Albert Gatt albert.gatt at um.edu.mt
Tue Apr 19 15:16:04 CEST 2011

EMNLP-11 Workshop on Language Generation and Evaluation


Language Generation and Evaluation (UCNLG+Eval) is a post-conference
workshop at EMNLP-2011, Edinburgh, on 31 July 2011.

Workshop aims
There are many branches of NLP research that involve the generation of
language (summarisation, MT, human-computer dialogue, application
front-ends, data-to-text generation, document  authoring, etc.). However, it
is not always easy to identify common ground among the generation components
of these application areas, which has sometimes made it difficult for
generic research in 'Natural Language Generation' (NLG) to engage with them
effectively. Increasingly common corpus-based approaches across these areas,
and in particular in NLG itself, offer a new perspective on this situation
and the opportunity to explore synergies and differences from the common
grounding of corpus data.

This workshop is the fourth in an occasional series seeking to provide a
forum for discussing NLG and its links with these closely related fields
from a corpus-oriented perspective. The workshops
have the general aims

* to provide a forum for reporting and discussing corpus-oriented methods
for generating language;
* to foster cross-fertilisation between NLG and other fields where language
is automatically generated; and
* to promote the sharing of data and methods for the purpose of system
building and comparative evaluation in all language generation research.

Each of these workshops has a special theme: at the first workshop (at
Corpus Linguistics in 2005) it was use of corpora in NLG; at the second
(UCNLG+MT at MT Summit XI in 2007) it was Language Generation and Machine
Translation; at the third it was Language Generation and Summarisation
(UCNLG+Sum at ACL-IJCNLP'09). The special theme of the 2011 workshop is
Language Generation and Evaluation, and the event will showcase recent
developments in methods for evaluating computationally generated language
across NLP, continue the discussion of future directions and host an invited
talk on shared-task evaluation campaigns.

Evaluation Special Theme
The past five years have seen big changes in NLG evaluation. The field has
moved from a situation where there were no comparative evaluation results
for independently developed alternative approaches to the present
increasingly rich diversity of data sets, methods and results for
comparative evaluation (intrinsic and extrinsic, human-assessed and
automatically computed). A distinctive and critical feature of these
developments has been the community-led approach to the establishment of
tasks, datasets and evaluation methods. The aim of the special evaluation
theme at UCNLG+Eval is to provide a forum for reporting cutting-edge
research on evaluation, taking stock of recent developments, discussing and
comparing alternative approaches to evaluation and exploring possible
directions for future development.

Call for Papers
The UCNLG+Eval Workshop organisers invite submissions addressing the special
theme of evaluating computationally generated text as well as submissions on
all aspects of using corpora in the generation of language. Specific topics
include, but are not limited to:

    * Statistical and machine learning approaches to language generation
    * Development and annotation of corpora for language generation research
    * Reuse of corpus resources developed for NLU (e.g. treebanks) in
language generation research
    * Domain-specific vs. general-purpose corpora for language generation
    * Evaluation of automatically generated language
    * Meta-evaluation of evaluation methods for language generation
    * Uses of corpora in the evaluation of automatically generated language
    * Proposals for new shared tasks in language generation

Note that by 'language generation research' we mean any field in which
language is automatically generated including research commonly coming under
the headings of NLG, MT, document summarisation and human-computer dialogue.

Papers should describe original and unpublished work, emphasizing actual
rather than intended work, and should indicate clearly the state of
completion of the reported work.  Wherever  appropriate, concrete evaluation
results should be included.  Papers that are being submitted to other
conferences or workshops should indicate this.

Submission information
Submissions should follow the two-column format of the ACL 2011 proceedings.
Each paper may consist of up to nine (9) pages of content, and any number of
additional pages with references only. Only submissions in pdf format will
be accepted. Submissions should describe original, unpublished work. Please
use the official ACL 2011 style files (at http://www.acl2011.org/call.shtml).
We reserve the right to reject submissions that do not conform to these
styles, including letter size and font size restrictions.  Submission is
electronic, using the START system, via a link on the workshop website.

The deadline for submission is 22 April 2011.

Each paper will be reviewed by at least three programme committee members.
Final decisions on the technical programme will be made by the workshop

As reviewing will be blind, papers should not include the authors' names and
affiliations.  Self-references that reveal the authors' identity, e.g., "We
previously showed (Smith, 1991) ...", should be
avoided. Instead, use citations such as "Smith previously showed (Smith,
1991) ...".  Acknowledgments sections should be removed before submission.

Papers will be reviewed according to the following criteria:
- Clarity
- Originality/Innovativeness
- Soundness/Correctness
- Meaningful comparison
- Substance
- Replicability
- Sharing of resources

A link to the evaluation form used by reviewers is available from the
workshop website.

The proceedings of the workshop will be edited by the workshop organisers
and published by the EMNLP 2011 conference organisers.

Important Dates
22 April 2011  Deadline for papers submissions
20 May 2011    Notification of acceptance to authors of workshop papers
03 June 2011   Camera-ready copies due
31 July 2011   UCNLG+EVAL workshop in Edinburgh

Workshop organisers
Anja Belz, NLTG, University of Brighton, UK
Roger Evans, NLTG, University of Brighton, UK
Albert Gatt, University of Malta, Malta
Kristina Striegnitz, Union College, USA

Programme committee
Aoife Cahill, Stuttgart University, Germany
Charlie Greenbacker, University of Delaware, USA
Emiel Krahmer, Tilburg University, NL
Mirella Lapata, University of Edinburgh, UK
Oliver Lemon, Heriot-Watt University, Edinburgh, UK
Daniel Marcu, ISI, University of Southern California, USA
Kathy McKeown, Columbia, USA
Karolina Owczarzak, NIST, USA
Ehud Reiter, Aberdeen, UK

Workshop website

Contact email
ucnlg at itri.brighton.ac.uk
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.elsnet.org/pipermail/elsnet-list/attachments/20110419/799d58b6/attachment-0001.htm 

More information about the Elsnet-list mailing list