Frédéric Blain

SHEF-NN: Translation Quality Estimation with Neural Networks

We describe our systems for Tasks 1 and 2 of the WMT15 Shared Task on Quality Estimation. Our submissions use (i) a continuous space language model to extract additional features for Task 1 (SHEFGP, SHEF-SVM), (ii) a continuous bagof-words model to produce word embeddings as features for Task 2 (SHEF-W2V) and (iii) a combination of features produced by QuEst++ and a feature produced with word embedding models (SHEFQuEst++). Our systems outperform the baseline as well as many other submissions. The results are especially encouraging for Task 2, where our best performing system (SHEF-W2V) only uses features learned in an unsupervised fashion.

The QT21/HimL Combined Machine Translation System

This paper describes the joint submission of the QT21 and HimL projects for the English→Romanian translation task of the ACL 2016 First Conference on Machine Translation (WMT 2016). The submission is a system combination which combines twelve different statistical machine translation systems provided by the different groups (RWTH Aachen University, LMU Munich, Charles University in Prague, University of Edinburgh, University of Sheffield, Karlsruhe Institute of Technology, LIMSI, University of Amsterdam, Tilde). The systems are combined using RWTH’s system combination approach. The final submission shows an improvement of 1.0 BLEU compared to the best single system on newstest2016.

Sheffield Systems for the English-Romanian WMT Translation Task.

This paper provides an overview of the submissions the University of Sheffield for the English-Romanian Translation Task of the ACL 2016 First Conference on Machine Translation (WMT16). The submitted translations were produced with a phrase-based system trained using the Moses toolkit, in two variants: (i) n-best rescoring using additional features from Quality Estimation (primary submission), and (ii) a novel weighted ranking optimisation approach (secondary submission).

USFD at SemEval-2016 Task 1: Putting different State-of-the-Arts into a Box

In this paper we describe our participa-tion in the STS Core subtask which is the determination of the monolingual seman-tic similarity between pair of sentences. In our participation we adapted state-of-the-art approaches from related work ap-plied on previous STS Core subtasks and run them on the 2016 data. We inves-tigated the performance of single meth-ods but also the combination of them. Our results show that Convolutional Neu-ral Networks (CNN) are superior to both the Monolingual Word Alignment and the Word2Vec approaches. The combination of all the three methods performs slightly better than using CNN only. Our results also show that the performance of our systems varies between the datasets.

USFD's Phrase-level Quality Estimation Systems.

We describe the submissions of the University of Sheffield (USFD) for the phraselevel Quality Estimation (QE) shared task of WMT16. We test two different approaches for phrase-level QE: (i) we enrich the provided set of baseline features with information about the context of the phrases, and (ii) we exploit predictions at other granularity levels (word and sentence). These approaches perform closely in terms of multiplication of F1-scores (primary evaluation metric), but are considerably different in terms of the F1scores for individual classes.

Continuous Adaptation to User Feedback for Statistical Machine Translation.

This paper gives a detailed experiment feedback of different approaches to adapt a statistical machine translation system towards a targeted translation project, using only small amounts of parallel in-domain data. The experiments were performed by professional translators under realistic conditions of work using a computer assisted translation tool. We analyze the influence of these adaptations on the translator productivity and on the overall post-editing effort. We show that significant improvements can be obtained by using the presented adaptation techniques.