Details

Viveta Gene, LangOptima, and Lead Semantics Announce Research Partnership

November 2025

We’re thrilled to announce a groundbreaking research partnership between Viveta Gene, PhD, LangOptima, and Lead Semantics, combining academic excellence with cutting-edge AI technology to transform terminology accuracy in AI-powered translation for regulated industries.

This collaboration unites practical localization expertise with advanced semantic technology to address one of the most critical challenges facing the translation industry: ensuring domain-specific accuracy when using large language models and knowledge graphs for post-editing.

Why This Partnership Matters

As AI continues to reshape the translation landscape, regulated industries demand precision and reliability. This partnership bridges the gap between theoretical research and real-world application, creating solutions that don’t just process text, but understand context, domain knowledge, and industry-specific terminology.

Viveta Gene, PhD brings extensive expertise as Head of Global Localization Solutions and Innovation at Intertranslations S.A., where she oversees cutting-edge localization applications while managing practical insights from 100M+ post-edited words in regulated industries. She is embarking on post-doctoral research at Ionian University under the supervision of Dr. Vilelmini Sosoni, Associate Professor in Translation Studies.

LangOptima brings proven expertise in language operations and marketing, helping companies grow through innovative AI-powered translation solutions. LangOptima’s Founder Edwin Trebels and Lead Semantics’ Knowledge Graph Mediated Translation (KGMT®) technology uses semantic knowledge graphs to improve domain-specific translation accuracy and fluency.

Lead Semantics brings deep AI and semantic technology expertise, helping popKnowledge Graphs (since 2019), Knowledge Graph RAG (KnowledgeGraphRAG, first discussed in 2023), and the TextDistil pipeline that transforms unstructured content into structured, meaningful knowledge.

The Research Focus

This research partnership explores:

“Evaluating the Impact of Prompt Engineering on Terminology Accuracy: A Comparative Analysis of Large Language Models and Knowledge Graphs for Post-Editing in Regulated Industries”

The research brings together LangOptima’s and Lead Semantics’ TextDistil pipeline and Knowledge Graph Mediated Translation (KGMT®) technology with practical post-editing insights, creating a synergy between university research, real-world translation operations, and advanced semantic AI solutions.

The synergy between Viveta Gene’s practical insights from Intertranslations‘ enterprise localization work (serving 100M+ post-edited words), practical insights from LangOptima’s expertise, and Lead Semantics’ cutting-edge and innovative semantic technology creates the perfect environment for breakthrough findings that will benefit:

• Academia: advancing translation technology research with empirical validation
• Regulated industries: improving localization accuracy for life sciences, legal, and financial sectors
• Translation professionals: enhancing post-editing workflows and quality
• The broader translation technology community: advancing best practices for AI-powered translation

From the Partners

“This research partnership perfectly complements my work objectives at Intertranslations, allowing me to explore cutting-edge AI applications while bringing practical insights back to our clients. Partnering with LangOptima and Lead Semantics unlocks the true potential of domain-specific, knowledge-driven AI for regulated industries.” —Viveta Gene, PhD, Head of Global Localization Solutions and Innovation at Intertranslations S.A.

“LangOptima is about helping companies grow and reach their largest audience possible. This research partnership with Viveta Gene and Lead Semantics advances our mission of unlocking domain-specific AI that truly understands context and terminology,  not just words.” — Edwin Trebels, Founder of LangOptima

“At Lead Semantics, our mission has always been to bring semantic computing to real-world enterprise problems. This partnership combines academic rigor with practical expertise to solve one of the most challenging problems in AI translation: ensuring accuracy in highly regulated domains where precision matters most.” — Prasad Yalamanchi, Founder of Lead Semantics

About the Partners

Viveta Gene, PhD is Head of Global Localization Solutions and Innovation at Intertranslations S.A., a leading international translation service provider serving regulated industries including life sciences, legal, and financial sectors. She is currently conducting post-doctoral research at Ionian University’s Department of Foreign Languages, Translation and Interpreting under the supervision of Dr. Vilelmini Sosoni, focusing on AI-powered translation technology for regulated industries.

LangOptima helps companies grow through marketing, knowledge graphs, translation operations, and custom software development. With expertise in workflow automation, fine-tuned LLMs, and Knowledge Graph Mediated Translation (KGMT), LangOptima applies objective-driven continuous improvement principles to help companies reach their largest audience possible.

Lead Semantics develops TextDistil, a language comprehension platform that transforms unstructured content into structured knowledge graphs. The company specializes in semantic computing, Knowledge Graph RAG, and AI-powered natural language processing for enterprise applications.

Edwin Trebels

Founder @LangOptima | Fractional Marketing and/or Localization Leader / Consultant | Business Intelligence | Knowledge Graph Mediated Translation (KGMT)

As the Founder of LangOptima, Edwin has over 20 years of marketing and translation industry experience and implement objective-driven continuous improvement (with a particular focus on ‘Toyota Kata’). He initiates and facilitates growth and development in a wide variety of global departments: marketing, translation, interpretation, source and target content strategy, multi-media production, human resources, training and vendor management.

Prasad Yalamanchi

Partner

Founder and CEO of Lead Semantics | Knowledge Graphs | Knowledge Graph Mediated Translation | Artificial Intelligence

Lead Semantics is the developer of TextDistil – a language comprehension platform solution that transforms unstructured content into structured knowledge graphs. Prasad brings his extensive expertise in computational linguistics and knowledge engineering to commercial applications of unstructured data. He possesses a Master’s degree in Computer Science and Computational Linguistics. In addition to his work at Lead Semantics, Prasad also serves as chair on select ontology engineering and natural language processing training programs and conferences.

Details

October 2022 (3 sessions), GALA ACADEMY, Localization project management training in MTPE

The Localization project management training in MTPE is an interactive training course consisting of 3 sessions, specifically designed for project managers, localization managers and vendor managers, but also anyone who is interested in best MTPE management practices. The purpose of this course is to set the specifications of a project eligible for MTPE and share insight into the know-how of the implementation of MTPE in the workflow by means of showcasing case studies and maintaining high interaction with the participants. Best practices on how to skyrocket post-editors’ performance in terms of quality and speed will be in the spotlight covering a wide range of aspects, from communication to guidelines, and applying a holistic approach.

GALA ACADEMY

Localization project management training in MTPE

Moderated by Viveta Gene and Maria do Campo

• Each event consists of 3-4 sessions
• Each session will be 90’.
• Interactive Format
• Target groups: project managers, localization managers, vendor managers

Session 1: 4^th October

When to apply a MTPE Workflow

What are the specifications that make a project eligible for MTPE? Are there specific use-cases that make it a good fit? How to weigh quality, cost and time and implement a successful workflow? And what about data confidentiality and matters of ethics in MTPE?

Join us to a quiz-like session to trigger discussion and train ourselves in the specifications that make a project a good fit for MTPE. Let’s lay together the foundation one step before the deep dive!

Session 2: 11^th October

How to apply a MTPE Workflow?

What are the MTPE stages and what is the PM role, proactive actions to take this journey and successfully reach the destination of positive feedback by the end customer?

We will make this an interactive case study session sharing the takings of our experience and guidelines on how to avoid the pitfalls and reap the benefits.

Contribute yourself to the training with your own challenging case, or simply listen and enjoy.

Session 3: 18^th October

Put Yourself in the Shoes of The Post-Editor. Get the training needed to have your team trained!

Have you ever thought that not all translators’ profiles are a good fit for MTPE?

What is the training that should be provided to the post-editors to ensure a positive attitude towards MTPE? And how will you be able to skyrocket post-editors’ performance in terms of quality and speed?

It is not all about guidelines… Join us to gain insight!

Details

MTPE has raised several issues over the past few years, on both a linguistic and business level. Are Post-Editors and LSPs up for the challenge? Viveta Gene joins Together 2022 to discuss MTPE challenges and to share training tips and tricks to make the most of this valuable service.

Register for Together 2022 and join the discussion on one of the services that is shaping the future of our industry. The early bird offer is still on, so make sure you secure your place now!

Details

As MTPE progressively finds its way into the workflows of more and more LSPs, a new model of professional ethics might be surfacing. While MT technology is just starting to reveal its true power, the human factor remains nonetheless significant, with the rise of the Post-Editor’s role. So, in these newly established workflows, what does MTPE stand for in terms of professional ethics? Are there any guiding principles or standards specific to MTPE work? And how can LSPs reap the benefits of MTPE while conforming to ethical standards? This webinar attempts to tackle all these questions in the context of Elia’s “Focus Forward Webinar Series”, which focuses on some of the most relevant issues faced by the language industry today.

You can watch a recording of the webinar here.

Details

The success of machine translation post-editing tasks depends heavily on the existence of clear post-editing guidelines. Despite this fact, there is currently no established protocol or international standard that spells out specific details regarding the content and scope of such guidelines.

With that in mind, and considering the confusion resulting from this important gap among linguists, clients and translation professionals, this workshop sets out to explore ways of effectively designing a set of post-editing guidelines.

Attendees were invited to share their opinions and experiences through live polls and surveys which will be used to draft a common template for post-editing guidelines. The resulting document will be published as a paper summarizing this process.

Stay tuned!

Details

The success of machine translation post-editing tasks depends heavily on the existence of clear post-editing guidelines. Despite this fact, there is currently no established protocol or international standard that spells out specific details regarding the content and scope of such guidelines.

With that in mind, and considering the confusion resulting from this important gap among linguists, clients and translation professionals, this workshop sets out to explore ways of effectively designing a set of post-editing guidelines.

Attendees were invited to share their opinions and experiences through live polls and surveys which will be used to draft a common template for post-editing guidelines. The resulting document will be published as a paper summarizing this process.

Stay tuned!

Details

This exclusive 2-part online workshop aims to highlight how efficient MTPE training can increase productivity and improve output quality, streamlining the post-editing process.

Prepared by Viveta Gene (Intertranslations) and Lucía Guerrero (CPSL), in collaboration with Memsource in the context of its “Machine Translation Workshop Series”, this initiative aspires to show the effectiveness of MTPE training using empirical data, while offering participants the chance to improve their skills, increasing their speed and learning to avoid common mistakes.

To this end, participants were invited to complete a short post-editing task prior to each of the workshop’s live sessions. Measuring the participants’ productivity rate before and after they receive MTPE training shed light on how training may impact both speed and accuracy in post-editing.

You can watch a recording of both Sessions here.

Details

Translating Europe Forum (TEF) is an annual event that has been bringing together translation professionals from all European countries for the past 8 years. This year’s edition, taking place entirely online, was dedicated to “Collaboration in Times of Automation”: translation is now far from the solitary profession it used to be and existing technologies may support its newfound collaborative aspects.

Intertranslations was proudly a part of TEF 2021, with Viveta Gene, Translation & Localization Industry Specialist, chairing the “Collaboration on Training: Industry and Academia” panel, where she presented a set of best practices for post-editing activities, based on recent findings in the field of MTPE training.

A recording of the “Collaboration on Training: Industry and Academia” panel can be found here.

Details

Intertranslations will be a part of this year’s virtual edition of TC42 (Translating and the Computer), a leading forum that brings together language professionals from various backgrounds as well as academic and business researchers to discuss latest trends and share insights on business practices. Our contribution to the conference aims at helping both businesses and linguists alike understand the scope of a Post-Editor’s role and examines current attitudes towards Post-Editing. Do negative perspectives outweigh the benefits, undermining the efficiency of this cost-effective and time-efficient new service? Make sure you register here to get a grip on industry perspectives by following our upcoming presentation at TC42!

Details

The transition from Translation Age to Post-Editing Age poses many challenges for the translation industry. LSPs need to build the role of the post-editor, define the post-editing process, and set accurate expectations for productivity and quality. This two-part webinar discusses the challenges of managing Machine Translation Post-Editing (MTPE) including changes in roles, workflows, and metrics, and the challenge of training for MTPE.

The first part of this interactive webinar focuses on the following trio of “how-to” questions:

• How do you create a common perspective between vendors and organizations?
• How do you manage vendor concerns?
• How do you build a training protocol?

All webinar participants will be given a post-webinar survey which will be analyzed and the results presented in the second webinar on 21 May 2020. That second webinar will zero in on how we can create a common perspective on post-editing between vendors and LSPs. The results will also be used to suggest a common post-editing training protocol.

Details

The transition from Translation Age to Post-Editing Age poses many challenges for the translation industry. LSPs need to build the role of the post-editor, define the post-editing process, and set accurate expectations for productivity and quality. This two-part webinar discusses the challenges of managing Machine Translation Post-Editing (MTPE) including changes in roles, workflows, and metrics, and the challenge of training for MTPE.

The first part of the webinar presented the management and training challenges of Machine Translation Post-Editing in an attempt to create a common perspective on post-editing for vendors and LSPs. The webinar concluded with a survey that was made available to the attendees. This second webinar analyzes the results of that survey and presents its findings with regard to develping a common and satisfactory Post-Editing Training Protocol for both vendors and LSPs.

Details

Much has been said and written about the pros and cons of using machine translation with post-editing (MTPE) to get text translated. Yet it is clear that machine translation is here to stay. Given that assumption, the most important question that arises from a business perspective is what evaluation methods can be used to measure MT output and the subsequent reduction in time and money. How is post-editing productivity measured? And how can post-editing quality assessment be applied in the modern localization pipeline?

This webinar presents a case-study that was undertaken by Intertranslations for a US-based software company in which time, cost, and quality were carefully analyzed using a practical evaluation model.

Details

This webinar is geared towards linguists who work in the localization industry and are moving from a translation to a post-editing role. It will discuss three topics:

1. Define what light versus full post-editing is from a theoretical point of view, supported by practical examples
2. Discuss the practical implications of neural and statistical MT for the post-editing process
3. Outline how linguists can move from a translator to a post-editor role by acquiring specific transition skills

The main focus of this webinar is on the transition from the “Translation Age” to the “Post-Editing Age”, and will shed more light on the grey zones.