Báo cáo khoa học: "Adapting Translation Models to Translationese Improves SMT" pot

... (FO) to refer to the subsets of the corpus that are translated from En- glish to French and from French to English, re- spectively. The translation tasks are English -to- French (E2F) and French -to- English ... given phrase to original language or to translationese. Motivated by this observation, we design two cross-entropy-based measures to assess how well each phrase tabl...

Ngày tải lên: 31/03/2014, 20:20

... as learning technique, it is possible to generate models that, when compared to models produced by machine learning algorithms conventionally applied to 41 WSD, are both more accurate for ... performance of a machine translation system (Carpuat et al., 2006). This paper focuses on the application of our ap- proach to the translation of verbs in English to Por- tuguese...

Ngày tải lên: 08/03/2014, 02:21

... match at all. But full NLP is too slow for this application, and partial matches do seem to be useful to its users, i.e. do seem to lead to licensing of photos. Using the output of the part-of-speech ... abbrevia- tions and other variant forms, e.g. Washington D.C.; Washington, DC; Washington, District of Columbia; Washington DC; Washington, D.C.; DC; and D.C. Some superunits...

Ngày tải lên: 08/03/2014, 05:21

... set. The simplest way to match query tokens to snip- pet tokens is to allow a query token to match any snippet token. This can be problematic when we have queries that have a token repeated with ... surprisingly pow- erful one – is to POS tag some relevant snippets for 238 a given query, and then to transfer the tags from the snippet tokens to matching query tokens. This “di- re...

Ngày tải lên: 16/03/2014, 20:20

... translations from the same MT system to those from different MT systems. Collaborative decoding (Li et al., 2009) scores the translation of a source span by its n-gram similarity to the translations ... produced the correct translation for the second sentence, but it failed to do so for the first one. If the translation of the first sentence could take into consideration th...

Ngày tải lên: 23/03/2014, 14:20

Ngày tải lên: 23/03/2014, 16:20

... can come to the old opera? to y: What? X: the old opera y: to the old opera; straight ahead, yes. Come on, I show X: yes, yes (I0 sec. pause) Y: it to you. ahead to the Kaufhof. To the X: ... components, in our case 5. The processing components need not be ATNs; it is up to the user of the formalism to choose the tool for the specific task that suits her/hlm best....

Ngày tải lên: 24/03/2014, 05:21

... for tree -to- string translation. The de- coder operates top-down and maintains a derivation history of translation rules encountered. The history is exactly the vertical chain of ancestors correspond- ing ... threshold. 3 Tree -to- string decoding with rule Markov models In this paper, we use our rule Markov model frame- work in the context of tree -to- string translation. Tree -to- s...

Ngày tải lên: 17/03/2014, 00:20

... transla- tion rules to large-scale Japanese-English transla- tion tasks. Finally, we conclude in Section 5. 2 Related Work 2.1 Tree -to- string and string -to- tree translations Tree -to- string translation ... koroshita parsing Bottom-up decoding tree -to- string string -to- tree Figure 1: Illustration of the training and decod- ing processes for tree -to- string and string -t...

Ngày tải lên: 20/02/2014, 04:20

... USA me@hal3.name Abstract We present a method to transliterate names in the framework of end -to- end statistical machine translation. The system is trained to learn when to transliterate. For Arabic to English MT, we developed ... apply it to any base SMT system, and to human translationsas well. Our goal in augment- ing abaseSMT systemis toincreasethis percentage. A secondary g...

Ngày tải lên: 20/02/2014, 09:20