Ann Arbor, Michigan, June 29-30, 2005

CoNLL in previous years: 1997 1998 1999 2000 2001 2002 2003 2004 | SIGNLL

Second call for papers

CoNLL is an international conference for discussion and presentation of research on natural language learning. We invite submission of papers about natural language learning topics, including, but not limited to:

  • Computational models of human language acquisition
  • Computational models of the evolution of language
  • Machine learning methods applied to natural language processing tasks (speech processing, phonology, morphology, syntax, semantics, discourse processing, language engineering applications)
  • Symbolic learning methods (Rule Induction and Decision Tree Learning, Lazy Learning, Inductive Logic Programming, Analytical Learning, Transformation-based Error-driven Learning)
  • Biologically-inspired methods (Neural Networks, Evolutionary Computing)
  • Statistical methods (Bayesian Learning, HMM, maximum entropy, SNoW, Support Vector Machines)
  • Reinforcement Learning
  • Active learning, ensemble methods, meta-learning
  • Computational Learning Theory analysis of language learning
  • Empirical and theoretical comparisons of language learning methods
  • Models of induction and analogy in Linguistics

We wish to encourage the submission of papers that address modeling of deeper linguistic phenomena than have typically been covered in the past. This could include sub-models of a learning system that address specific linguistic behaviors in order to improve overall performance on a general task. Particular examples might be modeling of specific syntactic constructs, such as coordination or ellipsis to improve parsing, or modeling separately different types of semantic constraints and preferences to improve word sense disambiguation.

CoNLL is the yearly conference organized by SIGNLL, the Association for Computational Linguistics Special Interest Group on Natural Language Learning. Previous CoNLL meetings were held in Madrid (1997), Sydney (1998), Bergen (1999), Lisbon (2000), Toulouse (2001), Taipei (2002), Edmonton (2003), and Boston (2004).

See and for more information about SIGNLL and CoNLL.

Shared Task: Semantic Role Labeling

Following last year's initiative, the CoNLL-2005 shared task will concern again the recognition of semantic roles for the English language. After exploring the use of partial parsing information in 2004, the main focus of interest for this edition is to increase the amount of syntactic and semantic input information, aiming to boost the performance of machine learning systems to the Semantic Role Labeling task. Participants are encouraged to propose novel learning architectures and ideas for using richer linguistic information.

As in previous editions, all participant groups will share the same experimental setting, and the evaluation will be done according to fixed criteria, thus allowing comparison between various learning strategies.

The shared task data and further information are available from:

Invited Speakers

Mark Steedman and Mark Johnson.


Main Session Submissions

A paper submitted to CoNLL-2005 must describe original, unpublished work. Submit a full paper of no more than 8 pages in PDF format by April 4, 2005 electronically through the web form at Only electronic submissions will be accepted. The submitted paper should be in two column format and follow the ACL style. Since reviewing will be blind, the paper should not include the authors' names and affiliations, and there should be no self-references that reveal the authors' identity. In the submission form, you will be asked for the following information: paper title, authors' names, affiliations, and email addresses, contact author's email address, a list of keywords, abstract, and an indication of whether the paper has been simultaneously submitted to other conferences (and if so which conferences). The contact author of an accepted paper under multiple submissions should inform the program co-chairs immediately whether he or she intends the accepted paper to appear in CoNLL-2005. A paper that appears in CoNLL-2005 must be withdrawn from other conferences.

CONLL will be taking place at the same time as the Workshop on Psychocomputational Models of Human Language Acquisition (, and the meetings will have a joint session for Mark Steedman's invited talk. If there is sufficient interest there will be a joint session for papers that are relevant to both audiences.

Authors of accepted submissions are to produce a final paper to be published in the proceedings of the conference, which will be available at the conference for participants, and distributed afterwards by ACL. Final papers must follow the ACL style and are due May 17, 2005.

Shared Task Submissions

Submit a paper of maximum 4 pages describing the learning approach, and your results on the development set by April 22, 2005 to the email address (see the shared task web page for submission instruction, concrete formats, and styles). A special section of the proceedings will be devoted to a comparison and analysis of the results and to a description of the approaches used.

Important Dates

Deadline for main session paper submission: April 4, 2005 (passed: 74 submissions; 19 accepted)
Deadline for shared task paper submission: April 22, 2005 (passed: 21 submissions)
Notification of acceptance: May 5, 2005
Deadline for camera-ready papers: May 17, 2005
Conference: June 29-30, 2005

Conference Organizers

Ido Dagan
Department of Computer Science
Bar Ilan University
Ramat-Gan, Israel
dagan @ cs biu ac il

Daniel Gildea
Department of Computer Science
University of Rochester
Rochester, New York
gildea @ cs rochester edu

Shared Task Organizers

Xavier Carreras and Lluis Marquez
Software Department (LSI),
Technical University of Catalonia (UPC)
Barcelona, Spain
{carreras,lluism} @ lsi upc edu

Program Committee

  • Ido Dagan, Bar Ilan University (Israel), program co-chair
  • Daniel Gildea, University of Rochester (USA), program co-chair
  • Xavier Carreras, UPC (Spain), shared task co-chair
  • Lluis Marquez, UPC (Spain), shared task co-chair
  • Steven Abney, University of Michigan (USA)
  • Eneko Agirre, University of the Basque Country (Spain)
  • Regina Barzilay, Massachusetts Institute of Technology (USA)
  • Claire Cardie, Cornell University (USA)
  • John Carroll, University of Sussex (UK)
  • Eugene Charniak, Brown University (USA)
  • James Cussens, University of York (UK)
  • Walter Daelemans, University of Antwerp (Belgium)
  • Radu Florian, IBM (USA)
  • Dayne Freitag, Fair Isaac (USA)
  • Rebecca Hwa, University of Pittsburgh (USA)
  • Hang Li, Microsoft (China)
  • Dekang Lin, University of Alberta (Canada)
  • Diane Litman, University of Pittsburgh (USA)
  • Diana McCarthy, University of Sussex (UK)
  • Rada Mihalcea, University of North Texas (USA)
  • Yuji Matsumoto, Nara Institute of Science and Technology (Japan)
  • John Nerbonne, University of Groningen (Netherlands)
  • Hwee-Tou Ng, National University of Singapore (Singapore)
  • Grace Ngai, The Hong Kong Polytechnic University (Hong Kong)
  • Miles Osborne, University of Edinburgh (UK)
  • Patrick Pantel, Information Sciences Institute (USA)
  • David Powers, Flinders University (Australia)
  • Dragomir Radev, University of Michigan (USA)
  • Ellen Riloff, University of Utah (USA)
  • Dan Roth, University of Illinois at Urbana-Champaign (USA)
  • Anoop Sarkar, Simon Fraser University (Canada)
  • Keh Yih Su, Behavior Design Corporation (ROC)
  • Suzanne Stevenson, University of Toronto (Canada)
  • Erik Tjong Kim Sang, University of Antwerp (Belgium)
  • Antal van den Bosch, Tilburg University (Netherlands)
  • Janyce Wiebe, University of Pittsburgh (USA)   Last update: May 24, 2005