English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

There is a growing interest in investigating what neural NLP models learnabout language. A prominent open question is the question of whether or not itis necessary to model hierarchical structure. We present a linguisticinvestigation of a neural parser adding insights to this question. We look attransitivity and agreement information of auxiliary verb constructions (AVCs)in comparison to finite main verbs (FMVs). This comparison is motivated bytheoretical work in dependency grammar and in particular the work of Tesni\`ere(1959) where AVCs and FMVs are both instances of a nucleus, the basic unit ofsyntax. An AVC is a dissociated nucleus, it consists of at least two words, andan FMV is its non-dissociated counterpart, consisting of exactly one word. Wesuggest that the representation of AVCs and FMVs should capture similarinformation. We use diagnostic classifiers to probe agreement and transitivityinformation in vectors learned by a transition-based neural parser in fourtypologically different languages. We find that the parser learns differentinformation about AVCs and FMVs if only sequential models (BiLSTMs) are used inthe architecture but similar information when a recursive layer is used. Wefind explanations for why this is the case by looking closely at howinformation is learned in the network and looking at what happens withdifferent dependency representations of AVCs. We conclude that there may bebenefits to using a recursive layer in dependency parsing and that we have notyet found the best way to integrate it in our parsers.

What Should/Do/Can LSTMs Learn When Parsing Auxiliary Verb Constructions?