Word Alignment via Quadratic Assignment

By Simon Lacoste-Julien, B. Taskar, Daniel Klein, Michael I. Jordan. 2006

100 citations recorded by the source corpus.

Recently, discriminative word alignment methods have achieved state-of-the-art accuracies by extending the range of information sources that can be easily incorporated into aligners. The chief advantage of a discriminative framework is the ability to score alignments based on arbitrary features of the matching word tokens, including orthographic form, predictions of other models, lexical context and so on. However, the proposed bipartite matching model of Taskar et al. (2005), despite being tractable and effective, has two important limitations. First, it is limited by the restriction that words have fertility of at most one. More importantly, first order correlations between consecutive words cannot be directly captured by the model. In this work, we address these limitations by enriching the model form. We give estimation and inference algorithms for these enhancements. Our best model achieves a relative AER reduction of 25% over the basic matching formulation, outperforming intersected IBM Model 4 without using any overly compute-intensive features. By including predictions of other models as features, we achieve AER of 3.8 on the standard Hansards dataset.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Natural Language Processing

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

In Computer Vision and Pattern Recognition

In Natural Language Processing

Citations in the corpus

SIGMa: simple greedy matching for aligning large knowledge bases (2012, background): “in [17], which encourages neighbor entities in one graph to”
Introduction to dual composition for inference (2011, background): “, 2010), Matchings (Lacoste-Julien et al., 2006; Duchi et al., 2007; Yarkony et al., 2010), Supermodular functions (Komodakis et al.”
Solving the multi-way matching problem by permutation synchronization (2013, unknown): “However, our approach is fully general and equally applicable to problems such as matching multiple graphs [10, 11].”
Squibs and Discussions: Measuring Word Alignment Quality for Statistical Machine Translation (2007, methodology): “F-Measure with an appropriate setting of α will be useful during the development process of new alignment models, or as a maximization criterion for discriminative training of alignment models (Cherry and Lin 2003; Ayan, Dorr, and Monz 2005; Ittycheriah and Roukos 2005; Liu, Liu, and Lin 2005; Frase”
1 Introduction to Dual Decomposition for Inference (2010, background): “, 2010), Matchings (Lacoste-Julien et al., 2006; Duchi et al., 2007; Yarkony et al., 2010), Supermodular functions (Komodakis et al.”
Database-Text Alignment via Structured Multilabel Classification (2007, background): “While most of the research focuses on unsupervised models [Brown et al., 1993], recently a number of supervised discriminative approaches have been proposed [Taskar et al., 2005; Lacoste-Julien et al., 2006].”
Structured Prediction, Dual Extragradient and Bregman Projections (2006, unknown): “These extensions are developed in Lacoste-Julien et al. (2006).”
On pairwise costs for network flow multi-object tracking (2014, background): “We present such an approach in this section, which generalizes the line of reasoning from [16].”