[Corpora-List] CFP: AMTA 2016 workshop on Interacting with Machine Translation (iMT 2016)

Simard, Michel Michel.Simard at cnrc-nrc.gc.ca
Fri Apr 29 03:57:37 CEST 2016


iMT Workshop - Interacting with Machine Translation ‎ at the AMTA 2016 Conference ‎ in Austin, Texas, October 26 - November 2, 2006‎

https://sites.google.com/site/imt2016workshop

------------------------------ Call For Abstracts ------------------------------

More and more people are interacting with MT, in an increasing variety of ways. They may be end users of machine-translated information, some of whom need high quality text, others who just need to obtain a gist of the meaning. Or they may be intermediaries, such as professional translators, in charge of post-editing the MT output, or volunteers who post-edit for a favourite cause. This Workshop focuses on the theme of interacting with machine translation.

In recent years, there has been a focus on one form of interaction with MT, post-editing, through various workshops (WPTP 1, 2, 3, 4) and publications, e.g. collected volumes (O'Brien et al 2014a, Carl et al 2016) and a special issue of the journal Machine Translation (O'Brien and Simard 2014). This has led to better understanding of interacting with MT through the lens of post-editing.

Historically, there has also been an interest in various forms of "interactive MT", e.g. involving the real-time updating of MT suggestions as the user edits (e.g. Foster et al. 1997, Langlais et al. 2000, Casacuberta et al. 2008, González-Rubio et al. 2013). However, we do not yet understand the impact (positive or negative) of this sort of technology on users such as translators.

Also in recent years, functionality has been developed in editing tools to log and model translator interaction with MT (e.g Casmacat, iOmegaT, Lilt, Matecat, Translog II, HandyCAT) and new interfaces are emerging as a result. Only limited research exists on the impacts of these UI innovations. Furthermore, very little research has been done from the point of view of readability, accessibility, acceptability, customer satisfaction, or usability of raw MT or post-edited MT on end users, or on how this compares with content translated by human translators, or written by native speakers.

The time is right to call for researchers, developers and users of MT to consider questions relating to interacting with MT. The workshop deliberately uses the term "interacting" to broaden the focus beyond post-editing and interactive MT, though both of these topics fit comfortably within the theme.

------------------------------ Interacting with MT: Questions of interest ------------------------------

We are particularly interested in the following themes, but also welcome other ideas which touch on human interaction with MT:‎

* How do organizations currently measure end-user interaction and

satisfaction with MT?

* What robust methods can be used to measure users' interaction with

MT?

* Do these measures correlate with traditional MT quality evaluation

metrics?

* How can developers best use end-user interaction data to improve MT

output quality and user experience of MT? ‎

* What new interactive features are being developed in CAT tools for

translators who interact with MT?

* What is the impact of features such as interactive editing,

highlighting of phrase alignment etc., on translators, the

translation process and product?

* How does translator interaction with MT (post-editing) differ from

translator interaction with human translation (revision)?‎

* How does raw or post-edited MT affect usability of products or

content?

* How does raw or post-edited MT affect user satisfaction?

* How does end-user satisfaction of MT output compare with

satisfaction of human translation output?

* How does satisfaction or usability of MT output compare with

satisfaction or usability of source language content?‎

* Do levels of satisfaction differ across target language, culture,

content type or domain?

------------------------------ Abstract Submission Information‎ ------------------------------

We call for *extended abstracts* that address any of the questions above, or related questions. Abstracts must be 500-600 words long, excluding references. Abstracts should be submitted as PDF files to the START system at https://www.softconf.com/amta2016/imt2016, no later than 11:59 pm PDT (GMT - 12 hours), Monday August 29, 2016. Authors will be notified of acceptance by September 8.

Submissions will be evaluated based on their relevance to the workshop theme, potential for a stimulating presentation, potential appeal to the target audience, and overall quality. Accepted abstracts will be presented as 20 minute presentations at the Workshop (with 10 for discussion). Note that at least one author must register for the Workshop.

Authors may subsequently be invited to submit extended papers to a Special Issue of the journal Machine Translation.‎

------------------------------ Multiple Submissions ------------------------------

Work presented at iMT must represent new work that has not been previously published. It is the responsibility of the author(s) to inform the organizers of any potential problem with respect to this requirement. Authors must inform the iMT organizers by email (michel.simard at cnrc-nrc.gc.ca, sharon.obrien at dcu.ie) specifying to which other conference or workshop they are submitting their work.‎

------------------------------ Important dates ------------------------------

* Submission of abstracts: August 29 * Notification of acceptance: September 8‎ * Workshop: October 29‎ or November 2, 2016 (TBD) ‎

------------------------------ Organisers ------------------------------

Sharon O'Brien ADAPT, CTTS DCU, Ireland sharon.obrien at dcu.ie

Michel Simard NRC Canada michel.simard at cnrc-nrc.gc.ca



More information about the Corpora mailing list