We present translation results on the shared task "Exploiting Parallel Texts for Statistical Machine Translation" generated by a chart parsing decoder operating on phrase tables augmented and generalized with target language syntactic categories. We use a target language parser to generate parse trees for each sentence on the target side of the bilingual training corpus, matching them with phrase table lattices built for the corresponding source sentence. Considering phrases that correspond to syntactic categories in the parse trees we develop techniques to augment (declare a syntactically motivated category for a phrase pair) and generalize (form mixed terminal and nonterminal phrases) the phrase table into a synchronous bilingual grammar. We present results on the French-to-English task for this workshop, representing significant improvements over the workshop's baseline system. Our translation system is available open-source under the GNU General Public License.

Probabilistic synchronous context-free grammar (PSCFG) translation models define weighted transduction rules that represent translation and reordering operations via nonterminal symbols. In this work, we investigate the source of the improvements in translation quality reported when using two PSCFG translation models (hierarchical and syntax-augmented), when extending a state-of-the-art phrasebased baseline that serves as the lexical support for both PSCFG models. We isolate the impact on translation quality for several important design decisions in each model. We perform this comparison on three NIST language translation tasks; Chinese-to-English, Arabic-to-English and Urdu-to-English, each representing unique challenges.

Statistical machine translation (SMT) is based on the ability to effectively learn word and phrase relationships from parallel corpora, a process which is considerably more difficult when the extent of morphological expression differs significant across the source and target languages. We present techniques that select appropriate word segmentations in the morphologically rich source language based on contextual relationships in the target language. Our results take advantage of existing word level morphological analysis components to improve translation quality above state-of-the-art on a limited-data Arabic to English speech translation task.

Decision rules that explicitly account for non-probabilistic evaluation metrics in machine translation typically require special training, often to estimate parameters in exponential models that govern the search space and the selection of candidate translations. While the traditional Maximum A Posteriori (MAP) decision rule can be optimized as a piecewise linear function in a greedy search of the parameter space, the Minimum Bayes Risk (MBR) decision rule is not well suited to this technique, a condition that makes past results difficult to compare. We present a novel training approach for non-tractable decision rules, allowing us to compare and evaluate these and other decision rules on a large scale translation task, taking advantage of the high dimensional parameter space available to the phrase based Pharaoh decoder. This comparison is timely, and important, as decoders evolve to represent more complex search space decisions and are evaluated against innovative evaluation metrics of translation quality.

Abstractis paper describes the open-source Syntax Augmented Machine Translation (SAMT) ¹on Hadoop toolkit-an end-to-end grammar based machine statistical machine translation framework running on the Hadoop implementation of the MapReduce programming model. We present the underlying methodology of the SAMT approach with detailed instructions that describe how to use the toolkit to build grammar based systems for large scale translation tasks.