2021.sigtyp-1.0 - ACL Anthology (PDF)

SIGTYP 2021 The 3rd Workshop on Research in Computational Typology and Multilingual NLP Proceedings of the Workshop June 10, 2021  ©2021 The Association for Computational Linguistics Order copies of this and other ACL proceedings from: Association for Computational Linguistics (ACL) 209 N. Eighth Street Stroudsburg, PA 18360 USA Tel: +1-570-476-8006 Fax: +1-570-476-0860 [email protected] ISBN 978-1-954085-34-3 ii  SIGTYP 2021 is the third edition of the workshop for typology-related research and its integration into multilingual Natural Language Processing (NLP). The workshop is co-located with the 2021 Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL 2021), which takes place virtually this year. Our workshop includes a shared task on robust language identification from speech. The final program of SIGTYP contains 4 keynote talks, 3 shared task papers, 10 archival papers, and 14 extended abstracts. This workshop would not have been possible without the contribution of its program committee, to whom we would like to express our gratitude. We should also thank Claire Bowern, Miryam de Lhoneux, Johannes Bjerva, and David Yarowsky for kindly accepting our invitation as invited speakers. The workshop is generously sponsored by Google. Please find more details on the SIGTYP 2021 website: https://sigtyp.github.io/ws2021. html iii   Organizing Committee: Ekaterina Vylomova, University of Melbourne Elizabeth Salesky, Johns Hopkins University Sabrina Mielke, Johns Hopkins University Gabriella Lapesa, University of Stuttgart Ritesh Kumar, Bhim Rao Ambedkar University Harald Hammarström, Uppsala University Ivan Vuli´c, University of Cambridge Anna Korhonen, University of Cambridge Roi Reichart, Technion – Israel Institute of Technology Edoardo M. Ponti, Mila Montreal and University of Cambridge Ryan Cotterell, ETH Zurich Program Committee: Željko Agi´c, 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, 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 Samardži´c, 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 M. Ponti, University of Cambridge Damián Blasi, Harvard University Adina Williams, Facebook Ivan Vuli´c, University of Cambridge Arturo Oncevay, University of Edinburgh Koel Dutta Chowdhury, Saarland University v  Elena Klyachko, National Research University Higher School of Economics Alexey Sorokin, Moscow State University Sylvain Kahane, Université Paris Nanterre Taraka Rama, University of North Texas Harald Hammarström, Max Planck Institute for the Science of Human History Olga Lyashevskaya, National Research University Higher School of Economics Kaushal Kumar Maurya, IIT Hyderabad Johann-Mattis List, Max Planck Institute for the Science of Human History Garrett Nicolai, University of British Columbia Yevgeni Berzak, Technion – Israel Institute of Technology Olga Zamaraeva, University of Washington Zoey Liu, Boston College Jeff Good, University at Buffalo Priya Rani, National University of Ireland Silvia Luraghi, University of Pavia Beata Trawinski, University of Vienna Miryam de Lhoneux, University of Copenhagen Kemal Kurniawan, University of Melbourne Andreas Shcerbakov, University of Melbourne Ritesh Kumar, Bhim Rao Ambedkar University Invited Speakers: Claire Bowern, Yale University Miryam de Lhoneux, Uppsala University / KU Leuven / University of Copenhagen Johannes Bjerva, Aalborg University David Yarowsky, Johns Hopkins University vi  Table of Contents OTEANN: Estimating the Transparency of Orthographies with an Artificial Neural Network Xavier Marjou . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1 Inferring Morphological Complexity from Syntactic Dependency Networks: A Test Guglielmo Inglese and Luca Brigada Villa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 A Universal Dependencies Corpora Maintenance Methodology Using Downstream Application Ran Iwamoto, Hiroshi Kanayama, Alexandre Rademaker and Takuya Ohko . . . . . . . . . . . . . . . . . . 23 Improving Cross-Lingual Sentiment Analysis via Conditional Language Adversarial Nets Hemanth Kandula and Bonan Min. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .32 Improving the Performance of UDify with Linguistic Typology Knowledge Chinmay Choudhary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 FrameNet and Typology Michael Ellsworth, Collin Baker and Miriam R. L. Petruck . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 Family of Origin and Family of Choice: Massively Parallel Lexiconized Iterative Pretraining for Severely Low Resource Text-based Translation Zhong Zhou and Alexander Waibel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 Measuring Prefixation and Suffixation in the Languages of the World Harald Hammarström . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Predicting and Explaining French Grammatical Gender Saumya Sahai and Dravyansh Sharma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 Morph Call: Probing Morphosyntactic Content of Multilingual Transformers Vladislav Mikhailov, Oleg Serikov and Ekaterina Artemova. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .97 SIGTYP 2021 Shared Task: Robust Spoken Language Identification Elizabeth Salesky, Badr M. Abdullah, Sabrina Mielke, Elena Klyachko, Oleg Serikov, Edoardo Maria Ponti, Ritesh Kumar, Ryan Cotterell and Ekaterina Vylomova . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 Language ID Prediction from Speech Using Self-Attentive Pooling Roman Bedyakin and Nikolay Mikhaylovskiy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130 A ResNet-50-Based Convolutional Neural Network Model for Language ID Identification from Speech Recordings Celano Giuseppe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136 Anlirika: An LSTM–CNN Flow Twister for Spoken Language Identification Andreas Scherbakov, Liam Whittle, Ritesh Kumar, Siddharth Singh, Matthew Coleman and Ekate- rina Vylomova . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 vii  Non-archival Abstracts Graph Convolutional Network for Swahili News Classification Alexandros Kastanos and Tyler Martin In this work, we demonstrate the ability of Text Graph Convolutional Network (Text GCN) to surpass the performance of traditional natural language processing benchmarks on the task of semi-supervised Swahili news categorisation. Our experiments highlight the more severely label-restricted context often facing low-resourced African languages. We build on this finding by presenting a memory-efficient vari- ant of Text GCN which replaces the naive one-hot node representation with a bag of words representation. Exploring Linguistic Typology Features in Multilingual Machine Translation Oscar Moreno and Arturo Oncevay We explore whether linguistic typology features can impact multilingual machine translation perfor- mance (many-to-English) by using initial pseudo-tokens and factored language-level embeddings. With 20 languages from different families or groups, we observed that the features of “Order of Subject (S), Object (O) and Verb (V)”, “Position of Negative Word with respect to S-O-V” and “Prefixing vs. Suf- fixing in Inflectional Morphology” provided slight improvements in low-resource language-pairs despite not overcoming the average performance for all languages. Multilingual Slot and Intent Detection (xSID) with Cross-lingual Auxiliary Tasks Rob van der Goot, Ibrahim Sharaf, Aizhan Imankulova, Ahmet Üstün, Marija Stepanovi´c, Alan Ramponi, Siti Oryza Khairunnisa, Mamoru Komachi and Barbara Plank Digital assistants are becoming an integral part of everyday life. However, commercial digital assistants are only available for a limited set of languages (as of March 2020, between 8 to around 20 languages). Because of this, a vast amount of people can not use these devices in their native tongue. In this work, we focus on two core tasks within the digital assistant pipeline: intent classification and slot detection. Intent classification recovers the goal of the utterance, whereas slot detection identifies important prop- erties regarding this goal. Besides introducing a novel cross-lingual dataset for these tasks, consisting of 13 languages, we evaluate a variety of models: 1) multilingually pretrained transformer-based models, 2) we supplement these models with auxiliary tasks to evaluate whether multi-task learning can be ben- eficial, and 3) annotation transfer with neural machine translation. Plugins for Structurally Varied Languages in XMG Framework Valeria Generalova This paper aims to suggest an XMG-based design of metagrammatical classes storing language-specific information on a multilingual grammar engineering project. It also presents a method of reusing the information from WALS. The principal claim is the hierarchy of features and the modular architecture of feature structures.  Modeling Linguistic Typology - A Probabilistic Graphical Models Approach Xia Lu In this paper, we propose to use probabilistic graphical models as a new theoretical and computational framework to study linguistic typology. The graphical structure of such a model represents a meta- language that consists of linguistic variables and the relationships between them while the parameters associated with each variable can be used to infer the strength of the relationships between the variables. Such models can also be used to predict feature values of new languages. Besides providing better solu- tions to existing problems in linguistic typology such a framework opens up to many new research topics that can help us to gain further insights into linguistic typology. Unsupervised Self-Training for Unsupervised Cross-Lingual Transfer Akshat Gupta, Sai Krishna Rallabandi and Alan W Black Labelled data is scarce, especially for low-resource languages. This beckons the need to come up with unsupervised methods for natural language processing tasks. In this paper, we introduce a general frame- work called Unsupervised Self-Training, capable of unsupervised cross-lingual transfer. We apply our proposed framework to a two-class sentiment analysis problem of code-switched data. We use the power of pre-trained BERT models for initialization and fine-tune them in an unsupervised manner, only us- ing pseudo labels produced by zero-shot predictions. We test our algorithm on multiple code-switched languages. Our unsupervised models compete well with their supervised counterparts, with their perfor- mance reaching within 1-7% (weighted F1 scores) when compared to supervised models trained for a two-class problem. Let-Mi: An Arabic Levantine Twitter Dataset for Misogynistic Language Hala Mulki and Bilal Ghanem Misogyny is one type of hate speech that disparages a person or a group having the female gender identity; it is typically defined as hatred of or contempt for women. Online misogyny has become an in- creasing worry for Arab women who experience gender-based online abuse on a daily basis. Such online abuse can be expressed through several misogynistic behaviors which reinforce and justify underestima- tion of women, male superiority, sexual abuse, mistreatment, and violence against women. Misogyny automatic detection systems can assist in the prohibition of anti-women Arabic toxic content. Developing these systems is hindered by the lack of the Arabic misogyny benchmark datasets. In this work, we intro- duce an Arabic Levantine Twitter dataset for Misogynistic language (LeT-Mi) to be the first benchmark dataset for Arabic misogyny. The proposed dataset consists of 6,550 tweets annotated either as neutral (misogynistic-free) or as one of seven misogyny categories: discredit, dominance, cursing/damning, sex- ual harassment, stereotyping and objectification, derailing, and the threat of violence. We further provide a detailed review of the dataset creation and annotation phases. The consistency of the annotations for the proposed dataset was emphasized through inter-rater agreement evaluation measures. Moreover, Let-Mi was used as an evaluation dataset through binary, multi-class, and target classification tasks which were conducted by several state-of-the-art machine learning systems along with Multi-Task Learning (MTL) configuration. The obtained results indicated that the performances achieved by the used systems are consistent with state-of-the-art results for languages other than Arabic, while employing MTL improved the performance of the misogyny/target classification tasks. Our dataset is available at https://github.com/bilalghanem/let-mi ix  Towards Figurative Language Generation in Afrikaans Imke van Heerden and Anil Bas This paper presents an LSTM-based approach to figurative language generation, which is an important step towards creative text generation in Afrikaans. Due to the scarcity of resources (in comparison to resource-rich languages), we train the proposed network on a single literary novel. This follows the same approach as Van Heerden and Bas (2021), however, we explicitly focus and expand on fully au- tomatic text generation, centring on figurative language in particular. The proposed model generates phrases that contain compellingly novel figures of speech such as metaphor, simile and personification. Improving Access to Untranscribed Speech by Leveraging Spoken Term Detection and Self-super- vised Learning of Speech Representations Nay San, Martijn Bartelds and Dan Jurafsky We summarise findings from our recent work showing that a large self-supervised model trained only on English speech provides a noise-robust and speaker-invariant feature extraction method that can be used for a speech information retrieval task with unrelated low resource target languages. A qualitative error analysis also revealed that the majority of the retrieval errors could be attributed to the differences in phonological inventories between English and the evaluation languages. With a longer-term aim of leveraging typological information to better adapt such models for the target languages, we also report on work in progress which examines the phonetic information encoded in these representations. On the Universality of Lexical Concepts Bradley Hauer and Grzegorz Kondrak We posit that lexicalized concepts are universal, and thus can be annotated cross-linguistically in parallel corpora. This is one of the implications of a novel theory that formalizes the relationship between words and senses in both monolingual and multilingual settings. The theory is based on a unifying treatment of the notions of synonymy and translational equivalence as different aspects of the relation of sameness of meaning within and across languages. Quantitative Detection of Cognacy in the Predictive Structure of Inflection Classes: Romance Ver- bal Conjugations against the Broader Typological Variation Borja Herce and Balthasar Bickel In recent years, Information Theory (with its core notion of entropy) has provided the theoretical back- ground for a lot of empirical research on inflectional systems, and has inspired various metrics to capture (different aspects of) their complexity. So far, however, entropy-based metrics have chiefly been used to assess synchronic states. Here we explore their potential for capturing patterns in language change and phylogenetic relatedness. Specifically, we probe different aspects of an inflectional system for their stability within one language family, Romance, and for the degree to which they distinguish this family from unrelated and less closely related languages. Based on most metrics, Romance appears to be dif- ferent from the control sample in the mean, variance, or both. The difference in variance is particularly interesting because it might suggest differences in relative diachronic stability and as phylogenetic sig- nals of relatedness. Subword Geometry: Picturing Word Shapes Olga Sozinova and Tanja Samardzic In this work in progress, we are investigating the structural properties of subwords in 20 languages by extracting word shapes, i.e. sequences of subword lengths. x  A Look to Languages through the Glass of BPE Compression Ximena Gutierrez-Vasques, Tanja Samardzic and Christian Bentz One of the predominant methods for subword tokenization is Byte-pair encoding (BPE). Originally, this is a data compression technique based on replacing the most common pair of consecutive bytes with a new symbol When applied to text, each iteration merges two adjacent symbols; this can be seen as a process of going from characters to subwords through iterations. Regardless of the language, the first merge operations tend to have a stronger impact on the compression of texts, i.e., they capture very frequent patterns that lead to a reduction of redundancy and to an incre- ment of the text entropy. However, the natural language properties that allow this compression are rarely analyzed, i.e., do all languages get compressed in the same way through BPE merge operations? We hypothesize that the type of recurrent patterns captured in each merge depends on the typology and even orthography and other corpus-related phenomena. For instance, for some languages, this compression might be related to frequent affixes or regular inflectional morphs, while for some others, it might be related to more idiosyncratic, irregular patterns or even related to orthographic redundancies. We propose a novel way to quantify this, inspired by the notion of morphological productivity. Information-Theoretic Characterization of Morphological Fusion Neil Rathi, Michael Hahn and Richard Futrell Traditionally, morphological typology divides synthetic languages into two broad groups (e.g. von Schlegel, 1808; von Humboldt, 1843). Agglutinative languages, such as Turkish, segment morphemes into inde- pendent features which can be easily split. On the other hand, fusional languages, such as Latin, “fuse” morphemes together phonologically (Bickel and Nichols, 2013). At the same time, there has long been recognition that the categories “agglutinative” and “fusional” are best thought of as a matter of degree, with Greenberg (1954) developing an “index of agglutination” metric for languages. Here, we propose an information-theoretic definition of the fusion of any given form in a language, which naturally delivers a graded measure of the degree of fusion. We use a sequence-to-sequence model to empirically verify that our measure captures typical linguistic classifications. xi