<html>
<head>
<style><!--
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 10pt;
font-family:Verdana
}
--></style>
</head>
<body class='hmmessage'>
<meta http-equiv="Content-Type" content="text/html; charset=unicode">
<meta name="Generator" content="Microsoft SafeHTML">
<style>
.hmmessage P
{margin:0px;padding:0px;}
body.hmmessage
{font-size:10pt;font-family:Verdana;}
</style>
<br><pre>Our apologies if you receive multiple postings of this CFP<br> <br>SemEval-2010 Task 2: Cross-Lingual Lexical Substitution<br><a href="http://lit.csci.unt.edu/index.php/Semeval_2010" target="_blank" onclick="onClickUnsafeLink(event);">http://lit.csci.unt.edu/index.php/Semeval_2010</a><br> <br> <br>The purpose of this e-mail is to encourage participation in the task 'Cross-Lingual Lexical Substitution' at the 5th International Workshop on Semantic Evaluations, SemEval-2010 <a href="http://semeval2.fbk.eu/semeval2.php" target="_blank" onclick="onClickUnsafeLink(event);">http://semeval2.fbk.eu/semeval2.php</a>.<br> <br> <br>TASK DESCRIPTION<br> <br>The goal of this task is to evaluate systems that attempt to solve the problem of cross-lingual lexical substitution. Given a paragraph and a target word, the goal of a participating system would be to provide several correct translations for that target word in a given fixed<br>language, with the constraint that the translations fit the given context in the source language. This is a follow-up of the English lexical substitution task from SemEval-2007 (McCarthy and Navigli, 2007,<br><a href="http://www.informatics.sussex.ac.uk/research/groups/nlp/mccarthy/task10index.html" target="_blank" onclick="onClickUnsafeLink(event);">http://www.informatics.sussex.ac.uk/research/groups/nlp/mccarthy/task10index.html</a>), but now the task is defined in a cross-lingual framework.<br> <br>This task is different from the traditional automatic machine translation task, in that:<br> <br>* we target one word at a time, not an entire sentence like machine translation<br>* we seek as many good translations as possible, not just one final answer<br> <br>A notable comparison between our task and the task of word sense disambiguation is that we do not restrict the translations to those in a specific dictionary or parallel corpus. We let the annotators and systems alike be free to choose translations from any resource(s) they might want<br>to use or consider optimal. Additionally, we do not assume a fixed grouping of translations.<br> <br>For the purpose of this evaluation, we restrict the source language to be English and the target language to be Spanish. We will use two different metrics to evaluate systems.<br> <br>ORGANIZERS<br> <br> * Rada Mihalcea [University of North Texas]<br> * Diana McCarthy [University of Sussex]<br> * Ravi Sinha [University of North Texas]<br> <br>IMPORTANT DATES<br> <br> * Test data availability: 1 March - 2 April, 2010<br> * Result submission deadline: within 7 days after downloading the TEST data<br> * Closing of the acceptance: 2 April<br> <br>VENUE<br> <br>The workshop will be held in conjunction with ACL July 11-16, Uppsala,<br><a href="http://acl2010.org/" target="_blank" onclick="onClickUnsafeLink(event);">http://acl2010.org</a><br> <br>For more information, please visit: <a href="http://lit.csci.unt.edu/index.php/Semeval_2010" target="_blank" onclick="onClickUnsafeLink(event);">http://lit.csci.unt.edu/index.php/Semeval_2010</a><br> </pre> <br /><hr />Hotmail: Free, trusted and rich email service. <a href='http://clk.atdmt.com/GBL/go/201469228/direct/01/' target='_new'>Get it now.</a></body>
</html>