Hierarchical Phrase-Based Translation with Weighted Finite-State Transducers and Shallow-<i>n</i> Grammars

Gispert, Adrià de; Iglesias, Gonzalo; Blackwood, Graeme; Banga, Eduardo Rodríguez; Byrne, William

doi:10.1162/coli_a_00006

Cited by 29 publications

(39 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The choice of the hypergraph representation is merely one of several alternatives. For example, we could have adopted a representation based on weighted finite state transducers (de Gispert, Iglesias, Blackwood, Banga, & Byrne, 2010) since our model describes a regular language both in terms of the PCFG and the surface level models we intersect it with. It is also possible to represent our grammar as a pushdown automaton (Iglesias, Allauzen, Byrne, de Gispert, & Riley, 2011) and intersect it with finite automata representing a language model and dependency-related information, respectively.…”

Section: Learningmentioning

confidence: 99%

A Global Model for Concept-to-Text Generation

Konstas¹,

Lapata²

2013

jair

View full text Add to dashboard Cite

Concept-to-text generation refers to the task of automatically producing textual output from non-linguistic input. We present a joint model that captures content selection ("what to say") and surface realization ("how to say") in an unsupervised domain-independent fashion. Rather than breaking up the generation process into a sequence of local decisions, we define a probabilistic context-free grammar that globally describes the inherent structure of the input (a corpus of database records and text describing some of them). We recast generation as the task of finding the best derivation tree for a set of database records and describe an algorithm for decoding in this framework that allows to intersect the grammar with additional information capturing fluency and syntactic well-formedness constraints. Experimental evaluation on several domains achieves results competitive with state-of-the-art systems that use domain specific constraints, explicit feature engineering or labeled data.

show abstract

Section: Learningmentioning

confidence: 99%

A Global Model for Concept-to-Text Generation

Konstas¹,

Lapata²

2013

jair

View full text Add to dashboard Cite

show abstract

“…In this section we describe HiFST [47,81], a hierarchical phrase-based decoder that uses a WFST. The HiFST decoder produces word lattices and applies the full language model using WFST operations.…”

Section: Hifstmentioning

confidence: 99%

“…In this section we describe the HiFST decoder [47,79,81]. As stated in Section 2.3.1 the CYK algorithm produces a large number of derivations that share sub-derivations.…”

Section: Hierarchical Phrase-based Decoding With Wfstsmentioning

confidence: 99%

The Geometry of Statistical Machine Translation

Waite¹,

Byrne²

2015

Proceedings of the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Langua

View full text Add to dashboard Cite

Most modern statistical machine translation systems are based on the linear model. There are many reasons for the prevalence of the linear model: other component models can be incorporated as features, there are many methods for estimating their parameters, and the resulting model scores can be easily used in finite-state representations.One popular method for estimating the parameters of a linear model is minimum error rate training (MERT). Galley and Quirk describe an optimal MERT algorithm that requires an exponential runtime for multiple sentences. We find that this form of MERT can be represented using convex geometry. Using this geometric representation we describe an optimal algorithm that runs in polynomial time with respect to the number of feasible solutions, and describe Projected MERT, a practical implementation of multidimensional MERT in low dimensions.Using this geometric representation of MERT we investigate whether the optimisation of linear models is tractable in general. It has been believed that the number of feasible solutions of a linear model is exponential with respect to the number of training sentences, however we show that the exponential explosion is due to feature dimension. A result that has important ramifications because of the current trend of building statistical machine translation systems around a large number of sparse features.We also show how these convex geometric descriptions of linear models can be neatly integrated into finite-state representations using tropical geometry. The resulting semirings provide a formulation for multidimensional MERT that can be applied to lattices and hypergraphs.In contrast to this theoretical work, we also present practical descriptions of tools and techniques used for fast parameter estimation and filtering of hierarchical phrase-based translation models and N-gram language models. These techniques allow us to quickly build rich models over large datasets. The resulting models are used as the basis of a machine translation system that performs competitively at international evaluations.

show abstract

“…Closer to our context, (de Gispert et al, 2010) propose to use Finite-State Transducers in the context of Hierarchical Phrase Based Translation. Their method is to iteratively construct and minimize the full "top-level lattice" representing the whole set of translations bottom-up.…”

Section: Related Workmentioning

confidence: 99%

Translation Rules with Right-Hand Side Lattices

Cromierès

Kurohashi

2014

Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP)

View full text Add to dashboard Cite

In Corpus-Based Machine Translation, the search space of the translation candidates for a given input sentence is often defined by a set of (cyclefree) context-free grammar rules. This happens naturally in Syntax-Based Machine Translation and Hierarchical Phrase-Based Machine Translation (where the representation will be the set of the target-side half of the synchronous rules used to parse the input sentence). But it is also possible to describe Phrase-Based Machine Translation in this framework. We propose a natural extension to this representation by using lattice-rules that allow to easily encode an exponential number of variations of each rules. We also demonstrate how the representation of the search space has an impact on decoding efficiency, and how it is possible to optimize this representation.

show abstract

Hierarchical Phrase-Based Translation with Weighted Finite-State Transducers and Shallow-n Grammars

Cited by 29 publications

References 22 publications

A Global Model for Concept-to-Text Generation

A Global Model for Concept-to-Text Generation

The Geometry of Statistical Machine Translation

Translation Rules with Right-Hand Side Lattices

Contact Info

Product

Resources

About