Discriminative Pruning for Discriminative ITG Alignment
نویسندگان
چکیده
While Inversion Transduction Grammar (ITG) has regained more and more attention in recent years, it still suffers from the major obstacle of speed. We propose a discriminative ITG pruning framework using Minimum Error Rate Training and various features from previous work on ITG alignment. Experiment results show that it is superior to all existing heuristics in ITG pruning. On top of the pruning framework, we also propose a discriminative ITG alignment model using hierarchical phrase pairs, which improves both F-score and Bleu score over the baseline alignment system of GIZA++.
منابع مشابه
Better Word Alignments with Supervised ITG Models
This work investigates supervised word alignment methods that exploit inversion transduction grammar (ITG) constraints. We consider maximum margin and conditional likelihood objectives, including the presentation of a new normal form grammar for canonicalizing derivations. Even for non-ITG sentence pairs, we show that it is possible learn ITG alignment models by simple relaxations of structured...
متن کاملIncorporating Constituent Structure Constraint into Discriminative Word Alignment
We introduce an approach to incorporate the constituent structure constraint into a discriminative word alignment model by presenting the constituent constraint in an explicit way and using three operations to ensure the constraint when search the best word alignment. In this way, we will be able to make use of the weak order constraint induced by the inversion transduction grammars (ITG), as w...
متن کاملImproved Discriminative ITG Alignment using Hierarchical Phrase Pairs and Semi-supervised Training
While ITG has many desirable properties for word alignment, it still suffers from the limitation of one-to-one matching. While existing approaches relax this limitation using phrase pairs, we propose a ITG formalism, which even handles units of non-contiguous words, using both simple and hierarchical phrase pairs. We also propose a parameter estimation method, which combines the merits of both ...
متن کاملSoft Syntactic Constraints for Word Alignment through Discriminative Training
Word alignment methods can gain valuable guidance by ensuring that their alignments maintain cohesion with respect to the phrases specified by a monolingual dependency tree. However, this hard constraint can also rule out correct alignments, and its utility decreases as alignment models become more complex. We use a publicly available structured output SVM to create a max-margin syntactic align...
متن کاملSemi-Supervised Block ITG Models for Word Alignment
Labeled training data for the word alignment task, in the form of word-aligned sentence pairs, is hard to come by for many language-pairs. Hence, it is natural to draw upon semi-supervised learning methods (Fraser and Marcu, 2006). We introduce a semisupervised learning method for word alignment using conditional entropy regularization (Grandvalet and Bengio, 2005) on top of a BITG-based discri...
متن کامل