Workshop Description
Encouraged by the 2019 and 2020 workshops, the aim of the third edition of SIGTYP workshop is to act as a platform and a forum for the exchange of information between typology-related research, multilingual NLP, and other research areas that can lead to the development of truly multilingual NLP methods. The workshop is specifically aimed at raising awareness of linguistic typology and its potential in supporting and widening the global reach of multilingual NLP, as well as at introducing computational approaches to linguistic typology. It will foster research and discussion on open problems, not only within the active community working on cross- and multilingual NLP but also inviting input from leading researchers in linguistic typology. Starting from 2021, the workshop will be dedicated to a shared theme, a central topic that most keynote talks and discussions will be focused on. For instance, in 2021 we would like to follow up a recent debate on linguistic diversity (p-linguistics, the study of individual languages) and universalism (g-linguistics, the study of Human Language), see Haspelmath (2020); Evans and Levinson (2009). The process of annotation of highly cross-lingual corpora (such as recently introduced Universal Dependencies (Nivre et al., 2016) and UniMorph (Sylak-Glassman, 2016)) requires distinguishing language-specific, historically accidental phenomena from truly universal phenomena such as the fact that all languages have demonstratives (Diessel, 2014). Our workshop will serve as a platform to enable fruitful discussions on the topic.



Main Topics
The workshop will provide focussed discussion on a range of topics, including (but not limited to) the following:

— Language-independence in training, architecture design, and hyperparameter tuning. Is it possible (and if yes, how) to unravel unknown biases that hinder the cross-lingual performance of NLP algorithms and to leverage the knowledge on such biases in NLP algorithms?

—Integration of typological features in language transfer and joint multilingual learning. In addition to established techniques such as “selective sharing”, are there alternative ways to encoding heterogeneous external knowledge in machine learning algorithms?

— New applications. The application of typology to currently uncharted territories, i.e. the use typological information in NLP tasks where such information has not been investigated yet.

— Automatic inference of typological features. The pros and cons of existing techniques (e.g. heuristics derived from morphosyntactic annotation, propagation from features of other languages, supervised Bayesian and neural models) and discussion on emerging ones.

— Typology and interpretability. The use of typological knowledge for interpretation of hidden representations of multilingual neural models, multilingual data generation and selection, and typological annotation of texts.

— Improvement and completion of typological databases. Combining linguistic knowledge and automatic data-driven methods towards the joint goal of improving the knowledge on cross-linguistic variation and universals.

— Linguistic diversity and universals. Challenges of cross-lingual annotation. Which linguistic phenomena or categories should be considered universal? How should they be annotated?



Submission Format
SIGTYP 2021 will consider both archival and non-archival work. We will issue a call for extended abstract submissions (non-archival) and general paper submissions (archival). The accepted submissions will be presented at the workshop, providing new insights and ideas. In terms of non-archival work, we will allow 2-page abstracts of already published work or work in progress. This way, we will not discourage researchers from preferring main conference proceedings, at the same time ensuring that interesting and thought-provoking research is presented at the workshop. In addition, we will consider general archival submissions (4-page and 8-page papers).



Program Committee
Željko Agić, Corti
Emily Ahn, University of Washington
Isabelle Augenstein, University of Copenhagen
Emily Bender, University of Washington
Johannes Bjerva, University of Copenhagen
Claire Bowern, Yale University
Miriam Butt, University of Konstanz
Giuseppe Celano, Leipzig University
Agnieszka Falenska, University of Stuttgart
Richard Futrell, University of California, Irvine
Elisabetta Ježek, University of Pavia
Gerhard Jäger, University of Tubingen
John Mansfield, The University of Melbourne
Paola Merlo, University of Geneva
Joakim Nivre, Uppsala University
Robert Östling, Stockholm University
Thomas Proisl, FAU Erlangen-Nurnberg
Michael Regan, University of New Mexico
Ella Rabinovich, University of Toronto
Tanja Samardžić, University of Zurich
Richard Sproat, Google Japan
Sabine Stoll, University of Zurich
Daan van Esch, Google AI
Giulia Venturi, ILC ``Antonio Zampolli''
Nidhi Vyas, Apple
Ada Wan, University of Zurich
Eleanor Chodroff, University of York
Elizabeth Salesky, Johns Hopkins University
Sabrina Mielke, Johns Hopkins University
Edoardo Ponti, University of Cambridge


Organizers
Ekaterina Vylomova Elizabeth Salesky Sabrina Mielke Gabriella Lapesa Ritesh Kumar Edoardo M. Ponti
Harald Hammarström Ivan Vulić Anna Korhonen Roi Reichart Ryan Cotterell