Learning Machine Translation
(edited volume, in preparation)

Editors:

Cyril Goutte
Nicola Cancedda
Marc Dymetman
George Foster

National Research Council Canada
Xerox Research Centre Europe
Xerox Research Centre Europe
National Research Council Canada

MLIA Workshop
Le Wiki (restricted)

NIPS Series

Following the NIPS 2006 workshop "Machine Learning for Multilingual Information Access", we are preparing an edited volume on the topic of Learning for Machine Translation, to be published by MIT Press in their "NIPS Series".

This volume will include the proceedings of the workshop contributions, but is also open to new contributions describing the development or use of Machine Learning techniques in the context of Machine Translation and related tasks.

Please read the Call for Contributions for more information.

Submission and formatting instructions:

Submissions should take the form of a 15-to-20 page chapter using the above format. They should be sent (in PDF format) to mlia (at) iltevents.org by October 1st, 2007.

Reviewer comments and final notification will be sent by November 15, 2007.

Tentative outline of the book:

I. Enabling Technologies

I.1: Mining parallel corpora from the web

I.2: Exploiting comparable corpora

I.3: Named entity translation

I.4: Word alignment

I.5: Language and translation models

I.6: Learning parallel syntax

I.7: Paraphrase

I.8: Evaluation of Machine Translation

II. Machine Translation

II.1: Syntax-based approaches

II.2: Large-scale discriminative approahces

II.3: Kernel-based Machine Translation

II.4: Semi-supervised learning for MT

II.5: (Inter)Active learning for MT

II.6: Adaptive MT

II.7: Learning to combine translation systems

II.8: Factored MT

Confirmed contributors: S. Bangalore, P. Haffner, A. Klementiev, D. Melamed, I. Muslea, D. Roth, A. Sarkar, J. Shawe-Taylor, R. Steinberger, J. Turian, N. Ueffing, Z. Wang, K. Yamada