NLP researcher working on multilingual LLM agents and LLM-based evaluation for personalized learning
2
papers here
2
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
—
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-09-20) 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 publication-volume data in this snapshot, so output is unscored. This axis would measure productivity over time — number of papers, venues, and yearly cadence — none of which are available here.
Impact—
No citation counts or influence indicators are available, so impact is unscored. This axis would measure how frequently and how widely the community cites the researcher's work, based on citation records not present here.
Focus0/100
The focus score is 0 out of 100, and the raw field distribution explains it: reviewed work splits exactly evenly — 50% natural language processing (multilingual LLM agents) and 50% NLP/AI in education (LLM evaluation). With only two catalog papers, this reads as breadth across two adjacent LLM-application areas rather than an established specialization; the score should stabilize as more publications accrue.
Collaboration61/100
Collaboration scores 61 out of 100, grounded in 20 distinct coauthors across the reviewed record and a recurring core of 12 frequent collaborators, including Yilun Liu, Shimin Tao, and Minggui He. Both catalog papers list Zhang as a middle author without corresponding status, consistent with a contributing role inside a moderately sized, stable team rather than solo or lead-author work.
Activity—
Recent publication-cadence data is insufficient, so activity is unscored. This axis would measure research tempo in recent years — how often new preprints or papers appear — beyond the two 2026 entries reviewed here.
Reproducibility78/100
Reproducibility scores 78, the mean over two scored papers. M-SQE carries high readiness — its gains are consistent and significance-tested — while PersonaPath is rated medium, rigorous in construction (2,000 personas, 347 textbooks) but simulation-only in evaluation. Evolve's reservations concern scorer-solver same-model circularity and unvalidated design constants rather than missing artifacts.
Research summary
Min Zhang is a natural language processing researcher whose reviewed work splits evenly between two strands: multilingual LLM agents and AI-in-education evaluation of LLMs. The Evolve catalog records two 2026 arXiv preprints, both with middle-author positioning and neither corresponding-authored. M-SQE introduces a two-view LLM scoring layer — intrinsic skill quality plus task fit — for multilingual agent-skill selection, including low-resource languages; its review notes consistent, significance-tested gains with unusually careful evaluation, tempered by scorer-solver same-model circularity and unvalidated design constants. PersonaPath contributes a knowledge-centric benchmark for personalized learning-path planning built from 2,000 personas and 347 textbooks, where the best model passes 29.5% under simulation-only evaluation. A reproducibility mean of 78 across two scored papers — one high, one medium readiness — points to unusually transferable preprint-stage artifacts. The collaboration score of 61 reflects 20 distinct coauthors and a recurring core of 12. The focus score of 0 mirrors the even 50/50 field split, while output, impact, and activity lack sufficient data to score. Because Min Zhang is a common name, this profile is name-resolved from public bibliographic data and limited to Evolve-reviewed papers.
Collaboration pattern
Min Zhang works in moderately sized teams: the reviewed record lists 20 distinct coauthors, with a stable core of 12 frequent collaborators including Yilun Liu, Shimin Tao, Minggui He, Chenxin Liu, and Xiaojun Meng. Collaboration scores 61/100. On both catalog papers Zhang is a middle author and not corresponding, suggesting a hands-on contributor within a recurring group rather than a solo investigator or group lead.
Research keywords
Highlights
M-SQE: two-view quality estimation for multilingual agent skills
PersonaPath: a knowledge-centric benchmark for personalized learning paths
Timeline
- 2026M-SQE preprint posted to arXiv, proposing multilingual skill quality estimation for agentic skill use (middle author, not corresponding).
- 2026PersonaPath benchmark released, built from 2,000 personas and 347 textbooks for personalized learning-path planning (middle author, not corresponding).
- 2026Both papers enter the Evolve review catalog, with reproducibility readiness rated high (M-SQE) and medium (PersonaPath); the two-paper reproducibility mean is 78.
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.
- The name Min Zhang is common; this profile is name-resolved from public bibliographic data, and no ORCID, institution, or country was provided to confirm identity.
- Only two papers — both 2026 arXiv preprints with middle-author, non-corresponding positions — are in the reviewed catalog, and their peer-review status is unknown.
- Radar scores come from a public bibliometric snapshot; the output, impact, and activity axes lack sufficient data and are unscored.
- Collaboration and reproducibility figures are computed only from the reviewed record and may not reflect the researcher's full publication history.