rather than post-hoc interpretation.

  • Transferable evidence reconstruction for longitudinal glucose data

    First-authored 2026 arXiv preprint on reconstructing transferable evidence from longitudinal glucose representations, connecting the tabular-model agenda to diabetes-adjacent health applications.

Timeline

  1. 2026First-authored arXiv preprint on support-compiled feature folding across tabular foundation models
  2. 2026First-authored study of in-context computation in tabular foundation models via attention-gated updates
  3. 2026First-authored preprint on transferable evidence reconstruction for longitudinal glucose representations; OpenAlex counts-by-year attributes all 3 cataloged works to 2026

Papers in the Evolve catalog

Frequent co-authors

Bingqing Peng×3Wenwei Wang×3Mengni Ye×3Beverly Jin×3Jinjie Gu×3Liang Sun×3Linxiao Yang×2Xue Wang×2Zuyi Zhu×1Linxiao Yang×1

Is this you? Claim this profile

Enter the email you published with. We'll send a one-time claim link; after claiming you can edit the introduction or hide this page. Requests from a mismatched address go to manual review.

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 scores and topic labels come from OpenAlex only; h-index and i10-index read as 0 and were treated as insufficient rather than scored.
  • No verified institution, country, or ORCID is available; the profile is name-resolved from public bibliographic data, and Tian Zhou is a common name.
  • All three cataloged works are 2026 arXiv preprints; no peer-reviewed venue, citation counts, or Evolve reproducibility verdicts exist yet.
  • Topic labels carry a count of one each, so field boundaries and topical concentration cannot be firmly established.
Tian Zhou — Researcher Profile | Evolve.Pub