CFP: Machine Translation Journal, Special Issue on “Pushing the frontier of Statistical Machine Translation


Special Issue on Pushing the frontier of Statistical Machine Translation


Guest editors:
Lucia Specia (University of Wolverhampton)
and Nicola Cancedda (Xerox Research Centre Europe)


Many current approaches to statistical machine translation are based on incremental improvements over a standard framework: word-alignments produced by IBM-models, rules extracted from word alignments using heuristics, a log-linear model combining commonly used features whose weights are optimized on a development set using reference translations,
and a beam-search decoder to generate final translations.

This special issue is dedicated to alternative approaches to statistical MT radically diverging in as many as possible of the aspects involved in the process, including (but not limited to) the

– Word / phrase alignment (including hierarchical alignment)
– Rule / phrase extraction
– Statistical translation models
– Features / factors and ways to combine them
– Combination of linguistic and statistical methods
– Language modelling
– Real-time adaptation of MT with user feedback
– Domain adaptation of MT
– Parameter optimization
– Decoding

Contributions on approaches which still achieve performance levels below the state of the art can be suitable for inclusion in this issue, provided they stem from sound directions with clear potential
for further developments.


Submission guidelines:

– Authors should follow the “Instructions for Authors” available on the MT Journal website:

– Submissions must be limited to 15 pages (including references)

– Papers should be submitted online directly on the MT journal’s submission website:, indicating this special issue in ‘article type’


Important dates

– Deadline for ‘Submission Intent’ e-mail: September 15, 2009 – contributors must send an email with the abstract of the submission to lspecia (at)

– Deadline for submissions: October 30, 2009

– Notification to the authors: December 20, 2009

– Deadline for camera-ready: January 30, 2009