{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T22:27:11Z","timestamp":1780352831846,"version":"3.54.1"},"reference-count":48,"publisher":"MIT Press","issue":"4","license":[{"start":{"date-parts":[[2022,7,6]],"date-time":"2022-07-06T00:00:00Z","timestamp":1657065600000},"content-version":"vor","delay-in-days":186,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"content-domain":{"domain":["direct.mit.edu"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2022,12,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>Dependency-based approaches to syntactic analysis assume that syntactic structure can be analyzed in terms of binary asymmetric dependency relations holding between elementary syntactic units. Computational models for dependency parsing almost universally assume that an elementary syntactic unit is a word, while the influential theory of Lucien Tesni\u00e8re instead posits a more abstract notion of nucleus, which may be realized as one or more words. In this article, we investigate the effect of enriching computational parsing models with a concept of nucleus inspired by Tesni\u00e8re. We begin by reviewing how the concept of nucleus can be defined in the framework of Universal Dependencies, which has become the de facto standard for training and evaluating supervised dependency parsers, and explaining how composition functions can be used to make neural transition-based dependency parsers aware of the nuclei thus defined. We then perform an extensive experimental study, using data from 20 languages to assess the impact of nucleus composition across languages with different typological characteristics, and utilizing a variety of analytical tools including ablation, linear mixed-effects models, diagnostic classifiers, and dimensionality reduction. The analysis reveals that nucleus composition gives small but consistent improvements in parsing accuracy for most languages, and that the improvement mainly concerns the analysis of main predicates, nominal dependents, clausal dependents, and coordination structures. Significant factors explaining the rate of improvement across languages include entropy in coordination structures and frequency of certain function words, in particular determiners. Analysis using dimensionality reduction and diagnostic classifiers suggests that nucleus composition increases the similarity of vectors representing nuclei of the same syntactic type.<\/jats:p>","DOI":"10.1162\/coli_a_00450","type":"journal-article","created":{"date-parts":[[2022,7,6]],"date-time":"2022-07-06T19:16:03Z","timestamp":1657134963000},"page":"849-886","update-policy":"https:\/\/doi.org\/10.1162\/mitpressjournals.corrections.policy","source":"Crossref","is-referenced-by-count":3,"title":["Nucleus Composition in Transition-based Dependency\n                    Parsing"],"prefix":"10.1162","volume":"48","author":[{"given":"Joakim","family":"Nivre","sequence":"first","affiliation":[{"name":"Uppsala University Department of Linguistics and Philology, RISE Research Institutes of Sweden. joakim.nivre@lingfil.uu.se"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ali","family":"Basirat","sequence":"additional","affiliation":[{"name":"Link\u00f6ping University, Department of Computer and Information Science. ali.basirat@liu.se"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Luise","family":"D\u00fcrlich","sequence":"additional","affiliation":[{"name":"Uppsala University, Department of Linguistics and Philology RISE Research Institutes of Sweden. luise.durlich@ri.se"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Adam","family":"Moss","sequence":"additional","affiliation":[{"name":"Uppsala University, Department of Linguistics and Philology. adam.moss@lingfil.uu.se"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"281","published-online":{"date-parts":[[2022,12,1]]},"reference":[{"key":"2022120516072525400_bib1","doi-asserted-by":"crossref","first-page":"257","DOI":"10.1007\/978-94-011-3474-3_10","article-title":"Parsing by chunks","volume-title":"Principle-Based Parsing","author":"Abney","year":"1991"},{"key":"2022120516072525400_bib2","first-page":"5","article-title":"Getting to the roots of dependency\n                        parsing","volume":"39","author":"Ballesteros","year":"2013"},{"key":"2022120516072525400_bib3","first-page":"13","article-title":"Dependency-based hybrid model of syntactic\n                        analysis for the languages with a rather free word order","volume-title":"Proceedings of the 16th Nordic Conference of Computational\n                        Linguistics (NODALIDA 2007)","author":"B\u0101rzdi\u0146\u0161","year":"2007"},{"key":"2022120516072525400_bib4","first-page":"1376","article-title":"Syntactic nuclei in dependency\n                        parsing\u2014a multilingual exploration","volume-title":"Proceedings of the 16th Conference of the European Chapter of the\n                        Association for Computational Linguistics: Main Volume","author":"Basirat","year":"2021"},{"issue":"1","key":"2022120516072525400_bib5","doi-asserted-by":"crossref","first-page":"1","DOI":"10.18637\/jss.v067.i01","article-title":"Fitting linear mixed-effects models using\n                        lme4","volume":"67","author":"Bates","year":"2015","journal-title":"Journal of Statistical Software"},{"key":"2022120516072525400_bib6","unstructured":"Belinkov,\n                                Yonatan\n          .\n                        2018. On Internal Language Representations in Deep\n                        Learning: An Analysis of Machine Translation and Speech\n                    Recognition. Ph.D. thesis, Massachusetts Institute of\n                        Technology."},{"key":"2022120516072525400_bib7","doi-asserted-by":"publisher","first-page":"77","DOI":"10.3115\/1697236.1697251","article-title":"Two stage constraint based hybrid approach\n                        to free word order language dependency parsing","volume-title":"Proceedings of the 11th International Conference on Parsing\n                        Technologies (IWPT)","author":"Bharati","year":"2009"},{"key":"2022120516072525400_bib8","doi-asserted-by":"publisher","first-page":"105","DOI":"10.3115\/981574.981589","article-title":"Parsing free word order languages in the\n                        Paninian framework","volume-title":"Proceedings of the 31st\n                        Annual Meeting of the Association for Computational Linguistics\n                        (ACL)","author":"Bharati","year":"1993"},{"key":"2022120516072525400_bib9","doi-asserted-by":"crossref","first-page":"103","DOI":"10.1007\/978-94-010-0201-1_7","article-title":"The Prague Dependency Treebank: A\n                        three-level annotation scenario","volume-title":"Treebanks: Building and Using Parsed Corpora","author":"B\u00f6hmov\u00e1","year":"2003"},{"key":"2022120516072525400_bib10","first-page":"239","article-title":"Cascaded grammatical relation\n                        assignment","volume-title":"Proceedings of the Joint SIGDAT\n                        Conference on Empirical Methods in NLP and Very Large Corpora","author":"Buchholz","year":"1999"},{"key":"2022120516072525400_bib11","doi-asserted-by":"publisher","first-page":"334","DOI":"10.3115\/v1\/P15-1033","article-title":"Transition-based dependency parsing with stack\n                        long short-term memory","volume-title":"Proceedings of the 53rd\n                        Annual Meeting of the Association for Computational Linguistics\n                        (ACL)","author":"Dyer","year":"2015"},{"key":"2022120516072525400_bib12","doi-asserted-by":"publisher","first-page":"117","DOI":"10.18653\/v1\/P19-1012","article-title":"The (non-)utility of structural features in BiLSTM-based\n                        dependency parsers","volume-title":"Proceedings of the 57th\n                        Annual Meeting of the Association for Computational Linguistics\n                        (ACL)","author":"Falenska","year":"2019"},{"key":"2022120516072525400_bib13","first-page":"1828","article-title":"Causal analysis of syntactic agreement\n                        mechanisms in neural language models","volume-title":"Proceedings\n                        of the 59th Annual Meeting of the Association for Computational Linguistics\n                        and the 11th International Joint Conference on Natural Language Processing\n                        (Volume 1: Long Papers)","author":"Finlayson","year":"2021"},{"key":"2022120516072525400_bib14","first-page":"99","article-title":"Annotation issues in Universal Dependencies for Korean and\n                        Japanese","volume-title":"Proceedings of the Fourth Workshop on\n                        Universal Dependencies (UDW 2020)","author":"Han","year":"2020"},{"key":"2022120516072525400_bib15","article-title":"Designing and interpreting probes with control\n                        tasks","author":"Hewitt","year":"2019","journal-title":"arXiv preprint\n                    arXiv:1909.03368"},{"key":"2022120516072525400_bib16","doi-asserted-by":"publisher","first-page":"907","DOI":"10.1613\/jair.1.11196","article-title":"Visualisation and \u2018diagnostic\n                        classifiers\u2019 reveal how recurrent and recursive neural networks\n                        process hierarchical structure","volume":"61","author":"Hupkes","year":"2018","journal-title":"Journal of\n                        Artificial Intelligence Research"},{"key":"2022120516072525400_bib17","first-page":"1","article-title":"Towards an implementable dependency\n                        grammar","volume-title":"Proceedings of the Workshop on\n                        Processing of Dependency-Based Grammars (ACL-COLING)","author":"J\u00e4rvinen","year":"1998"},{"key":"2022120516072525400_bib18","doi-asserted-by":"crossref","first-page":"20170213","DOI":"10.1098\/rsif.2017.0213","article-title":"The coefficient of determination r2 and\n                        intra-class correlation coefficient from generalized linear mixed-effects\n                        models revisited and expanded","volume":"14","author":"Johnson","year":"2017","journal-title":"Journal of the Royal\n                        Society Interface"},{"key":"2022120516072525400_bib19","first-page":"70","article-title":"Bubble trees and syntactic\n                        representations","volume-title":"Proceedings of the 5th Meeting\n                        of Mathematics of Language","author":"Kahane","year":"1997"},{"key":"2022120516072525400_bib20","doi-asserted-by":"publisher","first-page":"313","DOI":"10.1162\/tacl_a_00101","article-title":"Simple and accurate dependency parsing\n                        using bidirectional LSTM feature representations","volume":"4","author":"Kiperwasser","year":"2016","journal-title":"Transactions of the Association for Computational\n                        Linguistics"},{"key":"2022120516072525400_bib21","doi-asserted-by":"publisher","first-page":"63","DOI":"10.3115\/1118853.1118869","article-title":"Japanese dependency analysis using\n                        cascaded chunking","volume-title":"Proceedings of the 6th\n                        Workshop on Computational Language Learning (CoNLL)","author":"Kudo","year":"2002"},{"key":"2022120516072525400_bib22","first-page":"673","article-title":"Dynamic programming algorithms for\n                        transition-based dependency parsers","volume-title":"Proceedings\n                        of the 49th Annual Meeting of the Association for Computational Linguistics\n                        (ACL)","author":"Kuhlmann","year":"2011"},{"key":"2022120516072525400_bib23","first-page":"1566","article-title":"Recursive subtree composition in LSTM-based\n                        dependency parsing","volume-title":"Proceedings of the 2019\n                        Conference of the North American Chapter of the Association for\n                        Computational Linguistics: Human Language Technologies, Volume 1 (Long and\n                        Short Papers)","author":"de Lhoneux","year":"2019"},{"key":"2022120516072525400_bib24","doi-asserted-by":"crossref","first-page":"207","DOI":"10.18653\/v1\/K17-3022","article-title":"From raw text to Universal\n                        Dependencies\u2014Look, no tags!","volume-title":"Proceedings of\n                        the CoNLL 2017 Shared Task: Multilingual Parsing from Raw Text to Universal\n                        Dependencies","author":"de Lhoneux","year":"2017"},{"key":"2022120516072525400_bib25","first-page":"99","article-title":"Arc-hybrid non-projective dependency parsing\n                        with a static-dynamic oracle","volume-title":"Proceedings of the\n                        15th International Conference on Parsing Technologies","author":"de Lhoneux","year":"2017"},{"issue":"4","key":"2022120516072525400_bib26","doi-asserted-by":"crossref","first-page":"763","DOI":"10.1162\/coli_a_00392","article-title":"What should\/do\/can LSTMs learn when parsing\n                        auxiliary verb constructions?","volume":"46","author":"de Lhoneux","year":"2020","journal-title":"Computational Linguistics"},{"issue":"11","key":"2022120516072525400_bib27","first-page":"2579","article-title":"Visualizing data using\n                        t-SNE","volume":"9","author":"Van der Maaten","year":"2008","journal-title":"Journal of Machine Learning\n                    Research"},{"key":"2022120516072525400_bib28","doi-asserted-by":"publisher","first-page":"255","DOI":"10.1162\/coli_a_00402","article-title":"Universal dependencies","volume":"47","author":"de Marneffe","year":"2021","journal-title":"Computational Linguistics"},{"key":"2022120516072525400_bib29","doi-asserted-by":"crossref","first-page":"3278","DOI":"10.18653\/v1\/2020.findings-emnlp.294","article-title":"Graph-to-graph transformer for\n                        transition-based dependency parsing","volume-title":"Findings of\n                        the Association for Computational Linguistics: EMNLP 2020","author":"Mohammadshahi","year":"2020"},{"key":"2022120516072525400_bib30","first-page":"233","article-title":"Comparison of the SemTi-Kamols and\n                        Tesni\u00e8re\u2019s dependency grammars","volume-title":"Proceedings of the 4th International Conference on Human Language\n                        Technologies\u2014the Baltic Perspective","author":"Nespore","year":"2010"},{"key":"2022120516072525400_bib31","first-page":"149","article-title":"An efficient algorithm for projective\n                        dependency parsing","volume-title":"Proceedings of the 8th\n                        International Workshop on Parsing Technologies (IWPT)","author":"Nivre","year":"2003"},{"key":"2022120516072525400_bib32","doi-asserted-by":"crossref","first-page":"50","DOI":"10.3115\/1613148.1613156","article-title":"Incrementality in deterministic dependency\n                        parsing","volume-title":"Proceedings of the Workshop on\n                        Incremental Parsing: Bringing Engineering and Cognition Together\n                        (ACL)","author":"Nivre","year":"2004"},{"key":"2022120516072525400_bib33","doi-asserted-by":"publisher","first-page":"513","DOI":"10.1162\/coli.07-056-R1-07-027","article-title":"Algorithms for deterministic incremental\n                        dependency parsing","volume":"34","author":"Nivre","year":"2008","journal-title":"Computational\n                        Linguistics"},{"key":"2022120516072525400_bib34","doi-asserted-by":"publisher","first-page":"351","DOI":"10.3115\/1687878.1687929","article-title":"Non-projective dependency parsing in\n                        expected linear time","volume-title":"Proceedings of the Joint\n                        Conference of the 47th Annual Meeting of the ACL and the 4th International\n                        Joint Conference on Natural Language Processing of the AFNLP\n                        (ACL-IJCNLP)","author":"Nivre","year":"2009"},{"key":"2022120516072525400_bib35","first-page":"86","article-title":"Universal dependency evaluation","volume-title":"Proceedings of the NoDaLiDa 2017 Workshop on Universal Dependencies\n                        (UDW 2017)","author":"Nivre","year":"2017"},{"key":"2022120516072525400_bib36","first-page":"2216","article-title":"Maltparser: A data-driven parser-generator\n                        for dependency parsing","volume-title":"Proceedings of the 5th\n                        International Conference on Language Resources and Evaluation\n                        (LREC)","author":"Nivre","year":"2006"},{"key":"2022120516072525400_bib37","first-page":"1659","article-title":"Universal Dependencies v1: A multilingual\n                        treebank collection","volume-title":"Proceedings of the 10th\n                        International Conference on Language Resources and Evaluation\n                        (LREC)","author":"Nivre","year":"2016"},{"key":"2022120516072525400_bib38","first-page":"4034","article-title":"Universal Dependencies v2: An evergrowing\n                        multilingual treebank collection","volume-title":"Proceedings of\n                        the 12th International Conference on Language Resources and Evaluation\n                        (LREC)","author":"Nivre","year":"2020"},{"key":"2022120516072525400_bib39","doi-asserted-by":"publisher","first-page":"684","DOI":"10.3115\/992730.992745","article-title":"A statistical theory of dependency\n                        syntax","volume-title":"Proceedings of the 18th International\n                        Conference on Computational Linguistics (COLING)","author":"Samuelsson","year":"2000"},{"key":"2022120516072525400_bib40","first-page":"173","article-title":"An English dependency treebank \u00e0 la\n                        Tesni\u00e8re","volume-title":"Proceedings of the 8th Workshop\n                        on Treebanks and Linguistic Theories (TLT)","author":"Sangati","year":"2009"},{"key":"2022120516072525400_bib41","volume-title":"The Meaning of the Sentence in Its Pragmatic\n                        Aspects","author":"Sgall","year":"1986"},{"key":"2022120516072525400_bib42","doi-asserted-by":"publisher","first-page":"12","DOI":"10.18653\/v1\/D17-1002","article-title":"Fast(er) exact decoding and global training for\n                        transition-based dependency parsing via a minimal feature\n                        set","volume-title":"Proceedings of the Conference on Empirical\n                        Methods in Natural Language Processing (EMNLP)","author":"Shi","year":"2017"},{"key":"2022120516072525400_bib43","first-page":"113","article-title":"82 treebanks, 34 models: Universal\n                        dependency parsing with multi-treebank models","volume-title":"Proceedings of the 2018 CoNLL Shared Task: Multilingual Parsing from\n                        Raw Text to Universal Dependencies","author":"Smith","year":"2018"},{"key":"2022120516072525400_bib44","article-title":"Transition-based dependency parsing using\n                        recursive neural networks","volume-title":"NIPS Workshop on Deep\n                        Learning","author":"Stenetorp","year":"2013"},{"key":"2022120516072525400_bib45","volume-title":"\u00c9l\u00e9ments de syntaxe\n                        structurale","author":"Tesni\u00e8re","year":"1959"},{"key":"2022120516072525400_bib46","first-page":"195","article-title":"Statistical dependency analysis with\n                        support vector machines","volume-title":"Proceedings of the 8th\n                        International Workshop on Parsing Technologies (IWPT)","author":"Yamada","year":"2003"},{"key":"2022120516072525400_bib47","article-title":"Universal dependencies 2.8.1","author":"Zeman","year":"2021"},{"key":"2022120516072525400_bib48","first-page":"188","article-title":"Transition-based parsing with rich non-local\n                        features","volume-title":"Proceedings of the 49th Annual Meeting\n                        of the Association for Computational Linguistics (ACL)","author":"Zhang","year":"2011"}],"container-title":["Computational Linguistics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/direct.mit.edu\/coli\/article-pdf\/48\/4\/849\/2062072\/coli_a_00450.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/direct.mit.edu\/coli\/article-pdf\/48\/4\/849\/2062072\/coli_a_00450.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,12,5]],"date-time":"2022-12-05T16:08:17Z","timestamp":1670256497000},"score":1,"resource":{"primary":{"URL":"https:\/\/direct.mit.edu\/coli\/article\/48\/4\/849\/112170\/Nucleus-Composition-in-Transition-based-Dependency"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022]]},"references-count":48,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2022,12,1]]},"published-print":{"date-parts":[[2022,12,1]]}},"URL":"https:\/\/doi.org\/10.1162\/coli_a_00450","relation":{},"ISSN":["0891-2017","1530-9312"],"issn-type":[{"value":"0891-2017","type":"print"},{"value":"1530-9312","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2022]]},"published":{"date-parts":[[2022]]}}}