LINGUIST List 11.1431

Wed Jun 28 2000

Calls: SIGDIAL/Discourse/Dialogue, SIGDAT/NLP/Corpora

Editor for this issue: Jody Huellmantel <jodylinguistlist.org>


As a matter of policy, LINGUIST discourages the use of abbreviations or acronyms in conference announcements unless they are explained in the text.

Directory

  1. hasida, SIGDIAL Workshop on Discourse and Dialogue
  2. Priscilla Rasmussen, Empirical Methods in NLP & Very Large Corpora (ACL'2000 EMNLP/VLC)

Message 1: SIGDIAL Workshop on Discourse and Dialogue

Date: Wed, 28 Jun 2000 04:50:12 +0900
From: hasida <hasidacsl.sony.co.jp>
Subject: SIGDIAL Workshop on Discourse and Dialogue

			 CALL FOR PAPERS
		1ST SIGDIAL WORKSHOP ON DISCOURSE AND DIALOGUE
		Including Theme Session On Principles For
			Dialogue System Evaluation


			October 7,8, 2000
			 Hong Kong
 In conjunction with
 ACL-2000: The 38th Annual Meeting of the
 Association for Computational Linguistics

		http://www.pitt.edu/~dialcal/ACL2Ksymp.html

Description:

There has been a perceived need in the SigDIAL Community for a regular
workshop spanning the SigDIAL interest area of discourse and
dialogue. While there has been a lot of activity in this area, and
fairly frequent "specialty" workshops on various sub-topics, there has
not been a regular place for such research to be presented in a forum
to receive attention from the larger SigDIAL community. This workshop
is intended to be the first in a regular series. A general session,
open to the range of work in the area is to be combined with a
Specialty "Theme Session", in this case on Principles for Dialogue
System Evaluation.



TOPICS OF INTEREST FOR THE GENERAL SESSION

We welcome formal, corpus-based, implementational and analytical work
on discourse and dialogue, with a focus on the following three themes:

(i) Dialogue Systems
 Spoken, multi-modal, and text/web based dialogue systems
 including topics such as:
 - dialogue management models
 - task complexity and interfaces for less common and less
 expected tasks
 - repair, clarification and correction types
 - grounding and feedback strategies
 - user and user group modeling 
 - mixed initiative and user-adaptive dialogue
 - re-usable components for different systems
 - generic architectures and common toolkits for building
 dialogue systems
 - speech, text and graphics integration


(ii) Corpora and Corpus Tools
 Support for corpus-based work on discourse and dialogue, in
 particular
 - issues in discourse and dialogue annotation
 - tools and resources for discourse and dialogue studies
 - XML-based tools for dialogue access to internet information

(iii) Pragmatic and/or Semantic Modeling
 a. The pragmatics and/or semantics of discourse and dialogue
 (i.e., beyond a single sentence) including the following issues:
 - ellipsis and anaphora 
 - dependent reference
 - presupposition and accommodation
 - genres of discourse and dialogue
 - politeness

 b. Specific aspects of discourse and dialogue structure, in
 particular
 - dialogue acts 
 - rhetorical structure
 - prosody and discourse
 - temporal structure
 - topic-comment structure in discourse and dialogue
 - focus and the distribution of discourse referents
 - discourse structure and conversational implicatures


TOPICS OF INTEREST FOR THE THEME SESSION ON PRINCIPLES FOR EVALUATION
OF DIALOGUE SYSTEMS

As a special-theme session, we wish to discuss methods for evaluation
which promote fruitful research directions. Contributions in this
respect are solicited on topics including but not limited to:

- evaluation of task-oriented dialogue systems vs.
 self-oriented (amusement-oriented) dialogue systems
- how to evaluate the efficiency and/or comfort of dialogues
- objective, quantitative, synthetic evaluation vs.
 subjective, qualitative, analytic evaluation
- relation of evaluation of dialogue systems with evaluation of
 other parts of NLP
- common tools and infrastructures for evaluation
- how to assess/implement diversity of dialogues in evaluation

The last topic may need further description: A generally important
issue in evaluation of allegedly intelligent artifacts is how to
manipulate the diversity under the present state of the art. The
diversity of dialogues encompasses the vocabulary, syntactic
constructions, discourse structures, and so forth. A major source of
the diversity here is the gap between linguistic expressions and the
description of the world to talk about. Diverse dialogues in fact
arise in tasks, such as the Map Task, involving pattern recognition.
However, the visual pattern recognition in the Map Task makes it
utterly impossible to computationally implement with the current
technologies.



SUBMISSIONS 

To stimulate discussions, both the general and theme session will
feature both full paper presentations and short position/discussion
papers. Please indicate the submission format, as described below.

All papers should be sent electronically to dialcalpitt.edu with
subject line SigDIAL workshop submission. Papers should be received by
July 10th to insure full consideration.

FULL PAPER SUBMISSIONS are limited to original, unpublished work in
the areas of interest. Extended abstracts of papers may not exceed
3200 words (exclusive of title page, example pages, and
references). In addition to the regular text, 2 additional pages are
allowed (as an appendix) which may include examples of extended
discourse, graphical representation of discourse structure, or other
supporting material.

The style files for submission are the same as the ones
for ACL regular papers, which can be downloaded from
http://www.cs.ust.hk/acl2000/Latex/index.html (for latex)
http://www.cs.ust.hk/acl2000/work/ACL2000_submission.doc (for MS Word)


 The title page should include the following information:

		 Title: 
	 Authors' names, affiliations, and email addresses:
 Keywords: Up to 5 keywords specifying subject
			area (preferably from lists above)
 Which Session: General or Theme 
 Word Count, excluding title page and references: 
 Under Consideration for other Conferences (specify):
 Abstract: short summary (up to 5 lines) 



SHORT PAPER SUBMISSIONS

Short papers should be in the same format as long papers, but no more
than 2000 words (with similar final length). As well as original work
in progress, short papers may also involve positions on the topics
above, comparative analysis of other approaches, or new relevant
topics for discussion.



IMPORTANT DATES

Submissions		July 10th
Notification		August 10th
Final Submissions	September 1st
Workshop		October 7-8th


PROGRAM COMMITTEE: 
 Laila Dybkjaer (co-chair), Koiti Hasisa (co-chair), David Traum
(co-chair), Morena Danielli, Yasuharu Den, Barbara Di Eugenio,
Kristiina Jokinen, Pamela Jordan, Ian Lewin, Daniel Marcu, Katashi
Nagao, Akira Shimazu, Michael Strube, Jan van Kuppevelt, Marilyn
Walker (and others).

Contact Information:
Questions about submission:	 Pamela Jordan <dialcalpitt.edu>
Questions about General Session: Laila Dybkjaer <lailanis.sdu.dk>
Questions about Theme Session: Koiti Hasida <hasidaetl.go.jp>
Miscellaneous and Logistical Questions: David Traum <traumcs.umd.edu>
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue

Message 2: Empirical Methods in NLP & Very Large Corpora (ACL'2000 EMNLP/VLC)

Date: Tue, 27 Jun 2000 16:56:16 EDT
From: Priscilla Rasmussen <rasmussecs.rutgers.edu>
Subject: Empirical Methods in NLP & Very Large Corpora (ACL'2000 EMNLP/VLC)


 FINAL CALL FOR PAPERS

 (EMNLP/VLC-2000) JOINT SIGDAT CONFERENCE ON
 EMPIRICAL METHODS IN NATURAL LANGUAGE PROCESSING AND
 VERY LARGE CORPORA

 Sponsored by SIGDAT (ACL's Special Interest Group for Linguistic Data
 and Corpus-based Approaches to NLP)

 October 7-8, 2000
 Hong Kong University of Science and Technology

 In conjunction with
 ACL-2000: The 38th Annual Meeting of the
 Association for Computational Linguistics

 This conference aims to bring together academic researchers and
 industrial practitioners to discuss empirical and corpus-based natural
 language processing through technical paper sessions, invited talks,
 and panel discussions. Topics of interest include (but are not
 limited to):

 * statistical parsing
 * language and dialog models
 * machine translation
 * information retrieval
 * information extraction
 * comparative evaluation of empirical vs. rule- and knowledge-based
 technologies
 * lexical acquisition
 * statistical language understanding
 * phrase identification
 * noun phrase coreference
 * question answering
 * word sense disambiguation
 * word and term segmentation and extraction
 * alignment
 * bilingual lexicon extraction
 * text categorization

 This year, we are especially interested in papers discussing these
 topics in the context of web-oriented applications.

- -------------------------------------------------------------------------

 SUBMISSIONS

 Submissions are limited to original, unpublished and empirically
 evaluated work. Reviewing of papers will be blind. Electronic
 submissions are required; author instructions, stylesheets and
 a web-based submissions interface may be found at
 http://nlp.cs.jhu.edu/~sigdat2k . Submitted papers should conform to
 the colacl.sty style file (or the provided MSWord equivalent) for final
 2-column format, with the exception that name, affiliation and address
 should be replaced with 'XXX'. Full paper-length submissions are
 strongly encouraged, and should not exceed 9 pages in specified format.
 Submissions must be received on or before June 30, 2000.

- -------------------------------------------------------------------------

 Important Dates:
 June 30 Submission of full-length paper
 July 28 Acceptance notice
 August 29 Camera-ready paper due
 October 7-8 Conference date

- -------------------------------------------------------------------------

 Program Committee

 Chair: Hinrich Schuetze, GroupFire (hinrichgroupfire.com)
 Co-Chair: Keh-Yih Su, Behavior Design Corp. (kysubdc.com.tw)

 Einat Amitay, Macquarie Univ. & CSIRO
 Sophia Ananiadou, Univ. of Salford
 Susan Armstrong, Univ. of Geneva
 Thorsten Brants, Saarland Univ.
 Eric Brill, Microsoft Research
 Jason Chang, National Tsing Hua Univ.
 Rim-Hae Chang, Korea Univ.
 Key-Sun Choi, Korea Adv. Inst. of Science & Technology
 David Elworthy, Microsoft Research Cambridge
 Tomaz Erjavec, Institute Jozef Stefan
 Pascale Fung, Hong Kong Univ. of Science & Technology
 Eric Gaussier, Xerox Research Centre Europe
 Niyu Ge, Brown University
 Nancy Ide, Vassar College
 Martin Jansche, Ohio State Univ.
 Andy Kehler, UC San Diego
 Geunbae Lee, Pohang Univ. of Science & Technology
 Lillian Lee, Cornell Univ.
 Dekang Lin, Univ. of Alberta
 Kim-Teng Lua, National Univ. of Singapore
 Chris Manning, Stanford Univ.
 Yuji Matsumoto, Nara Inst. of Science & Technology
 Helen Meng, Chinese Univ. of Hong Kong
 Masaaki Nagata, NTT Cyber Space Labs
 Dragomir Radev, Univ. of Michigan
 Maosong Sun, Tsinghua Univ.
 Bing Swen, Peking Univ.
 Mark Wasson, Lexis-Nexis
 Yorick Wilks, University of Sheffield
 Dekai Wu, Hong Kong Univ. of Science & Technology
 Jakub Zavrel, University of Antwerp

 http://www-csli.stanford.edu/~schuetze/emnlp-vlc2000.html
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue