Knowledge Graph Mediated Translation (KGMT)
Machine translation grounded in your knowledge graph — accurate, consistent, and traceable, with near-zero hallucination.
For most language pairs, fluency is no longer the hard part. Control is: keeping domain terms, regulated language, and brand voice correct and consistent across every language, at volume. Knowledge Graph Mediated Translation (KGMT) is a method for giving a translation model that control — grounding it in a knowledge graph of your domain so the output stays faithful to verified knowledge, not just to what sounds plausible.
Where translation still breaks
In regulated and specialist content, the risk isn't clumsy phrasing. It's a model that quietly swaps in the wrong term, drops a qualifier that carries legal weight, or adds a detail that was never in the source. The longer and more domain-specific the text, the more often we've seen that happen. In a blog post it's an annoyance; in a label, a filing, or a contract it can be a liability.
How KGMT works
- Ground the model in your knowledge. A knowledge graph holds your domain's entities, terminology, and relationships as verified facts. KGMT supplies that to the model as it translates, so it draws on known knowledge rather than guessing — which constrains the fabricated output that ungrounded models tend to produce.
- Control the context, deliberately. Rather than hoping the model picks up the right context, KGMT delivers it on purpose. That helps most where plain models struggle: long-form and narrative text, abstract or specialist domains, and lower-resource languages.
- Keep provenance and version. KGMT can separate your organisation's own position from text it merely quotes, version content by date, and carry agreed brand characteristics through every language — so what ships is traceable, not only translated.
Optional add-ons
Automatic Post-Editing (APE) and COMET quality-estimation scores are available where you want an extra automated editing pass, or a measurable quality signal on every segment rather than a spot check.
Where it fits
KGMT runs standalone, or slots into what you already run — a translation management system, or another retrieval-augmented (RAG) workflow. It's additive: it makes the stack you have more controllable rather than replacing it.
Where it comes from
KGMT is a formal specification, developed by Edwin Trebels, Founder of LangOptima, with Lead Semantics, and implemented in their knowledge-transformation tool, TextDistil. The knowledge-graph grounding is the same approach LangOptima brings to enterprise data more broadly, here pointed specifically at translation.
See what it does to your own content.
One question for a first call: where does translation quality actually cost you today — and what would "controlled" look like across your languages?
What clients say about our development partner
“Guys at instea possess a rare talent in being able to grasp complex abstract concepts and turn them into working, scalable code, quickly and accurately.”
“I especially appreciate the structure and professionalism that instea has implemented in all development. I am pleased that we can bring new trends in the loyalty segment to gastronomy in Slovakia through technology.”
“We needed a special team to handle a complex topic with many analytical and development tasks. instea proved to be a very reliable partner for such kind of topics.”