The special challenge of the WMT 2007 shared task was domain adaptation. We took this opportunity to experiment with various ways of adapting a statistical machine translation systems to a special domain (here: news commentary), when most of the training data is from a different domain (here: European Parliament speeches). This paper also gives a description of the submission of the University of Edinburgh to the shared task. 1 Our framework: the Moses MT system The open source Moses (Koehn et al., 2007) MT system was originally developed at the University

This paper presents a simple and robust consensus decoding approach for combining multiple Machine Translation (MT) system outputs. A consensus network is constructed from an N-best list by aligning the hypotheses against an alignment reference, where the alignment is based on minimising the translation edit rate (TER). The Minimum Bayes Risk (MBR) decoding technique is investigated for the selection of an appropriate alignment reference. Several alternative decoding strategies proposed to retain coherent phrases in the original translations. Experimental results are presented primarily based on three-way combination of Chinese-English translation outputs, and also presents results for six-way system combination. It is shown that worthwhile improvements in translation performance can be obtained using the methods discussed.

Paraphrase patterns are useful in paraphrase recognition and generation. In this paper, we present a pivot approach for extracting paraphrase patterns from bilingual parallel corpora, whereby the English paraphrase patterns are extracted using the sentences in a foreign language as pivots. We propose a loglinear model to compute the paraphrase likelihood of two patterns and exploit feature functions based on maximum likelihood estimation (MLE) and lexical weighting (LW). Using the presented method, we extract over 1,000,000 pairs of paraphrase patterns from 2M bilingual sentence pairs, the precision of which exceeds 67%. The evaluation results show that: (1) The pivot approach is effective in extracting paraphrase patterns, which significantly outperforms the conventional method DIRT. Especially, the log-linear model with the proposed feature functions achieves high performance. (2) The coverage of the extracted paraphrase patterns is high, which is above 84%. (3) The extracted paraphrase patterns can be classified into 5 types, which are useful in various applications. 1

In this paper we describe a word reordering strategy for statistical machine translation that reorders the source side based on Part of Speech (POS) information. Reordering rules are learned from the word aligned corpus. Reordering is integrated into the decoding process by constructing a lattice, which contains all word reorderings according to the reordering rules. Probabilities are assigned to the different reorderings. On this lattice monotone decoding is performed. This reordering strategy is compared with our previous reordering strategy, which looks at all permutations within a sliding window. We extend reordering rules by adding context information. Phrase translation pairs are learned from the original corpus and from a reordered source corpus to better capture the reordered word sequences at decoding time. Results are presented for English → Spanish and

Phrase-based decoding produces state-of-theart translations with no regard for syntax. We add syntax to this process with a cohesion constraint based on a dependency tree for the source sentence. The constraint allows the decoder to employ arbitrary, non-syntactic phrases, but ensures that those phrases are translated in an order that respects the source tree’s structure. In this way, we target the phrasal decoder’s weakness in order modeling, without affecting its strengths. To further increase flexibility, we incorporate cohesion as a decoder feature, creating a soft constraint. The resulting cohesive, phrase-based decoder is shown to produce translations that are preferred over non-cohesive output in both automatic and human evaluations. 1

We describe a machine translation approach being designed at HKUST to integrate semantic processing into statistical machine translation, beginning with entity and word sense disambiguation. We show how integrating the semantic modules consistently improves translation quality across several data sets. We report results on five different IWSLT 2006 speech translation tasks, representing HKUST’s first participation in the IWSLT spoken language translation evaluation campaign. We translated both read and spontaneous speech transcriptions from Chinese to English, achieving reasonable performance despite the fact that our system is essentially text-based and therefore not designed and tuned to tackle the challenges of speech translation. We also find that the system achieves reasonable results on a wide range of languages, by evaluating on read speech transcriptions from Arabic, Italian, and Japanese into English. 1.

This paper presents a reordering framework for statistical machine translation (SMT) where source-side reorderings are integrated into SMT decoding, allowing for a highly constrained reordered search graph. The monotone search is extended by means of a set of reordering patterns (linguistically motivated rewrite patterns). Patterns are automatically learnt in training from word-to-word alignments and source-side Part-Of-Speech (POS) tags. Traversing the extended search graph, the decoder evaluates every hypothesis making use of a group of widely used SMT models and helped by an additional Ngram language model of sourceside POS tags. Experiments are reported on the Euparl task (Spanish-to-English and English-to-Spanish). Results are presented regarding translation accuracy (using human and automatic evaluations) and computational efficiency, showing significant improvements in translation quality for both translation directions at a very low computational cost. 1

Constrained decoding is of great importance not only for speed but also for translation quality. Previous efforts explore soft syntactic constraints which are based on constituent boundaries deduced from parse trees of the source language. We present a new framework to establish soft constraints based on a more natural alternative: translation boundary rather than constituent boundary. We propose simple classifiers to learn translation boundaries for any source sentences. The classifiers are trained directly on word-aligned corpus without using any additional resources. We report the accuracy of our translation boundary classifiers. We show that using constraints based on translation boundaries predicted by our classifiers achieves significant improvements over the baseline on large-scale Chinese-to-English translation experiments. The new constraints also significantly outperform constituent boundary based syntactic constrains. 1

This paper focuses on the task of inserting punctuation symbols into transcribed conversational speech texts, without relying on prosodic cues. We investigate limitations associated with previous methods, and propose a novel approach based on dynamic conditional random fields. Different from previous work, our proposed approach is designed to jointly perform both sentence boundary and sentence type prediction, and punctuation prediction on speech utterances. We performed evaluations on a transcribed conversational speech domain consisting of both English and Chinese texts. Empirical results show that our method outperforms an approach based on linear-chain conditional random fields and other previous approaches. 1

We apply machine learning to the Linear Ordering Problem in order to learn sentence-specific reordering models for machine translation. We demonstrate that even when these models are used as a mere preprocessing step for German-English translation, they significantly outperform Moses ’ integrated lexicalized reordering model. Our models are trained on automatically aligned bitext. Their form is simple but novel. They assess, based on features of the input sentence, how strongly each pair of input word tokens wi, wj would like to reverse their relative order. Combining all these pairwise preferences to find the best global reordering is NP-hard. However, we present a non-trivial O(n3) algorithm, based on chart parsing, that at least finds the best reordering within a certain exponentially large neighborhood. We show how to iterate this reordering process within a local search algorithm, which we use in training. 1