<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="novak-etal-2021-zero">
    <titleInfo>
        <title>Zero-shot cross-lingual Meaning Representation Transfer: Annotation of Hungarian using the Prague Functional Generative Description</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Attila</namePart>
        <namePart type="family">Novák</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Borbála</namePart>
        <namePart type="family">Novák</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Csilla</namePart>
        <namePart type="family">Novák</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2021-nov</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of The Joint 15th Linguistic Annotation Workshop (LAW) and 3rd Designing Meaning Representations (DMR) Workshop</title>
        </titleInfo>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Punta Cana, Dominican Republic</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>In this paper, we present the results of our experiments concerning the zero-shot cross-lingual performance of the PERIN sentence-to-graph semantic parser. We applied the PTG model trained using the PERIN parser on a 740k-token Czech newspaper corpus to Hungarian. We evaluated the performance of the parser using the official evaluation tool of the MRP 2020 shared task. The gold standard Hungarian annotation was created by manual correction of the output of the parser following the annotation manual of the tectogrammatical level of the Prague Dependency Treebank. An English model trained on a larger one-million-token English newspaper corpus is also available, however, we found that the Czech model performed significantly better on Hungarian input due to the fact that Hungarian is typologically more similar to Czech than to English. We have found that zero-shot transfer of the PTG meaning representation across typologically not-too-distant languages using a neural parser model based on a multilingual contextual language model followed by a manual correction by linguist experts seems to be a viable scenario.</abstract>
    <identifier type="citekey">novak-etal-2021-zero</identifier>
    <identifier type="doi">10.18653/v1/2021.law-1.1</identifier>
    <location>
        <url>https://aclanthology.org/2021.law-1.1</url>
    </location>
    <part>
        <date>2021-nov</date>
        <extent unit="page">
            <start>1</start>
            <end>11</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Zero-shot cross-lingual Meaning Representation Transfer: Annotation of Hungarian using the Prague Functional Generative Description
%A Novák, Attila
%A Novák, Borbála
%A Novák, Csilla
%S Proceedings of The Joint 15th Linguistic Annotation Workshop (LAW) and 3rd Designing Meaning Representations (DMR) Workshop
%D 2021
%8 nov
%I Association for Computational Linguistics
%C Punta Cana, Dominican Republic
%F novak-etal-2021-zero
%X In this paper, we present the results of our experiments concerning the zero-shot cross-lingual performance of the PERIN sentence-to-graph semantic parser. We applied the PTG model trained using the PERIN parser on a 740k-token Czech newspaper corpus to Hungarian. We evaluated the performance of the parser using the official evaluation tool of the MRP 2020 shared task. The gold standard Hungarian annotation was created by manual correction of the output of the parser following the annotation manual of the tectogrammatical level of the Prague Dependency Treebank. An English model trained on a larger one-million-token English newspaper corpus is also available, however, we found that the Czech model performed significantly better on Hungarian input due to the fact that Hungarian is typologically more similar to Czech than to English. We have found that zero-shot transfer of the PTG meaning representation across typologically not-too-distant languages using a neural parser model based on a multilingual contextual language model followed by a manual correction by linguist experts seems to be a viable scenario.
%R 10.18653/v1/2021.law-1.1
%U https://aclanthology.org/2021.law-1.1
%U https://doi.org/10.18653/v1/2021.law-1.1
%P 1-11

Markdown (Informal)

[Zero-shot cross-lingual Meaning Representation Transfer: Annotation of Hungarian using the Prague Functional Generative Description](https://aclanthology.org/2021.law-1.1) (Novák et al., LAW 2021)

• Zero-shot cross-lingual Meaning Representation Transfer: Annotation of Hungarian using the Prague Functional Generative Description (Novák et al., LAW 2021)

ACL

• Attila Novák, Borbála Novák, and Csilla Novák. 2021. Zero-shot cross-lingual Meaning Representation Transfer: Annotation of Hungarian using the Prague Functional Generative Description. In Proceedings of The Joint 15th Linguistic Annotation Workshop (LAW) and 3rd Designing Meaning Representations (DMR) Workshop, pages 1–11, Punta Cana, Dominican Republic. Association for Computational Linguistics.