Automatic word alignment plays a critical role in statistical machine translation. Unfortunately the relationship between alignment quality and statistical machine translation performance has not been well understood. In the recent literature the alignment task has frequently been decoupled from the translation task, and assumptions have been made about measuring alignment quality for machine translation which, it turns out, are not justified. In particular, none of the tens of papers published over the last five years has shown that significant decreases in Alignment Error Rate, AER (Och and Ney, 2003), result in significant increases in translation quality. This paper explains this state of affairs and presents steps towards measuring alignment quality in a way which is predictive of statistical machine translation quality. 1.

Statistical machine translation (SMT) treats the translation of natural language as a machine learning problem. By examining many samples of human-produced translation, SMT algorithms automatically learn how to translate. SMT has made tremendous strides in less than two decades, and many popular techniques have only emerged within the last few years. This survey presents a tutorial overview of state-of-the-art SMT at the beginning of 2007. We begin with the context of the current research, and then move to a formal problem description and an overview of the four main subproblems: translational equivalence modeling, mathematical modeling, parameter estimation, and decoding. Along the way, we present a taxonomy of some different approaches within these areas. We conclude with an overview of evaluation and notes on future directions.

This paper presents a novel training algorithm for a linearly-scored block sequence translation model. The key component is a new procedure to directly optimize the global scoring function used by a SMT decoder. No translation, language, or distortion model probabilities are used as in earlier work on SMT. Therefore our method, which employs less domain specific knowledge, is both simpler and more extensible than previous approaches. Moreover, the training procedure treats the decoder as a black-box, and thus can be used to optimize any decoding scheme. The training algorithm is evaluated on a standard Arabic-English translation task. 1

We introduce a semi-supervised approach to training for statistical machine translation that alternates the traditional Expectation Maximization step that is applied on a large training corpus with a discriminative step aimed at increasing word-alignment quality on a small, manually word-aligned sub-corpus. We show that our algorithm leads not only to improved alignments but also to machine translation outputs of higher quality. 1

This paper presents an extensive evaluation of five different alignments and investigates their impact on the corresponding MT system output. We introduce new measures for intrinsic evaluations and examine the distribution of phrases and untranslated words during decoding to identify which characteristics of different alignments affect translation. We show that precision-oriented alignments yield better MT output (translating more words and using longer phrases) than recalloriented alignments. 1

In this paper we present a novel approach for inducing word alignments from sentence aligned data. We use a Conditional Random Field (CRF), a discriminative model, which is estimated on a small supervised training set. The CRF is conditioned on both the source and target texts, and thus allows for the use of arbitrary and overlapping features over these data. Moreover, the CRF has efficient training and decoding processes which both find globally optimal solutions. We apply this alignment model to both French-English and Romanian-English language pairs. We show how a large number of highly predictive features can be easily incorporated into the CRF, and demonstrate that even with only a few hundred word-aligned training sentences, our model improves over the current state-ofthe-art with alignment error rates of 5.29 and 25.8 for the two tasks respectively. 1

For many years, statistical machine translation relied on generative models to provide bilingual word alignments. In 2005, several independent efforts showed that discriminative models could be used to enhance or replace the standard generative approach. Building on this work, we demonstrate substantial improvement in word-alignment accuracy, partly though improved training methods, but predominantly through selection of more and better features. Our best model produces the lowest alignment error rate yet reported on Canadian Hansards bilingual data. 1

This paper presents a maximum entropy machine translation system using a minimal set of translation blocks (phrase-pairs). While recent phrase-based statistical machine translation (SMT) systems achieve significant improvement over the original source-channel statistical translation models, they 1) use a large inventory of blocks which have significant overlap and 2) limit the use of training to just a few parameters (on the order of ten). In contrast, we show that our proposed minimalist system (DTM2) achieves equal or better performance by 1) recasting the translation problem in the traditional statistical modeling approach using blocks with no overlap and 2) relying on training most system parameters (on the order of millions or larger). The new model is a direct translation model (DTM) formulation which allows easy integration of additional/alternative views of both source and target sentences such as segmentation for a source language such as Arabic, part-of-speech of both source and target, etc. We show improvements over a state-of-the-art phrase-based decoder in Arabic-English translation. 1

State-of-the-art statistical machine translation is based on alignments between phrases – sequences of words in the source and target sentences. The learning step in these systems often relies on alignments between words. It is often assumed that the quality of this word alignment is critical for translation. However, recent results suggest that the relationship between alignment quality and translation quality is weaker than previously thought. We investigate this question directly, comparing the impact of highquality alignments with a carefully constructed set of degraded alignments. In order to tease apart various interactions, we report experiments investigating the impact of alignments on different aspects of the system. Our results confirm a weak correlation, but they also illustrate that more data and better feature engineering may be more beneficial than better alignment. 1

We offer a simple, effective, and scalable method for statistical machine translation parameter tuning based on the pairwise approach to ranking (Herbrich et al., 1999). Unlike the popular MERT algorithm (Och, 2003), our pairwise ranking optimization (PRO) method is not limited to a handful of parameters and can easily handle systems with thousands of features. Moreover, unlike recent approaches built upon the MIRA algorithm of Crammer and Singer (2003) (Watanabe et al., 2007; Chiang et al., 2008b), PRO is easy to implement. It uses off-the-shelf linear binary classifier software and can be built on top of an existing MERT framework in a matter of hours. We establish PRO’s scalability and effectiveness by comparing it to MERT and MIRA and demonstrate parity on both phrase-based and syntax-based systems in a variety of language pairs, using large scale data scenarios. 1