<html><head><meta http-equiv="content-type" content="text/html; charset=utf-8"></head><body style="overflow-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;"><div class="moz-forward-container" style="font-family: CourierNewPSMT;">[Apologies for cross-posting]<br><p>***************************<br>Second Call for Participation<br>****************************<br><br><b>TASK</b>: Assessing DIScourse COherence in Italian TEXts (DisCoTEX) @ EVALITA 2023 <br><b>Info</b>: <a class="moz-txt-link-freetext" href="https://sites.google.com/view/discotex/home" moz-do-not-send="true">https://sites.google.com/view/discotex/home</a><br><b>Final Workshop</b>: September 7th-8th, 2023, Parma, Italy<br><br></p><p>We invite interested parties from academia and industry to participate in the <b>DisCoTEX Task</b>, which will be held in the context of <a href="https://www.evalita.it/campaigns/evalita-2023/" moz-do-not-send="true">Evalita 2023</a>. <br></p><p>DisCoTEX is the first shared task focused on modeling discourse coherence for Italian real-word texts. Coherence is a key property of any well-organized text and it plays a fundamental role in human discourse processing as well as in a number of NLP applications. Inspired by previous literature on coherence modeling, the DisCoTEX task will be articulated into two subtasks:</p>1. Last sentence classification: this is conceived as a binary classification task. Specifically, given a short textual passage and an individual sentence, participants will be asked to predict whether the sentence follows or not, thus joining it to the passage gives out a coherent or incoherent passage. <br>2. Human score prediction: this is conceived as a regression task in which participants will be asked to predict the average coherence score assigned by human raters to short passages (either in their original or modified version). <br><p>Participants are free to participate in either one of them or both. <br></p><p>More details about the task definition, source data and evaluation are available at the task <a href="https://sites.google.com/view/discotex/home" moz-do-not-send="true">website</a>.</p><br>Given the novelty of the task and the crucial role that coherence modeling plays in a variety of application scenarios, we expect to attract groups from communities working on distinct fields, such as automatic essay scoring, readability assessment and document summarization. Moreover, we hope that beyond the competition, the task would rise the interest of scholars working on theoretical models of coherence from a linguistic and a cognitive perspective, as well as those involved in the interpretability of current language models based on deep learning networks. <br><p>-----------------------</p>Important Dates<br>-----------------------<br><strike>7th February 2023: training data available to participants</strike><br>2nd – 9 May 2023: test data available (evaluation window starts)<br>30th May 2023: results notification to participants<br>14th June 2023: technical report from participants due to task organizers<br>28th June 2023: final reports from task organizers due to EVALITA chairs<br>10th July 2023: review deadline<br>25th July 2023: camera ready version deadline<br>7th-8th September 2023: EVALITA workshop in Parma<br><br>Updates will be made available at the Evalita 2023 website, check it often.<br><br>----------------<br>Organizers<br>----------------<br>Dominique Brunato*<br>Davide Colla**<br>Felice Dell'Orletta*<br>Irene Dini*<br>Daniele Paolo Radicioni**<br>Andrea Amelio Ravelli***<br><br>* ItaliaNLP Lab, Istituto di Linguistica Computazionale "Antonio Zampolli" (ILC-CNR), Pisa<br>** Dipartimento di Informatica, Università degli Studi di Torino<br>*** Università di Bologna, Bologna<br><br>----------------------------------------------------------------<br><p>Contacts:<b> <a class="moz-txt-link-abbreviated moz-txt-link-freetext" href="mailto:discotex.evalita2023@gmail.com" moz-do-not-send="true">discotex.evalita2023@gmail.com</a></b></p><p>----------------------------------------------------------------<b><br></b></p><p><u></u>Feel free to contact us for further information!</p>The DisCoTEX organizing committee </div><span style="font-family: CourierNewPSMT;"></span><br class="Apple-interchange-newline" style="caret-color: rgb(0, 0, 0); color: rgb(0, 0, 0);"><div><br></div><div><br></div><br class="Apple-interchange-newline" style="caret-color: rgb(0, 0, 0); color: rgb(0, 0, 0);"><div>
<br>: : : : : : : : : : : : : : : : : : :<br><br>Daniele Radicioni, PhD<br>Department of Computer Science<br>University of Turin<br>Corso Svizzera, 185<br>10149 - Torino<br>phone: +39 011 6706802<br>fax: +39 011 751603<br>http://www.di.unito.it/~radicion<br><br><br><br>
</div>
<br></body></html>