<br><div class="gmail_quote">EMNLP-11 Workshop on Language Generation and Evaluation<br>
=======================================================<br>
<br>
ANNOUNCEMENT AND CALL FOR PAPERS<br>
<br>
Language Generation and Evaluation (UCNLG+Eval) is a post-conference<br>
workshop at EMNLP-2011, Edinburgh, on 31 July 2011.<br>
<br>
Workshop aims<br>
-------------<br>
There are many branches of NLP research that involve the generation<br>
of language (summarisation, MT, human-computer dialogue, application<br>
front-ends, data-to-text generation, document authoring, etc.).<br>
However, it is not always easy to identify common ground among<br>
the generation components of these application areas, which has<br>
sometimes made it difficult for generic research in 'Natural Language<br>
Generation' (NLG) to engage with them effectively. Increasingly<br>
common corpus-based approaches across these areas, and in particular<br>
in NLG itself, offer a new perspective on this situation and the<br>
opportunity to explore synergies and differences from the common<br>
grounding of corpus data.<br>
<br>
This workshop is the fourth in an occasional series seeking to<br>
provide a forum for discussing NLG and its links with these closely<br>
related fields from a corpus-oriented perspective. The workshops<br>
have the general aims<br>
<br>
* to provide a forum for reporting and discussing corpus-<br>
oriented methods for generating language;<br>
* to foster cross-fertilisation between NLG and other fields<br>
where language is automatically generated; and<br>
* to promote the sharing of data and methods for the purpose<br>
of system building and comparative evaluation in all language<br>
generation research.<br>
<br>
Each of these workshops has a special theme: at the first workshop<br>
(at Corpus Linguistics in 2005) it was use of corpora in NLG; at the<br>
second (UCNLG+MT at MT Summit XI in 2007) it was Language Generation<br>
and Machine Translation; at the third it was Language Generation and<br>
Summarisation (UCNLG+Sum at ACL-IJCNLP'09). The special theme of the<br>
2011 workshop is Language Generation and Evaluation, and the event will<br>
showcase recent developments in methods for evaluating computationally<br>
generated language across NLP, continue the discussion of future<br>
directions and host an invited talk on shared-task evaluation campaigns.<br>
<br>
Evaluation Special Theme<br>
------------------------<br>
The past five years have seen big changes in NLG evaluation. The field<br>
has moved from a situation where there were no comparative evaluation<br>
results for independently developed alternative approaches to the<br>
present increasingly rich diversity of data sets, methods and results<br>
for comparative evaluation (intrinsic and extrinsic, human-assessed and<br>
automatically computed). A distinctive and critical feature of these<br>
developments has been the community-led approach to the establishment<br>
of tasks, datasets and evaluation methods. The aim of the special<br>
evaluation theme at UCNLG+Eval is to provide a forum for reporting<br>
cutting-edge research on evaluation, taking stock of recent developments,<br>
discussing and comparing alternative approaches to evaluation and<br>
exploring possible directions for future development.<br>
<br>
Call for Papers<br>
---------------<br>
The UCNLG+Eval Workshop organisers invite submissions addressing the<br>
special theme of evaluating computationally generated text as well as<br>
submissions on all aspects of using corpora in the generation of<br>
language. Specific topics include, but are not limited to:<br>
<br>
* Statistical and machine learning approaches to language generation<br>
* Development and annotation of corpora for language generation research<br>
* Reuse of corpus resources developed for NLU (e.g. treebanks) in<br>
language generation research<br>
* Domain-specific vs. general-purpose corpora for language generation<br>
research<br>
* Evaluation of automatically generated language<br>
* Meta-evaluation of evaluation methods for language generation<br>
* Uses of corpora in the evaluation of automatically generated language<br>
* Proposals for new shared tasks in language generation<br>
<br>
Note that by 'language generation research' we mean any field in which<br>
language is automatically generated including research commonly coming under<br>
the headings of NLG, MT, document summarisation and human-computer dialogue.<br>
<br>
Papers should describe original and unpublished work, emphasizing actual<br>
rather than intended work, and should indicate clearly the state of<br>
completion of the reported work. Wherever appropriate, concrete<br>
evaluation results should be included. Papers that are being submitted<br>
to other conferences or workshops should indicate this.<br>
<br>
Submission information<br>
----------------------<br>
Submissions should follow the two-column format of EMNLP and ACL 2011<br>
proceedings. Each paper may consist of up to nine (9) pages of content,<br>
and any number of additional pages with references only. Submissions<br>
should describe original, unpublished work. Please use the official<br>
ACL 2011 style files (at <a href="http://www.acl2011.org/call.shtml" target="_blank">http://www.acl2011.org/call.shtml</a>). We reserve<br>
the right to reject submissions that do not conform to these styles,<br>
including letter size and font size restrictions. Submission will be<br>
electronic, using the START system, via a link that will soon be<br>
available on the workshop website. The deadline for submission is<br>
22 April 2011.<br>
<br>
Each paper will be reviewed by at least three programme committee<br>
members. Final decisions on the technical programme will be made by<br>
the workshop organisers.<br>
<br>
As reviewing will be blind, papers should not include the authors'<br>
names and affiliations. Self-references that reveal the authors'<br>
identity, e.g., "We previously showed (Smith, 1991) ...", should be<br>
avoided. Instead, use citations such as "Smith previously showed<br>
(Smith, 1991) ...". Acknowledgments sections should be removed before<br>
submission.<br>
<br>
Proceedings<br>
-----------<br>
The proceedings of the workshop will be edited by the workshop<br>
organisers and published by the EMNLP 2011 conference organisers.<br>
<br>
Important Dates<br>
---------------<br>
22 April 2011 Deadline for papers submissions<br>
20 May 2011 Notification of acceptance to authors of workshop papers<br>
03 June 2011 Camera-ready copies due<br>
31 July 2011 UCNLG+EVAL workshop in Edinburgh<br>
<br>
<br>
<br>
Workshop organisers<br>
-------------------<br>
Anja Belz, NLTG, University of Brighton, UK<br>
Roger Evans, NLTG, University of Brighton, UK<br>
Albert Gatt, University of Malta, Malta<br>
Kristina Striegnitz, Union College, USA<br>
<br>
Programme committee<br>
-------------------<br>
Aoife Cahill, Stuttgart University, Germany<br>
Charlie Greenbacker, University of Delaware, USA<br>
Emiel Krahmer, Tilburg University, NL<br>
Mirella Lapata, University of Edinburgh, UK<br>
Oliver Lemon, Heriot-Watt University, Edinburgh, UK<br>
Daniel Marcu, ISI, University of Southern California, USA<br>
Kathy McKeown, Columbia, USA<br>
Karolina Owczarzak, NIST, USA<br>
Ehud Reiter, Aberdeen, UK<br>
<br>
Workshop website<br>
----------------<br>
<a href="http://www.nltg.brighton.ac.uk/ucnlg/ucnlg11" target="_blank">http://www.nltg.brighton.ac.uk/ucnlg/ucnlg11</a><br>
<br>
Contact email<br>
-------------<br>
<a href="mailto:ucnlg@itri.brighton.ac.uk" target="_blank">ucnlg@itri.brighton.ac.uk</a><br clear="all"></div><br><br>