http://www.statmt.org/wmt08/[HOME] | [SHARED TRANSLATION TASK] | [SHARED EVALUATION TASK] | [RESULTS] | [BASELINE SYSTEM]
This workshop on statistical and hybrid methods for machine translation, and builds on the 2005 ACL Workshop on Parallel Text, the 2006 NAACL Workshop on Statistical Machine Translation, and the 2007 ACL Second Workshop on Statistical Machine Translation. The workshop will feature papers on topics related to MT, and will feature two shared tasks: a shared translation task for 12 pairs of European languages, and a shared evaluation task to test automatic evaluation metrics.
Topics of interest include, but are not limited to:
We encourage authors to evaluate their approaches to the above topics using the common data sets created for the shared translation task. In addition to scientific papers, we will also feature two shared tasks.
The first is a shared translation task which will examine translation between the following language pairs:
All participants who submit entries will have their translations evaluated. We will evaluate translation performance by human judgment. To facilitate the human evaluation we will require participants in the shared task to manually judge some of the submitted translations.
A more detailed description of the shared translation task (including information about the test and training corpora, a freely available MT system, and a number of other resources) is available from http://www.statmt.org/wmt08/shared-task.html. We also provide a baseline machine translation system, whose performance matches the best systems from last year's shared task.
The second task is a shared evaluation task. Participants in this task will submit automatic evaluation metrics for machine translation, which will be assessed on their ability to:
Participants in the shared translation task will submit translation results for a set of a few thousand sentences. Their system outputs will be distributed to participants in the shared evaluation task along with the reference translations. The translations will be ranked with automatic evaluation metrics. We will measure the correlation of automatic evaluation metrics with the human judgments.
More details of the shared evaluation task (including submission formats and the collected manual evaluations from last year's workshop) is available from http://www.statmt.org/wmt08/shared-evaluation-task.html.
We encourage individuals who are submitting research papers to evaluate their approaches using the training resources provided by this workshop and past workshops, so that their experiments can be repeated by others using these publicly available corpora.
|Regular paper submissions||March 14, 2008|
|(shared translation task) Results submissions||March 21, 2008|
|(shared evaluation task) Results submissions||April 4, 2008|
|(both shared tasks) Short paper submissions||April 4, 2008|
|Notification||April 12, 2008|
|Camera-ready papers||April 21, 2008|
Speaker: Daniel Marcu (Language Weaver/ISI)
Title: "Searching Efficiently for Solutions in Large-Scale NLP Application"
Abstract: As the fields of natural language processing and machine learning mature, the gap between the mathematical equations that we write when we model a problem statistically and the manner in which we implement these equations in NLP applications is widening. This talk reviews some of the challenges that we face when searching for best solutions in large-scale statistical applications, such as machine translation, and the effect that the ignoring of these challenges is having on end-to-end results. It also presents recent developments that have the potential to impact positively a wide range of applications where parameter estimation and search are critical.
|supported by the EuroMatrix project, P6-IST-5-034291-STP|
funded by the European Commission under Framework Programme 6