2010Q1 Reports: ACL 2010

From Adminwiki
Jump to: navigation, search

Each area of activity should be covered, either by the first three chairs directly, or in reports submitted by others, cf 2009Q1 Reports: ACL/IJCNLP-09. The Conference planning schedule includes checklists of tasks, and chairs are invited to correct the schedule based on current practice.


Contents

General Chair (Jan Hajič)

In summary, there have been and are no major problems encountered. Committees have been all appointed before September 2009, with the exception of the second SRW chair (Markéta Lopatková, Charles University, Prague), who has been added recently to coordinate the student travel awards.

Workshops (David Weir and Pushpak Bhattacharyya)

This year, proposals for workshops were jointly solicited for ACL, COLING, and NAACL/HLT. Review was conducted jointly by the workshop chairs of the respective conferences. In total, 55 proposals were received. Thirteen were selected for inclusion in the official ACL 2010 Workshop Program listed below. All workshops have put up their websites, committees and calls for papers and have their START accounts set up. The workshops have been fitted in the overall conference schedule (see this Excel spreadsheet or this internet browser (.htm) version).


WS1: SemEval-2010: 5th International Workshop on Semantic Evaluations

  • Date: July 15-16
  • Chairs: Katrin Erk, Carlo Strapparava


WS2: Fifth Workshop on Statistical Machine Translation and Metrics MATR

  • Date: July 15-16
  • Chairs: Chris Callison-Burch, Philipp Koehn, Christof Monz, Kay Peterson


WS3: The 4th Linguistic Annotation Workshop (The LAW IV)

  • Date: July 15-16
  • Chairs: Nianwen Xue, Massimo Poesio


WS4: BioNLP2010

  • Date: July 15
  • Chairs: K. Bretonnel Cohen, Dina Demner-Fushman, Sophia Ananiadou, John Pestian, Jun'ichi Tsujii, Bonnie Webber


WS5: Cognitive Modeling and Computational Linguistics

  • Date: July 15
  • Chairs: John Hale


WS6: NLP and Linguistics: Finding the Common Ground

  • Date: July 16
  • Chairs: Lori Levin, William Lewis, Fei Xia


WS7: 11th Meeting of ACL-SIGMORPHON

  • Date: July 15
  • Chairs: Jeffrey Heinz, Lynne Cahill and Richard Wicentowski


WS8: TextGraphs-5: Graph-based Methods for Natural Language Processing

  • Date: July 16
  • Chairs: Carmen Banea, Alessandro Moschitti, Swapna Somasundaran, Fabio Massimo Zanzotto


WS9: Named Entities Workshop (NEWS 2010)

  • Date: July 16
  • Chairs: A Kumaran and Haizhou Li


WS10: Applications of Tree Automata in Natural Language Processing

  • Date: July 16
  • Chairs: Frank Drewes, Marco Kuhlmann


WS11: Domain Adaptation for Natural Language Processing (DANLP)

  • Date: July 15
  • Chairs: Hal Daumé III, Tejaswini Deoskar, David McClosky, Barbara Plank, Jörg Tiedemann


WS12: Companionable Dialogue Systems

  • Date: July 15
  • Chairs: Yorick Wilks, Morena Danieli, Björn Gambäck


WS13: GEMS-2010 Geometric Models of Natural Language Semantics

  • Date: July 16
  • Chairs: Roberto Basili and Marco Pennacchiotti

Tutorials (Lluís Màrquez and Haifeng Wang)

We received 20 proposals before the deadline of January 8, 2010. Email was used for handling submissions with email acknowledgement of receipt of proposals as stated in the CFP. Following advice from previous tutorial chairs, we solicited 5 proposals. We were very careful to encourage these proposals but at the same time warn that there was no guarantee of acceptance because all proposals would be subject to the same reviewing process. 4 out of the 5 solicited proposals were accepted. All 20 proposals were reviewed by both co-chairs and a few external reviewers to avoid some conflicts of interest.

We selected 6 proposals based on the following criteria: novelty, scientific interest, presenter, proposal document and target audience.

The selected tutorials are:

T1: Annotation Presenter: Eduard Hovy

T2: From Structured Prediction to Inverse Reinforcement Learning Presenter: Hal Daumé III

T3: Wide-Coverage NLP with Linguistically Expressive Grammars Presenters: Josef van Genabith, Julia Hockenmaier and Yusuke Miyao

T4: Semantic Parsing: The Task, the State of the Art and the Future Presenters: Rohit J. Kate and Yuk Wah Wong

T5: Tree-based and Forest-based Translation Presenters: Yang Liu and Liang Huang

T6: Discourse Structure: Theory, Practice and Use Presenters: Bonnie Webber, Markus Egg and Valia Kordoni

All presenters are confirmed. They will provide descriptions of the tutorials (for publicity and tutorial proceedings) by April 2 and the complete tutorial materials by May 21, 2010.

Student Research Workshop (Seniz Demir, Jan Raab, Nils Reiter; Tomek Strzalkowski and Markéta Lopatková)

Program Committee Formation: The program committee list was approved by ACL board on January 8, 2010. The program committee consists of 54 members, of which 21 are from North America, 24 are from Europe, 2 are from Middle East, and 7 are from Asia/Pacific. They represent 11 broad areas of NLP and computational linguistics research.

Conference Setup and Procedures: The first call for papers was issued on October 8, 2009. The second call for papers was issued on January 11, 2010. The START account has also been readied for the submission process.

Reviewing Procedure: 37 valid submissions were received. Each submission was assigned to three reviewers. We started the reviewing process on March 6, 2010 and the reviews are due April 6, 2010.

Funding for Participants: We have been awarded travel funding for participants from National Science Foundation (NSF), ACL Walker Student Fund, and The European Chapter of the ACL (EACL). These funds are designated for all participants (from US, Europe, and Asia/Pacific). These will be coordinated by Markéta Lopatková, and a web page will be up soon with all the relevant information.

Mentoring Service (Björn Gambäck and Diana McCarthy)

This year, the mentoring service was extended to include student papers, system demonstrations, and all workshops associated with the conference as well as main conference papers. The mentoring deadline for each event was 6 weeks before each actual paper deadline and mentoring reviews were returned at least 2 weeks before that paper deadline. We recruited 21 mentors and each were asked to do 1 main conference or system demo paper and 1 other event at most. We only had 11 main conference papers and 1 CoNLL paper as mentoring submissions so the mentors were needed only once or not at all.

Publications (Jing-Shin Chang and Philipp Koehn)

The publication committee had submitted formatting instructions, author instructions and style files to the program committee and web master near the end of the last November before the first official Call for Paper.

The committe had also co-worked with Softconf for integrating ACLPUB into the START system. Before the camera-ready due date, we will continue to check what new features can be integrated, and make them implemented as complete as possible.

Although we are not going to publish paper copies of the proceedings this year, we still need to co-work with the local committee for the publication of CDROMs and program handbook (which includes local information as well as program schedules, and so on). Some such details will therefore be negotiated with the local committe in the near future.

Before the camera ready due, we will also continuously watch and coordinate the publication needs of the various co-events, including workshops, tutorials, demos, exhibitions, student research workshops and local information. And ensure the correctness and consistent style of the e-copies of the proceedings.

We will then start to compile the prceedings strting from the mid-May after the camera ready date.

The Statuts of ACLPUB integration to START

In the current state, ACLPUB in START is excellent enough for

  • well-formed input
  • for producing single-volume proceedings.

This means that some possible enhancements (according to Jin-Shin's "training set" for ACL-IJCNLP-2009) might be:

  • detection of various kinds of ill-formedness (& auto-correction or issuing warning when detected or at least ask users not to do something unusual on the submission page.) (the following are encountered during ACL-IJCNLP-09 but I guess ill-formedness cannot be exhaustively listed :-))
    • non-ascii text in metadata (particularly in Title & Author fields), e.g.,
    • inconsistent style of titles
    • inconsistent (or incorrect) spelling of the same author's name
    • unusual namees
  • multiple volume issues
    • to produce multiple volumes automatically
    • to merge multiple logical books into a single volume
  • cross-volume consistency
    • templates for main and workshops proceedings should be formatted as closely as possible: title page, copyright page, ISBN number, citation map.
  • other non-standard materials
    • prefaces (allows one) but sometimes we have more than one
    • list of members of special committees (we have templates for conference organization and PC, but not others)
    • abstracts of invited speakers (may be one or two, similar to the list of member case.)

The current changes had partially resolved the above list of major features or issues that the chairs think are important. But the integration also depends on how automatic we want ACLPUT+START to be. And some criteria (e.g., cross-volume consistency) may not be considered critical.

Demos (Sandra Kübler)

The demo session has had 65 submissions. For this reason, the demo chair extended the program committee from 10 to 27 members.

Reviewing is under way, reviews are due April 13, and the final decision will be announced April 20.

The deadline for camera-ready copies of the selected demos is May 16.

No problems expected as of now.

Exhibits (Jörg Tiedemann)

So far, no exhibitions have been registered yet. The call for exhibits has been sent around to all important channels and known sponsors have been contacted separately. The call has also been sent to a hand-picked list of publishers (about 15 companies) and a list of Swedish information/language technology companies (more than 30). I also had personal contact with several possible candidates and received some positive reactions (but no decisions yet). About the technical and other local arrangements: Everything seems to be in place and currently we don't see any problems with possible exhibitions.

Publicity (Koenraad De Smedt and Beáta Megyesi)

We have circulated the CFP for the ACL main conference, tutorials, software demonstrations and the student research workshop to majormailing lists including Corpora, CLUK, SIGANN, ELSNET, SIGIR, SIG Semitic, LR egroup, NoDaLi, Bio NLP, SentProc, AMLAP, ISCA Speech, SIGSEM, Connectionists list announcer, and to the ACL membership via Priscilla. All information can be found on the conference web.

We have sent out special notices about the widened scope of the conference and about the mentoring service.

We have circulated two newsletters, one in January, and one in February.

Upcoming things: Newsletter 3 in early April with info about e.g. tutorials, accomodation Newsletter 4 in early May about e.g. registration and program Circulate fliers on LREC Inform the press about the conference in local newspapers, TV, radio, etc

Sponsorship (Stephen Pulman, Frédérique Segond, Srinivas Bangalore, Christy Doran, Hercules Dalianis and Mats Wirén)

We have the following sponsors so far for ACL, at various levels:

  • Platinum:
    • Riksbankens Jubileumsfond (Bank of Sweden Tercentenary Foundation)
  • Gold:
    • Swedish National Graduate School of Language Technology
  • Silver:
    • CELI - Language & Information Technology (Italy)
    • ESTeam - Language Technology Software
    • Google Switzerland
    • Voice Provider (Sweden)
    • Uppsala City
    • Yahoo! Labs
  • Other
    • IBM Research (Student best paper award)
    • Language Weaver (conference bag sponsor)
  • Committed also:
    • Xerox Research Centre Europe
    • Textkernel

Program Chairs (Sandra Carberry and Stephen Clark)

(updated March 8, 2010)

ACL 2010 received 987 submissions, a record number for the conference. This was a surprise, since the submission deadline was identical for both long and short papers. Thus we expected that the total number of submissions would be lower than for ACL 2009, where authors of rejected long papers had the opportunity to resubmit them as revised short papers.

655 papers were submitted to ACL 2010 as long papers, and 332 papers were submitted as short papers. After filtering out those papers that either exceeded the length limitations or were not anonymous, 957 papers were distributed to the Area Chairs for reviewing.

In order to attract and appropriately review a wider variety of papers, we are experimenting this year with different review forms for the different categories of papers. For example, the review criteria (and thus the questions on the review form) for theoretical papers are different from the those for empirical papers. There are 10 different review categories (analysis, challenge, empirical, negative result, paradigms, position, resources, survey, systems, and theoretical); the set of review forms can be found at http://acl2010.org/reviewforms.html. Despite the submission page explicitly directing authors to examine the review forms posted on the ACL web site before selecting their review category, it was clear that authors did not do so. Thus the program chairs and area chairs examined every paper and changed the review category for papers where an inappropriate category had been selected. (Note that the Call for Papers stated that the Program Chairs reserved the right to change the review category.) This was an enormous amount of work. If this experiment is continued next year, we recommend that the submission web page do one of the following:

  1. contain a question associated with each category (such as "Does this paper present a system that has been deployed in an industrial or research setting and includes reports of tests with actual users?" for a systems paper), where the user must reply "Yes" in order to select that submission category.
  2. pop up the review form for the category that the author selects, along with the statement "I have read this entire review form and believe that it is appropriate for this submission". The author

would need to respond "Yes" in order to finalize the category selection.

We would like to note that Rich Gerber has been extremely helpful in modifying the START system to accommodate our needs. For example, the START system has been modified so that when a reviewer clicks on the review form for a paper, he or she gets the appropriate review form for that category of paper.

We also encountered problems with authors expecting that the submission deadline would not be enforced. We had several authors (some who are senior researchers and active in ACL) who were very unhappy that the START system refused their papers after the submission deadline. One spoke of a "traditional 1 hour or more grace period" for submissions. After consultation with the ACL Exec, we did not accept these late submissions since we felt it was the only way to be fair to all authors. (We did keep the START system open for 15 minutes after the deadline in order to avoid cutting off authors who were in the process of submitting prior to the deadline.) We most strongly recommend that ACL adopt the following (or something similar) as policy, that it be included in the Call for Papers in future years, and that it be added to the ACL conference handbook:

  "The ACL submission deadline will be extended only in the event
  that the START system crashes near the deadline.  The START system
  will automatically shut down at the deadline, and it is ACL policy
  that late submissions will not be allowed."

Without such a policy, exceptions that are informally granted by Program Chairs are unfair to other authors (who are not aware of the possibility of an exception) and cause problems for subsequent Program Chairs.

Our date for notification of acceptance/rejection has been moved up to Tuesday, April 20. This resulted from negotiations with the Coling chairs, in which the ACL notification deadline was moved earlier and the Coling submission deadline was moved later. This will enable authors of rejected ACL papers a few days to revise their papers and resubmit to Coling.

There will be three Best Paper Prizes at ACL 2010: Best Long Paper, Best Short Paper, and Best Student Paper. We will first select the Best Long and Short Papers; then from among the remaining papers where a student is first author, we will select the Best Student Paper. We have decided to award only one prize in each category, since we feel that the prize is more prestigious if it is not shared. A special Best Paper Prize panel is being formed to make the selections.

One problem that arises from the unexpectedly large number of submissions is accommodating enough papers in the technical program so as to maintain at least a 20% acceptance rate. The tentative schedule appears at the end of this report.

  1. The schedule will accommodate 109 long talks (including the Best Long Paper talk). Given that we have 655 long paper submissions, that represents only a 16.6% acceptance rate for long papers with long oral talks. Thus we will need to accept some long papers as 10 minute talks plus poster presentations.
  2. The tentative schedule will accommodate 64 short talks (10 minutes) with associated poster presentations. In order to maintain a 20% acceptance rate for long paper submissions, 22 of the long paper submissions must be accepted as long papers with short talks and associated poster presentations. Our examination of the requests by authors of long papers indicates that overwhelmingly they have elected a long (25 minute) presentation, so this is likely to cause some displeasure.
  3. That leaves only 42 slots for accepting short papers as 10 minute talks plus associated poster presentations. Given that there were 332 short paper submissions, that would represent an acceptance rate of only 12.7% which is clearly undesirable.
  4. Thus under the current schedule, we would need to accept 25 short paper submissions as poster presentations without oral talks. This fits with the existing schedule that allows for 90 posters.
  5. Ufortunately, this represents a very low overall acceptance rate for ACL (only 20%). Consideration should be given to adding an extra parallel track. Joakim says that there is a 100-seat room that could be used for a small extra parallel session.

ACL 2010 will have two outstanding invited talks:

  • Andrei Broder: vice-president of Yahoo and both an ACM Fellow and an IEEE Fellow. He has been asked to present a talk on the emerging field of computational advertising, with an emphasis on issues relevant to computational linguistics and natural language processing.
  • Zenzi Griffin: professor of psychology at the University of Texas at Austin. She has been asked to present a talk on psycholinguistic research relevant to computational linguistics. She recently presented a talk entitled "Towards a Psycholinguistics of Social Interaction: Issues in Direct and Indirect Address" at the CUNY Conference on Sentence Processing that was extremely well-received.

Our goal was to select invited speakers whose talks would be related to computational linguistics, but would broaden the perspective of the conference attendees. In addition, we sought individuals who had a reputation for excellent presentational skills. We believe that both of these individuals satisfy these criteria and will present excellent and exciting invited talks at ACL. One issue that was raised by an Area Chair is that neither of the invited speakers is from Europe but the conference is being held in Europe. We consulted with Jan and Joakim and neither had a problem with our sending invitations to the above two individuals, given that they both satisfied our criteria so much better than the other candidates. In addition, they brought to our attention that Zenzi Griffin studied in Sweden --- she is a graduate of the International Baccalaureate program at Kungsholmens Gymnasium in Stockholm, Sweden and studied psychology at Stockhom University for one year before transferring to Michigan State University.

We anticipate an exciting program at ACL 2010!

Tentative Schedule

See the tabs "Monday, Tuesday, Wednesday" in this Excel spreadsheet or this internet browser (.htm) version).

Comments

  1. It is **essential** that lunch be quickly available in a room adjacent to or very near the poster room; otherwise, it is unclear whether attendees will have the time to visit the posters.
  2. The SRW is scheduled as a fifth parallel session, alongside regular short poster talks. The SRW talks would be in the 100 seat auditorium in the main building that Joakim mentioned being available.
  3. The Demos are scheduled in parallel with the afternoon talks on the second day. They would be in the extra small rooms that Joakim mentioned as being available for demos.
  4. The best long paper will be presented at the end of the conference. The best short paper will be the first short talk at one of the short talk/poster sessions.
  5. The schedule runs until 17:35 on Tuesday. However, given that the hotels are only a couple of blocks from the conference center and that the banquet site is between the conference center and the hotels, we feel that this still allows enough time to return to the hotels and change clothes for the banquet.

Local Arrangements Chair (Joakim Nivre)

The Local Arrangements Committee has met all the milestones listed in the Conference Handbook so far:

  • Finalize location and dates, sign contracts meetings and accommodations [T-18]
  • Select Exhibits Chair (Jörg Tiedemann) and Publicity Chairs (Koenraad de Smedt, Beáta Megyesi) [T-18]
  • Identify Local Arrangements Committee [T-15]
  • Make contact with ACL Sponsorship Committee, identify Local Sponsorship Chairs (Mats Wirén, Hercules Dalianis) to serve on this committee [T-15; delayed to T-10 because of delay with international sponsorship committee]
  • Collect preliminary local info (describe meeting place and accommodations, travel info, visas, weather, etc.) [T-15; completed at T-24]
  • Establish website [T-15]
  • Set approximate registration deadlines and costs [T-12]
  • Design conference logo for program, posters; print and distribute posters and fliers [T-12; completed T-14 to enable distribution at 2009 conferences]
  • Confirm venues: confirm arrangements with meeting venue: rooms, spaces, lunches, etc; confirm arrangements with accommodation venue: rooms [T-7]
  • Arrange conference social program: confirm arrangements with banquet and other social event providers; arrange reception; arrange banquet [T-7]
  • Plan registration process [T-6]
  • Plan on-site process, staff needs, etc. [T-6; completed T-5]
  • Organize conference bag or other gimicks [T-6; basically completed T-5 although bag design is still in progress]
  • Liaise with Sponsorship Chairs for corporate logos [T-6; started already at T-10]
  • Appoint Student Volunteer Coordinator (Mattias Nilsson) [T-6; completed T-10]

In addition, we have completed the following tasks:

  • Set up email addresses for all major functions (general@acl2010.org, program@acl2010.org, local@acl2010.org, etc.) [May 20, 2009; continually updated]
  • Made a detailed room allocation for main conference, tutorials, workshops and other meetings [November 6, 2009; revised December 16, 2009]
  • Made a detailed budget for local costs [November 10, 2009; continually updated]
  • Issued two Newsletters [January 15, 2010; February 19, 2010]
  • Issued a Call for Exhibits [January 19, 2010]
  • Confirmed capacity of wireless internet and electricity [January 29, 2010]
  • Set up a web page for accommodation booking [February 11, 2010]
  • Made a detailed plan for the Conference Handbook (reusing the design from ACL-IJCNLP 2009) [February 19, 2010]

There are currently no outstanding problems to report from the Local Arrangements Committee.