n-Best Reranking for the Efficient Integration of Word Sense Disambiguation and Statistical Machine Translation

Specia, Lucia; Sankaran, Baskaran; Nunes, Maria das Graças Volpe

doi:10.1007/978-3-540-78135-6_34

Cited by 12 publications

(15 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While phrase-based SMT models incorporate the one sense per collocation hypothesis by attempting to translate phrases rather than single words , the one sense per discourse hypothesis has not been explicitly used in SMT modeling. Even the recent generation of SMT models that explicitly use WSD modeling to perform lexical choice rely on sentence context rather than wider document context and translate sentences in isolation (Carpuat and Wu, 2007;Chan et al, 2007;Giménez and Màrquez, 2007;Stroppa et al, 2007;Specia et al, 2008). Other context-sensitive SMT approaches (Gimpel and Smith, 2008) and global lexical choice models (Bangalore et al, 2007) also translate sentences independently.…”

Section: Related Workmentioning

confidence: 99%

One translation per discourse

Carpuat

2009

Proceedings of the Workshop on Semantic Evaluations: Recent Achievements and Future Directions - DEW '09

View full text Add to dashboard Cite

We revisit the one sense per discourse hypothesis of Gale et al. in the context of machine translation. Since a given sense can be lexicalized differently in translation, do we observe one translation per discourse? Analysis of manual translations reveals that the hypothesis still holds when using translations in parallel text as sense annotation, thus confirming that translational differences represent useful sense distinctions. Analysis of Statistical Machine Translation (SMT) output showed that despite ignoring document structure, the one translation per discourse hypothesis is strongly supported in part because of the low variability in SMT lexical choice. More interestingly, cases where the hypothesis does not hold can reveal lexical choice errors. A preliminary study showed that enforcing the one translation per discourse constraint in SMT can potentially improve translation quality, and that SMT systems might benefit from translating sentences within their entire document context. * The author was partially funded by GALE DARPA Contract No. HR0011-06-C-0023. Any opinions, findings and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the Defense Advanced Research Projects Agency.

show abstract

Section: Related Workmentioning

confidence: 99%

One translation per discourse

Carpuat

2009

Proceedings of the Workshop on Semantic Evaluations: Recent Achievements and Future Directions - DEW '09

View full text Add to dashboard Cite

show abstract

“…2 Other WSD for SMT approaches require at least one of the PSD generalizations Several independent studies in WSD modeling for SMT have been proposed after our initial disappointing results (in particular Cabezas and Resnik (2005); Chan et al (2007); Giménez and Màrquez (2007); Stroppa et al (2007); Gimpel and Smith (2008); Specia et al (2008)). While those evaluations were conducted on different data sets, different language pairs, different baseline SMT systems and except in Giménez and Màrquez (2008), only evaluated translation quality in terms of BLEU or NIST scores, it is interesting to note that they are all based on at least one of the three PSD generations and that consistent gains across a wide variety of metrics are only obtained when all three generalizations are applied.…”

Section: From Conventional Wsd To Phrasementioning

confidence: 99%

Word sense disambiguation for statistical machine translation

Carpuat

View full text Add to dashboard Cite

We present the first known empirical test of an increasingly common speculative claim, by evaluating a representative Chinese-to-English SMT model directly on word sense disambiguation performance, using standard WSD evaluation methodology and datasets from the Senseval-3 Chinese lexical sample task. Much effort has been put in designing and evaluating dedicated word sense disambiguation (WSD) models, in particular with the Senseval series of workshops. At the same time, the recent improvements in the BLEU scores of statistical machine translation (SMT) suggests that SMT models are good at predicting the right translation of the words in source language sentences. Surprisingly however, the WSD accuracy of SMT models has never been evaluated and compared with that of the dedicated WSD models. We present controlled experiments showing the WSD accuracy of current typical SMT models to be significantly lower than that of all the dedicated WSD models considered. This tends to support the view that despite recent speculative claims to the contrary, current SMT models do have limitations in comparison with dedicated WSD models, and that SMT should benefit from the better predictions made by the WSD models.

show abstract

“…In an English-to-Portuguese translation task, Specia et al (2008) work with a syntactically motivated PB-SMT system (Quirk et al 2005), which they enrich by a WSD model limited to disambiguating ten highly frequent and ambiguous verbs.…”

Section: English As Source Languagementioning

confidence: 99%

Integrating source-language context into phrase-based statistical machine translation

Haque

Naskar

Bosch

et al. 2011

Machine Translation

View full text Add to dashboard Cite

The translation features typically used in Phrase-Based Statistical Machine Translation (PB-SMT) model dependencies between the source and target phrases, but not among the phrases in the source language themselves. A swathe of research has demonstrated that integrating source context modelling directly into log-linear PB-SMT can positively influence the weighting and selection of target phrases, and thus improve translation quality. In this contribution we present a revised, extended account of our previous work on using a range of contextual features, including lexical features of neighbouring words, supertags, and dependency information. We add a number of novel aspects, including the use of semantic roles as new contextual features in PB-SMT, adding new language pairs, and examining the scalability of our research to larger amounts of training data. While our results are mixed across feature selections, classifier hyperparameters, language pairs, and learning curves, we observe that including contextual features of the source sentence in general produces improvements. The most significant improvements involve the integration of long-distance contextual features, such as dependency relations in combination with part-of-speech tags in Dutch-to-English subtitle translation, the combination of dependency parse and semantic role information in English-to-Dutch parliamentary debate translation, or supertag features in English-to-Chinese translation.

show abstract

n-Best Reranking for the Efficient Integration of Word Sense Disambiguation and Statistical Machine Translation

Cited by 12 publications

References 11 publications

One translation per discourse

One translation per discourse

Word sense disambiguation for statistical machine translation

Integrating source-language context into phrase-based statistical machine translation

Contact Info

Product

Resources

About