Edinburgh’s Syntax-Based Systems at WMT 2014
By Philip Williams, Rico Sennrich, Maria Nadejde, Matthias Huck, Philipp Koehn. 2014
26 citations recorded by the source corpus.
This paper describes the string-to-tree systems built at the University of Edin-burgh for the WMT 2014 shared translation task. We developed systems for English-German, Czech-English, French-English, German-English, Hindi-English, and Russian-English. This year we improved our English-German system through target-side compound splitting, morphosyntactic constraints, and refine-ments to parse tree annotation; we addressed the out-of-vocabulary problem us-ing transliteration for Hindi and Russian and using morphological reduction for Russian; we improved our German-English system through tree binarization; and we reduced system development time by filtering the tuning sets.
Subfields: Artificial Intelligence, Machine Translation
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Philipp Koehn
- Rico Sennrich
- Felix Stahlberg
- Graham Neubig
- Yonatan Belinkov
- Ondrej Bojar
- Nadir Durrani
- Barry Haddow
In Machine Translation
- Philipp Koehn
- Rico Sennrich
- Barry Haddow
- Nadir Durrani
- Matthias Huck
- Yonatan Belinkov
- Philip Williams
- Maria Nadejde
Citations in the corpus
- Findings of the 2014 Workshop on Statistical Machine Translation (2014, neutral): “UEDIN-SYNTAX University of Edinburgh (Williams et al., 2014)”
- Edinburgh’s Syntax-Based Systems at WMT 2015 (2014, methodology): “For German → English we built a tree-to-string system with similar setup as last year’s (Williams et al., 2014).”
- Edinburgh’s Statistical Machine Translation Systems for WMT16 (2016, methodology): “We used hard constraints and re-used the baseline weights (re-tuning did not appear to give additional gains Although the gains in B LEU were small, previous analysis for German showed that B LEU lacks sensitivity to grammatical improvements when compared to human evaluators (Williams, 2014).”
- A Joint Dependency Model of Morphological and Syntactic Structure for Statistical Machine Translation (2015, methodology): “For comparison with the state-of-the-art, we train a full system on our restructured representation, which incorporates all models and settings of our WMT 2015 submission system (Williams et al., 2015).”
- Combined spoken language translation (2014, methodology): “UEDIN-C is a string-to-tree translation system with similar features as the ones described in [40].”
- Edinburgh SLT and MT system description for the IWSLT 2014 evaluation (2014, methodology): “The contrastive 1 system is a string-to-tree translation system with similar features as the ones described in [15].”