Many prediction tasks in NLP involve assigning values to mutually dependent variables. For example, when designing a model to automatically perform linguistic analysis of a sentence or a document (e.g., parsing, semantic role labeling, or discourse analysis), it is crucial to model the correlations between labels. Many other NLP tasks, such as machine translation, textual entailment, and information extraction, can be also modeled as structured prediction problems.

In order to tackle such problems, various structured prediction approaches have been proposed, and their effectiveness has been demonstrated. Studying structured prediction is interesting from both NLP and machine learning (ML) perspectives. From the NLP perspective, syntax and semantics of natural language are clearly structured and advances in this area will enable researchers to understand the linguistic structure of data. From the ML perspective, the large amount of available text data and complex linguistic structures bring challenges to the learning community. Designing expressive yet tractable models and studying efficient learning and inference algorithms become important issues.

This workshop follows the two previous successful editions in 2017 and 2016 on Structured Prediction for NLP, as well as the closely related ICML 2017 Workshop on Deep Structured Prediction. It is very timely, as there has been a renewed interest in structured prediction among NLP researchers due to recent advances in methods using continuous representations, able to learn with task-level supervision, or modelling latent linguistic structure.

Topics will include, but are not limited to, the following:

  • Efficient learning and inference algorithms
  • Joint inference and learning approaches
  • Reinforcement learning and imitation learning for structured learning in NLP
  • Multi-task learning for structured output tasks
  • Latent structured variable models
  • Structured deep generative models
  • Neural graph learning approaches for NLP
  • Integer linear programming and other modeling techniques
  • Approximate inference for structured prediction
  • Structured training for non-linear models
  • Deep learning and neural network approaches for structured prediction
  • Structured prediction software
  • Structured prediction applications in NLP

Invited Speakers

  • Mirella Lapata, University of Edinburgh, UK
  • Jason Eisner, Johns Hopkins University, USA
  • Andrew McCallum, University of Massachusetts Amherst, USA
  • Claire Cardie, Cornell University, USA
  • Chris Dyer, DeepMind, UK
  • He He, Stanford University, USA

Organizers

Program Committee

  • Wilker Aziz, University of Amsterdam, Netherlands
  • Joost Bastings, University of Amsterdam, Netherlands
  • Hal Daumé III, Microsoft & University of Maryland, USA
  • Hiko Schamoni, Heidelberg University, Germany
  • Stefan Riezler, Heidelberg University, Germany
  • Artem Sokolov, Amazon, Germany
  • Xilun Chen, Cornell University, USA
  • Arzoo Katiyar, Cornell University, USA
  • Tianze Shi, Cornell University, USA
  • Sebastian Mielke, Johns Hopkins University, USA
  • Parisa Kordjamshidi, Tulane University, USA
  • Vivek Srikumar, University of Utah, USA
  • Yoon Kim, Harvard University, USA
  • Ivan Titov, University of Edinburgh, Scotland
  • Yoav Artzi, Cornell University, USA
  • Roi Reichart, Technion - Israel Institute of Technology, Israel
  • Amir Globerson, Tel Aviv University, Israel
  • Alexander Schwing, UIUC, USA
  • Kevin Gimpel, TTI Chicago, USA
  • Waleed Ammar, Allen AI Institute, USA
  • Matt Gormley, CMU, USA
  • Luke Zettlemoyer, University of Washington, USA
  • Pranava Madhyastha, Imperial College London, UK
  • Trevor Cohn, University of Melbourne, Australia
  • Shay Cohen, University of Edinburgh, UK
  • Marek Rei, University of Cambridge, UK
  • Amandla Mabona, University of Cambridge, UK
  • Noah Smith, University of Washington, USA

Submissions

We invite submissions of the following kinds:

  • Research papers
  • Position papers
  • Tutorial/overview papers

Long/short papers should consist of eight/four pages of content plus unlimited pages for bibliography. Submissions must be in PDF format following the NAACL 2019 templates, anonymized for review. Papers can be submitted as non-archival, so that their content can be reused for other venues. Add “(NON-ARCHIVAL)” to the title of the submission. Non-archival papers that are accepted will be linked from this webpage if their authors request so. Previously published work can also be submitted as non-archival in the same way, with the additional requirement to state on the first page the original publication.

Reviewing will be double-blind, and thus no author information should be included in the papers; self-reference should be avoided as well.

Submission is electronic and is managed by the START conference management system at https://www.softconf.com/emnlp2019/SPNLP/

Each submission will be reviewed by at least 2 program committee members.

Important Dates

  • Submission deadline: Wednedsay, March 6, 2019
  • Notification of acceptance: Wednedsay, March 27, 2019
  • Camera-ready papers due: Friday, April 5, 2019
  • Workshop date: June 6 or 7, 2019

Time is in GMT-12. Deadline is 11:59pm of the date indicated.