<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="belinkov-2022-probing">
    <titleInfo>
        <title>Probing Classifiers: Promises, Shortcomings, and Advances</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Yonatan</namePart>
        <namePart type="family">Belinkov</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2022-03</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <genre authority="bibutilsgt">journal article</genre>
    <relatedItem type="host">
        <titleInfo>
            <title>Computational Linguistics</title>
        </titleInfo>
        <originInfo>
            <issuance>continuing</issuance>
            <publisher>MIT Press</publisher>
            <place>
                <placeTerm type="text">Cambridge, MA</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">periodical</genre>
        <genre authority="bibutilsgt">academic journal</genre>
    </relatedItem>
    <abstract>Probing classifiers have emerged as one of the prominent methodologies for interpreting and analyzing deep neural network models of natural language processing. The basic idea is simple—a classifier is trained to predict some linguistic property from a model’s representations—and has been used to examine a wide variety of models and properties. However, recent studies have demonstrated various methodological limitations of this approach. This squib critically reviews the probing classifiers framework, highlighting their promises, shortcomings, and advances.</abstract>
    <identifier type="citekey">belinkov-2022-probing</identifier>
    <identifier type="doi">10.1162/coli_a_00422</identifier>
    <location>
        <url>https://aclanthology.org/2022.cl-1.7</url>
    </location>
    <part>
        <date>2022-03</date>
        <detail type="volume"><number>48</number></detail>
        <detail type="issue"><number>1</number></detail>
        <extent unit="page">
            <start>207</start>
            <end>219</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Journal Article
%T Probing Classifiers: Promises, Shortcomings, and Advances
%A Belinkov, Yonatan
%J Computational Linguistics
%D 2022
%8 March
%V 48
%N 1
%I MIT Press
%C Cambridge, MA
%F belinkov-2022-probing
%X Probing classifiers have emerged as one of the prominent methodologies for interpreting and analyzing deep neural network models of natural language processing. The basic idea is simple—a classifier is trained to predict some linguistic property from a model’s representations—and has been used to examine a wide variety of models and properties. However, recent studies have demonstrated various methodological limitations of this approach. This squib critically reviews the probing classifiers framework, highlighting their promises, shortcomings, and advances.
%R 10.1162/coli_a_00422
%U https://aclanthology.org/2022.cl-1.7
%U https://doi.org/10.1162/coli_a_00422
%P 207-219

Markdown (Informal)

[Probing Classifiers: Promises, Shortcomings, and Advances](https://aclanthology.org/2022.cl-1.7) (Belinkov, CL 2022)

• Probing Classifiers: Promises, Shortcomings, and Advances (Belinkov, CL 2022)

ACL

• Yonatan Belinkov. 2022. Probing Classifiers: Promises, Shortcomings, and Advances. Computational Linguistics, 48(1):207–219.