Prompt Engineering for Multilingual Communication

Language professionals such as translators, post-editors, reviewers, project managers, marketers and professionals from other industries who need to generate and manage multilingual content in a company or organisation.

• Learn several types of neural machine translation (NMT) and artificial intelligence (AI) tools with the aim of producing multilingual texts.
• Understand how NMT/AI tools are programmed and how this impacts their appropriate use.
• Be aware of the unpredictability of NMT/AI output.
• Frame what strategies and workflows are needed to produce multilingual texts with NMT/AI tools.
• Develop self-tailored evaluation criteria and be able to apply them.
• Have a space to experiment with different tools and writing tasks, merge the results and be critical with them.

Theoretical framework of neural machine translation (NMT) and large language models (LLMs) as a type of artificial intelligence (AI), critical and ethical use of language technology, legal and data privacy questions.

NMT and AI prompting for multilingual communication. Theoretical introduction to the similarities and differences between NMT and LLMs in multilingual text production, with a focus on various workflow examples for different settings and text types.

Short exercises with:

• NMT tools
• AI tools and prompting
• Experimenting with various kinds of texts and tools
• Analysis of the text production task taking into account text type, genre, audience and purpose
• Prompting, output evaluation and fine-tuning as an iterative process
• Evaluation criteria and quality assessment

Presentation of a multilingual production task, the process and the criteria created for assessing the results.