B8536 - Computer-Aided Human Translation: CAT, MT and AI (CL1)

Learning outcomes

The student is familiar with key digital tools for file management, computer-assisted translation (CAT) and AI-enhanced translation, including Machine Translation (MT); they can use CAT tools effectively, apply basic prompting techniques for AI-driven translation tools, and work with MT engines; they can vouchsafe translation quality and post-edit texts efficiently and accurately across various text types and domains, adapting them to different audiences, as needed; they can set up, manage, and evaluate projects involving other professionals that combine CAT, MT, post-editing, and AI tools; they can expand and refine their knowledge of CAT, AI and post-editing autonomously, applying it to new AI-driven practices, in line with professional ethics and industry standards.

Course contents

The "Computer-aided human translation: CAT, MT and AI " (CatMAI) module is delivered in the second semester and is one of the two modules that make up the "Technologies for Translation" course, together with "Information Mining and Terminology", which is delivered in the first semester and which is held by Prof. Adriano Ferraresi.

The first part offers a theoretical introduction to the notion of computer-assisted translation. Subsequently some of the most common CAT tools are presented (Trados Studio, MemoQ, MateCat), proprietary and free. In particular, the module focuses on the translation of the main file formats, the creation and managing of translation memories and the quality assurance of target texts (QA check). The basis for managing complex translation projects involving different professionals (Project Management) through CAT are also provided. Finally, the latest CAT tool functionalities that integrate generative artificial intelligence (AI) to optimise workflows in both translation and revision will be presented.

The second part, which is linked to the first one, starts with a presentation of the obvious advantages of using AI-enhanced translation, including machine translation (MT) accompanied by post-editing. It will also focus on the possibilities of integration between CAT and MT systems and AI-based generative translation functionalities included in CAT software.

With regard to PE, different ways of intervention (light, full, etc.) are discussed in relation to variables such as the specific conditions of the revision task, the post-editor profile (bilingual, monolingual of the target language, expert in the field, etc.), the type of translation, the publication venue and the circulation methods planned for the revised target text, its potential readers and users, etc. Various PE strategies allowing for the improvement of the raw output provided by MT and generative AI systems are also presented. The aim of these strategies is to obtain a target text which meets the specific requirements of the translation context.

Basic prompting techniques that can be used within AI-based translation tools, both to optimise the raw output of a machine-translated text and to post-edit it, will also be introduced.

Readings/Bibliography

During lessons based on theoretical aspects, the teacher will be using the following bibliographical references:

Bersani Berselli, G. (a cura di) (2011) "Usare la Traduzione Automatica". Bologna: CLUEB.

Bowker, L., (2002). “Computer-Aided Translation Technology. A Practical Introduction”. University of Ottawa Press.

Bowker, Lynne, (2005). “Productivity vs. Quality? A pilot study on the impact of translation memory systems”. Localisation Focus 4:1. 13-20.

Cevoli, M. & S. Alasia, (2012). “Guida completa a OmegaT”. Badalona: Qabiria.

Declercq, C. (2023). Editing in translation technology. In Routledge encyclopedia of translation technology (pp. 551-564). Routledge.

Fagbohun, O., Harrison, R. M., & Dereventsov, A. (2024). An empirical categorization of prompting techniques for large language models: A practitioner's guide. arXiv preprint arXiv:2402.14837.

He, S. (2024). Prompting ChatGPT for translation: A comparative analysis of translation brief and persona prompts. arXiv preprint arXiv:2403.00127.

Koponen, M. (2016) "Is machine translation post-editing worth the effort? A survey of research into post-editing and effort". The Journal of Specialised Translation. Disponibile online: https://www.jostrans.org/issue25/art_koponen.pdf

Lecci, C. & E. Di Bello, (2012). “Usare la traduzione assistita”. Bologna: CLUEB.

Poulis, Alexandros and David Kolovratnik (2012) "To Post-edit or not to Post-edit? Estimating the Benefits of MT Post-editing for a European Organization". Proceedings of the AMTA 2012 Workshop on Post-editing Technology and Practice (WPTP 2012). The Tenth Biennial Conference of the Association for Machine Translation in the Americas, October 28-November 1 2012, San Diego, CA, USA. Disponibile online: http://amta2012.amtaweb.org/AMTA2012Files/html/9/9_paper.pdf

Qian, M., & Kong, C. (2024, May). Enabling Human-Centered Machine Translation Using Concept-Based Large Language Model Prompting and Translation Memory. In International Conference on Human-Computer Interaction (pp. 118-134). Cham: Springer Nature Switzerland.

Toto P. (2021). “Flipped classrooms and translation technology teaching: a case study”. In Wang C., Zheng B., Empirical studies of translation and interpreting. London: Routledge.

Vela M., Pal S., Zampieri M., Kumar Naskar S. (2019). “Improving CAT Tools in the Translation Workflow: New Approaches and Evaluation”. Saarland University, Germany, University of Wolverhampton, UK, Jadavpur University, India. Online: https://arxiv.org/pdf/1908.06140.pdf

Vieira, L. N. (2019). Post-editing of machine translation. "The Routledge handbook of translation and technology". 319-336. London: Routledge.

Zhang, B., Haddow, B., & Birch, A. (2023). Prompting large language model for machine translation: A case study. In International Conference on Machine Learning (pp. 41092-41110). PMLR.

Teaching methods

Lessons take the form of workshops covering theoretical aspects as well as devoting substantial space to practical exercises.

Theoretical contents are acquired through presentations by the lecturer and, when relevant, readings assigned to the students during the course.

The applied part consists of hands-on practice in the lab and homework exercises. These are discussed during troubleshooting sessions in the following class, so as to constantly monitor progress in the development of the technological skills that make the object of the course.

As concerns the teaching methods of this course unit, all students must attend Module 1, 2 on Health and Safety online.

Attendance is compulsory (at least 70% of lessons must be attended).

Assessment methods

The final exam lasts overall two hours and consists of a practical test on computer-assisted translation - e.g. the creation of translation memories, terminology databases and translation projects/packages using one of the CAT tools presented in class - lasting one hour and thirty minutes, and a theoretical test on MT, post-editing and AI - an open-ended question - lasting approximately thirty minutes and covering the principles and methodologies covered in class during the module and their applications in professional translation, with a critical analysis of their potential and applications.

ESAME PER STUDENTESSE E STUDENTI InTeCo/InConf

Only for students of the InTeCo and InConf curricula (Master's degree in Interpretation), for whom the course is an elective, the final exam consists of two open-ended questions - lasting approximately sixty minutes - concerning the principles and methodologies covered in class during the module and their applications in professional translation, with a critical analysis of their potential and applications.

Evaluation

30 – 30L excellent results, demonstrating an excellent understanding of the course content, as well as a good awareness and ability to evaluate different CAT systems to be adopted for different needs and workflows.
27 – 29 above average results, with minor errors or balanced by a good knowledge of fundamental concepts and applications.
24 – 26 good results, with some errors or knowledge gaps that show a partial understanding of contents and required skills.
21 – 23 sufficient results, but with notable gaps in knowledge or skills acquired in the course contents.
18 – 20 results that only prove minimal knowledge of the course contents.
< 18 insufficient, basic concepts have not been understood or demonstrated, the students has to take again the test.

 

Students with specific learning difficulties (SpLD) or with disabilities that can affect their ability to attend courses are invited to contact the University service for students with disabilities and SLD at the earliest opportunity -- ideally before the start of the course: https://site.unibo.it/studenti-con-disabilita-e-dsa/en/for-students. The University service will suggest possible adjustments to the course work and/or exam, which must then be submitted to the course leader so they can assess their feasibility, in line with the learning objectives of the course. Please note that adjustments to the exam must be requested at least two weeks in advance.

Teaching tools

Lessons are held in a computer lab with internet connection and beamer.

Since lessons take the form of workshops, with substantial time devoted to pratical hands-on exercises, students have the possibility to become acquainted with the main software programs used in the fields of CAT, both proprietary and open-source/free.

Support materials (sample texts, slides, project files, instructions etc.) are made available through the Moodle e-learning platform.

Links to further information

https://moodle.dipintra.it/course/view.php?id=67

Office hours

See the website of Claudia Lecci