Early-career NLP researcher at Van Lang University: small-language-model benchmarking and fairness 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: the OpenAlex snapshot shows h-index 0 and no verified venue list, so output volume cannot be scored yet. This axis would measure total publications, venue types, and yearly output.
Impact—
Insufficient data: citation counts are not yet available (h-index and i10-index both 0 in the snapshot), so research impact cannot be scored. This axis would measure citations, citation trajectory, and influence on follow-up work.
Focus—
Insufficient data: the five indexed topics each appear once and reviewed work spans essay scoring and knowledge tracing, so a stable focus score cannot be computed. This axis would measure topical concentration versus spread over time.
Collaboration32/100
Scores 32/100, reflecting a small early network: the OpenAlex snapshot lists 4 distinct coauthors, with recurring partners Nguyen Dung Son, Dang Quang Minh, Nguyen Huu Loi, and Nguyen Thai Anh. Both reviewed papers are multi-author arXiv preprints with the researcher in a middle author position, consistent with team-based applied ML work rather than solo or corresponding-author leadership. Network scale should grow as the publication record matures.
Activity61/100
Scores 61/100 on strong recent momentum: the snapshot records 7 works in the last three years, with yearly indexed output rising from 1 work in 2025 to 6 in 2026. Both Evolve-reviewed papers appeared in 2026, suggesting an intense current phase concentrated on arXiv-first releases in NLP and educational data mining rather than a long accumulated record.
Reproducibility—
Insufficient data: no systematic artifact or code-release record is available yet; both reviewed papers carry a medium repro-readiness rating from Evolve review, which alone cannot support a score. This axis would measure code, data, and checkpoint availability across the portfolio.
Research summary
Truong Viet Vu is an early-career applied machine learning researcher at Van Lang University in Vietnam whose work concentrates on natural language processing and educational data mining. The profile is name-resolved from public bibliographic data (OpenAlex identity; ORCID 0009-0004-3941-1306), and the record is young: the snapshot shows an h-index of 0, 7 works over the last three years, and indexed output rising from 1 work in 2025 to 6 in 2026. Two Evolve-reviewed 2026 arXiv papers define the current research line. The first benchmarks sub-3B open language models for zero-shot automated essay scoring on a single 8 GB consumer GPU, using rubric-decomposed prompting; Evolve credits a candid, statistically careful benchmark while noting the best quadratic weighted kappa (0.388) still loses to a length-only baseline. The second, 'Accuracy Is Not Enough,' is a cross-architecture audit of demographic bias in deep knowledge tracing using ABROCA, praised as statistically careful and honest even though headline p-values sit at the permutation floor. Topic records — topic modeling, neural network applications, medical image segmentation, epilepsy research, and NLP — suggest broad applied interests beyond the education focus. Collaboration is small but stable: 4 distinct coauthors, with Nguyen Dung Son, Dang Quang Minh, Nguyen Huu Loi, and Nguyen Thai Anh recurring, and the researcher in middle author positions without corresponding role. Activity scores 61/100 on recent momentum; collaboration scores 32/100 on network size. Output, impact, focus, and reproducibility lack sufficient data to score.
Collaboration pattern
Collaboration is compact and stable: the OpenAlex snapshot lists 4 distinct coauthors, and Evolve's catalog shows the same recurring group — Nguyen Dung Son, Dang Quang Minh, Nguyen Huu Loi, and Nguyen Thai Anh — around both reviewed papers. The researcher sits in a middle author position without corresponding-author role, consistent with contributing specialist work inside a small team at Van Lang University. The low collaboration score (32/100) reflects network breadth, not absence of partners.
Research keywords
Highlights
Rubric-decomposed prompting for local zero-shot essay scoring
First cross-architecture audit of demographic bias in deep knowledge tracing
Rapid 2026 output surge across varied topics
OpenAlex records 6 works indexed in 2026 versus 1 in 2025, with 7 works over the last three years; indexed topics span topic modeling, neural network applications, medical image segmentation, epilepsy research, and NLP alongside the education-focused line.
Timeline
- 2025First indexed work appears in the OpenAlex snapshot, with 1 publication recorded for the year.
- 2026Indexed output rises to 6 works, the most active year on record in the snapshot.
- 2026Co-authors the Evolve-reviewed arXiv benchmark of sub-3B open language models for zero-shot essay scoring on an 8 GB GPU, in a middle author position.
- 2026Co-authors the cross-architecture audit of demographic bias in deep knowledge tracing on arXiv, reviewed by Evolve the same year.
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.
- Bibliometrics come from the OpenAlex snapshot only (identity: openalex); h-index 0 and i10-index 0 reflect thin indexing, not necessarily zero real citations.
- Only 2 Evolve-reviewed papers (both 2026 arXiv preprints, middle author, non-corresponding) inform this profile; career facts such as role history, education, and awards are unavailable.
- Output, impact, focus, and reproducibility dimensions are unscored due to insufficient data; no estimates were made.
- Identity is name-resolved from public bibliographic data via OpenAlex and ORCID 0009-0004-3941-1306; Vietnamese names can be shared, so attribution may over- or under-count same-name authors.