TECHNOLOGIES FOR TRANSLATION AND POST-EDITING MOD.2

Translation Technologies and Post-editing (Module 2) is a core subject within the MA in Translation and Interpreting and is offered to students in both the “English/Spanish/Italian Sign Language” and “English/Chinese” tracks.
Technologies are not presented simply as a set of software applications to learn, but as tools that support translators in linguistic analysis, research and decision-making. Through practical activities, students will develop a critical approach to the use of language technologies in professional translation. Module 2 will focus on the current regulatory framework governing the use of AI, the post-editing of machine translation, and the informed use of AI in translation and language services.

Knowledge and understanding
- knowledge of the AI Act (EU Regulation 2024/1689)
- knowledge of ISO 18587 on post-editing
- knowledge of the skills and competences required of post-editors under ISO 18587
- knowledge of prompting techniques

Applying knowledge and understanding
- be able to perform post-editing with different levels of intervention on raw output
- be able to use AI in an informed, critical and ethically responsible manner
- be able to use prompting techniques

Making judgements
- be able to assess the extent to which machine translation is appropriate and/or requires human translator intervention
- be able to assess the advantages and disadvantages of the technological tools used during the course
- be able to verify and assess the quality, reliability and appropriateness of AI-generated output

Communication skills
- be able to use appropriate metalanguage relating to translation technologies
- be able to work effectively in a group
- be able to communicate difficulties, doubts and critical issues

Learning skills
- be able to independently expand their knowledge and skills in the use of translation technologies

- Excellent command of basic computer skills (file and folder management, software installation, file-format management and basic system settings)
- C1-level proficiency in English
- Excellent command of Italian (at least B2 level for international students)
- Willingness to use technological tools and engage in collaborative work
- Willingness to work in groups
- Aptitude for independent problem solving
- Attention to detail

The course will cover the following topics:
1. Machine translation: development, key features, types, tools and critical analysis
2. Post-editing (PE vs translation vs revision, ISO 18587, competences, types of PE, editing actions, etc.)
3. Artificial intelligence: AI Act (EU Regulation 2024/1689), prompting, and ethical and informed use in translation and language services
4. Practical activities (in class and at home, individually and in groups)

All the needed content will be in the slides provided by the lecturer.
Some readings and other useful resources will be made available on the course Moodle space.

The final assessment must be completed in groups and will consist of:

1. Project work on post-editing (submission via Moodle)
This project work is designed to assess students’ command of machine translation and post-editing in both theory and practice.

2. Project work on AI and prompting applied to language services (submission via Moodle)

3. Oral discussion of the work completed
The oral examination, based on both projects, is designed to assess students’ command of subject-specific metalanguage and their decision-making and judgement in relation to the assigned tasks.

NB: Compulsory requirement for completion of the examination
As an integral part of the examination, all students must complete the “RWS Linguistic AI Certification Training” (https://www.trados.com/training/certification-rws-linguistic-ai-certification-training/ ), provided free of charge by RWS, following the Course Instructor’s instructions. The training is entirely online, free of charge, consists of a single module and leads to certification. Full instructions will be provided on Moodle during the course. Link to the course:
https://www.trados.com/training/certification-rws-linguistic-ai-certification-training/

written and oral

The instructor is responsible for ensuring the authenticity and originality of all examinations and coursework. In cases of suspected academic misconduct, an additional on-site assessment may be required during the exams, which may differ from the standard format.

1. Project work 1: Post-editing (15 points)
a) Critical analysis of raw output (0–5)
0 = none; output accepted without comparison with the source text; 1–2 = partial, with relevant problems missed and/or inappropriate interventions; 3–4 = adequate, identifying the main problems but with some inaccuracies in assessing the intervention required; 5 = systematic and accurate, with informed assessment of the need for and extent of intervention.

b) Linguistic evidence supporting PE decisions (0–5)
0 = none or based on irrelevant/unreliable/inadequately documented sources; 1–2 = limited or not always relevant, with several inaccuracies; 3–4 = relevant and adequately documented, with some inaccuracies or room for improvement in source selection; 5 = accurate, relevant, documented and critically used to support PE decisions.

c) Compliance with the PE brief (0–5)
0 = failure to meet mandatory requirements, making the work non-compliant with the assignment; 1–2 = partial compliance, with numerous unnecessary/insufficiently justified interventions; 3–4 = generally consistent, with some unnecessary/insufficiently justified interventions; 5 = fully consistent, with no unnecessary or insufficiently justified interventions.

Score: ___/15

2. Project work 2: AI and prompting applied to translation (15 points)
a) Prompting techniques (0–5)
0 = generic, ambiguous, inaccurate or contradictory prompts; 1–2 = minimal command, limited to one prompt type with no improvement in subsequent prompts; 3–4 = good command, with improvements to subsequent prompts; 5 = full command, using different prompt types and/or improving prompts based on the output obtained.

b) Critical analysis of output in relation to prompting (0–5)
0 = repeated discrepancies/contradictions between prompt objectives and analysis; 1–2 = confused or ambiguous analysis and inaccurate metalanguage; 3–4 = generally consistent with the prompt, with some inaccuracies/ambiguities; 5 = fully consistent with the prompt.

c) Switch between AI interaction and human intervention (0–5)
0 = uninformed, with random/unjustified manual interventions; 1–2 = limited awareness, with numerous unnecessary/inconsistent interventions; 3–4 = generally informed, with some unnecessary/not fully consistent interventions; 5 = fully informed, with relevant interventions consistent with the intended objective.

Score: ___/15

3. Oral examination
Discussion of the project work assessing the ability to justify choices, relate them to the theoretical and methodological principles covered, critically assess output and strategies, and use subject-specific metalanguage. The individual score may be confirmed, increased or reduced in relation to the group score.

The final grade will be the average of the grades obtained in Modules 1 and 2 (including any Trados certification bonus awarded in Module 1).

The Italian lode (with distinction) may be awarded with a final grade of 30/30 for full command of the required knowledge and skills, high operational autonomy, strong critical and decision-making skills, and particularly high awareness in the use of translation technologies.

The course will be taught in Italian and will include both lectures and interactive sessions. Active participation in practical activities is an integral part of the learning process.
Students must enrol in the Moodle course and refer to the materials available there when preparing for the examination. Materials will be made available on a weekly basis.

The course content may be adjusted according to the needs of the class and the availability of tools provided by the University.

Important:
Given the complexity of the topics covered, students who are unable to attend for various reasons (Erasmus mobility, prolonged illness, work commitments, etc.) are encouraged to contact the Course Instructor (mariaelisa.fina@unive.it) in good time to receive guidance aimed at minimising any disadvantages resulting from non-attendance.

This subject deals with topics related to the macro-area "Cities, infrastructure and social capital" and contributes to the achievement of one or more goals of U. N. Agenda for Sustainable Development