Early-career NLP researcher working on resource-efficient essay scoring and fairness audits in educational AI
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—
Insufficient data: this axis would measure cumulative publication volume, per-year output, and authorship roles across the record; the OpenAlex snapshot currently lacks the aggregated coverage needed to compute a reliable score.
Impact—
Insufficient data: this axis would capture citation counts, h-index and i10-index trajectory, and uptake; the current record reports h-index 0 and i10-index 0, too early to score fairly.
Focus—
Insufficient data: this axis would measure how concentrated the portfolio is across topics and venues; the five OpenAlex topics each carry a single work, too sparse to characterize a focus.
Collaboration32/100
The OpenAlex record lists 4 distinct co-authors, matching the recurring collaborators Nguyen Dung Son, Dang Quang Minh, Truong Viet Vu, and Nguyen Thai Anh, and both catalog papers are middle-position team efforts rather than solo or corresponding-author work. The collaboration score of 32 reflects this small, tight-knit circle: an early-stage pattern of repeated joint writing within one core group, with no evidence yet of a broad external network.
Activity32/100
Activity data from OpenAlex shows one indexed work in 2021 and two in 2026, with 2 works in the last three years, giving an activity score of 32. This suggests an intermittent, early-stage publishing rhythm with a clear recent burst: both catalog papers appeared in 2026, indicating current momentum rather than sustained high-volume output.
Reproducibility—
Insufficient data: this axis would assess released code, datasets, and replication feasibility; only qualitative medium repro-readiness flags exist for the two catalog papers, with no systematic artifact evidence.
Research summary
Nguyen Huu Loi is an early-career researcher whose profile is name-resolved from public OpenAlex bibliographic data; the record carries no institutional affiliation, country, or ORCID. The Evolve catalog holds two 2026 arXiv papers with this researcher as a middle-position co-author, both at the intersection of natural language processing and educational technology. The first asks how far sub-3B open language models can go in zero-shot automated essay scoring on an 8 GB consumer GPU; Evolve's review finds rubric-decomposed prompting best rescues local scoring, while the best quadratic weighted kappa of 0.388 still loses to a length-only baseline. The second, 'Accuracy Is Not Enough,' is reviewed as the first cross-architecture audit of demographic bias in deep knowledge tracing, applying ABROCA-based fairness analysis, and is judged statistically careful and honest. OpenAlex topics — topic modeling, NLP techniques, AI ethics and social impacts, adversarial robustness, and explainable AI — each carry a single work. Quantitative signals remain thin: h-index and i10-index are both 0, countsByYear lists one 2021 work against two in 2026 (recent3yWorks 2), and 4 distinct co-authors, with recurring collaborators Nguyen Dung Son, Dang Quang Minh, Truong Viet Vu, and Nguyen Thai Anh. Collaboration and activity each score 32, matching a small core team that has become recently active.
Collaboration pattern
All catalog work is collaborative: both 2026 papers list the researcher as a middle-position co-author, and the OpenAlex record shows 4 distinct co-authors. The recurring circle — Nguyen Dung Son, Dang Quang Minh, Truong Viet Vu, and Nguyen Thai Anh — suggests one stable core team writing across both NLP-for-education and fairness-audit projects, with no corresponding-author or solo papers visible yet.
Research keywords
Highlights
Rubric-decomposed prompting for local zero-shot essay scoring
First cross-architecture audit of demographic bias in deep knowledge tracing
Timeline
- 2026Co-authored a zero-shot essay scoring benchmark testing sub-3B open language models on an 8 GB consumer GPU against the ASAP-AES dataset.
- 2026Co-authored 'Accuracy Is Not Enough,' the first cross-architecture audit of demographic bias in deep knowledge tracing using ABROCA.
- 2026OpenAlex indexes 2 works for this year with 4 distinct co-authors — the most active point in the record so far.
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 figures (h-index 0, i10-index 0, countsByYear) come from an OpenAlex record resolved by name; the name is common, so the record may merge or miss works.
- No institutional affiliation, country, or ORCID is attached to this profile; identity rests on public bibliographic data only.
- Quantitative coverage is thin — one 2021 work and two 2026 works — so the output, impact, focus, and reproducibility axes are left unscored.
- Both catalog papers are 2026 arXiv preprints reviewed by Evolve; no peer-reviewed venue information is present in the data.