International Workshop on

Using Linguistic Information

for Hybrid Machine Translation

LIHMT

Shared Task on

Applying Machine Learning

Techniques to Optimise the

Division of Labour in Hybrid

Machine Translation

Barcelona

November 2011

LIHMT 2011

International Workshop on

Using Linguistic Information

for Hybrid Machine Translation

18th November 2011

Universitat Politècnica de Catalunya

Barcelona

Contents

Introductions, programme, and About the OpenMT-2 project

Rich morphology and what can we expect from hybrid approaches to MT [abstract].

Ondřej Bojar

Statistical MT with syntax and morphology: challenges and some solutions [abstract]

Alon Lavie

Linguistic indicators for quality estimation of machine translation [abstract]

Lucia Specia

Improved statistical machine translation using multiword expressions.

Dhouha Bouamor, Nasredine Semmar, and Pierre Zweigenbaum

VERTa: exploring a multidimensional linguistically-motivated metric.

Elisabet Comelles, Jordi Atserias, Victoria Arranz, Irene Castellón, and Olivier Hamon

Using Apertium linguistic data for tokenization to improve Moses SMT performance.

Santiago Cortés Vaíllo and Sergio Ortiz Rojas

Reordering by parsing

Jakob Elming and Martin Haulrich

Comparing corpus-based MT approaches using restricted resources.

Monica Gavrila and Natalia Elita

Deep evaluation of hybrid architectures: simple metrics correlated with human judgments

Gorka Labaka, Arantza Díaz de Ilarraza, Cristina España-Bonet, Lluís Màrquez and Kepa Sarsola

A radically simple, effective annotation and alignment methodology for semantic frame based SMT and MT evaluation.

Chi-kiu Lo and Dekai Wu

Word translation disambiguation without parallel texts.

Erwin Marsi, André Lynum, Lars Bungum, and Björn Gambäck

A new hybrid machine translation approach using cross-language information retrieval and only target text corpora.

Nasredine Semmar and Dhouha Bouamor

ML4HMT

Shared Task on

Applying Machine Learning

Techniques to Optimise the

Division of Labour in Hybrid

Machine Translation

19th November

Barcelona Media

Barcelona

Contents

Introduction, programme and about META-NET

Machine translation system combination with MANY for ML4HMT

LoïcBarrault & Patrik Lambert

DCU confusion network-based system combination for ML4HMT

Tsuyoshi Okita and Josef van Genabith

DFKI system combination with sentence ranking at ML4HMT-2011.

Eleftherios Avramidis

DFKI system combination using syntactic information at ML4HMT-2011

Christian Federmann, Sabine Hunsicker, Yu Chen, and Rui Wang

Results from the ML4HMT shared task on applying machine learning techniques to optimise the division of labour in hybrid MT

Christian Federmann