Document-Level Machine Translation as a Re-translation Process
Eva Martínez García, Cristina España Bonet, Lluís Màrquez Villodre · Procesamiento del lenguaje natural · 2014
Most of the current Machine Translation systems are designed to trans- late a document sentence by sentence ignoring discourse information and producing incoherencies in the nal translations. In this paper we present some document- level-oriented post-processes to improve translations' coherence and consistency. In- coherences are detected and new partial translations are proposed. The work focuses on studying two phenomena: words with inconsistent translations throughout a text and also, gender and number agreement among words. Since we deal with specic phenomena, an automatic evaluation does not reect signicant variations in the