Translation Model Interpolation for Domain Adaptation in TectoMT
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F15%3A10318159" target="_blank" >RIV/00216208:11320/15:10318159 - isvavai.cz</a>
Result on the web
<a href="https://ufal.mff.cuni.cz/events/deep-machine-translation-workshop#proceedings" target="_blank" >https://ufal.mff.cuni.cz/events/deep-machine-translation-workshop#proceedings</a>
DOI - Digital Object Identifier
—
Alternative languages
Result language
angličtina
Original language name
Translation Model Interpolation for Domain Adaptation in TectoMT
Original language description
We present an implementation of domain adaptation by translation model interpolation in the TectoMT translation system with deep transfer. We evaluate the method on six language pairs with a 1000-sentence in-domain parallel corpus, and obtain improvements of up to 3 BLEU points. The interpolation weights are set uniformly, without employing any tuning.
Czech name
—
Czech description
—
Classification
Type
D - Article in proceedings
CEP classification
IN - Informatics
OECD FORD branch
—
Result continuities
Project
<a href="/en/project/LM2010013" target="_blank" >LM2010013: LINDAT-CLARIN: Institute for analysis, processing and distribution of linguistic data</a><br>
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)<br>S - Specificky vyzkum na vysokych skolach
Others
Publication year
2015
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Article name in the collection
Proceedings of the 1st Deep Machine Translation Workshop
ISBN
978-80-904571-7-1
ISSN
—
e-ISSN
—
Number of pages
8
Pages from-to
89-96
Publisher name
ÚFAL MFF UK
Place of publication
Praha, Czechia
Event location
Praha, Czechia
Event date
Sep 3, 2015
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—