uu.seUppsala University Publications
Change search
ReferencesLink to record
Permanent link

Direct link
Climbing Mount BLEU: The Strange World of Reachable High-BLEU Translations
Uppsala University, Disciplinary Domain of Humanities and Social Sciences, Faculty of Languages, Department of Linguistics and Philology. Convertus AB, Uppsala, Sweden..
Uppsala University, Disciplinary Domain of Humanities and Social Sciences, Faculty of Languages, Department of Linguistics and Philology.
Univ Helsinki, FIN-00014 Helsinki, Finland..
2016 (English)In: Baltic Journal of Modern Computing, ISSN 2255-8942, E-ISSN 2255-8950, Vol. 4, no 2, 269-281 p.Article in journal (Refereed) PublishedText
Abstract [en]

We present a method for finding oracle BLEU translations in phrase-based statistical machine translation using exact document-level scores. Experiments are presented where the BLEU score of a candidate translation is directly optimised in order to examine the properties of reachable translations with very high BLEU scores. This is achieved by running the document-level decoder Docent in BLEU-decoding mode, where proposed changes to the translation of a document are only accepted if they increase BLEU. The results confirm that the reference translation cannot in most cases be reached by the decoder, which is limited by the set of phrases in the phrase table, and demonstrate that high-BLEU translations are often of poor quality.

Place, publisher, year, edition, pages
2016. Vol. 4, no 2, 269-281 p.
Keyword [en]
Statistical machine translation, oracle decoding, BLEU, Docent
National Category
Language Technology (Computational Linguistics)
Identifiers
URN: urn:nbn:se:uu:diva-299620ISI: 000377706600016OAI: oai:DiVA.org:uu-299620DiVA: diva2:949838
Available from: 2016-07-25 Created: 2016-07-25 Last updated: 2016-07-25Bibliographically approved

Open Access in DiVA

No full text

Search in DiVA

By author/editor
Smith, AaronHardmeier, Christian
By organisation
Department of Linguistics and Philology
In the same journal
Baltic Journal of Modern Computing
Language Technology (Computational Linguistics)

Search outside of DiVA

GoogleGoogle Scholar
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 54 hits
ReferencesLink to record
Permanent link

Direct link