Skip to main content

Research & Academic Partnerships

We build alongside universities and researchers

LangOptima works with academic institutions to advance knowledge-guided AI and translation: joint research, a university mentorship program, and open, peer-reviewed publications. We welcome new collaborations.

Working with academia

How we collaborate

Joint research

Our work on Knowledge Graph Mediated Translation is published with the Ionian University, led by Viveta Gene, PhD, with Dr. Vilelmini Sosoni of the Ionian University. Viveta is Associate Researcher at LangOptima, a postdoctoral researcher and adjunct professor at the Ionian University, and Head of Global Localization Solutions at Intertranslations S.A.

Read the papers ↓

University mentorship

We run a mentorship program with the University of Warwick and Monash University, connecting students and early-career researchers with applied industry work.

See the mentorship program →

Partner with us

We welcome further partnerships with universities, research groups, and individual researchers: joint studies, shared data, co-authored publications. Tell us what you're working on.

Start a conversation →

The research

Two peer-reviewed papers, free to download

Both 2026 papers by Viveta Gene, PhD, and Dr. Vilelmini Sosoni on how a knowledge layer makes machine translation dependable in regulated, high-risk industries. Enter your details once and we send both.

Convergence 2026Peer-reviewedProceedings in press

Knowledge-guided machine translation for regulatory compliance in high-risk industries

Viveta Gene, PhD, and Dr. Vilelmini Sosoni

Sets out how a knowledge layer guides machine translation for regulatory compliance in high-risk industries — where a mistranslated instruction, dosage, or safety warning carries real regulatory and human cost — and why grounding translation in structured domain knowledge, rather than relying on a larger language model alone, is what makes regulated translation dependable at scale.

Venue:
Proceedings of Convergence 2026: Human-AI Integration for Multilingual and Accessible Communication
Where:
Guildford, UK (University of Surrey)
When:
June 17–19, 2026
Pages:
pp. 67–82
DOI:
pending (proceedings in press)
NeTTIT 2026Peer-reviewed

Dual-Metric Compliance and Quality Evaluation of Knowledge Graph Mediated Translation in Regulated Domains: An Enhanced Architectural Framework

Viveta Gene, PhD, and Dr. Vilelmini Sosoni

Introduces a dual-metric evaluation — compliance and quality scored separately — for knowledge-graph-mediated translation in regulated domains, with an enhanced architectural framework. On a controlled 30-error regulatory corpus (a 500-word retinol product description, EN→Greek/French), KGMT detected 100% of source-compliance violations while two LLM baselines detected 0%. The controlled-corpus result is peer-reviewed; how it generalizes to any given corpus is directional.

Venue:
Proceedings of the 3rd International Conference on New Trends in Translation and Interpreting Technology (NeTTIT)
Where:
Dubrovnik, Croatia
When:
June 24–27, 2026
Pages:
pp. 109–118
DOI:
10.26615/issn.2815-4711.2026_015

Both papers are delivered together: one form, two PDFs. Full citations are at the foot of this page.

Get both papers

Enter your details and we'll send both peer-reviewed papers to your inbox.

No spam. We'll only contact you about these papers, unless you want to talk.

See it applied

These papers back our Translation Compliance calculator

The 100%-vs-0% detection result from the NeTTIT paper is the one external figure behind our Translation Compliance Exposure calculator, a tool that models what a compliance violation costs as it propagates through translation to every market.

Open the calculator →

The method behind the papers

Knowledge graphs, across your industry

The knowledge layer these papers evaluate is the same one that connects an organization's data for AI. Explore how it applies to your data landscape and your sector.

Knowledge Graphs

Prepare your data for AI with a connected knowledge layer.

KGMT

Knowledge Graph Mediated Translation, the method these papers evaluate.

Context Graphs

The product that puts a knowledge graph to work on your data.

By industry

Content & MediaCustomer SupportCybersecurityEnergy & UtilitiesFinancial ServicesGovernment & Public SectorHealthcareHuman ResourcesInsuranceLegalLife SciencesManufacturingResearch & AcademiaRetail & E-commerceTelecommunications

Credits & Citations

Please cite the papers as follows. Both are the work of Viveta Gene, PhD, and Dr. Vilelmini Sosoni.

  1. Gene, V., & Sosoni, V. (2026). Knowledge-guided machine translation for regulatory compliance in high-risk industries. Proceedings of Convergence 2026: Human-AI Integration for Multilingual and Accessible Communication, pp. 67–82. Guildford, UK, June 17–19, 2026.
  2. Gene, V., & Sosoni, V. (2026). Dual-Metric Compliance and Quality Evaluation of Knowledge Graph Mediated Translation in Regulated Domains: An Enhanced Architectural Framework. Proceedings of the 3rd International Conference on New Trends in Translation and Interpreting Technology, pp. 109–118. June 24–27, 2026, Dubrovnik, Croatia. ©NeTTIT 2026. https://doi.org/10.26615/issn.2815-4711.2026_015

Go deeper

See knowledge-guided translation in practice

These papers evaluate Knowledge Graph Mediated Translation (KGMT), which checks compliance at source so a violation never propagates through your language pipeline. See how KGMT makes regulated translation auditable.

See Knowledge Graph Mediated Translation (KGMT) →