Tree-to-string alignment template for statistical machine translation

Liu, Yang; Li, Qun; Lin, Sheng

doi:10.3115/1220175.1220252

Cited by 193 publications

(180 citation statements)

References 21 publications

(23 reference statements)

Supporting

Mentioning

180

Contrasting

Order By: Relevance

“…Huang et al (2006), Liu et al (2006), Zollmann and Venugopal (2006), , , Almaghout et al (2010), Almaghout et al (2012), Li et al (2012). In terms of labeling Hiero rules, SAMT (Zollmann and Venugopal 2006;Mylonakis and Sima'an 2011) exploits a 'softer' notion of syntax by fitting the CCG-like syntactic labels to non-constituent phrases.…”

Section: Syntax-based Labelsmentioning

confidence: 99%

Labeling hierarchical phrase-based models without linguistic resources

Wenniger

Sima’an

2015

Machine Translation

View full text Add to dashboard Cite

Long-range word order differences are a well-known problem for machine translation. Unlike the standard phrase-based models which work with sequential and local phrase reordering, the hierarchical phrase-based model (Hiero) embeds the reordering of phrases within pairs of lexicalized context-free rules. This allows the model to handle long range reordering recursively. However, the Hiero grammar works with a single nonterminal label, which means that the rules are combined together into derivations independently and without reference to context outside the rules themselves. Follow-up work explored remedies involving nonterminal labels obtained from monolingual parsers and taggers. As of yet, no labeling mechanisms exist for the many languages for which there are no good quality parsers or taggers. In this paper we contribute a novel approach for acquiring reordering labels for Hiero grammars directly from the word-aligned parallel training corpus, without use of any taggers or parsers. The new labels represent types of alignment patterns in which a phrase pair is embedded within larger phrase pairs. In order to obtain alignment patterns that generalize well, we propose to decompose word alignments into trees over phrase pairs. Beside this labeling approach, we contribute coarse and sparse features for learning soft, weighted label-substitution as opposed to standard substitution. We report extensive experiments comparing our model to two baselines: Hiero and the known syntax augmented machine translation (SAMT) variant, which labels Hiero rules with nonterminals extracted from monolingual syntactic parses. We also test a simplified labeling scheme based on inversion transduction grammar (ITG English task we obtain performance improvement up to 1 BLEU point, whereas for the German-English task, where morphology is an issue, a minor (but statistically significant) improvement of 0.2 BLEU points is reported over SAMT. While ITG labeling does give a performance improvement, it remains sometimes suboptimal relative to our proposed labeling scheme.

show abstract

Section: Syntax-based Labelsmentioning

confidence: 99%

Labeling hierarchical phrase-based models without linguistic resources

Wenniger

Sima’an

2015

Machine Translation

View full text Add to dashboard Cite

show abstract

“…This algorithm is referred to as GHKM (Galley et al, 2004) and is widely used in SSMT systems (Galley et al, 2006;Liu et al, 2006;Huang et al, 2006). The word alignment used in GHKM is usually computed independent of the syntactic structure, and as DeNero and Klein (2007) and May and Knight (2007) have noted,…”

Section: Baseline Approach: Tts Templates Obeying Word Alignmentmentioning

confidence: 99%

“…The first is synchronous parsing (Galley et al, 2006;May and Knight, 2007), where TTS templates are used to construct synchronous parse trees for an input sentence, and the translations will be generated once the synchronous trees are built up. The other way is the TTS transducer (Liu et al, 2006;Huang et al, 2006), where TTS templates are used just as their name indicates: to transform a source parse tree (or forest) into the proper target string. Since synchronous parsing considers all possible synchronous parse trees of the source sentence, it is less constrained than TTS transducers and hence requires more computational power.…”

Section: Extracting Phrasal Tts Templatesmentioning

confidence: 99%

“…Because such approaches directly learn a generative model over phrase pairs, they are theoretically preferable to the standard heuristics for extracting the phrase pairs from the many-to-one word-level alignments produced by the IBM series models (Brown et al, 1993) or the Hidden Markov Model (HMM) (Vogel et al, 1996). We wish to apply this direct, Bayesian approach to learn better translation rules for syntaxbased statistical MT (SSMT), by which we specifically refer to MT systems using Tree-to-String (TTS) translation templates derived from syntax trees (Liu et al, 2006;Huang et al, 2006;Galley et al, 2006;May and Knight, 2007), as opposed to formally syntactic systems such as Hiero (Chiang, 2007). The stumbling block preventing us from taking this approach is the extremely large space of possible TTS templates when no word alignments are given.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Bayesian learning of phrasal tree-to-string templates

Liu

Gildea

2009

Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing Volume 3 - EMNLP '09

View full text Add to dashboard Cite

We examine the problem of overcoming noisy word-level alignments when learning tree-to-string translation rules. Our approach introduces new rules, and reestimates rule probabilities using EM. The major obstacles to this approach are the very reasons that word-alignments are used for rule extraction: the huge space of possible rules, as well as controlling overfitting. By carefully controlling which portions of the original alignments are reanalyzed, and by using Bayesian inference during re-analysis, we show significant improvement over the baseline rules extracted from word-level alignments.

show abstract

“…(Liu et al, 2006;Huang et al, 2006;Zollmann and Venugopal, 2006;Wu and Hkust, 1998). In terms of labeling Hiero rules, SAMT (Zollmann and Venugopal, 2006;Mylonakis and Sima'an, 2011) exploits a "softer notion" of syntax by fitting the CCG-like syntactic labels to non-constituent phrases.…”

Section: Hierarchical Models and Related Workmentioning

confidence: 99%

Bilingual Markov Reordering Labels for Hierarchical SMT

Wenniger

Sima’an

2014

Proceedings of SSST-8, Eighth Workshop on Syntax, Semantics and Structure in Statistical Translation

View full text Add to dashboard Cite

Earlier work on labeling Hiero grammars with monolingual syntax reports improved performance, suggesting that such labeling may impact phrase reordering as well as lexical selection. In this paper we explore the idea of inducing bilingual labels for Hiero grammars without using any additional resources other than original Hiero itself does. Our bilingual labels aim at capturing salient patterns of phrase reordering in the training parallel corpus. These bilingual labels originate from hierarchical factorizations of the word alignments in Hiero's own training data. In this paper we take a Markovian view on synchronous top-down derivations over these factorizations which allows us to extract 0 th -and 1 st -order bilingual reordering labels. Using exactly the same training data as Hiero we show that the Markovian interpretation of word alignment factorization offers major benefits over the unlabeled version. We report extensive experiments with strict and soft bilingual labeled Hiero showing improved performance up to 1 BLEU points for ChineseEnglish and about 0.1 BLEU points for German-English.

show abstract

Tree-to-string alignment template for statistical machine translation

Cited by 193 publications

References 21 publications

Labeling hierarchical phrase-based models without linguistic resources

Labeling hierarchical phrase-based models without linguistic resources

Bayesian learning of phrasal tree-to-string templates

Bilingual Markov Reordering Labels for Hierarchical SMT

Contact Info

Product

Resources

About