Current statistical machine translation systems usually extract rules from bilingual corpora annotated with 1-best alignments. They are prone to learn noisy rules due to alignment mistakes. We propose a new structure called weighted alignment matrix to encode all possible alignments for a parallel text compactly. The key idea is to assign a probability to each word pair to indicate how well they are aligned. We design new algorithms for extracting phrase pairs from weighted alignment matrices and estimating their probabilities. Our experiments on multiple language pairs show that using weighted matrices achieves consistent improvements over using n-best lists in significant less extraction time. 1

In this paper we describe the statistical machine translation system of the Universität Karlsruhe developed for the translation task of the Fourth Workshop on Statistical Machine Translation. The state-ofthe-art phrase-based SMT system is augmented with alternative word reordering and alignment mechanisms as well as optional phrase table modifications. We participate in the constrained condition of German-English and English-German as well as in the constrained condition of French-English and English-French. 1

This paper describes the phrase-based SMT systems developed for our participation in the WMT11 Shared Translation Task. Translations for English↔German and English↔French were generated using a phrase-based translation system which is extended by additional models such as bilingual and fine-grained POS language models, POS-based reordering, lattice phrase extraction and discriminative word alignment. Furthermore, we present a special filtering method for the English-French Giga corpus and the phrase scoring step in the training is parallelized.

In past Evaluations for Machine Translation of European Languages, it could be shown that the translation performance of SMT systems can be increased by integrating a bilingual language model into a phrase-based SMT system. In the bilingual language model, target words with their aligned source words build the tokens of an n-gram based language model. We analyzed the effect of bilingual language models and show where they could help to better model the translation process. We could show improvements of translation quality on German-to-English and Arabic-to-English. In addition, for the Arabic-to-English task, training an extra bilingual language model on the POS tags instead of the surface word forms led to further improvements. 1

We present a word alignment framework that can incorporate partial manual alignments. The core of the approach is a novel semi-supervised algorithm extending the widely used IBM Models with a constrained EM algorithm. The partial manual alignments can be obtained by human labelling or automatically by high-precision-low-recall heuristics. We demonstrate the usages of both methods by selecting alignment links from manually aligned corpus and apply links generated from bilingual dictionary on unlabelled data. For the first method, we conduct controlled experiments on Chinese-English and Arabic-English translation tasks to compare the quality of word alignment, and to measure effects of two different methods in selecting alignment links from manually aligned corpus. For the second method, we experimented with moderate-scale Chinese-English translation task. The experiment results show an average improvement of 0.33 BLEU point across 8 test sets. 1

Word alignment is an important preprocessing step for machine translation. The project aims at incorporating manual alignments from Amazon Mechanical Turk (MTurk) to help improve word alignment quality. As a global crowdsourcing service, MTurk can provide flexible and abundant labor force and therefore reduce the cost of obtaining labels. An easyto-use interface is developed to simplify the labeling process. We compare the alignment results by Turkers to that by experts, and incorporate the alignments in a semi-supervised word alignment tool to improve the quality of the labels. We also compared two pricing strategies for word alignment task. Experimental results show high precision of the alignments provided by Turkers and the semi-supervised approach achieved 0.5 % absolute reduction on alignment error rate. 1

This paper proposes a novel semisupervised word alignment technique called EMDC that integrates discriminative and generative methods. A discriminative aligner is used to find high precision partial alignments that serve as constraints for a generative aligner which implements a constrained version of the EM algorithm. Experiments on small-size Chinese and Arabic tasks show consistent improvements on AER. We also experimented with moderate-size Chinese machine translation tasks and got an average of 0.5 point improvement on BLEU scores across five standard NIST test sets and four other test sets. 1

PCFGs can grow exponentially as additional annotations are added to an initially simple base grammar. We present an approach where multiple annotations coexist, but in a factored manner that avoids this combinatorial explosion. Our method works with linguisticallymotivated annotations, induced latent structure, lexicalization, or any mix of the three. We use a structured expectation propagation algorithm that makes use of the factored structure in two ways. First, by partitioning the factors, it speeds up parsing exponentially over the unfactored approach. Second, it minimizes the redundancy of the factors during training, improving accuracy over an independent approach. Using purely latent variable annotations, we can efficiently train and parse with up to 8 latent bits per symbol, achieving F1 scores up to 88.4 on the Penn Treebank while using two orders of magnitudes fewer parameters compared to the naïve approach. Combining latent, lexicalized, and unlexicalized annotations, our best parser gets 89.4 F1 on all sentences from section 23 of the Penn Treebank. 1