Leveraging the advantages of associative alignment methods for PB-SMT systems

Baosong Yang, Yves Lepage*

*この研究の対応する著者

研究成果: Conference contribution

抄録

Training statistical machine translation systems used to require heavy computation times. It has been shown that approximations in the probabilistic approach could lead to impressing improvements (Fast align). We show that, by leveraging the advantages of the associative approach, we achieve similar, even faster, training times, while keeping comparable BLEU scores. Our contributions are of two types: of the engineering type, by introducing multi-processing both in sampling-based alignment and hierarchical sub-sentential alignment; of modeling type, by introducting approximations in hierarchical sub-sentential alignment that lead to important reductions in time without affecting the alignments produced. We test and compare our improvements on six typical language pairs of the Europarl corpus.

本文言語English
ホスト出版物のタイトルHuman Language Technology. Challenges for Computer Science and Linguistics - 7th Language and Technology Conference, LTC 2015, Revised Selected Papers
編集者Zygmunt Vetulani, Marek Kubis, Joseph Mariani
出版社Springer Verlag
ページ214-228
ページ数15
ISBN(印刷版)9783319937816
DOI
出版ステータスPublished - 2018
イベント7th Language and Technology Conference: Human Language Technologies as a Challenge for Computer Science and Linguistics, LTC 2015 - Poznan, Poland
継続期間: 2015 11月 272015 11月 29

出版物シリーズ

名前Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
10930 LNAI
ISSN(印刷版)0302-9743
ISSN(電子版)1611-3349

Other

Other7th Language and Technology Conference: Human Language Technologies as a Challenge for Computer Science and Linguistics, LTC 2015
国/地域Poland
CityPoznan
Period15/11/2715/11/29

ASJC Scopus subject areas

  • 理論的コンピュータサイエンス
  • コンピュータ サイエンス(全般)

フィンガープリント

「Leveraging the advantages of associative alignment methods for PB-SMT systems」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル