Translated levels up adaptive MT with expanded document context

The company’s machine translation (MT) system now leverages up to 100,000 words of document content to provide translation suggestions, drastically reducing the time to edit MT suggestions. The upgrade is available at no additional cost to translators and clients.

ROME, July 12, 2023 – Translated, a leading provider of AI-powered language solutions, today announced a significant expansion to its document-level adaptation, a standout feature of ModernMT, the company’s award-winning adaptive machine translation (MT) system. ModernMT can now leverage up to 100,000 words of document content, 4 times more than the best large language models, to preserve style and terminology preferences, providing unparalleled document-specific accuracy in MT suggestions and significantly reducing the time to edit MT suggestions (TTE).

Expanded document-level adaptation enhances the translation of complex, detailed, or regulatory content, such as annual reports, legal documents, marketing materials, and technical documentation. The upgrade is available immediately at no additional cost to all ModernMT customers, users of Translated’s CAT tool Matecat who have an active ModernMT subscription, and translators working on projects for Translated’s clients.

Document-level adaptation in machine translation, pioneered by ModernMT in 2017, initially involved referencing similar sentences in translation memories and some translated segments. Despite the pioneering nature of this approach, early implementations faced challenges: translators struggled with issues such as gender bias and inconsistent terminology due to the distance between the segment they were working on and its related context. By taking into account all edits in the document, even those in completely different or distant segments, the MT model is now able to provide document-specific translation suggestions. This development significantly reduces the need for repeated corrections of elements such as pronouns.

“Since the very beginning, Translated has been committed to combining the sensitivity of translators with the speed of machines,” said Marco Trombetti, CEO of Translated. “With the introduction of expanded document-level adaptation, translators can now translate more words per hour and deliver higher overall quality, which leads to increased earnings. As a result, Translated’s customers now enjoy more accurate translations delivered in record time.”

Expanded document-level adaptation is now available in ModernMT. It is also integrated into directly supported CAT tools (Matecat, Trados, and memoQ) via ModernMT’s API. Translated’s clients will indirectly benefit from this upgrade, as the company’s translation services are fully based on Matecat.

For all translators, Matecat is available for unlimited use free of charge at, while ModernMT is offering a free one-month trial at

About Translated

Translated is a leading language services provider and pioneer of artificial intelligence to support professional translators. Rated by IDC as the world leader in machine translation (IDC MarketScape 2022), the company was established in 1999 by linguist Isabelle Andrieu and computer scientist Marco Trombetti to allow everyone to understand and be understood in their own language. In pursuit of its mission, Translated relies on a symbiotic combination of human creativity and machine intelligence, using 300,000+ vetted, native-speaking professional translators combined with an award-winning and patented context-adaptive machine translation engine, ModernMT. Translated provides fast, consistent, high-quality translations to 240,000+ clients in 200 languages and over 40 areas of expertise.

MultiLingual Staff
MultiLingual creates go-to news and resources for language industry professionals.


Weekly Digest

Subscribe to stay updated

MultiLingual Media LLC