Is Preference Alignment Always the Best Option to Enhance LLM-Based Translation? An Empirical Analysis

Is Preference Alignment Always the Best Option to Enhance LLM-Based Translation? An Empirical Analysis Hippolyte Gisserot-Boukhlef author Ricardo Rei author Emmanuel Malherbe author Céline Hudelot author Pierre Colombo author Nuno M Guerreiro author 2024-11 text Proceedings of the Ninth Conference on Machine Translation Barry Haddow editor Tom Kocmi editor Philipp Koehn editor Christof Monz editor Association for Computational Linguistics Miami, Florida, USA conference publication gisserot-boukhlef-etal-2024-preference 10.18653/v1/2024.wmt-1.127 https://aclanthology.org/2024.wmt-1.127/ 2024-11 1373 1392