@article{wang-etal-2023-authorship,
title = "Can Authorship Representation Learning Capture Stylistic Features?",
author = "Wang, Andrew and
Aggazzotti, Cristina and
Kotula, Rebecca and
Soto, Rafael Rivera and
Bishop, Marcus and
Andrews, Nicholas",
journal = "Transactions of the Association for Computational Linguistics",
volume = "11",
year = "2023",
address = "Cambridge, MA",
publisher = "MIT Press",
url = "https://aclanthology.org/2023.tacl-1.80",
doi = "10.1162/tacl_a_00610",
pages = "1416--1431",
abstract = "Automatically disentangling an author{'}s style from the content of their writing is a longstanding and possibly insurmountable problem in computational linguistics. At the same time, the availability of large text corpora furnished with author labels has recently enabled learning authorship representations in a purely data-driven manner for authorship attribution, a task that ostensibly depends to a greater extent on encoding writing style than encoding content. However, success on this surrogate task does not ensure that such representations capture writing style since authorship could also be correlated with other latent variables, such as topic. In an effort to better understand the nature of the information these representations convey, and specifically to validate the hypothesis that they chiefly encode writing style, we systematically probe these representations through a series of targeted experiments. The results of these experiments suggest that representations learned for the surrogate authorship prediction task are indeed sensitive to writing style. As a consequence, authorship representations may be expected to be robust to certain kinds of data shift, such as topic drift over time. Additionally, our findings may open the door to downstream applications that require stylistic representations, such as style transfer.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wang-etal-2023-authorship">
<titleInfo>
<title>Can Authorship Representation Learning Capture Stylistic Features?</title>
</titleInfo>
<name type="personal">
<namePart type="given">Andrew</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Cristina</namePart>
<namePart type="family">Aggazzotti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rebecca</namePart>
<namePart type="family">Kotula</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rafael</namePart>
<namePart type="given">Rivera</namePart>
<namePart type="family">Soto</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marcus</namePart>
<namePart type="family">Bishop</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nicholas</namePart>
<namePart type="family">Andrews</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<genre authority="bibutilsgt">journal article</genre>
<relatedItem type="host">
<titleInfo>
<title>Transactions of the Association for 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>Automatically disentangling an author’s style from the content of their writing is a longstanding and possibly insurmountable problem in computational linguistics. At the same time, the availability of large text corpora furnished with author labels has recently enabled learning authorship representations in a purely data-driven manner for authorship attribution, a task that ostensibly depends to a greater extent on encoding writing style than encoding content. However, success on this surrogate task does not ensure that such representations capture writing style since authorship could also be correlated with other latent variables, such as topic. In an effort to better understand the nature of the information these representations convey, and specifically to validate the hypothesis that they chiefly encode writing style, we systematically probe these representations through a series of targeted experiments. The results of these experiments suggest that representations learned for the surrogate authorship prediction task are indeed sensitive to writing style. As a consequence, authorship representations may be expected to be robust to certain kinds of data shift, such as topic drift over time. Additionally, our findings may open the door to downstream applications that require stylistic representations, such as style transfer.</abstract>
<identifier type="citekey">wang-etal-2023-authorship</identifier>
<identifier type="doi">10.1162/tacl_a_00610</identifier>
<location>
<url>https://aclanthology.org/2023.tacl-1.80</url>
</location>
<part>
<date>2023</date>
<detail type="volume"><number>11</number></detail>
<extent unit="page">
<start>1416</start>
<end>1431</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Journal Article
%T Can Authorship Representation Learning Capture Stylistic Features?
%A Wang, Andrew
%A Aggazzotti, Cristina
%A Kotula, Rebecca
%A Soto, Rafael Rivera
%A Bishop, Marcus
%A Andrews, Nicholas
%J Transactions of the Association for Computational Linguistics
%D 2023
%V 11
%I MIT Press
%C Cambridge, MA
%F wang-etal-2023-authorship
%X Automatically disentangling an author’s style from the content of their writing is a longstanding and possibly insurmountable problem in computational linguistics. At the same time, the availability of large text corpora furnished with author labels has recently enabled learning authorship representations in a purely data-driven manner for authorship attribution, a task that ostensibly depends to a greater extent on encoding writing style than encoding content. However, success on this surrogate task does not ensure that such representations capture writing style since authorship could also be correlated with other latent variables, such as topic. In an effort to better understand the nature of the information these representations convey, and specifically to validate the hypothesis that they chiefly encode writing style, we systematically probe these representations through a series of targeted experiments. The results of these experiments suggest that representations learned for the surrogate authorship prediction task are indeed sensitive to writing style. As a consequence, authorship representations may be expected to be robust to certain kinds of data shift, such as topic drift over time. Additionally, our findings may open the door to downstream applications that require stylistic representations, such as style transfer.
%R 10.1162/tacl_a_00610
%U https://aclanthology.org/2023.tacl-1.80
%U https://doi.org/10.1162/tacl_a_00610
%P 1416-1431
Markdown (Informal)
[Can Authorship Representation Learning Capture Stylistic Features?](https://aclanthology.org/2023.tacl-1.80) (Wang et al., TACL 2023)
ACL