Nguyen Thai Anh · Van Lang University — LLM benchmarking and fairness auditing for educational measurement
2
papers here
0
reviewed by Evolve
0
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—
Output scoring is not yet available: the citation-linked record is too sparse for a reliable score, with h-index and i10-index both reported as 0. This axis would measure publication volume, multi-year output trends, and author-role distribution once a fuller bibliographic history accumulates.
Impact—
Impact scoring is unavailable because citation data for this author has not accumulated — both citation indices currently stand at 0. This axis would measure citation counts, h-index trajectory, and the scholarly influence of the work over time.
Focus—
A focus score cannot be computed yet: OpenAlex topic labels for this record are heterogeneous (brain connectivity, epilepsy, EEG interfaces, topic modeling, NLP) and do not map cleanly onto the reviewed papers. This axis would measure topical concentration versus spread.
Collaboration32/100
Evolve's collaboration score of 32 reflects a compact network: the OpenAlex record lists only 4 distinct co-authors, and the recurring cluster — Nguyen Dung Son, Dang Quang Minh, Nguyen Huu Loi, Truong Viet Vu — points to a stable intra-institutional team at Van Lang University. Both reviewed 2026 papers place the researcher in the last author position without corresponding-author status, consistent with senior supervisory involvement in a small, locally concentrated group.
Activity57/100
An activity score of 57 sits in the moderate band, matching the raw record: 6 works in the last three years, all concentrated in 2026 in the OpenAlex counts-by-year data. The pattern points to an emerging researcher who is currently active, with the 2026 cluster including both Evolve-reviewed arXiv studies, rather than a long-running publishing cadence.
Reproducibility—
Reproducibility is not yet scored at the profile level. Evolve's individual reviews flag both 2026 papers as medium repro-readiness, but the axis would aggregate code release, artifact availability, and dataset access across the full record once more works are reviewed.
Research summary
Nguyen Thai Anh is a researcher at Van Lang University in Vietnam whose visible work sits at the intersection of natural language processing, educational data mining, and machine-learning fairness, applied to educational measurement. The OpenAlex-resolved record (no ORCID linked) shows 6 works in the last three years, all dated 2026, with an h-index and i10-index of 0 — the signature of an early-stage profile whose citation footprint has not yet developed. Evolve's catalog includes two 2026 arXiv papers, both with the researcher in the last author position and neither as corresponding author. The first benchmarks sub-3B open language models for zero-shot automated essay scoring on an 8 GB consumer GPU; the review finds rubric-decomposed prompting the strongest rescue for small local models, even though the best quadratic weighted kappa (0.388) still trails a length-only baseline. The second is the first cross-architecture audit of demographic bias in deep knowledge tracing, judged statistically careful and honest, though its headline p-values sit at the permutation floor. OpenAlex topic labels on the record are heterogeneous — spanning functional brain connectivity, EEG and brain-computer interfaces, epilepsy, topic modeling, and NLP — which likely reflects topical noise in name-based disambiguation rather than the reviewed research agenda. Collaboration stays concentrated: 4 distinct co-authors form a recurring Van Lang University cluster, and the moderate activity score of 57 alongside a collaboration score of 32 depicts an active but early, locally rooted researcher.
Collaboration pattern
Collaboration is tightly concentrated: the OpenAlex record lists 4 distinct co-authors, matching the recurring cluster of Nguyen Dung Son, Dang Quang Minh, Nguyen Huu Loi, and Truong Viet Vu. Both Evolve-reviewed 2026 papers place the researcher last among the authors but not as corresponding author, consistent with a senior supervisory role inside a small Van Lang University team. The collaboration score of 32 reflects an intra-institutional network with little visible cross-institution or international reach so far.
Research keywords
Highlights
Sub-3B local LLMs for zero-shot essay scoring on consumer hardware
First cross-architecture audit of demographic bias in deep knowledge tracing
Timeline
- 2026OpenAlex records 6 works for this author in 2026, the only year with recorded output in the current counts-by-year data.
- 2026Last-author arXiv study benchmarks sub-3B open language models for zero-shot automated essay scoring on an 8 GB consumer GPU.
- 2026Co-authors a cross-architecture audit of demographic bias in deep knowledge tracing, Evolve-reviewed as statistically careful and honest.
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 indicators (h-index, i10-index, topics, co-author counts) come from an OpenAlex record without a linked ORCID; 'Nguyen Thai Anh' is a common Vietnamese name, so this profile is name-resolved from public bibliographic data and may mix homonymous records.
- OpenAlex topic labels on the record include biomedical themes (functional brain connectivity, epilepsy, EEG and brain-computer interfaces) that do not appear in the two reviewed papers, so topical coverage may be noisy.
- Output and impact dimensions are unscored because citation-linked data is insufficient (h-index 0 and i10-index 0 as reported).
- Qualitative assessment rests on only two Evolve-reviewed papers, both 2026 arXiv preprints, so verdicts reflect non-peer-reviewed versions of the work.