Which Pieces Does Unigram Tokenization Really Need?

Sander Land; Yuval Pinter

Which Pieces Does Unigram Tokenization Really Need?

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 <b>...</b> for bold, <i>...</i> for italic, and <url>...</url> for URLs.

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

The Unigram tokenization algorithm offers a probabilistic alternative to the greedy heuristics of Byte-Pair Encoding. Despite its theoretical elegance, its implementation in practice is complex, limiting its adoption to the SentencePiece package and adapters thereof. We bridge this gap between theory and practice by providing a clear guide to implementation and parameter choices. We also identify a simpler algorithm that accepts slightly higher training loss in exchange for improved compression.

Anthology ID:: 2026.findings-acl.316
Volume:: Findings of the Association for Computational Linguistics: ACL 2026
Month:: July
Year:: 2026
Address:: San Diego, California, United States
Editors:: Maria Liakata, Viviane P. Moreira, Jiajun Zhang, David Jurgens
Venue:: Findings
SIG:
Publisher:: Association for Computational Linguistics
Note:
Pages:: 6351–6360
Language:
URL:: https://aclanthology.org/2026.findings-acl.316/
DOI:
Bibkey:
Cite (ACL):: Sander Land and Yuval Pinter. 2026. Which Pieces Does Unigram Tokenization Really Need?. In Findings of the Association for Computational Linguistics: ACL 2026, pages 6351–6360, San Diego, California, United States. Association for Computational Linguistics.
Cite (Informal):: Which Pieces Does Unigram Tokenization Really Need? (Land & Pinter, Findings 2026)
Copy Citation:
PDF:: https://aclanthology.org/2026.findings-acl.316.pdf
Checklist:: 2026.findings-acl.316.checklist.pdf

PDF Cite Search Checklist Fix data

Export citation

BibTeX
MODS XML
Endnote
Preformatted

@inproceedings{land-pinter-2026-pieces,
    title = "Which Pieces Does Unigram Tokenization Really Need?",
    author = "Land, Sander  and
      Pinter, Yuval",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2026.findings-acl.316/",
    pages = "6351--6360",
    ISBN = "979-8-89176-395-1",
    abstract = "The Unigram tokenization algorithm offers a probabilistic alternative to the greedy heuristics of Byte-Pair Encoding. Despite its theoretical elegance, its implementation in practice is complex, limiting its adoption to the SentencePiece package and adapters thereof. We bridge this gap between theory and practice by providing a clear guide to implementation and parameter choices. We also identify a simpler algorithm that accepts slightly higher training loss in exchange for improved compression."
}

Download as File

<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="land-pinter-2026-pieces">
    <titleInfo>
        <title>Which Pieces Does Unigram Tokenization Really Need?</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Sander</namePart>
        <namePart type="family">Land</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yuval</namePart>
        <namePart type="family">Pinter</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-07</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Findings of the Association for Computational Linguistics: ACL 2026</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Maria</namePart>
            <namePart type="family">Liakata</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Viviane</namePart>
            <namePart type="given">P</namePart>
            <namePart type="family">Moreira</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Jiajun</namePart>
            <namePart type="family">Zhang</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">David</namePart>
            <namePart type="family">Jurgens</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">San Diego, California, United States</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-395-1</identifier>
    </relatedItem>
    <abstract>The Unigram tokenization algorithm offers a probabilistic alternative to the greedy heuristics of Byte-Pair Encoding. Despite its theoretical elegance, its implementation in practice is complex, limiting its adoption to the SentencePiece package and adapters thereof. We bridge this gap between theory and practice by providing a clear guide to implementation and parameter choices. We also identify a simpler algorithm that accepts slightly higher training loss in exchange for improved compression.</abstract>
    <identifier type="citekey">land-pinter-2026-pieces</identifier>
    <location>
        <url>https://aclanthology.org/2026.findings-acl.316/</url>
    </location>
    <part>
        <date>2026-07</date>
        <extent unit="page">
            <start>6351</start>
            <end>6360</end>
        </extent>
    </part>
</mods>
</modsCollection>

Download as File

%0 Conference Proceedings
%T Which Pieces Does Unigram Tokenization Really Need?
%A Land, Sander
%A Pinter, Yuval
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Findings of the Association for Computational Linguistics: ACL 2026
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-395-1
%F land-pinter-2026-pieces
%X The Unigram tokenization algorithm offers a probabilistic alternative to the greedy heuristics of Byte-Pair Encoding. Despite its theoretical elegance, its implementation in practice is complex, limiting its adoption to the SentencePiece package and adapters thereof. We bridge this gap between theory and practice by providing a clear guide to implementation and parameter choices. We also identify a simpler algorithm that accepts slightly higher training loss in exchange for improved compression.
%U https://aclanthology.org/2026.findings-acl.316/
%P 6351-6360

Download as File

Markdown (Informal)

[Which Pieces Does Unigram Tokenization Really Need?](https://aclanthology.org/2026.findings-acl.316/) (Land & Pinter, Findings 2026)

Which Pieces Does Unigram Tokenization Really Need? (Land & Pinter, Findings 2026)

ACL

Sander Land and Yuval Pinter. 2026. Which Pieces Does Unigram Tokenization Really Need?. In Findings of the Association for Computational Linguistics: ACL 2026, pages 6351–6360, San Diego, California, United States. Association for Computational Linguistics.