Recently, confusion network decoding has been applied in machine translation system combination. Due to errors in the hypothesis alignment, decoding may result in ungrammatical combination outputs. This paper describes an improved confusion network based method to combine outputs from multiple MT systems. In this approach, arbitrary features may be added log-linearly into the objective function, thus allowing language model expansion and re-scoring. Also, a novel method to automatically select the hypothesis which other hypotheses are aligned against is proposed. A generic weight tuning algorithm may be used to optimize various automatic evaluation metrics including TER, BLEU and METEOR. The experiments using the 2005 Arabic to English and Chinese to English NIST MT evaluation tasks show significant improvements in BLEU scores compared to earlier confusion network decoding based methods. 1

This paper presents a new hypothesis alignment method for combining outputs of multiple machine translation (MT) systems. An indirect hidden Markov model (IHMM) is proposed to address the synonym matching and word ordering issues in hypothesis alignment. Unlike traditional HMMs whose parameters are trained via maximum likelihood estimation (MLE), the parameters of the IHMM are estimated indirectly from a variety of sources including word semantic similarity, word surface similarity, and a distance-based distortion penalty. The IHMM-based method significantly outperforms the state-of-the-art TER-based alignment model in our experiments on NIST benchmark datasets. Our combined SMT system using the

Currently there are several approaches to machine translation (MT) based on different paradigms; e.g., phrasal, hierarchical and syntax-based. These three approaches yield similar translation accuracy despite using fairly different levels of linguistic knowledge. The availability of such a variety of systems has led to a growing interest toward finding better translations by combining outputs from multiple systems. This paper describes three different approaches to MT system combination. These combination methods operate on sentence, phrase and word level exploiting information from-best lists, system scores and target-to-source phrase alignments. The word-level combination provides the most robust gains but the best results on the development test sets (NIST MT05 and the newsgroup portion of GALE 2006 dry-run) were achieved by combining all three methods. 1

In this article we describe HiFST, a lattice-based decoder for hierarchical phrase-based translation and alignment. The decoder is implemented with standard Weighted Finite-State Transducer (WFST) operations as an alternative to the well-known cube pruning procedure. We find that the use of WFSTs rather than k-best lists requires less pruning in translation search, resulting in fewer search errors, better parameter optimization, and improved translation performance. The direct generation of translation lattices in the target language can improve subsequent rescoring procedures, yielding further gains when applying long-span language models and Minimum Bayes Risk decoding. We also provide insights as to how to control the size of the search space defined by hierarchical rules. We show that shallow-n grammars, low-level rule catenation, and other search constraints can help to match the power of the translation system to specific language pairs.

This paper presents an empirical study on how different selections of input translation systems affect translation quality in system combination. We give empirical evidence that the systems to be combined should be of similar quality and need to be almost uncorrelated in order to be beneficial for system combination. Experimental results are presented for composite translations computed from large numbers of different research systems as well as a set of translation systems derived from one of the bestranked machine translation engines in the 2006 NIST machine translation evaluation. 1

We describe refinements to hierarchical translation search procedures intended to reduce both search errors and memory usage through modifications to hypothesis expansion in cube pruning and reductions in the size of the rule sets used in translation. Rules are put into syntactic classes based on the number of non-terminals and the pattern, and various filtering strategies are then applied to assess the impact on translation speed and quality. Results are reported on the 2008 NIST Arabic-to-English evaluation task. 1

We describe an approach to improve Statistical Machine Translation (SMT) performance using multi-lingual, parallel, sentence-aligned corpora in several bridge languages. Our approach consists of a simple method for utilizing a bridge language to create a word alignment system and a procedure for combining word alignment systems from multiple bridge languages. The final translation is obtained by consensus decoding that combines hypotheses obtained using all bridge language word alignments. We present experiments showing that multilingual, parallel text in Spanish, French, Russian, and Chinese can be utilized in this framework to improve translation performance on an Arabic-to-English task. 1

A novel and robust approach to improving statistical machine translation fluency is developed within a minimum Bayesrisk decoding framework. By segmenting translation lattices according to confidence measures over the maximum likelihood translation hypothesis we are able to focus on regions with potential translation errors. Hypothesis space constraints based on monolingual coverage are applied to the low confidence regions to improve overall translation fluency. 1

The success of kernel methods including support vector networks (SVMs) strongly depends on the design of appropriate kernels. While initially kernels were designed in order to handle fixed-length data, their extension to unordered, variable-length data became more than necessary for real pattern recognition problems such as object recognition and bioinformatics. We focus in this paper on object recognition using a new type of kernel referred to as “context-dependent”. Objects, seen as constellations of local features (interest points, regions, etc.), are matched by minimizing an energy function mixing (1) a fidelity term which measures the quality of feature matching, (2) a neighborhood criteria which captures the object geometry and (3) a regularization term. We will show that the fixed-point of this energy is a “contextdependent” kernel (“CDK”) which also satisfies the Mercer condition. Experiments conducted on object recognition show that when plugging our kernel in SVMs, we clearly outperform SVMs with “context-free ” kernels. 1.

Current system combination methods usually use confusion networks to find consensus translations among different systems. Requiring one-to-one mappings between the words in candidate translations, confusion networks have difficulty in handling more general situations in which several words are connected to another several words. Instead, we propose a lattice-based system combination model that allows for such phrase alignments and uses lattices to encode all candidate translations. Experiments show that our approach achieves significant improvements over the state-ofthe-art baseline system on Chinese-to-English translation test sets. 1