2019/06/04 by Samuel Läubli, Läubli, Samuel, Chantal Amrhein +10
Computer Science · #Computation and Language (cs.CL) #FOS: Computer and information sciences #Natural Language Processing Techniques #Software Engineering Research #Topic Modeling
paper · pdf · doi:10.48550/arxiv.1906.01685
openalex publication_date 2019/06/04 · openalex created_date 2022/07/29 · openalex updated_date 2026/07/28
Neural machine translation (NMT) has set new quality standards in automatic\ntranslation, yet its effect on post-editing productivity is still pending\nthorough investigation. We empirically test how the inclusion of NMT, in\naddition to domain-specific translation memories and termbases, impacts speed\nand quality in professional translation of financial texts. We find that even\nwith language pairs that have received little attention in research settings\nand small amounts of in-domain data for system adaptation, NMT post-editing\nallows for substantial time savings and leads to equal or slightly better\nquality.\n