STEB: Style Text Embedding Benchmark

Rafael Alberto Rivera Soto; Anna Wegmann; Cristina Aggazzotti

STEB: Style Text Embedding Benchmark

Rafael Alberto Rivera Soto, Anna Wegmann, Cristina Aggazzotti

Correct Metadata for

Use this form to create a GitHub issue with structured data describing the correction. You will need a GitHub account. Once you create that issue, the correction will be reviewed by a staff member.

⚠️ Mobile Users: Submitting this form to create a new issue will only work with github.com, not the GitHub Mobile app.

Important: The Anthology treat PDFs as authoritative. Please use this form only to correct data that is out of line with the PDF. See our corrections guidelines if you need to change the PDF.

Title Adjust the title. Retain tags such as <fixed-case>.

Authors Adjust author names and order to match the PDF.

Abstract Correct abstract if needed. Retain XML formatting tags such as <tex-math>. You may use ... for bold, ... for italic, ... for underline, <sc>...</sc> for small-caps, <tt>...<tt> for typewriter text, <url>...</url> for URLs, <a href=...> for hyperlinks, and <par/> for paragraph breaks.

Verification against PDF Ensure that the new title/authors match the snapshot below. (If there is no snapshot or it is too small, consult the PDF.)

Authors concatenated from the text boxes above:

ALL author names match the snapshot above—including middle initials, hyphens, and accents.

Abstract

While semantic embeddings are rigorously evaluated on the Massive Text Embedding Benchmark, the evaluation of style embeddings remains fragmented, with each work relying on their own set of tasks and datasets. To bridge this gap, we introduce the Style Text Embedding Benchmark, a comprehensive open-source benchmark intended to standardize the evaluation of style embeddings. STEB encompasses 96 datasets across 7 languages, spanning applications such as authorship verification, authorship retrieval, AI-text detection, probing of llinguistic features, and others. We find that semantic embeddings consistently fail in stylistic tasks, and that there is no style embedding that is universally superior across all tasks evaluated. STEB comes with an open-source code base and leaderboard at https://github.com/rrivera1849/STEB.

Anthology ID:: 2026.emnlp-main.572
Volume:: Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
Month:: October
Year:: 2026
Address:: Budapest, Hungary
Editors:: Hung-Yi Lee, Sunipa Dev, Jessy Li, Desmond Elliott
Venue:: EMNLP
SIG:: SIGDAT
Publisher:: Association for Computational Linguistics
Note:
Pages:: 13109–13145
Language:
URL:: https://aclanthology.org/2026.emnlp-main.572/
DOI:
Bibkey:
Cite (ACL):: Rafael Alberto Rivera Soto, Anna Wegmann, and Cristina Aggazzotti. 2026. STEB: Style Text Embedding Benchmark. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 13109–13145, Budapest, Hungary. Association for Computational Linguistics.
Cite (Informal):: STEB: Style Text Embedding Benchmark (Rivera Soto et al., EMNLP 2026)
Copy Citation:
PDF:: https://aclanthology.org/2026.emnlp-main.572.pdf

PDF Cite Search Fix data

Export citation

BibTeX
MODS XML
Endnote
Preformatted

@inproceedings{rivera-soto-etal-2026-steb,
    title = "{STEB}: Style Text Embedding Benchmark",
    author = "Rivera Soto, Rafael Alberto  and
      Wegmann, Anna  and
      Aggazzotti, Cristina",
    editor = "Lee, Hung-Yi  and
      Dev, Sunipa  and
      Li, Jessy  and
      Elliott, Desmond",
    booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
    month = oct,
    year = "2026",
    address = "Budapest, Hungary",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2026.emnlp-main.572/",
    pages = "13109--13145",
    ISBN = "979-8-89176-459-0",
    abstract = "While semantic embeddings are rigorously evaluated on the Massive Text Embedding Benchmark, the evaluation of style embeddings remains fragmented, with each work relying on their own set of tasks and datasets. To bridge this gap, we introduce the Style Text Embedding Benchmark, a comprehensive open-source benchmark intended to standardize the evaluation of style embeddings. STEB encompasses 96 datasets across 7 languages, spanning applications such as authorship verification, authorship retrieval, AI-text detection, probing of llinguistic features, and others. We find that semantic embeddings consistently fail in stylistic tasks, and that there is no style embedding that is universally superior across all tasks evaluated. STEB comes with an open-source code base and leaderboard at https://github.com/rrivera1849/STEB."
}

Download as File

<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="rivera-soto-etal-2026-steb">
    <titleInfo>
        <title>STEB: Style Text Embedding Benchmark</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Rafael</namePart>
        <namePart type="given">Alberto</namePart>
        <namePart type="family">Rivera Soto</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Anna</namePart>
        <namePart type="family">Wegmann</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>
    <originInfo>
        <dateIssued>2026-10</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Hung-Yi</namePart>
            <namePart type="family">Lee</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Sunipa</namePart>
            <namePart type="family">Dev</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Jessy</namePart>
            <namePart type="family">Li</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Desmond</namePart>
            <namePart type="family">Elliott</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Budapest, Hungary</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-459-0</identifier>
    </relatedItem>
    <abstract>While semantic embeddings are rigorously evaluated on the Massive Text Embedding Benchmark, the evaluation of style embeddings remains fragmented, with each work relying on their own set of tasks and datasets. To bridge this gap, we introduce the Style Text Embedding Benchmark, a comprehensive open-source benchmark intended to standardize the evaluation of style embeddings. STEB encompasses 96 datasets across 7 languages, spanning applications such as authorship verification, authorship retrieval, AI-text detection, probing of llinguistic features, and others. We find that semantic embeddings consistently fail in stylistic tasks, and that there is no style embedding that is universally superior across all tasks evaluated. STEB comes with an open-source code base and leaderboard at https://github.com/rrivera1849/STEB.</abstract>
    <identifier type="citekey">rivera-soto-etal-2026-steb</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-main.572/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>13109</start>
            <end>13145</end>
        </extent>
    </part>
</mods>
</modsCollection>

Download as File

%0 Conference Proceedings
%T STEB: Style Text Embedding Benchmark
%A Rivera Soto, Rafael Alberto
%A Wegmann, Anna
%A Aggazzotti, Cristina
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F rivera-soto-etal-2026-steb
%X While semantic embeddings are rigorously evaluated on the Massive Text Embedding Benchmark, the evaluation of style embeddings remains fragmented, with each work relying on their own set of tasks and datasets. To bridge this gap, we introduce the Style Text Embedding Benchmark, a comprehensive open-source benchmark intended to standardize the evaluation of style embeddings. STEB encompasses 96 datasets across 7 languages, spanning applications such as authorship verification, authorship retrieval, AI-text detection, probing of llinguistic features, and others. We find that semantic embeddings consistently fail in stylistic tasks, and that there is no style embedding that is universally superior across all tasks evaluated. STEB comes with an open-source code base and leaderboard at https://github.com/rrivera1849/STEB.
%U https://aclanthology.org/2026.emnlp-main.572/
%P 13109-13145

Download as File

Markdown (Informal)

[STEB: Style Text Embedding Benchmark](https://aclanthology.org/2026.emnlp-main.572/) (Rivera Soto et al., EMNLP 2026)

STEB: Style Text Embedding Benchmark (Rivera Soto et al., EMNLP 2026)

ACL

Rafael Alberto Rivera Soto, Anna Wegmann, and Cristina Aggazzotti. 2026. STEB: Style Text Embedding Benchmark. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 13109–13145, Budapest, Hungary. Association for Computational Linguistics.