Elamur Research⌕
☰
home / papers / 2026 / Teaching Machine Translation Technologies with MTUUOC

Teaching Machine Translation Technologies with MTUUOC

Summary

Translators now need practical skills with neural machine translation and LLMs. However, the open-source software for building such systems is often poorly documented and hard for non-programmers to use. The authors, from UOC (a fully online university) and UAB, describe how they teach these technologies with their open-source MTUOC toolkit. MTUOC is a modular, mostly Python-based, cross-platform collection of tools that covers the whole MT lifecycle:

  • gathering, aligning and cleaning corpora;
  • wrappers and configuration templates for training and fine-tuning NMT models and LLMs;
  • a server that hosts multiple engines and connects to CAT tools;
  • evaluation through automatic metrics and a post-editing environment that records effort.

The paper reports on its use in a Bachelor's course, a Master's course and a pilot MOOC. Key obstacles were students' unfamiliarity with the terminal and the dominance of Windows. The authors addressed these with scaffolded, cross-platform activities. They then present a redesigned four-module course, starting October 2026, shared by all three tracks. The findings are qualitative teaching observations, not controlled measurements. The value lies in an openly licensed, reusable curriculum that builds technical autonomy in future translators.

Key points

  • Toolkit components:
    • MTUOC-server, which can run several NMT or LLM engines at once.
    • Desktop and web translators, plus a post-editing interface with quality control and quality estimation.
    • Add-ons that link the toolkit with Trados Studio, Okapi Tools and OmegaT.
    • Training interfaces for Marian, OpenNMT, Eole and Axolotl.
    • Evaluation with sacreBLEU and COMET.
  • Formal teaching:
    • In the Bachelor's Machine Translation and Post-editing course, the basics are mandatory and advanced tasks are optional.
    • In the Master's Translation and Technologies course, fine-tuning and technical evaluation are required for all students.
  • Pilot MOOC: it attracted more than 100 participants.
  • Planned 2026 design:
    • Four modules, run twice a year starting mid-October and mid-March, on a single calendar shared by all tracks.
    • Module 3 covers Marian, Eole and fairseq2.
    • Module 4 covers in-context learning with translation memories and termbases, plus full fine-tuning and PEFT (LoRA).
    • Course communication moves from Google Groups to Discord.
  • Materials: all are open source, maintained in a GitHub Wiki, and available at mtuoc.github.io.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

Fields: Linguistics & languages · Machine translation

Topics: Machine translation

Cite

The paper itself, not this page. BibTeX from the library's record of the work.

@inproceedings{oliver2026teaching,
  title = {{Teaching Machine Translation Technologies with MTUUOC}},
  author = {Oliver, Antoni and Alvarez-Vidal, Sergi},
  booktitle = {Proceedings of the 1st International Workshop on Teaching AI-Based Translation and Technologies (TAITT 2026)},
  year = {2026},
  url = {https://aclanthology.org/2026.taitt-1.2/}
}

References

10 references, 2 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Cid, Clara Ginovart, Carme Colominas, and Antoni Oliver. 2020. Language industry views on the profile of the post-editor. Translation Spaces, 9(2):283– 313. find
  2. EMT Board. 2022. European graduate competences framework in translation. Technical report, European Commission, Directorate-General for Translation. find
  3. Kenny, Dorothy and Stephen Doherty. 2014. Statistical machine translation in the translation curriculum: Overcoming obstacles and empowering translators. The Interpreter and Translator Trainer, 8(2):276– 294. find
  4. Kenny, Dorothy. 2022. Machine Translation for Everyone: Empowering Users in the Age of Artificial Intelligence. Language Science Press, Berlin. find
  5. Mellinger, Christopher D. 2018. Problem-based learning in computer-assisted translation pedagogy. HERMES-Journal of Language and Communication in Business, (57):195–208. find
  6. Oliver, Antoni and Sergi Álvarez. 2023. Training and integration of neural machine translation with MTUOC. In Proceedings of the 1st Workshop on Open Community-Driven Machine Translation, pages 5–13. find
  7. Oliver, Antoni, Sergi Alvarez, and Toni Badia. 2020. PosEdiOn: Post-editing assessment in Python. In Proceedings of the 22nd Annual Conference of the European Association for Machine Translation, pages 403–410. find
  8. Olohan, Maeve. 2007. Economic trends and developments in the translation industry: What relevance for translator training? The Interpreter and Translator Trainer, 1(1):37–63. find
  9. Post, Matt. 2018. A call for clarity in reporting BLEU scores. In Proceedings of the Third Conference on Machine Translation: Research Papers, pages 186– 191. find our card
  10. Rei, Ricardo, Craig Stewart, Ana C Farinha, and Alon Lavie. 2020. COMET: A neural framework for MT evaluation. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 2685–2702. find our card

This page in Markdown → https://md.research.elamur.ai/free/en/papers/oliver-2026-teaching-machine-translation-technologies-mtuuoc/