Trustworthy AI at Technical University of Cluj-Napoca: fact-checking, misinformation detection, health-record integrity
3
papers here
0
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
0
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-17) 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 output score is available: the snapshot reports an h-index and i10-index of 0 and no fuller career publication volume, so OpenAlex coverage appears incomplete. This axis would measure total works, output trajectory, and venue spread over time.
Impact—
No impact score is available. Citation counts and venue-level influence are absent from the snapshot, likely because the three 2026 arXiv preprints are too recent to accrue citations. This axis would measure citation-based scholarly influence over time.
Focus—
No focus score is available. Topic labels exist — Topic Modeling, Misinformation, and Spam and Phishing Detection each appear twice — but the sample is too small to score topical concentration. This axis measures stability and consistency of research themes.
Collaboration32/100
Evolve's radar scores collaboration at 32, based on four distinct co-authors in the OpenAlex snapshot. The recurring cluster — Marius Nicusor Pantea, Nora Girda, Amelia Petrenciuc, and Alexandru Lecu — points to a small, stable local team at the Technical University of Cluj-Napoca. On all three cataloged 2026 papers Groza appears in the last author position without corresponding-author flags, consistent with supervising junior researchers rather than wide international co-authorship.
Activity40/100
Activity is scored at 40. The snapshot lists 3 works in the last three years, all concentrated in 2026 — countsByYear shows exactly 3 publications in 2026 and none in earlier covered years. This is a compact but recent burst: three arXiv preprints in a single year on closely related misinformation and health-record integrity problems, indicating an active but newly visible publication stream rather than sustained multi-year volume.
Reproducibility—
No reproducibility score is available. None of the three cataloged papers carries a reproReadiness assessment or an Evolve verdict yet, and code or dataset availability is unknown. This axis would measure artifact sharing, evaluation transparency, and replication readiness.
Research summary
Adrian Groza is a researcher at the Technical University of Cluj-Napoca in Romania whose recent work targets trustworthy AI for information integrity: detecting misinformation, verifying claims, and gating AI-assisted health records. The Evolve catalog lists three 2026 arXiv preprints, on all of which Groza holds the last author position: a source-grounded integrity-gate framework for AI-assisted personal health records; a memory-first, knowledge-graph-grounded multi-agent system for misinformation detection; and an explainable coherence score for detecting temporal inconsistencies in political news. OpenAlex topic assignments reinforce this focus — Topic Modeling, Misinformation and Its Impacts, and Spam and Phishing Detection each appear twice, alongside Electronic Health Records Systems and Access Control and Trust. Bibliometric coverage is thin: the snapshot reports an h-index and i10-index of 0, three works in the last three years, and four distinct co-authors, yielding radar scores of 32 for collaboration and 40 for activity. Output, impact, focus, and reproducibility are not yet scored. This profile is name-resolved from public bibliographic data (identity: openalex).
Collaboration pattern
The OpenAlex snapshot records 4 distinct co-authors, all recurring and apparently affiliated with the Technical University of Cluj-Napoca: Marius Nicusor Pantea, Nora Girda, Amelia Petrenciuc, and Alexandru Lecu. Groza's consistent last-author position on all three 2026 papers, without corresponding-author flags, fits a senior-supervisor pattern within a small local group. No international or industry collaborations appear in the available data.
Research keywords
Highlights
Memory-first fact-checking with knowledge-graph grounding
Per the 2026 arXiv paper 'Memory-First Fact-Checking: A Knowledge-Graph-Grounded Multi-Agent System for Misinformation Detection', this work combines a knowledge-graph-grounded architecture with a multi-agent design for misinformation detection, consulting stored knowledge first when verifying claims.
Explainable coherence score for political news
Per the 2026 arXiv paper 'An Explainable Coherence Score for Detecting Temporal Inconsistencies in Political News', this work introduces an explainable scoring approach that surfaces temporal inconsistencies in political news coverage.
Source-grounded integrity gates for AI-assisted health records
Per the 2026 arXiv paper 'Review Before Trust: Source-Grounded Integrity Gates for AI-Assisted Personal Health Records', this work proposes source-grounded integrity gates that require review before trusting AI-assisted personal health record content.
Timeline
- 2026Released 'Review Before Trust: Source-Grounded Integrity Gates for AI-Assisted Personal Health Records' on arXiv, in the last author position.
- 2026Released 'Memory-First Fact-Checking: A Knowledge-Graph-Grounded Multi-Agent System for Misinformation Detection' on arXiv, in the last author position.
- 2026Released 'An Explainable Coherence Score for Detecting Temporal Inconsistencies in Political News' on arXiv, in the last author position; all three cataloged works fall in this single 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 are computed from an OpenAlex snapshot only; no ORCID was provided, so the profile is name-resolved from public bibliographic data.
- The snapshot reports an h-index and i10-index of 0, likely reflecting incomplete coverage rather than an empty career record.
- All cataloged works are 2026 arXiv preprints with no Evolve verdicts yet, so citation impact cannot be assessed at this stage.
- Field labels, keywords, and reproReadiness are null for every cataloged paper, limiting field and reproducibility scoring.