LINGUIST List 7.427

Fri Mar 22 1996

Calls: Machine Translation, Semantic Lexicons

Editor for this issue: T. Daniel Seely <>


  1. Eduard Hovy, AMTA-96 Machine Translation Conference
  2. Evelyne Viegas, ACL'96 Workshop on BREADTH & DEPTH of SEMANTIC LEXICONS

Message 1: AMTA-96 Machine Translation Conference

Date: Mon, 18 Mar 1996 15:00:01 EST
From: Eduard Hovy <hovyISI.EDU>
Subject: AMTA-96 Machine Translation Conference

 Association for Machine Translation in the Americas

 AMTA-96 Conference, Montreal, October 2-5, 1996

 Accelerating Machine Translation Development:
 Research, Business, and Personal Use

Following the very successful AMTA conference near Washington, DC in
October 1994, the second conference of the Association for Machine
Translation in the Americas will be held in Montreal, Canada, at the
Radisson Hotel, on October 2-5, 1996.

Once again there will be something for everyone! Retaining the pattern
established by its predecessor, AMTA-96 will offer a blend of invited
talks, panel discussions, research papers, system demonstrations and
descriptions, tutorials, workshops, book exhibits, and social events.
The three days of the conference will also facilitate gatherings of
the Special Interest Groups on topics ranging from interlinguas and
ontologies, lexicons, standards and data exchange, MT on PCs, and MT

The overall intent of the conference is to bring together MT developers,
researchers, and users, to share the latest information on MT and to
forge partnerships for addressing the challenge of language barriers
that impede communication on the Information Highway. Participation
by members of AMTA's sister organizations in Europe and Asia is
strongly encouraged.

Invited talks and panel discussions will highlight topical and contro-
versial questions, encouraging lively interactions, as they did at
AMTA-94. In the theory sessions, technical papers will address a
wide range of topics, while in practical sessions, the problems of
developing and bringing MT systems to market will be discussed, with
online demonstrations. In addition, booths can be rented to display
systems and products.

AMTA will also hold its General Members' Meeting during the conference.
Program committee:
 Susan Armstrong, ISSCO (Geneva) 
 Scott Bennett, Logos Inc. (NJ)
 Colin Brace, LIM (Amsterdam)
 Lynn Carlson, Department of Defense (Washington)
 Ken Church, AT&T (NJ)
 David Clements, Globalink (Washington)
 Bonnie Dorr, University of Maryland
 David Farwell, CRL, New Mexico State University
 Mary Flanagan, CompuServe (Boston)
 Laurie Gerber, SYSTRAN Inc. (San Diego)
 Eduard Hovy, USC/ISI (Los Angeles) (chair)
 Pierre Isabelle, CITI (Montreal)
 Hitoshi Isahara, MITI (Tokyo)
 Kevin Knight, USC/ISI (Los Angeles)
 Marjorie Leon, Pan American Health Org (Washington)
 Kazunori Muraki, NEC (Tokyo)
 Virginia Teller, CUNY (New York)
 Muriel Vasconcellos, AMTA (San Diego)
 John White, PRC (Washington)


Authors/system developers are invited to submit two kinds of presentations:
1. Theoretical papers: Unpublished papers are requested about original
 work on all aspects of Machine Translation. Papers should be in
 English, not longer than 10 pages, with minimum character font size
 of 11 pt.
2. System descriptions with optional system demonstrations: Approx. 30
 minutes will be allocated per system description/demo. Submissions
 should be in English, not longer than 4 pages. If a system demon-
 stration is included, please provide the following information:
 - hardware platform,
 - operating system,
 - name and contact information of system operations specialist.

First page: Both types of submission should include an additional first
page with the following information:
- paper title,
- author(s)' name(s), address(es), telephone and fax numbers, email
- one-paragraph abstract,
- for theoretical papers: subject area keyword(s)
 for system descriptions/demos: the words "System description/demo".

Submissions are due at either address below on April 15, 1996.

Softcopy submissions (papers that do not print will be returned to the
 email address:
 subject line: AMTA-96 submission
 paper encoding:
 - ASCII plain text
 - Microsoft Word (RTF format)
 - PostScript

Hardcopy submissions (please send four (4) copies):
 AMTA-96: Eduard Hovy
 USC Information Sciences Institute
 4676 Admiralty Way
 Marina del Rey, CA 90292-6695

Organization and further details.
Local arrangements chair:
 Elliott Macklovitch, CITI, Montreal (email:
Program chair:
 Eduard Hovy, USC/ISI, Marina del Rey (email:
- --------------------------------------------------------------------------
Eduard Hovy
email: USC Information Sciences Institute
tel: 310-822-1511 ext 731 4676 Admiralty Way
fax: 310-823-6714 Marina del Rey, CA 90292-6695
project homepage:
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue

Message 2: ACL'96 Workshop on BREADTH & DEPTH of SEMANTIC LEXICONS

Date: Mon, 18 Mar 1996 17:50:48 MST
From: Evelyne Viegas <>



June 28, 1996
Santa Cruz, California, USA.	

Building semantic lexicons is a very time consuming task. Efficient
large-scale acquisition and representation of lexical knowledge will
be greatly aided by capturing regularities in the lexicon.

Two main issues present themselves:

a) treatment of lexical ambiguity and
b) lexical rules as a conceptual tool for controlled proliferation of

Whereas the former has been regarded as a topical issue for quite
some time, the latter is only now receiving its due attention. This
workshop will concentrate on lexical rules as a regulator of breadth
and depth of the lexicons. Lexical rules are known under a variety of
names, e.g., Leech's (1991) "semantic transfer rules," "lexical
inference rules" of Ostler and Atkins (1991) and others. They are also
addressed in the framework of such theories as the generative lexicon
of Pustejovsky (1995). Such linguistic frameworks as LFG and HPSG have
also used the concept, albeit in a different sense and for a different
purpose. At the same time, theoretical accounts of the use of lexical
rules (such as, for instance, preemption or blocking) are rather too
general and underspecified to support actual processing. The workshop
will stress issues connected with the practical application of lexical
rules: when to apply the rules, how the rules influence system design,
how to reexamine and adjust the theoretically posited rules in view of
practical needs and evidence. Another central issue for the workshop
will be large-scale acquisition of computational-semantic lexicons.
We are mainly interested in examining the following trade-offs: the
coverage vs.the depth of existing semantic lexicons vs. the effort
involved in building them.

The workshop is intended for researchers in computational linguistics,
artificial intelligence, psycholinguistics or other fields who have
been working in lexical semantics and large-scale lexical knowledge

Some (though not necessarily all) specific questions suggested for
discussion include:

1) What are the different types of lexical rules which should be
considered in the building of computational lexicons (inflectional and
derivational morphology, verbal diatheses, regular word-sense shifts,

2) When should the rules be applied (run-time, load-time,
acquisition, other)

3) How to evaluate the cost-efficiency of the acquisition effort
against the utility of the resulting lexicons. How could we
characterize an NLP system along the dimensions of size, corpus
coverage, and depth.

4) Analyses of appropriate types of inheritance for different lexical

5) The use of lexical underspecification (and contextual word-use
grounding) as a partial alternative to lexical rules.

Computational and descriptive case studies are welcome. However,
submissions should centrally address one of the above issues rather
than simply describe a system or a theory. We greatly encourage the
submission of original and unplished work.

Presentations will last for 30 minutes, of which at least 5 minutes
(preferably, more) must be allocated for discussion. Papers will be
organized around themes. A summary general discussion will be
scheduled at the end of the day.
Attendees are required to register for the main ACL-96 conference.

Submissions must include a descriptive abstract of about 200 words and
should not exceed 3,000 words, excluding the references. Electronic
submissions are encouraged and should be submitted either directly or
by email as described below. The title page should include Title of
the paper, names, adresses, email, telephone and fax number of all
authors, and the abstract. Any correspondance will be adressed to the
first author. Questions concerning the workshop should be sent to

For electronic submissions, please name your files with the name of
the first author. For instance, Geraldine Lavaud, first author, will
place there the following files:		the .ps version of her paper
	lavaud.ascii		the .ascii version of her paper only
				 if postcript is not available		the .ascii file of the title page
				 (title, authors names, adresses, abstract)

Directions to submit directly:

	login: anonymous
	password: <your email address>
	cd lex-incoming
	binary (only if your paper is not in ascii format)
	put <names of your paper>

If submitted by e-mail, use:

Directions to submit by mail:

If hard-copy submission is inevitable, send 5 copies of the paper to:

Evelyne Viegas
Computing Research Laboratory
New Mexico State University
Las Cruces, NM 88003

tel: 505 646 5757
fax: 505 646 6218


In order to facilitate interaction and focus the discussion,
a pre-workshop mailing list will be established; please indicate
whether or not you would like to be included by sending e-mail to
Participants will also be able to look at other participants' papers a
month before the workshop via anonymous ftp to
The directions to look at other papers are:

login: anonymous
password: <your email address>
cd lex-rule
binary (only if the paper you want is not ascii)
get <name of paper>

submission: March 29th, 1996
notification: April 26th, 1996
final version due: May 22th, 1996

Asap:		Mosaic home page for the workshop set at

April 30:	Beginning of e-mail discussion

Evelyne Viegas (Chair)	New Mexico State University, CRL, USA
Sergei Nirenburg	New Mexico State University, CRL, USA
Boyan Onyshkevych	Carnegie Mellon University, USA
Nicholas Ostler Linguacubun Ltd, UK
Victor Raskin		Purdue University, USA
Antonio Sanfilippo	Sharp Laboratories of Europe, UK

Philip Resnik		Sun Microsystems Laboratories, USA
Frederique Segond	Rank Xerox Research Centre, France
Evelyne Tzoukermann	ATT Bell Laboratories, USA

Final texts will be published in the Workshop Proceedings. Depending
on the quality of papers, publication in book form will be pursued.
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue