<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="meloni-etal-2021-ab">
    <titleInfo>
        <title>Ab Antiquo: Neural Proto-language Reconstruction</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Carlo</namePart>
        <namePart type="family">Meloni</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Shauli</namePart>
        <namePart type="family">Ravfogel</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yoav</namePart>
        <namePart type="family">Goldberg</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2021-06</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies</title>
        </titleInfo>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Online</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>Historical linguists have identified regularities in the process of historic sound change. The comparative method utilizes those regularities to reconstruct proto-words based on observed forms in daughter languages. Can this process be efficiently automated? We address the task of proto-word reconstruction, in which the model is exposed to cognates in contemporary daughter languages, and has to predict the proto word in the ancestor language. We provide a novel dataset for this task, encompassing over 8,000 comparative entries, and show that neural sequence models outperform conventional methods applied to this task so far. Error analysis reveals a variability in the ability of neural model to capture different phonological changes, correlating with the complexity of the changes. Analysis of learned embeddings reveals the models learn phonologically meaningful generalizations, corresponding to well-attested phonological shifts documented by historical linguistics.</abstract>
    <identifier type="citekey">meloni-etal-2021-ab</identifier>
    <identifier type="doi">10.18653/v1/2021.naacl-main.353</identifier>
    <location>
        <url>https://aclanthology.org/2021.naacl-main.353</url>
    </location>
    <part>
        <date>2021-06</date>
        <extent unit="page">
            <start>4460</start>
            <end>4473</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Ab Antiquo: Neural Proto-language Reconstruction
%A Meloni, Carlo
%A Ravfogel, Shauli
%A Goldberg, Yoav
%S Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies
%D 2021
%8 June
%I Association for Computational Linguistics
%C Online
%F meloni-etal-2021-ab
%X Historical linguists have identified regularities in the process of historic sound change. The comparative method utilizes those regularities to reconstruct proto-words based on observed forms in daughter languages. Can this process be efficiently automated? We address the task of proto-word reconstruction, in which the model is exposed to cognates in contemporary daughter languages, and has to predict the proto word in the ancestor language. We provide a novel dataset for this task, encompassing over 8,000 comparative entries, and show that neural sequence models outperform conventional methods applied to this task so far. Error analysis reveals a variability in the ability of neural model to capture different phonological changes, correlating with the complexity of the changes. Analysis of learned embeddings reveals the models learn phonologically meaningful generalizations, corresponding to well-attested phonological shifts documented by historical linguistics.
%R 10.18653/v1/2021.naacl-main.353
%U https://aclanthology.org/2021.naacl-main.353
%U https://doi.org/10.18653/v1/2021.naacl-main.353
%P 4460-4473

Markdown (Informal)

[Ab Antiquo: Neural Proto-language Reconstruction](https://aclanthology.org/2021.naacl-main.353) (Meloni et al., NAACL 2021)

• Ab Antiquo: Neural Proto-language Reconstruction (Meloni et al., NAACL 2021)

ACL

• Carlo Meloni, Shauli Ravfogel, and Yoav Goldberg. 2021. Ab Antiquo: Neural Proto-language Reconstruction. In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 4460–4473, Online. Association for Computational Linguistics.