Return-Path: Received: from CS.CMU.EDU by A.GP.CS.CMU.EDU id aa08845; 30 Mar 94 16:20:37 EST Received: from Csli.Stanford.EDU by CS.CMU.EDU id aa20961; 30 Mar 94 16:19:47 EST Received: from localhost.Stanford.EDU by CSLI.Stanford.EDU (4.1/25-CSLI-eef) id AA11789; Wed, 30 Mar 94 10:23:07 PST Message-Id: <9403301823.AA11789@CSLI.Stanford.EDU> From: kwc@research.att.com (Ken Church) To: empiricists@CSLI.Stanford.EDU Subject: INTERNATIONAL WORKSHOP ON DIRECTIONS OF LEXICAL RESEARCH Date: Wed, 30 Mar 1994 10:23:05 -0800 Sender: roscheis@CSLI.Stanford.EDU From: kwc@research.att.com (Ken Church) (by way of yarowsky@unagi.cis.upenn.edu (David Yarowsky)) xxxxxxxxxxxxxxxxxxxxxPOST-COLING 94 WORKSHOP ANNOUNCEMENTxxxxxxxxxxxxxxxxxxxx INTERNATIONAL WORKSHOP ON DIRECTIONS OF LEXICAL RESEARCH 15-17th of August, 1994 in Beijing Co-Chairs Nicoletta Calzolari Istituto di Linguistica Computazionale del CNR Via della Faggiola 32, 56100 PISA, ITALIA Cheng-ming Guo China National Laboratory of AI Technology and Systems Tsinghua University BEIJING 100084 CHINA 1. Intentions The scholarly and scientific tradition in lexical research was given an engineering edge by three outbursts of massive lexical work that began in the last decade, the CYC project in the United States, the EDR project in Japan, and the Grosseto Workshop whose principle of reusability generated a series of EC projects including Acquilex, ET-7, Multilex, and Genelex. These projects led to, and were equalled by, world wide interest in statistics on very large corpora. The close inter-relationships between Lexicon development and Corpus analysis are increasingly accentuated. This international drive on the lexicon over a period of a decade has done the field an immense service. However, recent reflections on the lexical research over the last ten years are not always as favourable as one might expect. Wilks (1993) made a criticism of IBM's approach to Machine Translation. The main point was that there is a natural ceiling of success to PURE statistical methods. Ide and Veronis (1993) went so far as to ask if we have wasted our time over the last decade on extracting knowledge from machine-readable dictionaries. Another area of concern nowadays is the development of common lexical specifications, or lexical standards. Up to which level of lexical description can standards be proposed, to ensure data reusability? Furthermore, one important development in AI and Cognitive Science in recent years warrants the attention of lexical researchers. It involves the trend for the integration of NLP with various subareas of AI, e.g. computer vision (see Dennett, 1991; Mc Kevitt, 1994). The need for unified representation scheme incorporating both perceptual information and common sense knowledge poses new challenges. The proposed workshop attempts to clarify issues in current lexical research in terms of further research directions as an answer to recent challenges. Suggested topics are as follows: a) notes comparing between well-known lexical projects, particulary the EDR project, the CYC project, and the EC projects --what have we learned ? b) lexical needs for unified representations of common sense knowledge and perceptual knowledge, visual or audio --brainstorming on the design and construction of the lexicons for such integrated systems; c) lexical needs of very large knowledge bases for nuclear lexicons as the core for knowledge acquisition -- speculations and practice concerning the design and construction of such nuclei. 2. Format This workshop is intended as an opportunity provided for the exchange of views on issues of common concern to the area of lexical research. Panel sessions and discussions are stressed rather than formal speeches. All activities at the workshop are intended as reactions to recent challenges. The workshop will take place on the beautiful Tsinghua University campus on the outskirt of Beijing. Day 1: the state-of-the-art discussion -- achievements, issues and concerns Day 2: lexical needs for integrated systems Day 3: lexical needs for knowledge acquisition for very large knowledge systems Each attendee of the workshop pays $100 to cover registration, preprints,local transportation from and to Beijing Airport, and hotel + food expenses for the 3-day workshop duration. Air fare to and from Beijing rests with all attendees themselves. Registation forms will be sent out to workshop attendees upon request. All correspondence concerning workshop registration should be directed to Chengming Guo by fax, e-mail, or postal mail at the Computer Science Department, Tsinghua University, Beijing, 100084, China. To ensure proper accomodation, special pre-arrangements have to be made with Chengming Guo, should the attendee wish to arrive earlier than August 14th and/or stay a little longer after the workshop ends on the 17th of August. 3. Submission requirements Papers of no more than 6 pages should be submitted by e-mail to the co-chairs of the workshop at "glottolo@icnucevm.cnuce.cnr.it" for Nicoletta Calzolari, and at "chengming%bepc2@scs.bitnet" for Chengming Guo. Postal mail of three hard copies of the paper to Nicoletta Calzolari and Chengming Guo are also acceptable. Papers must be printed to 8 1/2 to 11" size. Workshop preprints will be made available to all attendees. Paper submission by the 15th of May, 1994 Notification of acceptance by 5th June, 1994 Camera-ready copy by 30th of June, 1994 4. Sponsors Istituto di Linguistica Computazionale del CNR, Pisa, ITALY China National Laboratory of AI Technology and Systems, Tsinghua University, Beijing China 5. Co-Chairs Nicoletta Calzolari Istituto di Linguistica Computazionale del CNR Via della Faggiola 32, 56100 PISA, ITALIA Phone: +39 50 56 04 81 Fax:+39 50 58 90 55 Email:glottolo@icnucevm.cnuce.cnr.it Cheng-ming Guo Computer Science Department Tsinghua University Beijing 100084 CHINA Phone: +86 1 2594895 Fax:+86 1 2562768 Email:chengming%bepc2@scs.bitnet 6. Program committee: Sue Atkins Oxford University Press, UK Nicoletta Calzolari Institute of Computational Linguistics, CNR, Italy Kenneth Church AT&T Bell Laboratory, USA Cheng-Ming Guo Tsinghua University, China Judith Klavans Columbia University, USA Paul Mc Kevitt University of Sheffield, UK Yoshihiko Nitta Hitachi Advanced Laboratory, Japan Yorick Wilks University of Sheffield, UK Toshio Yokoi EDR, Japan Antonio Zampolli Institute of Computational Linguistics, CNR, Italy Uri Zernik GE, USA