2019Q1 Reports: ACL 2019

2019Q1 Reports: ACL 2019

General Chair

Lluís Màrquez, Amazon

The 57th annual meeting of the Association for Computational Linguistics (ACL 2019) will take place in Florence, Italy, at the 'Fortezza da Basso' from July 28th to August 2nd, 2019. I am proud to say that I have a very strong team of Chairs, both in terms of expertise and human quality, for putting together the conference. For the organizing positions that I had primary control over (Workshops, Tutorials, Demos, Publications, Student Research Workshop), I attempted to recruit a balanced set of people across several dimensions. Regarding gender, there are 17 female organizers and 19 male organizers in all the committees, as listed in the conference website.

We have been using Slack for most intra-committee communication, including some coordination with the NAACL 2019 General Chair, Jill Burstein. This is working well. However, some communications are still predominantly on email, especially the discussions with the local organizers, PCO, website, etc. I have to say that this duality has brought some confusions and information loses over the last weeks, for which I recommend to stick to one medium as much as possible. Finally, we have also used some video conferences for the topics that are too complex or tedious over email.

In my opinion, the main challenge of this year is to cope with the abrupt increase of the conference size, which was totally unanticipated at the time of bidding, venue selection, and signature of contracts. After the large EMNLP in late 2018, we increased our initial estimates substantially (by ~50%) and we started re-planning the conference for a max. size of 3,000 attendees, which is very close to the limit accepted by the venue. The number of submissions received (over 2700 valid papers; 75% increase with respect to 2018), not only confirmed the sharp growth but exceeded again our expectations. This situation poses new challenges on the reviewing/acceptance process but also on the conference planning and setting. At this moment, there is a coordinated effort among the PC Co-Chairs, the Local Organizers, the PCO, Priscilla and myself, in order to adapt to the situation, and to finalize the conference setting (including new contracts with the venue for extra space and services). In summary, I see this success of the ACL 2019 call with excitement, but with some worries too. My priority is to serve the community by ensuring the best possible scientific program, and, at the same time, the best possible experience to all conference attendees. These two dimensions are not easy to optimize jointly.

This year, we also want to pay a tribute to the memory of Jan Wiebe, who sadly passed away in December 2018. We are coordinating this effort with NAACL 2019, which is the first conference in the year. Among other things, NAACL plans to have a best paper award dedicated to her. In ACL, we plan to commemorate Jan’s memory in some way at the beginning of the Lifetime Achievement award session (LTA), which is plenary and highly attended. Most likely, we will have a person close to Jan giving a short commemorative speech.

Marti Hearst also suggested to use some time on the LTA session to acknowledge the new ACL Fellows from 2018. I believe this is a great idea, and something that could be adopted as the standard procedure in the future.

The conference website http:/www.acl2019.org, which is managed by the PCO, contains updated information on calls, committees, workshops, tutorials, site and travel information, etc.

The rest of this report is structured around the activities associated with each Chair position, each with a separate section below.

Program Co-Chairs

Anna Korhonen, University of Cambridge

David Traum, University of Southern California

Source: https://docs.google.com/document/d/13dvoyDyfBY2HZfdZMH0P8WCphfrk37YL_63zKEwbAas/edit

We have focussed our efforts on improving various aspects of the review process, in particular where previous practices are no longer effective, given the rapid growth of the field / conference (e.g. we saw 75% increase in submissions from last year - more on this below) and the shortage of high quality reviewers that the field as a whole is now suffering from. In particular, our goal has been to help the program committee make good decisions and provide good feedback to authors, while minimizing the amount of extra work, and the number of cases where people have to cope with short turn-around times. Thus we have backed away from some features of recent previous conferences, which individually have some merit but also some problematic aspects, particularly involving short deadlines. In particular,

• We have continued and extended the NAACL-2018 practice of differentiating “Senior Area Chairs” (SACs) who assign papers to reviewers and compile final area rankings, from “Area Chairs” (ACs), who manage a smaller set of papers within an area, leading the discussion and writing a meta-review (which will be shared with authors) and recommendations. This has resulted in a program committee structure where the roles and responsibilities of SACs and ACs are clearer, similar to that in conferences that have a “senior program committee”. We have also increased the number of SACs (44) and ACs (over 176). Each AC is in charge of only 10-20 papers in their area. This number is sufficiently small that they can, when needed, also help and review papers in topic areas where we have a lack of high quality reviewers.

• We have dropped reviewer bidding for papers and (like NAACL:) author response this year. We dropped the bidding partly because it was time consuming and felt it made sense to invest the time in improving the reviewing process instead (which is now realistic, as we have a large senior program committee) and partly because we felt reviewers should be reviewing papers that they are qualified to review, rather than cherry-picking a few that they want to review. However, we do still allow reviewers to report if they are incapable of reviewing a paper or have a conflict of interest with it. We dropped the review response because it takes a lot of time and has not shown to make a great deal of difference (e.g. last years PC chairs have a paper accepted in this years NAACL showing exactly that).

• We have moved away from the elaborate review forms from NAACL-2018 and ACL 2018, back to a more streamlined form, adapted from EMNLP-2018. This is also because the elaborate forms took a lot of time, with little evidence that they actually helped the review process.

In terms of how we have run things (apologies our points will follow in a random order!):

• We have been active in Twitter and in making news announcements on the conference website but haven’t used a conference blog. This is partly because of the many changes we have made (e.g. a big senior programme committee and corresponding changes in START) and the huge number of submissions that have kept us busy. Also initially no conference blog was available. The publicity chairs have now set it up, but as far as we can see no one has posted anything yet? For the conference blog to make sense we would need someone (other than us) in charge, inviting contributions and also other conference chairs contributing to it. In the meanwhile, we will keep posting to Twitter and to the News section of the conference website and we believe this will meet the needs. Also, we are committed to providing timely and helpful responses to queries that we receive by email.

• Following the example of ACL-2018 and NAACL-2019, we have requested nominations from the community for senior area chairs, area chairs, reviewers and invited speakers. These nominations included 113 Unique Area Chairs, 28 Unique Senior Area Chairs, 11 Unique Invited Speakers and 751 Unique Reviewers. We will publish this data as soon as we can also report how many of these nominations resulted in invitations / accepts to the program committee (soon, but right now out numbers are still changing). The basic requirements for nominations were:
  • Reviewers:
    • As a minimum requirement, nominated reviewers must have a good publication record in NLP/CL. Apart from that, the most important factors are thoroughness and reliability.
  • Area Chairs:
    • Nominated ACs must have a PhD in an area related to NLP/CL, a strong publication record in leading NLP/CL venues, and an extensive experience in reviewing for such venues.
  • Senior Area Chairs:
    • Nominated SACs must be experienced NLP/CL researchers with an impressive research and publication record in leading NLP/CL venues. Previous Area Chairing experience from *CL and EMNLP conferences is highly desirable.
  • Invited Speakers:
    • It is expected that the Invited Speaker is not only an exceptional researcher but also a community influencer and someone who can deliver an inspiring talk. Nominations from outside the mainstream NLP/CL community are also welcome!
• We have 22 areas, adding “Applications”, which has been used in NAACL, to the 21 used by ACL-2018.
• We have 44 SACs (2 per area) and 176 ACs (this number is still changing). The SAC and AC selection was performed according to the criteria above (see nominations) and on the basis of a mix of previous experience and excellence as well as according to gender and regional diversity). Current SACs and ACs are listed on the conference webpage: http://www.acl2019.org/EN/senior-program-committee.xhtml
• The START system did not fully support our PC approach, with the distinct roles for PC chairs, SACs, ACs, and PC members. ACs could either have access to a whole track (as “track managers”, as has been done in the past), or treated as regular reviewers. Our interim solution, worked out with SoftConf was for them to beef up the “meta-reviewer” functionality, so that this could be a declared role that is supported in assignment boxes. If the conference continues in this direction, it would be beneficial to add further support for this role.
• PC members were chosen by SACs for each area, given the nominations and volunteers mentioned above, as well as the previous year’s committee. This did result in multiple invites for some people, and a number of cases of people reviewing for more than one area. As of now we have 2000 PC members (excluding SACs and ACs), and 2676 total assignments of PC members to areas.
• 2913 submissions were initially received in the START system, from 62 different countries, by contact author. 114 submissions were deleted or have so far been withdrawn. At the time of the submission deadline we had 2820 submissions. This is an increase of c. 75% over ACL 2018 and (as far as we are aware of) an all time record for ACL-related conferences.
• All papers were carefully inspected for anonymity, formatting, and plagiarism issues. We used assistants for this process (as well as some others), which was absolutely vital as it would have taken us for 1-2 weeks to do this ourselves. All issues that were found by the assistants were examined by one of the PC chairs. As a result, 88 submissions have been “desk rejected” for violating policies (plagiarism, anonymity, page length, or major formatting changes compressing more content in the pages than the format allows).
• The remaining submissions were also checked by SACs to identify COI and area mismatches. Of the 2711 submissions that are being sent for review, there are 1622 long and 1089 short papers. Here is the current breakdowns by area, listing total, long, and shot papers for each. The biggest surprise is the generation area, with an increase from 59 to 156 submissions from 2018 to 2019 (much larger than the overall growth rate), also Linguistic theories, jumped from 24 to 60.
  • Applications 138 65 73
  • Dialogue and Interactive Systems 182 125 57
  • Discourse and Pragmatics 57 33 24
  • Document Analysis 82 48 34
  • Generation 156 98 58
  • Information Extraction and Text Mining 249 156 93
  • Linguistic Theories Cognitive Modeling and Psycholinguistics 60 39 21
  • Machine Learning 222 149 73
  • Machine Translation 210 104 106
  • Multidisciplinary and Area Chair COI 116 71 45
  • Multilinguality 74 43 31
  • Phonology Morphology and Word Segmentation 43 25 18
  • Question Answering 154 99 55
  • Resources and Evaluation 130 70 60
  • Sentence-level Semantics 111 70 41
  • Sentiment Analysis and Argument Mining 151 91 60
  • Social Media 94 51 43
  • Summarization 83 48 35
  • Tagging Chunking Syntax and Parsing 99 50 49
  • Textual Inference and Other Areas of Semantics 81 50 31
  • Vision Robotics Multimodal Grounding and Speech 82 59 23
  • Word-level Semantics 137 78 59
• Summary of program deadlines:
  • Mar. 12-15: [SAC] allocate papers to reviewers and to ACs
  • March 17- April 7: Review period
  • April 8-23: [AC] emergency reviews, discussion
  • April 24-25: [AC] finalise meta-reviews and recommendations
  • April 26-29: [SAC] rank papers and make initial decisions and recommendations
  • April 30- May 12: [PC chairs] make decisions
  • May 13: [PC chairs] notifications
  • June 3:Camera ready
• Other upcoming tasks:
  • Determine maximum space available with Conference and Local chairs, implications for acceptance rate
  • Decide on Invited speakers
  • Select the best paper panel
  • Select the conference program (coordinate with TACL)
  • Select the Session chairs
  • Produce stats and reports
• Issues to discuss:
  • dual submission -where is the line of “archival” vs non-archival workshops?
  • issues with dual submission leading to lack of anonymity if made public during the review period?
  • problematic asking authors to indicate future submission to other meetings - they don’t know yet and it’s a pain to add this info later.
  • sharing of titles, abstracts, and/or anonymous digests (DUDE) with other conferences to prevent dual submission violation?
  • start changes to support area chairs/metareviewers - more needed in future
  • worries about capacity/acceptance rate
  • how to do coi management efficiently (who has to put in long lists in advance, or multiple rounds of reconciliation)
  • where to draw the line on format violations? Any way to help authors avoid these?
• List of Area Chairs:
  • Dialogue and Interactive Systems:
    • SACs: Kallirroi Georgila, Ryuichiro Higashinaka,
    • ACs: Michel Galley, Zhou Yu, Milica Gasic, Rebecca J. Passonneau, Gabriel Skantze, Matthew Marge, Helen Hastie, Kazunori Komatani, Yun-Nung Chen, Pascale Fung
  • Discourse and Pragmatics:
    • SACs: Annie Louis, Andrew Kehler
    • ACs: Benamara Farah, Giuseppe Carenini, Michael Strube, Bonnie Webber, Smaranda Muresan, Manfred Stede
  • Document Analysis:
    • SACs: Bracha Shapira, Eugene Agichtein
    • ACs: Michael Bendersky, Dilek Hakkani-Tur, Anton Leuski, Andrei Popescu-Belis, Peng Zhang, Xiang Ren, Sujian Li
  • Generation:
    • SACs: Cecile Paris, Kees van Deemter
    • ACs: Stephanie M. Lukin, Matthew Stone, Nina Dethlefs, John Kelleher, Paul Piwek, Yoav Goldberg
  • Information Extraction and Text Mining:
    • SACs: Alessandro Moschitti, Heng Ji
    • ACs: Isabelle Augenstein, Hannaneh Hajishirzi, Nazli Goharian, Ruihong Huang, Kevin Cohen, Siddharth Patwardhan, Sumithra Velupillai, Yunyao Li, Gerard de Melo, Mark Stevenson, Avi Sil, Aurélie Névéol, Kenneth Church, Mausam, Alan Ritter, Hoifung Poon, Nut Limsopatham
  • Linguistic Theories, Cognitive Modeling and Psycholinguistics:
    • SACs: Frank Keller, Aline Villavicencio
    • ACs: Afra Alishahi, Yevgeni Berzak, Shuly Wintner, Vera Demberg, Emily Prud'hommeaux
  • Machine Learning:
    • SACs: Chris Dyer, Ariadna Quattoni
    • ACs: Ashish Vaswani, Kai-Wei Chang, Fei Sha, Barbara Plank, William Yang, Tim Rocktäschel, Le Sun, Jason Naradowsky, Alice Oh, Amir Globerson, Pontus Stenetorp, Andreas Vlachos
  • Machine Translation:
    • SACs: Trevor Cohn, Yang Liu
    • ACs: Dekai Wu, Kevin Duh, Jörg Tiedemann, Deyi Xiong, Taro Watanabe, Philipp Koehn, Marine Carpuat, Arianna Bisazza, Alexander Fraser, Zhaopeng Tu, Qun Liu, Yvette Graham, Daniel Cer, Minh-Thang Luong
  • Multidisciplinary (also for AC COI):
    • SACs: Patrick Pantel, Julia Hockenmaier
    • ACs: Yoav Artzi, Bowen Zhou, Grzegorz Chrupała, Dong Nguyen, Simone Paolo, Sara Rosenthal\
  • Multilinguality:
    • SACs: Joakim Nivre, Timothy Baldwin
    • ACs: Anders Søgaard, Jonathan May, Christian Hardmeier
  • Phonology, Morphology and Word Segmentation:
    • SACs: Graham Neubig, Hinrich Schütze
    • ACs: Ryan Cotterell, Manaal Faruqui, Hai Zhao, Kemal Oflazer, Miikka Silfverberg
  • Question Answering:
    • SACs: Sanda Harabagiu, Zornitsa Kozareva
    • ACs: Kang Liu, Yansong Feng, Shafiq Joty, Eric Nyberg, Preslav Nakov, Giovanni Da San Martino, Jennifer Chu-Carroll, Idan Szpektor
  • Resources and Evaluation:
    • SACs: Sara Tonelli, Ron Artstein
    • ACs: Gina-Anne Levow, Thierry Declerck, Nancy Ide, Kenji Sagae, Udo Kruschwitz, Beata Megyesi, Roberto Navigli, Owen Rambow
  • Sentence-level Semantics:
    • SACs: Mona Diab, Ivan Titov
    • ACs: Wei Xu, Siva Reddy, Steven Bethard, Eduardo Blanco, Wenpeng Yin, Liang Huang, Edward Grefenstette, Michael Roth, Mehrnoosh Sadrzadeh, Anette Frank
  • Sentiment Analysis and Argument Mining:
    • SACs: Marie-Francine Moens, Bing Liu
    • ACs: Saif Mohammad, Els Lefever, Liang-Chih Yu, Yulan He, Oren Tsur, Claire Cardie, Yue Zhang, Swapna Somasundaran, Jinho D. Choi
  • Social Media:
    • SACs: Kalina Bontcheva, Cristian Danescu-Niculescu-Mizil
    • ACs: Nigel Collier, Jacob Eisenstein, Dirk Hovy, David Jurgens, Tim Finin, Diyi Yang, Wei Gao, Wei Wei
  • Summarization:
    • SACs: Mirella Lapata, Chin-Yew Lin
    • ACs: Wenjie Li, Xiaojun Wan, Jackie Chi Kit Cheung, Shashi Narayan, Xiaodan Zhu, Fei Liu
  • Tagging, Chunking, Syntax and Parsing:
    • SACs: Phil Blunsom, Noah A. Smith
    • ACs: Roi Reichart, Marek Rei, Daisuke Kawahara, Emily Pitler, Omri Abend, Weiwei Sun
  • Textual Inference and Other Areas of Semantics:
    • SACs: Sabine Schulte im Walde, Raffaella Bernardi
    • ACs: Omer Levy, Angeliki Lazaridou, Jonathan Berant, Vivek Srikumar, Dimitri Kartsaklis, Christopher Potts, Roy Schwartz
  • Vision, Robotics, Multimodal, Grounding and Speech:
    • SACs: Louis-Philippe Morency, Michael Johnston
    • ACs: Catharine Oertel, Matthias Scheutz, Sakriani Sakti, Elia Bruni, Manny Rayner, Douwe Kiela, Yonatan Bisk, Yale Song
  • Word-level Semantics:
    • SACs: Eneko Agirre, Diana McCarthy
    • ACs: Mohammad Taher Pilehvar, Ekaterina Shutova, Ivan Vulić, Laura Rimell, Paul Cook, Chris Biemann, Marianna Apidianaki, Diarmuid Ó Séaghdha, Jose Camacho-Collados, Aitor Soroa
  • Applications:
    • SACs: Joel Tetreault, Karin Verspoor
    • ACs: Sarvnaz Karimi, Filip Ginter, Vincent Ng, Beth Ann, Jens Edlund, Maria Liakata

Local Organisation Co-Chairs

Bernardo Magnini, FBK, Trento

Simonetta Montemagni, ILC-CNR, Pisa

Alessandro Lenci, University of Pisa

• Conference setup

We have planned for ~3,000 attendants overall. Here follows the main conference/workshop/tutorial current room allocation. Please note that, besides monumental rooms, all other spaces (named Halls below) will be built on demand within the Spadolini building at the Fortezza: as a consequence there is some flexibility for what what concerns the needed rooms and their capacity. For instance, we could add a spill over for the plenaries for 300 pax and/or another big room for workshops (see below).

 Main Conference:
 Plenaries
    - 2100 (Cavaniglia), plus 500 (Hall2 as spill-over), plus 500 (Hall3 as spill-over). Hall4-300 could also be used as spill-over if needed

 Six parallel sessions
    - 1 x 2100 (Cavaniglia - movable partitions can be used to limit the capacity of the room to e.g. 800/1000 seats)
    - 1 x 800 (Hall2-500 + Hall4-300 as spill-over)
    - 3 x 500 (Hall3, Hall10, Hall13)
    - Arsenale pavilion for posters if the final decision is to have them in parallel 

 Workshops:
 Sunday 28/7
    - 1 x 250 (monumental room Teatrino Lorenese)
 Thursday and Friday 1-2/8 (11 workshops on 1/8, 10 on 2/8)
    - 1 x 800 (Hall2-500 + Hall4-300 as spill-over)
    - 3 x 500 (Hall3, Hall10, Hall13)
    - 2 x 300 (Hall9, Hall12)
    - 1 x 250 (monumental room Teatrino Lorenese)
    - 1 x 200 (monumental room Basilica or sala Attico Spadolini)
    - 2 x 140 (monumental rooms Volta and Scherma)
    - 1 x  90  (monumental room Polveriera)
    - if needed, we might also use the Cavaniglia room (2100 seats)

 Tutorials:
 Sunday 28/7 (9 tutorials half-day each)
    - 1 x 800 (Hall2-500 + Hall4 as spill-over)
    - 3 x 500 (Hall3, Hall10, Hall13)
    - 1 x 300 (Hall 9)
    - if needed, we might also use the Cavaniglia room (2100 seats)

• Publicity

Dissemination of the news about the conference on the ACL 2019 social media channels: the publicity co-chairs created a new Twitter and Facebook account for the conference. The Instagram account remained the same of the previous editions:

    Facebook: https://www.facebook.com/acl2019Italy/
    Twitter: https://twitter.com/acl2019_italy
    Instagram: https://www.instagram.com/aclnlp/

All news and updates were also disseminated through mailing lists relevant to NLP and machine learning. ACL 2019 Chair Blog: The structure of the Blog has be prepared. Currently, the website is hosted at University of Pisa (protected by password). URL: http://acl2019pcblog.fileli.unipi.it.

• Social Event and Recognition Dinner

Since there are not many locations that can host 3,000 people in Florence, we are considering the following options:

    1. The easiest option is the Fortezza area, which is a good place both in terms of capacity and in terms of cost. At Fortezza, traditional Tuscanian food, different types of music (e.g. live opera arias etc.), and other entertainments (e.g. traditional Florence “Sbandieratori”) can be offered. Being right in the historical center, people will then be free to go around the city. 
    2. The Certosa Abbey (http://www.certosadifirenze.it/), which requires bus transportation. Dinner will be organized at the Certosa
    3. A historical location in the city center, very close to the Fortezza area, including dinner. We are currently exploring this solution with Florence Municipality.

Soon we will be able to provide more details to choose among the different options. Recognition dinner: to be defined - please confirm how many people will be attending.

• Sponsorship

Local organizers and the PCO have been receiving sponsorship inquiries and routinely responding with (a) the Sponsorship brochure provided by Priscilla, and (b) the contact information for the ACL sponsorship team. Roberto Basili and Giovanni Semeraro are responsible for Italian companies.

• Student volunteers

The Student Volunteers Coordinators are setting up the Joint call for Student Scholarship Applications and Student Volunteers, and communication with ACL Office (Priscilla Rasmussen) is underway. They are establishing the selection criteria in order to provide a suitable number of scholarships/volunteers also considering the expected large attendance. They have prepared a provisional call which will be published by mid April.

• Hotels and room block

The conference period will be high season in Florence. A platform for reservation is already active since 2017 in the conference website, but very few rooms have already been booked by prospective participants. We would like to recommend all ACL participants to book their accommodation as soon as possible; it is really important otherwise no room will be granted, if not prepaid from ACL. In the meanwhile, we are trying to get more rooms from Student hotels.

• Registration

We are already receiving several inquiries about registrations. It is necessary to know asap when online registrations will be available, registration fees, etc., in order to reply to the growing number of requests for this information.

• Visas

The PCO is taking care of visa requests; a short text is available in the website. After paper acceptance, the PCO will provide invitation letters, if requested.

• Childcare

Following the site inspection done with Priscilla, we would suggest to arrange a kids area at the Palazzina Lorenese, very close to the Spadolini Pavilion. Please confirm if this service needs to be arranged, in order to contact a childcare service agency, define the setting up and costs.

• Exhibits

Nothing specific to report at this stage; depends on final sponsorship numbers and commitments. Appropriate space (ground floor of the Spadolini building) has been individuated for at least 7 big booths and 20 small booths. The suggested setting up date for booths & exhibitors is July 28th. Dismantling is to be done on July 31st late afternoon or August 1st. The PCO and the venue manager will provide setting-up listings on request. The PCO and Priscilla will collect display of logos and define the final exhibitors map.

4. Workshop Co-Chairs

Barbara Plank, IT University of Copenhagen

Sebastian Riedel, University College London

The ACL/EMNLP-IJCNLP/NAACL joint call for workshop proposals received 84 workshop proposals, which is a substantial increase compared to previous year (2018: 58 proposals, 4 venues). Out of the 84 proposals, 59 were accepted for the three venues. ACL 2019 features the following 21 workshops and co-located events:

  - 19 selected workshops to be held post-conference (August 1 and 2, 2019)
  - one co-located two-day event (WMT) to be held immediately following the ACL 2019 conference.
  - one pre-conference workshop: WiNLP (Widening NLP) workshop, to be held in parallel with the tutorials, following earlier years
  - the Student Research Workshop (SRW), to be held during the main conference. 

The 19 ACL 2019 workshops were selected via a joint call and review procedure by the ACL workshop committee comprised of all the seven workshop chairs of the 2019 editions of ACL, NAACL, and EMNLP-IJCNLP. The workshop review process followed the procedure of the previous year, namely: Each proposal was reviewed independently by at least two committee members via softconf. Each committee member reviewed around 24 proposals this year. To aid the review process, we followed previous years’ process and attempted to categorize the proposals into categories, to help align proposals with areas of expertise of the committee members. This proved quite difficult due to the broadness of disciplines, but it helped to identify similar proposals.

After the reviewing phase, a joint final acceptance/rejection decision was taken, where each proposal was discussed individually via online meetings.

Before considering the bulk of the 84 submissions, we note that there are a five workshops and co-located events that the ACL organization agrees to pre-admit: CoNLL,*SEM, SemEval, Widening NLP and WMT. In contrast to previous years, only 60% of these were placed at their first-choice venue. First choice allocation turned out to be particularly difficult, due to the very preference for ACL as first choice venue (74% of all workshop proposals indicated ACL as first choice, see details below).

The overall high number of submission resulted in extra work for local organizers and general chairs across all three major venues, who tried to get additional workshop rooms, while keeping a healthy growth rate. Additionally, an online survey was carried out and received more than 1,700 responses from past conference and workshop attendees. The workshop program at each of the three conferences was designed to optimize workshop location preferences as much as possible, as well as diversity of topics and organizers. This information was used solely for workshop size allocation.

Details on venue preference out of 84 submissions:

    First choice:
    10% EMNLP-IJCNLP 2019
    74% ACL 2019
    13% NAACL 2019
    4% No Preference

2nd choice:

    14% ACL 2019
    27% No Preference
    20% NAACL 2019
    38% EMNLP-IJCNLP 2019

Here are the 21 selected workshops / colocated conferences for ACL 2019. All links to the workshops webpages can be found in: http://www.acl2019.org/EN/workshops.xhtml Co-located Conferences

• Fourth Conference on Machine Translation (WMT19) (August 1-2)
  • Barry Haddow, Philipp Koehn, Matthias Huck, Lucia Specia, Yvette Graham, Christof Monz, Christian Federmann and Ondřej Bojar

Special Workshops

• Student Research Workshop (SRW) (July 29-31)
  • Fernando Alva-Manchego, Eunsol Choi and Daniel Khashabi
• Widening NLP 2019 (July 28)
  • Zeerak Waseem, Amittai Axelrod, Erika Doggett, Rossana Cunha Silva, Samira Shaikh, Diyi Yang and Kristen Johnson

Workshops on August 1

• BioNLP 2019
  • Dina Demner-Fushman, Kevin Cohen, Sophia Ananiadou and Jun'ichi Tsujii
• TyP-NLP, Typology for Polyglot NLP
  • Yevgeni Berzak, Ryan Cotterell, Manaal Faruqui, Eitan Grossman, Anna Korhonen, Roi Reichart, Haim Dubossarsky, Arya D. McCarthy, Edoardo Maria Ponti, Ivan Vulić and Ekaterina Vylomova
• BlackboxNLP 2019: Analyzing and interpreting neural networks for NLP
  • Tal Linzen, Grzegorz Chrupała, Yonatan Belinkov and Dieuwke Hupkes
• The Fourth Arabic Natural Language Processing Workshop
  • Wassim El-Hajj, Lamia Hadrich-Belguith, Fethi Bougares, Walid Magdy, Imed Zitouni, Nadi Tomeh, Mahmoud El-Haj, Wajdi Zaghouani, Nizar Habash, Hend Al-Khalifa, Houda Bouamor, Kareem Darwish and Mona Diab
• ArgMining 2019: The 6th International Workshop on Argument Mining
  • Benno Stein and Henning Wachsmuth
• The Thirteenth Linguistic Annotation Workshop (LAW XIII)
  • Annemarie Friedrich and Deniz Zeyrek
• NLP for Conversational AI
  • Tsung-Hsien Wen, Pei-Hao Su, Yun-Nung Chen, Dilek Hakkani-Tur, Mike Lewis, Minh-Thang Luong, Anuj Kumar, and Tania Bedrax-Weiss
• The Third Workshop on Abusive Language Online
  • Zeerak Waseem, Vinodkumar Prabhakaran, Joel Tetreault and Sarah Roberts
• Second Workshop on Storytelling (StoryNLP)
  • Francis Ferraro, Ting-Hao Huang, Stephanie M. Lukin and Margaret Mitchell
• The First International Workshop on Designing Meaning Representations (DMR)
  • Nianwen Xue, Jan Hajic, Chu-Ren Huang, Stephan Oepen, Martha Palmer, James Pustejovsky and William Croft

Workshops on August 2

• Joint Workshop on Multiword Expressions and WordNet (MWE-WN 2019)
  • Verginica Barbu Mititelu, Francis Bond, Jelena Mitrovic, Carla Parra Escartín and Agata Savary
• 1st International Workshop on Computational Approaches to Historical Language Change
  • Nina Tahmasebi, Adam Jatowt, Lars Borin and Yang Xu
• The 14th Workshop on Innovative Use of NLP for Building Educational Applications (BEA)
  • Ekaterina Kochmar, Claudia Leacock, Helen Yannakoudakis, Ildikó Pilán, Nitin Madnani and Torsten Zesch
• 4th Workshop on Representation Learning for NLP (RepL4NLP-2019)
  • Isabelle Augenstein, Spandana Gella, Sebastian Ruder, Katharina Kann, Burcu Can, Johannes Welbl, Alexis Conneau, Xiang Ren and Marek Rei
• The Sixteenth SIGMORPHON Workshop on Computational Research in Phonetics, Phonology, and Morphology
  • Garrett Nicolai and Ryan Cotterell
• Deep Learning & Formal Languages: Building Bridges
  • Jason Eisner, Matthias Gallé, Jeffrey Heinz, Ariadna Quattoni and Guillaume Rabusseau
• Gender Bias in Natural Language Processing
  • Marta R. Costa-jussà, Christian Hardmeier, Kellie Webster and Will Radford
• BSNLP-2019: The 7th Biennial Workshop on Balto-Slavic NLP Sponsored by ACL Special Interest Group on Slavic NLP
  • Tomaz Erjavec, Preslav Nakov, Jakub Piskorski, Lidia Pivovarova, Jan Snajder, Josef Steinberger and Roman Yangarber
• Social Media Mining for Health Research and Applications (#SMM4H)
  • Graciela Gonzalez-Hernandez, davy weissenbacher, Michael J. Paul, Abeed Sarker, Ari Klein and Karen O'Conno

5. Student Research Workshop (SRW)

• SRW Organisers

  Fernando Alva-Manchego, University of Sheffield
  Eunsol Choi, University of Washington
  Daniel Khashabi, University of Pennsylvania

• Faculty Advisors to the Student Research Workshop

  Hannaneh Hajishirzi, University of Washington
  Aurelie Herbelot, University of Trento
  Scott Yih, Allen Institute for Artificial Intelligence
  Yue Zhang, Westlake University

Information about the Student Research Workshop (SRW) for ACL 2019 has been posted at the following URL: https://sites.google.com/view/acl19studentresearchworkshop/. The SRW has applied for a NSF grant of $18,000. This includes travel stipends for up to 12 student participants and organisers. Contact has been made with Priscilla to investigate other funding opportunities, as well as the Student Volunteer Program, which helps students cover registration fee to the main conference. Submission deadlines for the SRW are as follows:

  Pre-submission mentoring deadline: March 5, 2019
  Pre-submission mentoring feedback: April 2, 2019
  Paper submission deadline: April 26, 2019
  Review deadline:  May 7, 2019
  Acceptance notification: May 24, 2019
  Camera-ready deadline: June 3 , 2019
  Travel grant application deadline: June 12, 2019
  Travel grant notification: June 15, 2019

For the program committee, we have recruited 55 PC members so far. We plan on inviting more people considering that the number of submissions is likely to be larger than originally estimated. Pre-submission and post-acceptance mentors are recruited mostly from the PC members. We have 15 pre-submission mentors and 14 post-acceptance mentors. We have received 63 submissions for pre-mentoring. The SRW START site has been set up. The SRW CFP has been distributed to the following lists: ACL member portal, MT-list, and Corpora, as well as on our official Twitter account (@acl_srw).

Tutorial Co-Chairs

Preslav Nakov, Qatar Computing Research Institute, HBKU

Alexis Palmer, University of North Texas

In total, we received 45 proposals for the joint ACL/NAACL/EMNLP-IJCNLP call, and we accepted 22 of them. 34 tutorials had ACL as the preferred conference: 8 were selected, 19 were rejected, 7 were selected for their alternative choice. A total of 9 tutorials were selected for ACL. See the titles, instructors and abstracts below.

Morning Tutorials

T1: Latent Structure Models for Natural Language Processing

André F. T. Martins, Tsvetomila Mihaylova, Nikita Nangia and Vlad Niculae

Latent structure models are a powerful tool for modeling compositional data, discovering linguistic structure, and building NLP pipelines. They are appealing for two main reasons: they allow incorporating structural bias during training, leading to more accurate models; and they allow discovering hidden linguistic structure, which provides better interpretability.

This tutorial will cover recent advances in discrete latent structure models. We discuss their motivation, potential, and limitations, then explore in detail three strategies for designing such models: gradient approximation, reinforcement learning, and end-to-end differentiable methods. We highlight connections among all these methods, enumerating their strengths and weaknesses. The models we present and analyze have been applied to a wide variety of NLP tasks, including sentiment analysis, natural language inference, language modeling, machine translation, and semantic parsing. Examples and evaluation will be covered throughout. After attending the tutorial, a practitioner will be better informed about which method is best suited for their problem.

T2: Graph-Based Meaning Representations: Design and Processing

Alexander Koller, Stephan Oepen and Weiwei Sun

The last several years have seen extensive interest in encoding and processing sentence meaning in the form of labeled directed graphs. Frameworks instantiating this line of research include e.g. Abstract Meaning Representation, graph-based rendering of Minimal Recursion Semantics, Bilexical Semantic Dependency Graphs, and Universal Conceptual Cognitive Annotation. Complementary to advanced vector-based representations of meaning, parsing to such hierarchically structured and discrete semantic representations has been a cornerstone of Natural Language Understanding since the early days and will continue to make essential contributions to ‘making sense’ of natural language.

This tutorial will (a) briefly review relevant background in formal and linguistic semantics; (b) semi-formally define a unified abstract view on different flavors of semantic graphs and associated terminology; (c) survey common frameworks for graph-based meaning representation and available graph banks; and (d) offer a technical overview of a representative selection of different parsing approaches. The ultimate goal is to provide a unified view on different semantic graph banks and associated parsing work and, thus, to reduce the barrier to entry for NLP developers and users to benefit from recent successes and best practices in this exciting field.

T3: Discourse Analysis and Its Applications

Shafiq Joty, Giuseppe Carenini, Raymond Ng and Gabriel Murray

Discourse processing is a suite of Natural Language Processing (NLP) tasks to uncover linguistic structures from texts at several levels, which can support many text mining applications. This involves identifying the topic structure, the coherence structure, the coreference structure, and the conversation structure for conversational discourse. Taken together, these structures can inform text summarization, essay scoring, sentiment analysis, machine translation, information extraction, question answering, and thread recovery.

The tutorial starts with an overview of basic concepts in discourse analysis -- monologue vs. conversation, synchronous vs. asynchronous conversation, and key linguistic structures in discourse analysis. It then covers traditional machine learning methods along with the most recent works using deep learning, and compares their performances on benchmark datasets. For each discourse structure we describe, we show its applications in downstream text mining tasks. Methods and metrics for evaluation are discussed in detail. We conclude the tutorial with an interactive discussion of future challenges and opportunities.

T4: Computational Analysis of Political Texts: Bridging Research Efforts Across Communities

Goran Glavaš, Federico Nanni and Simone Paolo Ponzetto

The usage of computational methods for the study of political texts has drastically expanded in scope, allowing for a sustained growth of the text-as-data community in political science. NLP methods have been extensively used for a number of analyses and tasks, including inferring policy positions of actors from textual evidence, detecting topics in political documents, and analyzing stylistic aspects of political communication (e.g., assessing the role of language ambiguity in framing the political agenda). Political scientists created resources and used available NLP methods to process textual data largely in isolation from the NLP community. At the same time, NLP researchers addressed closely related tasks such as election prediction, ideology classification, and stance detection. These two communities still remain largely agnostic of one another, with NLP researchers mostly unaware of interesting applications and use cases in political science and political scientists lagging behind in applying cutting-edge NLP methods to their problems.

This tutorial will provide a comprehensive overview of the body of work on computational analysis of political texts. We first look at the role that textual data play in political analyses and then proceed to examine the concrete resources and tasks addressed by the text-as-data political science community. Next, we present the research efforts carried out so far by the NLP community with a focus on methods for the topical analysis of political texts, covering both unsupervised topic induction and supervised topic classification studies. Finally, we conclude the tutorial by focusing on political text scaling, a challenging task on ideology detection from textual data, which is at the center of quantitative political science and has recently also attracted attention from NLP scholars.

T5: Wikipedia as a Resource for Text Analysis and Retrieval

Marius Pasca

Articles within Wikipedia collectively form what might be the largest, publicly-available, decentralized resource of unstructured or semi-structured knowledge, reflecting an ever-growing number of topics of interest to people, in general, and Web users, in particular. This tutorial examines the role of Wikipedia in tasks related to text analysis and retrieval. Text analysis tasks, which take advantage of Wikipedia, include coreference resolution, word sense and entity disambiguation and information extraction. In information retrieval, a better understanding of the structure and meaning of queries helps in matching queries against documents, clustering search results, answer and entity retrieval and retrieving knowledge panels for queries asking about popular entities. The tutorial reviews characteristics, advantages and limitations of Wikipedia relative to other existing, human-curated resources of knowledge; derivative resources, created by converting semi-structured content in Wikipedia into structured data; and the role of Wikipedia and its derivatives in text analysis and in enhancing information retrieval.

Afternoon Tutorials

T6: Deep Bayesian Natural Language Processing

Jen-Tzung Chien

This introductory tutorial addresses the advances in deep Bayesian learning for natural language with ubiquitous applications ranging from speech recognition to document summarization, text classification, text segmentation, information extraction, image caption generation, sentence generation, dialogue control, sentiment classification, recommendation system, question answering and machine translation, to name a few. Traditionally, "deep learning" is taken to be a learning process where the inference or optimization is based on the real-valued deterministic model. The "semantic structure" in words, sentences, entities, actions and documents drawn from a large vocabulary may not be well expressed or correctly optimized in mathematical logic or computer programs. The "distribution function" in discrete or continuous latent variable model for natural language may not be properly decomposed or estimated.

This tutorial addresses the fundamentals of statistical models and neural networks, and focus on a series of advanced Bayesian models and deep models including hierarchical Dirichlet process, Chinese restaurant process, hierarchical Pitman-Yor process, Indian buffet process, recurrent neural network, long short-term memory, sequence-to-sequence model, variational auto-encoder, generative adversarial network, attention mechanism, memory-augmented neural network, skip neural network, stochastic neural network, policy neural network, and Markov recurrent neural network. We present how these models are connected and why they work for a variety of applications on symbolic and complex patterns in natural language. The variational inference and sampling method are formulated to tackle the optimization for complicated models. The word and sentence embeddings, clustering and co-clustering are merged with linguistic and semantic constraints. A series of case studies are presented to tackle different issues in deep Bayesian learning and understanding. At last, we will point out a number of directions and outlooks for future studies.

T7: Unsupervised Cross-Lingual Representation Learning

Sebastian Ruder, Anders Søgaard and Ivan Vulić

In this tutorial, we provide a comprehensive survey of the exciting recent work on cutting-edge weakly-supervised and unsupervised cross-lingual word representations. After providing a brief history of supervised cross-lingual word representations, we focus on: 1) how to induce weakly-supervised and unsupervised cross-lingual word representations in truly resource-poor settings where bilingual supervision cannot be guaranteed; 2) critical examinations of different training conditions and requirements under which unsupervised algorithms can and cannot work effectively; 3) more robust methods for distant language pairs that can mitigate instability issues and low performance for distant language pairs; 4) how to comprehensively evaluate such representations; and 5) diverse applications that benefit from cross-lingual word representations (e.g., MT, dialogue, cross-lingual sequence labeling and structured prediction applications, cross-lingual IR).

T8: Advances in Argument Mining

Katarzyna Budzynska and Chris Reed

Argument and debate form cornerstones of civilised society and of intellectual life. Processes of argumentation run our governments, structure scientific endeavour and frame religious belief. As our understanding of how arguments are assembled, are interpreted and have impact has improved, so it has become possible to frame computational questions about how it might be possible for machines to model and replicate the processes involved in identifying, reconstructing, interpreting and evaluating reasoning expressed in natural language arguments.

This course aims to introduce students to an exciting and dynamic area that has witnessed remarkable growth over the past 36 months. Argument mining builds on opinion mining, sentiment analysis and related to tasks to automatically extract not just what people think, but why they hold the opinions they do. From being largely beyond the state of the art barely five years ago, there are now many hundreds of papers on the topic and millions of dollars of commercial and research investment. This tutorial provides a synthesis of the major advances in the area over the past three years.

T9: Storytelling from Structured Data and Knowledge Graphs : An NLG Perspective

Abhijit Mishra, Anirban Laha, Karthik Sankaranarayanan, Parag Jain and Saravanan Krishnan

In this tutorial, we discuss the foundational, methodological, and system development aspects of translating structured data (such as data in tabular form) and knowledge bases (such as knowledge graphs) into natural language discourses. The tutorial covers challenges and approaches for Natural Language Generation (NLG), with a primary focus on the (structured) data-to-text paradigm. Our attendees will be able to take home the following: (1) the basic as well as trending ideas around how modern NLP and NLG techniques could be applied to describe and summarize textual data that is non-linguistic in nature or has some structure, and (2) a few interesting open-ended questions, which could lead to significant research contributions in future.

We will provide an overview of diverse approaches ranging from data representation techniques to domain adaptable solutions for the data-to-text problem setting. Various solutions, starting from traditional rule-based/heuristic-driven, modern data-driven and ultra-modern deep-neural style architectures will be discussed, followed by a brief discussion on evaluation and quality estimation. Since large scale domain independent labelled (parallel) data is rarely available for data-to-text problems, a significant portion of the tutorial will be dedicated towards unsupervised, scalable, and domain-adaptable approaches.

System Demonstration Co-Chairs

Enrique Alfonseca, Google

Marta R. Costa-jussà, Technical University of Catalonia

The call for papers for system demonstrations is online at the conference website: http://www.acl2019.org/EN/call-for-papers.xhtml.

The technical Program Committee is in place, with 65 reviewers who have so far accepted the invitation to participate.

The deadline for submissions will be on April 11th. We currently have four submissions.

Important Dates

  Paper submission deadline: April 11th, 2019
  Notification of acceptance: May 17th, 2019
  Camera-ready submission: June 10th, 2019

Publication Co-Chairs

Douwe Kiela, Facebook

Ivan Vulić, University of Cambridge

Shay Cohen, University of Edinburgh (Advisory)

Kevin Gimpel, Toyota Technological Institute at Chicago (Advisory)

Starting from the style files from ACL 2019 and NAACL 2019, we have produced new tex style files for ACL 2019. The main improvement compared to the previous style files is concerned with fixing the frequent and non-deterministically occurring tex bug which would happen when hyperref is used under pdftex, and a citation spills across a page boundary. As a general comment addressed to the entire community, we have also been advised to balance the gender disparity in references provided in the style files. Since the suggestion came too close to the submission deadline, we have decided not to update the style files so close to the deadline, but we strongly suggestion to integrate the proposed change into all future style files for major conferences. Other activities from the publication chairs included compiling a list of all workshops collocated with the main ACL event along with the information to speed up communication with all book chairs. Finally, senior publication chairs have finalised an FAQ for future publication chairs, which will be immensely useful for less experienced chairs at the upcoming conferences, available here: https://docs.google.com/document/d/1D41tt7Zj0_xjbxEaNWT6Y9NUcoBu0Lkp08IwlwX155E/edit#heading=h.nu0qu1d3gvr6

Publicity Co-Chairs

Felice Dell'Orletta - CNR-ILC, Pisa

Lucia Passaro - University of Pisa

Sara Tonelli - Bruno Kessler Foundation

Dissemination:

Dissemination of the news about the conference on the ACL 2019 social media channels. We created a new Twitter and Facebook account for the conference, the Instagram account remained the same of the previous editions (we just added the email lucia.passaro@fileli.unipi.it to forward emails sent to instagram@aclweb.org).

  Facebook: https://www.facebook.com/acl2019Italy/
  Twitter: https://twitter.com/acl2019_italy
  Instagram: https://www.instagram.com/aclnlp/

All news and updates were disseminated also through mailing lists relevant to NLP and machine learning. ACL 2019 Chair Blog:

Prepared the structure of the Blog. At the moment the website is hosted at University of Pisa (protected by password).

  URL: http://acl2019pcblog.fileli.unipi.it 
  PSD: $acl2019!protected!$

TBD: If we would like to make the website accessible to a different link, the best choice could be to define, from the desired domain (e.g. < pcblog.acl2019.org >) a CNAME to < vm-131-114-72-235.unipi.it. > In this case, it could be appropriate to have a valid SSL (i.e. signed by a recognized C.A.).

Conference Handbook Chair

Elena Cabrio, University Côte d'Azur

Local Sponsorship Co-Chairs

Roberto Basili, University of Rome Tor Vergata

Giovanni Semeraro, University of Bari

Business Office

Priscilla Rasmussen, ACL

The ACL Office will manage the registrations as usual, including developing the registration statistics we always have and assure that the anti-harassment policy is included on the registration form itself. I will act as an advisor to the PCO on finances (including interim deposits for local operating costs), exhibits, issuing visa invitation letters and overall space, catering and audio visual setup along with the Local Arrangements team who are really quite good and dedicated. The sponsorships have been slowly coming in. Commitments are only now being made by some of the large and continuing sponsor companies. And the Local Sponsorship team of Roberto Basili (University of Rome Tor Vergata) and Giovanni Semeraro (University of Bari) are actively reaching out to firms throughout Italy. To date, these are our commitments: