Terminology-Aware Machine Translation for Accessible Science (TamTAS)

Start date

February 2026

End date

January 2029

About the project

Summary

This project addresses the urgent need for accurate and accessible scientific communication by enabling multilingual access to scientific knowledge. It challenges the dominance of English in scientific dissemination and aims to empower both researchers and the general public to engage with science in their native languages. To achieve this, we propose a terminology-aware machine translation (MT) framework tailored to scientific texts. The core technology is based on Large Reasoning Models (LRMs)—an advanced class of Large Language Models (LLMs) that treat translation as a reasoning task. LRMs incorporate chain-of-thought prompting, self-correction, and document-level understanding, ensuring terminological consistency and coherence across longer texts.

To further improve translation robustness, we will develop a tightly integrated pipeline combining Quality Estimation (QE) and Automatic Post-Editing (APE). QE models will detect and assess terminology-related errors, guiding APE modules—using reinforcement learning techniques like direct preference optimization—to refine the translations. Feedback from QE will also be used to improve the LRM itself. The training of these components will be supported by specialized corpora annotated with terminology errors. The project also explores post-translation text augmentation to adapt outputs for different audiences. This includes simplification and explanation strategies to make scientific content more accessible and reusable in education and public engagement.

We focus on five languages—English, Spanish, Catalan, Estonian, and Irish—covering both well-resourced and under-resourced contexts, and apply our approach to the Life Sciences domain. Pilot collaborations with the Centre de Recerca Genòmica (CRG) in Barcelona, the Institute of Family Medicine and Public Health in Tartu, and Conradh na Gaeilge in Ireland will ensure real-world validation and user-driven refinement (both from the scientific and the general public perspective). Bridging MT, NLP, and scientific expertise, the project will be evaluated using both adapted automatic metrics and human assessments by domain experts and general users. The final system, validated at TRL 5–6, will significantly enhance the accuracy, inclusivity, and trustworthiness of scientific communication across languages—supporting a more equitable and globally connected research ecosystem.

People

Principal investigator

Professor Constantin Orasan

Professor of Language and Translation Technologies

Biography

I am Professor of Language and Translation Technologies at the Centre of Translation Studies, University of Surrey, UK, and a Fellow of the Surrey Institute for People-Centred Artificial Intelligence. Before starting this role, I was Reader (Associate Professor) in Computational Linguistics at the University of Wolverhampton, UK, and the deputy head of the Research Group in Computational Linguistics at the same university. I hold a PhD in computational linguistics and a BSc in computer science.

With over 25 years of experience in the fields of Natural Language Processing, Artificial Intelligence, and Linguistics, I have established myself as a leading researcher in the development of technologies that facilitate access to information. My PhD was in automatic summarisation, and I have led projects on question answering, text simplification, and translation technologies. Notable projects that I have led are EmpASR, an AHRC-funded project focused on training interpreters on how to benefit from the latest developments in artificial intelligence; HarnessingNLP4Court, a UKRI-funded project focused on facilitating access to legal information; the EXPERT project, an Initial Training Network (ITN) funded under the EU’s FP7 to train the next generation of world-class researchers in the field of data-driven translation technology; and the FIRST project, which developed language technologies for making texts more accessible to people with autism.

My current research is interdisciplinary, focusing on the intersection of AI, NLP, and translation studies. In recent years, I have increasingly focused on the practical application of NLP to support translators and interpreters. My recent publications explore reference-less translation evaluation, the processing of multilingual content in low-resource settings, the use of automatic speech recognition to support interpreters, and the use of large language models in text accessibility. My research is well known as a result of over 150 peer-reviewed articles in journals, books, and international conferences.

I am currently leading an EPSRC-funded project focused on making science accessible, and I am Co-Director of the ADA Leverhulme Doctoral Scholarships Network. More information about my work can be found at https://dinel.org.uk/.

See profile