this article the problem of finding the word alignment of a bilingual sentence-aligned corpus by using language-independent statistical methods. There is a vast literature on this topic, and many different systems have been suggested to solve this problem. Our work follows and extends the methods introduced by Brown, Della Pietra, Della Pietra, and Mercer (1993) by using refined statistical models for the translation process. The basic idea of this approach is to develop a model of the translation process with the word alignment as a hidden variable of this process, to apply statistical estimation theory to compute the "optimal" model parameters, and to perform alignment search to compute the best word alignment

In this paper, we present and compare various single-word based alignment models for statistical machine translation. We discuss the five IBM alignment models, the Hidden-Markov alignment model, smoothing techniques and various modifications.

This article presents methods for biasing statistical translation models to reflect these properties. Evaluation with respect to independent human judgments has confirmed that translation models biased in this fashion are significantly more accurate than a baseline knowledge-free model. This article also shows how a statistical translation model can take advantage of preexisting knowledge that might be available about particular language pairs. Even the simplest kinds of languagespecific knowledge, such as the distinction between content words and function words, are shown to reliably boost translation model performance on some tasks. Statistical models that reflect knowledge about the model domain combine the best of both the rationalist and empiricist paradigms

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.

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

The parameters of statistical translation models are typically estimated from sentence-aligned parallel corpora. We show that significant improvements in the alignment and translation quality of such models can be achieved by additionally including wordaligned data during training. Incorporating wordlevel alignments into the parameter estimation of the IBM models reduces alignment error rate and increases the Bleu score when compared to training the same models only on sentence-aligned data. On the Verbmobil data set, we attain a 38 % reduction in the alignment error rate and a higher Bleu score with half as many training examples. We discuss how varying the ratio of word-aligned to sentencealigned data affects the expected performance gain. 1

In this article, we show how a bilingual text-translation alignment method can be adapted to deal with more than two versions of a text. Experiments on a trilingual corpus demonstrate that this method yields better bilingual alignments than can be obtained with bilingual text-alignment methods. Moreover, for a given number of texts, the computational complexity of the multilingual method is the same as for bilingual alignment.

We present a phrasal inversion transduction grammar as an alternative to joint phrasal translation models. This syntactic model is similar to its flatstring phrasal predecessors, but admits polynomial-time algorithms for Viterbi alignment and EM training. We demonstrate that the consistency constraints that allow flat phrasal models to scale also help ITG algorithms, producing an 80-times faster inside-outside algorithm. We also show that the phrasal translation tables produced by the ITG are superior to those of the flat joint phrasal model, producing up to a 2.5 point improvement in BLEU score. Finally, we explore, for the first time, the utility of a joint phrasal translation model as a word alignment method. 1

Recent years have seen a few serious attempts to develop methods and measures for the evaluation of word alignment systems, notably the Blinker project (Melamed, 1998) and the ARCADE project (Vronis and Langlais, forthcoming). In this paper we discuss different approaches to the problem and report on results from a project where two word alignment systems have been evaluated. These results include methods and tools for the generation of reference data and a set of measures for system performance. We note that the selection and sampling of reference data can have a great impact on scoring results.