Enhancing Translation Quality: A Comparative Study of Fine-Tuning and Prompt Engineering in Dialog-Oriented Machine Translation Systems. Insights from the MULTITAN-GML Team
Lichao Zhu, Maria Zimina, Behnoosh Namdarzadeh, Nicolas Ballier, Jean-Baptiste Yunès · 2024
For this shared task, we have used several machine translation engines to produce translations (en ⇔ fr) by fine-tuning a dialog-oriented NMT engine and having NMT baseline translations post-edited through prompt engineering.Our objectives are to assess the effectiveness of a fine-tuning strategy with a robust NMT model, to advance towards a comprehensive pipeline that covers the entire translation process (from fine-tuning and machine translation to automatic post-editing (APE)), and to evaluate the strengths and weaknesses of NMT systems.