CRAC 2020 was the Third Workshop on Computational Models of Reference, Anaphora and Coreference, held at COLING 2020 (Dec 12, online).
Background: The end of Discourse Anaphora and Anaphor Resolution Colloquium series in 2011 scattered the research papers on anaphora/coreference resolution among very different fora until a common event in Computational Linguistics entirely dedicated to this area was revived in 2016 with the Coreference Beyond OntoNotes (CORBON 2016) workshop co-located with NAACL and CORBON 2017 co-located with EACL. In 2018 its focus, perceived as too narrow, was broadened to cover all cases of computational modelling of reference, anaphora, and coreference with CRAC 2018 and CRAC 2019 workshops held again at NAACL. Following the recent advances in application of word embeddings and deep neural networks to various NLP tasks, we believe that the task of cross-lingual coreference resolution can also benefit from the new perspective.
Objectives: The aim of the workshop is to provide a forum where work on all aspects of computational work on anaphora resolution and annotation, including both coreference and types of anaphora such as bridging references resolution and discourse deixis, can be presented.
The workshop welcomed submissions describing both theoretical and applied computational work on anaphora/coreference resolution, including on languages other than English, and less-researched types of anaphora such as bridging references. Topics of interest include, but are not limited to:
Coreference resolution for less-researched languages
Annotation and interpretation of anaphoric relations, including relations other than identity coreference (e.g., bridging references, reference to abstract entities)
Investigation of difficult cases of anaphora/coreference and their resolution (e.g., zero anaphora, ellipsis, event coreference)
Anaphora/coreference resolution in noisy data (e.g. in speech, social media)
Use of anaphora/coreference resolution for discourse processing
Annotation tools and scheme for anaphora/coreference resolution
Cross-lingual anaphora/coreference resolution
Cross-domain anaphora/coreference resolution
Applications of anaphora/coreference resolution (e.g., summarization, machine translation)
Evaluation methodology for anaphora/coreference resolution
Workshop papers due: Oct 8, 2020
Notification of acceptance: Oct 23, 2020
Camera-ready papers due: Nov 1, 2020
Pre-recorded videos due: Nov 10, 2020
Workshop date: Dec 12, 2020
Welcome and Invited Talk
14:00–14:05: Introduction (Maciej Ogrodniczuk, Sameer Pradhan, Yulia Grishina and Vincent Ng)
14:05–15:00: Anaphora Resolution beyond OntoNotes (Juntao Yu)
Short Break
Paper Summary Session 1: Mention Detection and Deep Learning Approaches (Chair: Vincent Ng)
15:10–15:20: Anaphoric Zero Pronoun Identification: A Multilingual Approach (Abdulrahman Aloraini and Massimo Poesio)
15:20–15:30: Partially-Supervised Mention Detection (Lesly Miculicich and James Henderson)
15:30–15:40: E.T.: Entity-Transformers. Coreference Augmented Neural Language Model for Richer Mention Representations via Entity-Transformer Blocks (Nikolaos Stylianou and Ioannis Vlahavas)
15:40–15:50: Neural Coreference Resolution for Arabic (Abdulrahman Aloraini, Juntao Yu and Massimo Poesio)
15:50–16:00: Sequence-to-Sequence Networks Learn the Meaning of Reflexive Anaphora (Robert Frank and Jackson Petty)
16:00–16:10: Sequence to Sequence Coreference Resolution (Gorka Urbizu, Ander Soraluze and Olatz Arregi)
Short Break
Paper Summary Session 2: Resources, Evaluation and Beyond the Identity of Reference (Chair: Maciej Ogrodniczuk)
16:20–16:30: A Benchmark of End-to-End and Deterministic Coreference Resolution of Dutch Novels and News (Corbèn Poot and Andreas van Cranenburgh)
16:30–16:40: A Dataset for Anaphora Analysis in French E-mails (Hani Guenoune, Cédric Lopez, Kevin Cousot, Melissa Mekaoui and Mathieu Lafourcade)
16:40–16:50: Integrating Knowledge Graph Embeddings to Improve Mention Representation for Bridging Anaphora Resolution (Onkar Pandit, Pascal Denis and Liva Ralaivola)
16:50–17:00: Reference to Discourse Topics: Introducing “Global” Shell Nouns (Fabian Simonjetz)
17:00–17:10: TwiConv: A Coreference-annotated Corpus of Twitter Conversations (Berfin Aktaş and Annalena Kohnert)
17:10–17:20: Predicting Coreference in Abstract Meaning Representations (Tatiana Anikina, Alexander Koller and Michael Roth)
Short Break
Paper Summary Session 3: Applications (Chair: Sameer Pradhan)
17:30–17:40: It’s Absolutely Divine! Can Fine-Grained Sentiment Analysis Benefit from Coreference Resolution? (Orphee De Clercq and Veronique Hoste)
17:40–17:50: Enhanced Labelling in Active Learning for Coreference Resolution (Vebjørn Espeland, Beatrice Alex and Benjamin Bach)
17:50–18:00: Coreference Strategies in English-German Translation (Ekaterina Lapshinova-Koltunski, Marie-Pauline Krielke and Christian Hardmeier)
18:00–18:10: Reference in Team Communication for Robot-Assisted Disaster Response: An Initial Analysis (Natalia Skachkova and Ivana Kruijff-Korbayova)
18:10–18:20: Resolving Pronouns in Twitter Streams: Context can Help! (Anietie Andy, Chris Callison-Burch and Derry Tanti Wijaya)
Short Break
Plenary Session on Universal Anaphora (Chair: Massimo Poesio)
18:30–18:45: Universal Anaphora 1.0 (Massimo Poesio and Maciej Ogrodniczuk)
18:45–18:50: Coreference at Georgetown: GUM, ANNIS and Amalgum (Amir Zeldes)
18:50–18:55: Revisiting the OntoNotes Sweet-Spot Enroute to Universal Anaphora (Sameer Pradhan, Amir Zeldes, Yilun Zhu, Massimo Poesio and Juntao Yu)
18:55–19:00: Coreference in Prague dependency corpora and the UA perspective (Anna Nedoluzhko)
19:00–19:05: CODI Shared Task: Anaphora Resolution in Dialogues (Michael Strube, Massimo Poesio, Carolyn Rose, Sopan Khosla, Ramesh Manuvinakurike, Vincent Ng)
19:05–19:10: Discussion
Best Paper Award Session
19:10–19:15: Best Paper Award and Closing of the Workshop (Maciej Ogrodniczuk, Sameer Pradhan, Yulia Grishina and Vincent Ng)
Oral presentations (for all accepted papers) have been pre-recorded and are still available at the workshop Web page in Underline. Videos take up to 15 mins for long papers and 10 mins for short papers. They were available for viewing by all attendees before the start of the workshop. We believe that this new virtual format should allow more time for the participants to watch the presentations at their leisure and convenience before each session and more interaction between attendees during the Q&A sessions.
On December 12 an unofficial pre-workshop morning session (between 9:00 and 13:00, all times CET) was organized on Zoom and contained all pre-recorded talks played one by one to let latecomers catch up with the content.
The workshop started at 14:00 CET and was composed of a one-hour invited talk, three one-hour paper summary sessions and the final plenary session on Universal Anaphora. Every paper session contained six 10-minute slots for each paper split into 5-minute live talk by paper authors (the same amount of time for long and short papers) summarizing their results, and a 5-minute Q&A session (immediately after each paper).
Workshop proceedings are available in the ACL Anthology.
Antonio Branco, University of Lisbon
Dan Cristea, Alexandru Ioan Cuza University of Iasi
Stephanie Dipper, University of Bochum
Yulia Grishina, Amazon
Veronique Hoste, Ghent University
Sandra Kübler, Indiana University
Sobha Lalitha Devi, AU-KBC Research Center, Anna University of Chennai
Emmanuel Lassalle, Machina Capital, Paris
Costanza Navaretta, University of Copenhagen
Anna Nedoluzhko, Charles University in Prague
Michal Novak, Charles University in Prague
Constantin Orasan, University of Surrey
Massimo Poesio, Queen Mary University of London
Marta Recasens, Google
Yannick Versley, Amazon
Heike Zinsmeister, University of Hamburg
Maciej Ogrodniczuk, Institute of Computer Science, Polish Academy of Sciences
Sameer Pradhan, University of Pennsylvania and cemantix
Yulia Grishina, Amazon
Vincent Ng, University of Texas at Dallas