NLP researcher at NTU Athens studying machine translation evaluation, gender bias in metrics, and LLM interpretability
2
papers here
2
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
4
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-15) 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—
An output score is not yet available: this snapshot lacks the verified publication-volume data the axis requires. It would measure total works and productivity over time; current yearly counts (29 works in 2026) await full indexing.
Impact—
No impact score exists yet: citation-side data is insufficient in this snapshot. The axis would measure scholarly influence via citations, h-index, and uptake; only a preliminary h-index of 4 and i10-index of 3 are recorded, too thin to score.
Focus0/100
The focus score of 0 reflects an evenly split, broad portfolio: OpenAlex assigns 0.5 to natural language processing (machine translation evaluation) and 0.5 to NLP/model interpretability. Indexed topics spread across topic modeling (34 works), NLP techniques (15), sentiment analysis and opinion mining (12), multimodal machine learning (12), and explainable AI (9) — no dominant cluster.
Collaboration46/100
A collaboration score of 46 rests on 9 distinct coauthors in the OpenAlex record. Recurring partners include Orfeas Menis Mastromichalakis, Wafaa Mohammed, Giuseppe Attanasio, Chrysoula Zerva, Giannis Kalyvas, Vassilis Lyberatos, and Giorgos Stamou. Both Evolve-reviewed papers list him in a middle author position without corresponding role.
Activity100/100
Activity reaches the maximum score of 100. Indexed works per year climb from 1-3 in 2020-2023 to 12 in 2024, 16 in 2025, and 29 in 2026 (year still in progress), with 57 works in the last three years — a steep, ongoing acceleration.
Reproducibility55/100
Reproducibility scores 55, a reproMean computed over 2 assessed papers. Both Evolve-reviewed works carry medium repro-readiness: the gender-bias benchmark builds on an occupation-balanced GAMBIT+ subset and WMT26 evaluator runs, while the counterfactual self-explanations study reports aggregate-level statistics that temper direct replication of its causal claim.
Research summary
Giorgos Filandrianos is a researcher at the National Technical University of Athens working across two complementary strands of natural language processing: machine translation evaluation and model interpretability. The OpenAlex record, resolved from public bibliographic data, shows a steep productivity ramp: from 1-3 indexed works per year in 2020-2023 to 12 in 2024, 16 in 2025, and 29 in 2026 to date, with 57 works over the last three years and a maximum activity score of 100. His field distribution splits evenly (0.5/0.5) between machine translation evaluation and NLP/model interpretability, and indexed topics — topic modeling (34 works), NLP techniques (15), sentiment analysis and opinion mining (12), multimodal machine learning (12), and explainable AI (9) — confirm the breadth behind a focus score of 0. Two 2026 arXiv papers reviewed by Evolve illustrate both strands: a benchmark of WMT26 machine translation evaluators on an occupation-balanced GAMBIT+ subset, finding a masculine preference in 126 of 131 significant runs, and an empirical study linking counterfactual self-explanation quality to model scale (0.87 correlation with flip rate). Both carry medium repro-readiness (reproMean 55) and middle authorship. With an h-index of 4 and 9 distinct coauthors, this is an early-stage, fast-scaling profile whose citation impact is not yet scoreable.
Collaboration pattern
OpenAlex records 9 distinct coauthors, and the co-occurrence list points to Orfeas Menis Mastromichalakis as the most frequent partner, alongside Wafaa Mohammed, Giuseppe Attanasio, Chrysoula Zerva, Giannis Kalyvas, Vassilis Lyberatos, and Giorgos Stamou — a network anchored on his National Technical University of Athens affiliation. In both Evolve-reviewed papers he sits in a middle author position without corresponding role, suggesting a contributor pattern within small NLP teams rather than sole or senior leadership.
Research keywords
Highlights
Masculine preference in 126 of 131 significant WMT26 evaluation runs
Counterfactual self-explanations strengthen with model scale (ρ = 0.87)
Publication output accelerates after 2023
Indexed works jump from 1-3 per year (2020-2023) to 12 in 2024, 16 in 2025, and 29 in 2026 to date — 57 works in the last three years.
Timeline
- 2020OpenAlex record begins with 1 indexed work (2 more follow in 2021).
- 2023Output rises to 3 indexed works, the pre-2024 peak.
- 2024Productivity accelerates sharply to 12 indexed works.
- 202516 indexed works; sustained growth continues.
- 202629 indexed works so far; both Evolve-reviewed arXiv papers appear — the WMT26 gender-bias evaluation benchmark and the counterfactual self-explanations study.
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 derive from OpenAlex only (identity: openalex); coverage gaps or identity splits are possible.
- Output and impact dimensions lack sufficient data and are reported as missing, not scored.
- Qualitative verdicts rest on only two Evolve-reviewed 2026 arXiv papers, both with middle authorship.
- The 2026 count of 29 works covers an incomplete year and may change.