Difference between revisions of "2019Q3 Reports: Workshop Chairs"

From Admin Wiki
Jump to navigation Jump to search
(Created page with "Workshop co-chairs: * Barbara Plank (IT University of Copenhagen) * Sebastian Riedel (Facebook AI Research & University College London) We accepted 59 workshops out of 84 wor...")
 
(added suggestions and observations)
 
(One intermediate revision by the same user not shown)
Line 3: Line 3:
 
* Sebastian Riedel (Facebook AI Research & University College London)
 
* Sebastian Riedel (Facebook AI Research & University College London)
  
We accepted 59 workshops out of 84 workshop proposal submission through the process of a joint cross-conference NLP workshops call, which for 2019 was shared between ACL, NAACL and EMNLP (Link to CFP email announcement). Workshop proposals were accompanied with the organizers' top two venue preferences. Submissions and reviews were coordinated through START, and reviewed and discussed by a joint cross-conference committee including the workshop chairs of all three conferences, including the two of us, plus Smaranda Muresan, Swapna Somasundaran and Elena Volodina (NAACL), Vera Demberg and Naoaki Okazaki (EMNLP). We largely followed the COLING model, for which there is a blog post describing more of this process (except that the reviewer load was higher due to the larger number of submissions, roughly 24 proposals/reviewer). During the joint selection process, we tried to allocate accepted workshops to their  first or second choice of venue, however, this was impossible due to the overwhelming preference for ACL this year.  Of the 84 submissions, 74% had ACL as their first choice, and 14% as second choice. The ACL workshops were assigned to the two days based on room availability, projected registrations, and in limited cases to maintain topical balance.
+
We accepted 59 workshops out of 84 workshop proposal submission through the process of a joint cross-conference NLP workshops call, which for 2019 was shared between ACL, NAACL and EMNLP ([Link to CFP email announcement](https://www.aclweb.org/portal/content/first-joint-call-workshop-proposals-acl-emnlp-ijcnlp-naacl-2019)). Workshop proposals were accompanied with the organizers' top two venue preferences. Submissions and reviews were coordinated through START, and reviewed and discussed by a joint cross-conference committee including the workshop chairs of all three conferences, including the two of us, plus Smaranda Muresan, Swapna Somasundaran and Elena Volodina (NAACL), Vera Demberg and Naoaki Okazaki (EMNLP). We largely followed the COLING model, for which there is [a blog post](https://coling2018.org/workshop-review-process-for-acl-coling-emnlp-and-naacl-2018/) describing more of this process (except that the reviewer load was higher due to the larger number of submissions, roughly 24 proposals/reviewer). During the joint selection process, we tried to allocate accepted workshops to their  first or second choice of venue, however, this was impossible due to the overwhelming preference for ACL this year.  Of the 84 submissions, 74% had ACL as their first choice, and 14% as second choice. The ACL workshops were assigned to the two days based on room availability, projected registrations, and in limited cases to maintain topical balance.
 
The Student Research Workshop, WMT, CoNLL, *SEM, SemEval and Widening NLP were "pre-admitted" workshops (or  co-located events in case of WMT) which submitted a proposal, but were not reviewed through any of this process.  
 
The Student Research Workshop, WMT, CoNLL, *SEM, SemEval and Widening NLP were "pre-admitted" workshops (or  co-located events in case of WMT) which submitted a proposal, but were not reviewed through any of this process.  
In general for the organization and coordination process, we followed the excellent workshop chairs duties guide. We would like to emphasize that we followed earlier recommendations and underline the importance of a "master file" shared coordination spreadsheet with up-to-date information about all the workshops, including organizers' email address, workshop URLs, and later in the process, proceedings URLs, ISBNs, poster scheduling and layout details, and possibly other one-off information such as organizers' requests for name changes, adding new organizer contacts, and other matters.  
+
In general for the organization and coordination process, we followed the excellent [workshop chairs duties guide](https://aclweb.org/adminwiki/index.php?title=Workshop_chair_duties). We would like to emphasize that we followed earlier recommendations and underline the importance of a "master file" shared coordination spreadsheet with up-to-date information about all the workshops, including organizers' email address, workshop URLs, and later in the process, proceedings URLs, ISBNs, poster scheduling and layout details, and possibly other one-off information such as organizers' requests for name changes, adding new organizer contacts, and other matters.  
 +
 
 +
Points for discussion and suggestions:
 +
* Reviewing: We followed the 2018 model (where 9 workshop chairs reviewed on average 15 proposals). Given the increased number of submissions (84 vs 58 in 2018) which seems to follow the general growing trend of *ACL conferences, this current model is unlikely to be sustainable in the future. In particular, we were 7 workshop chairs (as there were 3 conferences in 2019 vs 4 in 2018, including COLING) which amounted to 24 proposals per reviewer (with 2 reviews/proposal). The final decision and venue allocation required care and amounted to several online calls between all 7 chairs. We would like to propose to increase the number of workshop chairs per conference to at least 3 each, which helps both the reviewing process and to split up workshop chair duties. An alternative is to establish a PC for workshop proposals reviewal and solicit 3 reviews/proposal.
 +
* Communication regarding schedule for handbook etc: Access to schedule information for the conference handbook, website etc can be retrieved from the order files in Softconf. However, we no longer had access to the individual Softconf pages to retrieve the latest information (and only some workshops mailed their program when their proceedings were due). We initially asked Softconf for full admin access to START to create all workshop START pages and this access was revoked in December on purpose. It might be worth to establish a dedicated workshop schedule co-chair that interacts with all workshop organizers, handbook chairs and publication chairs and keeps track of all last minute changes regarding the schedule. Moreover, we would encourage to go digital for the handbook in future.
 +
 
  
 
The workshops and co-located events at ACL 2019 are:
 
The workshops and co-located events at ACL 2019 are:

Latest revision as of 00:25, 27 July 2019

Workshop co-chairs:

  • Barbara Plank (IT University of Copenhagen)
  • Sebastian Riedel (Facebook AI Research & University College London)

We accepted 59 workshops out of 84 workshop proposal submission through the process of a joint cross-conference NLP workshops call, which for 2019 was shared between ACL, NAACL and EMNLP ([Link to CFP email announcement](https://www.aclweb.org/portal/content/first-joint-call-workshop-proposals-acl-emnlp-ijcnlp-naacl-2019)). Workshop proposals were accompanied with the organizers' top two venue preferences. Submissions and reviews were coordinated through START, and reviewed and discussed by a joint cross-conference committee including the workshop chairs of all three conferences, including the two of us, plus Smaranda Muresan, Swapna Somasundaran and Elena Volodina (NAACL), Vera Demberg and Naoaki Okazaki (EMNLP). We largely followed the COLING model, for which there is [a blog post](https://coling2018.org/workshop-review-process-for-acl-coling-emnlp-and-naacl-2018/) describing more of this process (except that the reviewer load was higher due to the larger number of submissions, roughly 24 proposals/reviewer). During the joint selection process, we tried to allocate accepted workshops to their first or second choice of venue, however, this was impossible due to the overwhelming preference for ACL this year. Of the 84 submissions, 74% had ACL as their first choice, and 14% as second choice. The ACL workshops were assigned to the two days based on room availability, projected registrations, and in limited cases to maintain topical balance. The Student Research Workshop, WMT, CoNLL, *SEM, SemEval and Widening NLP were "pre-admitted" workshops (or co-located events in case of WMT) which submitted a proposal, but were not reviewed through any of this process. In general for the organization and coordination process, we followed the excellent [workshop chairs duties guide](https://aclweb.org/adminwiki/index.php?title=Workshop_chair_duties). We would like to emphasize that we followed earlier recommendations and underline the importance of a "master file" shared coordination spreadsheet with up-to-date information about all the workshops, including organizers' email address, workshop URLs, and later in the process, proceedings URLs, ISBNs, poster scheduling and layout details, and possibly other one-off information such as organizers' requests for name changes, adding new organizer contacts, and other matters.

Points for discussion and suggestions:

  • Reviewing: We followed the 2018 model (where 9 workshop chairs reviewed on average 15 proposals). Given the increased number of submissions (84 vs 58 in 2018) which seems to follow the general growing trend of *ACL conferences, this current model is unlikely to be sustainable in the future. In particular, we were 7 workshop chairs (as there were 3 conferences in 2019 vs 4 in 2018, including COLING) which amounted to 24 proposals per reviewer (with 2 reviews/proposal). The final decision and venue allocation required care and amounted to several online calls between all 7 chairs. We would like to propose to increase the number of workshop chairs per conference to at least 3 each, which helps both the reviewing process and to split up workshop chair duties. An alternative is to establish a PC for workshop proposals reviewal and solicit 3 reviews/proposal.
  • Communication regarding schedule for handbook etc: Access to schedule information for the conference handbook, website etc can be retrieved from the order files in Softconf. However, we no longer had access to the individual Softconf pages to retrieve the latest information (and only some workshops mailed their program when their proceedings were due). We initially asked Softconf for full admin access to START to create all workshop START pages and this access was revoked in December on purpose. It might be worth to establish a dedicated workshop schedule co-chair that interacts with all workshop organizers, handbook chairs and publication chairs and keeps track of all last minute changes regarding the schedule. Moreover, we would encourage to go digital for the handbook in future.


The workshops and co-located events at ACL 2019 are:

  • Joint Workshop on Multiword Expressions and WordNet (MWE-WN 2019)
  • BioNLP 2019
  • TyP-NLP: The Workshop on Typology for Polyglot NLP
  • The Second BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP at ACL 2019
  • 1st International Workshop on Computational Approaches to Historical Language Change
  • The Fourth Arabic Natural Language Processing Workshop
  • 6th Workshop on Argument Mining
  • The 14th Workshop on Innovative Use of NLP for Building Educational Applications (BEA)
  • 4th Workshop on Representation Learning for NLP (RepL4NLP-2019)
  • The Sixteenth SIGMORPHON Workshop on Computational Research in Phonetics, Phonology, and Morphology
  • NLP for Conversational AI
  • The Thirteenth Linguistic Annotation Workshop (LAW XIII)
  • Deep Learning & Formal Languages: Building Bridges
  • Gender Bias in Natural Language Processing
  • The 7th Workshop on Balto-Slavic Natural Language Processing
  • Widening NLP 2019
  • The Third Workshop on Abusive Language Online
  • Second Workshop on Storytelling (StoryNLP)
  • The First International Workshop on Designing Meaning Representations (DMR)
  • 4th Workshop on Social Media Mining for Health (#SMM4H) Research and Applications
  • Fourth Conference on Machine Translation (WMT19)