@inproceedings{carpuat-etal-2010-reordering,
    title = "Reordering Matrix Post-verbal Subjects for {A}rabic-to-{E}nglish {SMT}",
    author = "Carpuat, Marine  and
      Marton, Yuval  and
      Habash, Nizar",
    editor = "Langlais, Philippe  and
      Gagnon, Michel",
    booktitle = "Actes de la 17e conf{\'e}rence sur le Traitement Automatique des Langues Naturelles. Articles longs",
    month = jul,
    year = "2010",
    address = "Montr{\'e}al, Canada",
    publisher = "ATALA",
    url = "https://preview.aclanthology.org/fix-sig-urls/2010.jeptalnrecital-long.30/",
    pages = "292--301",
    abstract = "We improve our recently proposed technique for integrating Arabic verb-subject constructions in SMT word alignment (Carpuat et al., 2010) by distinguishing between matrix (or main clause) and non-matrix Arabic verb-subject constructions. In gold translations, most matrix VS (main clause verb-subject) constructions are translated in inverted SV order, while non-matrix (subordinate clause) VS constructions are inverted in only half the cases. In addition, while detecting verbs and their subjects is a hard task, our syntactic parser detects VS constructions better in matrix than in non-matrix clauses. As a result, reordering only matrix VS for word alignment consistently improves translation quality over a phrase-based SMT baseline, and over reordering all VS constructions, in both medium- and large-scale settings. In fact, the improvements obtained by reordering matrix VS on the medium-scale setting remarkably represent 44{\%} of the gain in BLEU and 51{\%} of the gain in TER obtained with a word alignment training bitext that is 5 times larger."
}