Date: 04 Sep 1994 ----------------- ====================================================================== TOPICS: 1. Call for Participation: AAAI95 Spring Symposium Empirical Methods in Discourse Interpretation and Generation Deadline: October 28, 1994 (M. Walker) 2. Job offer: Department of Linguistics, University of Melbourne Lecturer in Computational Linguistics (Limited Tenure) (R. Dale) 3. Question on Aggregation in Text Planning (H. Dalianis) 4. Call for Papers: Special Issue of the Artificial Intelligence Journal Devoted to Empirical Artificial Intelligence. (D. Hart) Deadline: Jan. 10, 1995. 5. Call for Papers: Fifth International Workshop on Natural Language Understanding and Logic Programming (NLULP5) (They also expect Generation papers despite the name...) Deadline: 19 December 1994. (gpl@fct.unl.pt) ====================================================================== [Administrative Notes: Hello again after a long summer interruption. The SIGGEN list is opened to questions, opinions, announcements of publications by the members, availability of tech-reports, research lab descriptions etc. If you have material you want to make available to the other members of the list, we maintain an ftp archive for that purpose: black.bgu.ac.il:pub/siggen. We also keep there an archive of the mailing list. Post your contributions to siggen@black.bgu.ac.il We are in the process of gathering a list of "keywords" in the field, that could serve as a classification tool. It would help if every member could send a list of keywords that characterize his/her own work. You can post to the same address: siggen@black.bgu.ac.il The list has grown to 220 members during the summer. ] ====================================================================== From: walker Call for Participation EMPIRICAL METHODS IN DISCOURSE INTERPRETATION AND GENERATION AAAI 1995 Spring Symposium Series March 27 - 29, 1995 Stanford University, California Computational theories of communicative action (discourse interpretation and generation) provide the basis for the design of many types of AI systems (e.g., intelligent tutors, believable agents, intelligent spoken language systems, intelligent software agents (softbots), etc.) Developing the robust, broad coverage, theories of discourse that are needed in today's systems requires an empirical basis. However, there are no shared methods, tools or resources for the discourse community. To date, much work in computational theories of discourse has focused on specifying the mechanisms underlying a particular discourse phenomenon. It is often difficult to tell how prevalent that phenomenon is, whether it is related to other observed and studied phenomena, and what percentage of the cases a particular theory covers. This symposium has two foci: (1) an investigation of the empirical methods that can be used in the development and evaluation of computational theories of discourse, (2) the development of a set of shared resources for the computational discourse community. We solicit papers that focus on empirical methods and their benefits, e.g., by presenting an empirical method and a result derived using that method, including but not limited to the following topics: o Corpus-Based methods as applied to theories of discourse. o Methods for evaluating dialogue or discourse modules in implemented systems. o Simulation tools or testbeds used in developing and evaluating theories of discourse. o Coding schemes developed and tested for the quantitative study of some discourse phenomenon, particularly papers that show that multiple judges can use the coding scheme with replicable results. o Tools that support (semi-)automatic or empirical studies of discourse phenomena. o Papers that apply or extend methods used in traditionally empirical disciplines (e.g., psychology or sociolinguistics) to computational theories of discourse. o Empirical analyses using any method that distinguishes between claims made by different computational discourse theories. Interested participants should submit 6 copies of either (1) a paper (in 12 pt font, not to exceed 3000 words) addressing one of the above points, or (2) a statement of interest briefly describing the author's relevant work in this area and listing recent relevant publications. Send contributions--plain ascii or postscript--by October 28, 1994 to walker@merl.com If electronic submission is impossible, please send six copies to Dr. Marilyn Walker, Mitsubishi Electric Research Laboratories, 201 Broadway, Cambridge, MA 02139 (phone: 617-621-7505; fax:617-621-7550). Notification of acceptance will be given by November 30, 1994. Material to be included in the working notes of the symposium must be received by January 20, 1995. Organizing Committee: Marilyn Walker (co-chair), Mitsubishi Electric Research Labs, walker@merl.com. Johanna Moore (co-chair), University of Pittsburgh, jmoore@cs.pitt.edu; Marti Hearst, Xerox Palo Alto Research Center; Lynette Hirschman, MITRE; Aravind Joshi, University of Pennsylvania; ====================================================================== From: Robert Dale Department of Linguistics, University of Melbourne Lecturer in Computational Linguistics (Limited Tenure) The Department of Linguistics is a dynamic new Department with growing strengths in discourse, semantics, cognitive science and speech synthesis. We wish to develop new teaching and research programs in computational linguistics at undergraduate and postgraduate levels, and are seeking a lecturer in Computational Linguistics, for a three year appointment beginning in 1995. The position is funded by University Development funds and the Microsoft Institute of Advanced Software Technology. Subject to the success of the program in attracting students and establishing research in the area, it is hoped to advertise a continuing position in computational linguistics at the end of the three-year term. You should have expertise in Computational Linguistics, a good background in general linguistics, and additional specializations in an area such as Machine Translation, Text Generation, Natural Language Processing, Computational Phonology or Computational Morphology, Linguistic Knowledge Representation. You will contribute to the development of teaching and research in Computational Linguistics through the teaching of undergraduate and postgraduate courses in the Linguistics and Cognitive Science programs, as well as some teaching in general linguistics; through supervision of postgraduate research in Linguistics and in relevant parts of the Cognitive Science program. You will pursue research in one or more of the above areas, and help coordinate the Linguistics contribution to the Master's in Cognitive Science. Salary: $41,574 - $49,370 p.a. according to experience and qualifications. Further information and a position description: Ms Indra Kurzeme, Administrative Assistant, Department of Linguistics (03) 344 5488; fax (03) 344 4980; email Indra_Kurzeme@muwayf.unimelb.edu.au Applications close: 3 October 1994 Reference number: G0001481 Applications should be sent, in duplicate, quoting reference number and three referees (including facsimile numbers) to the Director, Personnel Services, The University of Melbourne, Parkville, Victoria, 3052; fax (03) 344 4694. The University of Melbourne is an equal opportunity employer and has a smoke free work-place policy. ===================================================================== From: Hercules Dalianis Dear Collegues ! My name is Hercules Dalianis and I am a PhD student at the Department of Computer and Systems Science, Royal Institute of Technology Stockholm, Sweden. I am working on Aggregation in Natural Language Generation. Aggregation, a process carried out after the deep generation or text planning but before the surface generation stage, is the removal of redundant portions of the semantic input in order to make the text more compact. For example, aggregation would compress "Mr. Smith drove to Dublin. Mrs. Smith drove to Dublin. Mr. Brown drove to London. Mr. Jones sailed to Amsterdam. Mr. Black sailed to Amsterdam." into "Mr. and Mrs. Smith drive to Dublin. Mr. Brown drove to London. Mr. Jones and Mr. Black sailed to Amsterdam." While this example may seem contrived, the need for aggregation occurs frequently in database output and other examples of computer-generated text. Together with Dr. Eduard Hovy, Information Sciences Institute/University of Southern California (ISI/USC), I have found 8 aggregation rules in an earlier study. I am now interested in subjecting these rules to a wider test than previously. Therefore,I would very much appreciate it if you could to send me any text plans together with corresponding texts, if you have any available, so I can apply my aggregation rules and perhaps improve the texts somewhat. Your assistance in my work will be invaluable; it is hard for me to find a large and varied enough set of text plans alone. I would be very glasd if you could send me the text plans and corresponding texts by e-mail or paper mail. Naturally, unless you give me permission, all text plans will be held strictly confidential, and I will inform you of all results before I publish any papers. Yours Sincerly Hercules Dalianis Dep. of Computer and Systems Sciences Royal Institute of Technology Ringv{gen 39c 118 63 Stockholm Sweden ph. 08-668 90 98 mob. ph. 010-668 13 59 ====================================================================== Reply-to: dhart@cs.umass.edu Call for Papers Special Issue of the Artificial Intelligence Journal Devoted to Empirical Artificial Intelligence Editors: Paul Cohen (cohen@cs.umass.edu) and Bruce Porter (porter@cs.utexas.edu). We are looking for papers that characterize and explain the behaviors of systems in task environments. Papers should report results of studies of AI systems, or new techniques for studying systems. The studies should be empirical, by which we mean "based on observation" (not exclusively "experimental," and certainly not exclusively statistical hypothesis testing). Examples (some of which are already in the AI literature) include: A report of performance comparisons of message-understanding systems, explaining why some systems perform better than others in some task environments A study of commonly-used benchmarks or test sets, explaining why a simple algorithm performs well on many of them A study of the empirical time and space complexity of an important algorithm or sample of algorithms Results of corpus-based machine-translation projects A paper that introduces a feature of a task that suggests why some task instances are easy and others difficult, and tests this claim Theoretical explanations (with appropriate empirical backing) of unexpected empirical results, such as constant-time performance on the million-queens problem A statistical procedure for comparing performance profiles such as learning curves A resampling method for confidence intervals for statistics computed from censored data (e.g., due to cutoffs on run times) A paper that postulates (on empirical or theoretical grounds) an equivalence class of systems that appeared superficially different, providing empirical evidence that, on some important measures, members of the class are more similar to each other than they are to nonmembers. The empirical orientation will not preclude theoretical articles; it is often difficult to explain and generalize results without a theoretical framework. However, the overriding criterion for papers will be whether they attempt to characterize, compare, predict, explain and generalize what we observe when we run AI systems. This is an atypical special issue because many of us think there is nothing special about empirical AI. It isn't a subfield or a particular topic, but rather a methodology that applies to many subfields and topics. We are concerned, however, that despite the scope of empirical AI, it might be underrepresented in the pages of the Artificial Intelligence Journal. This special issue is an experiment to find out: if the number of submitted, publishable papers is high, then we may conclude that the Journal could publish a higher proportion of such papers in the future, and this issue might be inaugural rather than special. Three principles will guide reviewers: Papers should be interesting, they should be convincing, and in most cases they should pose a question or make a claim. A paper might be unassailable from a methodological standpoint, but if it is an unmotivated empirical exercise (e.g., "I wonder, for no particular reason, which of these two algorithms is faster"), it won't be accepted. In the other corner, we can envision fascinating papers devoid of convincing evidence. Different interpretations of "convincing" are appropriate at different stages of projects and for different kinds of projects; for example, the standards for hypothesis testing are stricter than those for exploratory studies, and the standards for new empirical methods are of a different kind, pertaining to power and validity. If, however, the focus of a paper is a claim, then convincing evidence must be provided. Deadline: Jan. 10, 1995. Please contact either of the editors as soon as possible to tell us whether you intend to submit a paper, and include a few lines describing the paper, so we can gauge the level of interest and the sorts of work we'll be receiving. Request: Due to the broad nature of this call, it will be difficult to reach all potential contributors. So, please tell a friend... The Editorial Board for this issue includes: B. Chandrasekaran, Eugene Charniak, Mark Drummond, John Fox, Steve Hanks, Lynette Hirschman, Adele Howe, Rob Holte, Steve Minton, Jack Mostow, Martha Pollack, Ross Quinlan, David Waltz, Charles Weems. Dave Hart UMass, Amherst dhart@cs.umass.edu ====================================================================== From: gpl@fct.unl.pt Fifth International Workshop on Natural language Understanding and Logic Programming NLULP5 CALL FOR PAPERS Conference dates: May 29 (Mon) --- 31 (Wed), 1995 Conference place: Fundacao Calouste Gulbenkian, Lisbon, Portugal Chairman: Gabriel Pereira Lopes Department of Computer Science Faculty of Science and Technology Universidade Nova de Lisboa, Quinta da Torre, 2825 Monte da Caparica, Portugal Phon. +351-1-295 3220 Fax. +351-1-295 56 41 Email. gpl@fct.unl.pt Program Committee: Hozumi Tanaka (Tokyo) Yuji Matsumoto (Nara) Harvey Abramson (Aizu) Charles Grant Brown (Stockholm) Veronica Dahl (Vancouver) Sandiway Fong (NEC, Princeton) Mark Johnson (Providence, RI) Martin Kay (Xerox, Palo Alto) Gregers Koch (Copenhagen) Michael C. McCord (IBM Yorktown) Fred Popowich (Vancouver) Patrick Saint-Dizier (Toulouse) Luis Moniz Pereira (Lisbon) Helder Coelho (Lisbon) Ryoichi Sugimura(Matsushita Electric Ind.,Osaka) Gabriel P. Lopes (Lisbon) (chair) The Organizing Committee of the Fifth International Workshop on Natural Language Processing and Logic Programming invites the submission of papers for NLULP5, in Lisbon, Portugal. TOPICS OF INTEREST: Papers are invited on substantial, original, and unpublished research on all aspects of Logic Programming applications to computational linguistics, including, but not limited to, the followings. - syntax - parsing - semantics - generation - phonetics - language understanding - phonology - speech analysis/synthesis - morphology - computational lexicons - discourse - electronic dictionaries - pragmatics - terminology - quantitative/qualitative linguistics - text database and retrieval - mathematical linguistics - documentation - contrastive linguistics - machine translation - cognitive linguistics - machine aids for translation - large text corpora - natural language interface - text processing - dialogue systems - hardware/software for NLP - multimedia systems It will be nice to have in Lisbon Practical Applications of PROLOG in the area of NLU, for demonstrating Prolog's attractiveness for this area. REQUIREMENTS FOR SUBMISSION: Papers should have a maximum fifteen pages in final format, should be written in English, and describe original work. They should emphasize completed work rather than intended work, and they should indicate clearly the state of completion of the reported results. FORMAT FOR SUBMISSION: Authors should submit four copies of preliminary versions of their papers with the page limits above, on A4 paper with the title, author(s), addresses (including email if possible), affiliation across the page top, a short (five to ten line) summary, and a specification of the topic area preferably drawn from the list above. As well, authors are strongly urged to email the title page information by the deadline date. Send the papers and emails to the chair. IMPORTANT DATES: Preliminary paper submission due: December 19, 1994 Notification of paper arrival: December 31, 1994 Inquiries for lost papers: January 20, 1995 Acceptance notification: March 1, 1995 Camera-ready copies due: April 1, 1995 REVIEW SCHEDULE: Preliminary papers are due by 19 December 1994. Papers received after that date will be returned unopened. Notification of receipt will be mailed to the first author (or designated author) soon after receipt by 31 December 1994. All inquiries regarding lost papers must be made by January 20, 1995. Designated authors will be notified of acceptance by March 1, 1995. Camera-ready copies of final papers prepared according to the publisher requirements, must be received by 1 April 1995 by the chair along with a signed copyright release statement. Papers received after that date may not be included in the proceedings. OTHER ACTIVITIES: (1) Invited talks and panels will be included in the program. Proposals and suggestions for invited talks and panels should be sent to the chair as soon as possible. (2) Anyone wishing to arrange an exhibit or present a demonstration should send a brief description, together with a specification of physical requirements (space, power, telephone connections, tables, etc.) to the chair. (3) An attractive social program will be organized for people willing to know more about Portugal, Lisbon and its surroundings. (4) At that time of the year the weather is quite worm and unpolluted beaches are nearby Lisbon, at Costa da Caparica.