MET workshops
Delving into GenAI as a resource for language professionals
This workshop explores the role of generative AI in translation and editing, and illustrates ways language professionals can effectively harness its potential. Participants will gain a high-level understanding of GPT technology, including key concepts such as temperature, Top-k and Top-p settings. The workshop also covers prompt engineering techniques, including persona prompting, prompt chaining and prompt reverse engineering. Common issues with AI-generated text, such as repetition, redundancy, hallucinations and lack of complexity, will also be addressed. Finally, participants will analyse synthetic texts and discuss editing techniques for refining AI-generated output to ensure it meets professional standards.
Facilitator: Michael Farrell
Purpose: To illustrate how language professionals can incorporate GenAI into their workflows.
Description: In the first part of the workshop, participants will learn a few basic concepts of GPT models, including how temperature, Top-k and Top-p settings affect AI-generated outputs. They will be introduced to prompt engineering techniques such as persona prompting, few-shot prompting, prompt chaining and prompt reverse engineering, with examples tailored to translation and editing tasks. The session will also include a discussion on the limitations of AI and its potential pitfalls.
In the second part of the workshop, participants will engage in hands-on activities to analyse AI-generated outputs. These exercises will focus on identifying and addressing issues such as redundancy and verbosity, with the goal of refining the text to make it more engaging and give it a natural, human-like quality.
We will conclude by reviewing the results of the activities, reflecting on best practices and exploring how participants can apply these techniques in their own workflows.
Participant profile: Both editors and translators are likely to be interested. A few examples will use languages other than English. Most of these will be in Italian. However, the facilitator will provide either verbal or written English translations.
Outcome: Participants will leave the workshop equipped with implementable strategies for integrating generative AI into their workflow to improve efficiency while maintaining professional quality standards.
Preparation: Participants are encouraged to read:
Farrell, Michael. 2024. Survey on the use of generative artificial intelligence by professional translators. Presented at the 46th Conference Translating and the Computer, Luxembourg, November 18-20, 2024. Luxembourg. Peer-reviewed version to be published shortly in the proceedings.
Farrell, Michael. 2024. Editing synthetic text from generative artificial intelligence: two exploratory case studies. Presented at the 46th Conference Translating and the Computer, Luxembourg, November 18-20, 2024. Luxembourg. Peer-reviewed version to be published shortly in the proceedings.
About the facilitator: Michael Farrell is primarily a freelance translator and transcreator with extensive experience in the cultural tourism field and the transcreation of advertising copy and press releases, chiefly for the promotion of technology products. He also lectures on post-editing, machine translation, artificial intelligence and computer tools for translators at the IULM University in Milan, Italy. He is the developer of the terminology search tool IntelliWebSearch and a qualified member of the Italian Association of Translators and Interpreters (AITI); he is currently serving as MET’s webmaster.
