Machine learning researcher spanning LLM pretraining and NLP-based educational assessment
2
papers here
2
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
—
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-16) 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 publication-count or venue data underlies this axis yet. Output would measure total publications, venues, and yearly productivity; beyond the two 2026 preprints reviewed here, the available data cannot characterize it.
Impact—
Citation-based indicators are not available for this profile. Impact would measure citations, citing works, and influence on follow-up research; with no such metrics in the input data, no score is estimated.
Focus0/100
The focus score is 0, derived from a field distribution split exactly 50/50 between two distinct areas — AI in education / educational data mining (NLP for assessment) and machine learning (large language models). The two 2026 preprints sit one in each field, so the current catalog shows no concentration in a single domain; the score reflects this even split rather than measured quality of either line of work.
Collaboration71/100
The collaboration score is 71, based on 34 distinct co-authors in the underlying bibliometric record. The frequent-coauthor list is dominated by a recurring cluster — Junkai Tong, Mingjia Li, Haoran Chen, Yaoyu Jiang, Hanjie Ge, Hong Qian, Jiaqi Cao, Chiyu Chen, Shuang Cheng, Xu Cheng — plus the group credit 'The Intern-NCP Team', consistent with team-based, large-author-list work on both 2026 preprints, on which Yixuan Wang appears as a middle author.
Activity—
Recent-activity data is insufficient. This axis would track publishing cadence and year-by-year engagement; with papers confined to a single year (2026) and no longitudinal records, activity cannot be scored.
Reproducibility78/100
Reproducibility averages 78 across the 2 scored papers. NCP-ArchPreview carries a high reproducibility-readiness label, ProIQA a medium one; both Evolve verdicts note missing significance tests (ProIQA on small datasets, NCP-ArchPreview on Stage-1/Stage-2 gains). The score therefore reflects generally strong reproducibility readiness with open statistical-rigor caveats rather than verified replication.
Research summary
Yixuan Wang is an early-stage machine learning researcher whose reviewed work spans two adjacent areas: large language model pretraining and NLP-based educational assessment. The evidence base is a full profile assembled from public bibliographic data plus two 2026 arXiv preprints reviewed by Evolve, on both of which Wang appears as a middle author rather than first or corresponding author. The first, ProIQA, applies LLM-constructed reasoning trees and graph neural networks to fine-grained quality assessment of math test items, improving over stem-only baselines; Evolve notes the gains lack significance tests on small datasets. The second, the NCP-ArchPreview technical report, explores latent-space language modeling through next concept prediction, claiming 1.95× token-efficient pretraining at 8.9B parameters, with Stage-1 gains (+2.45) shrinking in Stage-2 (+0.59). The field distribution splits exactly 50/50 between AI in education / educational data mining and machine learning (large language models), and the focus score of 0 reflects that even split. Collaboration is the strongest populated signal: a collaboration score of 71 built on 34 distinct co-authors, with a recurring cluster including Junkai Tong, Mingjia Li, Haoran Chen, Yaoyu Jiang, Hanjie Ge, and Hong Qian. Reproducibility averages 78 across the 2 scored papers. Output, impact, and activity lack sufficient data and are left unscored.
Collaboration pattern
Yixuan Wang works in sizeable teams: the bibliometric record lists 34 distinct co-authors, and both cataloged 2026 papers carry middle authorship without a corresponding-author role. A stable cluster recurs across the record — Junkai Tong, Mingjia Li, Haoran Chen, Yaoyu Jiang, Hanjie Ge, Hong Qian, Jiaqi Cao, Chiyu Chen, Shuang Cheng, Xu Cheng — alongside the group credit 'The Intern-NCP Team' on the NCP report, indicating lab- or intern-program-scale collaboration rather than solo or duo work.
Research keywords
Highlights
ProIQA: process-based, fine-grained quality assessment of math test items
NCP-ArchPreview: next concept prediction toward latent-space language models
Timeline
- 2026Co-authored ProIQA, a process-based framework for fine-grained math item quality assessment (arXiv preprint; middle authorship).
- 2026Contributed to the NCP-ArchPreview technical report on next concept prediction for latent-space language models (arXiv preprint; middle authorship).
- 2026Catalog record spans both AI-in-education assessment and LLM pretraining, with 34 distinct co-authors and a mean reproducibility score of 78 across the two papers.
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 scores are derived from public bibliographic data (OpenAlex-sourced) only; no citation counts, h-index, or venue-tier data were available.
- The name 'Yixuan Wang' is common; this profile is name-resolved from public bibliographic data, and institution, country, and ORCID are unconfirmed.
- Only two 2026 arXiv preprints reviewed by Evolve underlie the paper-level evidence; output, impact, and activity dimensions lack sufficient data and are not scored.
- One entry in the frequent-coauthor list is a formatting artifact and was excluded from collaboration analysis.