Procrustes Analysis for Improving Language Model Merging

Olivier Ferret

doi:10.63317/3ywqnrqzviyt

Procrustes Analysis for Improving Language Model Merging

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 external publication .)

Authors concatenated from the text boxes above:

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

Abstract

The availability of many fine-tuned neural language models for different tasks naturally leads to the question of whether it is worthwhile to combine them, particularly through parameter merging, which is the least resource-intensive option. Among the many existing methods, some focus on parameter alignment before actual merging. In this article, we propose a new method within this research area, based on Procrustes analysis. We evaluate this method for merging fine-tuned models for the same task, derived from the same encoder-based model. Considering nine tasks from the GLUE benchmark, three Named Entity Recognition tasks, and six reference merging methods, we show that our proposal can improve upon existing merging methods in most tested configurations.

Anthology ID:: 2026.lrec-1.783
Volume:: Proceedings of the Fifteenth Language Resources and Evaluation Conference
Month:: May
Year:: 2026
Address:: Palma de Mallorca, Spain
Editors:: Stelios Piperidis, Núria Bel, Henk van den Heuvel, Nancy Ide, Simon Krek, Antonio Toral
Venue:: LREC
SIG:
Publisher:: ELRA Language Resource Association
Note:
Pages:: 9988–9998
Language:
External URL:: https://lrec.elra.info/lrec2026-main-783
DOI:: 10.63317/3ywqnrqzviyt
Bibkey:
Cite (ACL):: Olivier Ferret. 2026. Procrustes Analysis for Improving Language Model Merging. In Proceedings of the Fifteenth Language Resources and Evaluation Conference, pages 9988–9998, Palma de Mallorca, Spain. ELRA Language Resource Association.
Cite (Informal):: Procrustes Analysis for Improving Language Model Merging (Ferret, LREC 2026)
Copy Citation:

External Cite Search Fix data

Export citation

BibTeX
MODS XML
Endnote
Preformatted

@inproceedings{ferret-2026-procrustes,
    title = "{P}rocrustes Analysis for Improving Language Model Merging",
    author = "Ferret, Olivier",
    editor = "Piperidis, Stelios  and
      Bel, N{\'u}ria  and
      van den Heuvel, Henk  and
      Ide, Nancy  and
      Krek, Simon  and
      Toral, Antonio",
    booktitle = "Proceedings of the Fifteenth Language Resources and Evaluation Conference",
    month = may,
    year = "2026",
    address = "Palma de Mallorca, Spain",
    publisher = "ELRA Language Resource Association",
    url = "https://aclanthology.org/2026.lrec-1.783/",
    doi = "10.63317/3ywqnrqzviyt",
    pages = "9988--9998",
    abstract = "The availability of many fine-tuned neural language models for different tasks naturally leads to the question of whether it is worthwhile to combine them, particularly through parameter merging, which is the least resource-intensive option. Among the many existing methods, some focus on parameter alignment before actual merging. In this article, we propose a new method within this research area, based on Procrustes analysis. We evaluate this method for merging fine-tuned models for the same task, derived from the same encoder-based model. Considering nine tasks from the GLUE benchmark, three Named Entity Recognition tasks, and six reference merging methods, we show that our proposal can improve upon existing merging methods in most tested configurations."
}

Download as File

<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ferret-2026-procrustes">
    <titleInfo>
        <title>Procrustes Analysis for Improving Language Model Merging</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Olivier</namePart>
        <namePart type="family">Ferret</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-05</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the Fifteenth Language Resources and Evaluation Conference</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Stelios</namePart>
            <namePart type="family">Piperidis</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Núria</namePart>
            <namePart type="family">Bel</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Henk</namePart>
            <namePart type="family">van den Heuvel</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Nancy</namePart>
            <namePart type="family">Ide</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Simon</namePart>
            <namePart type="family">Krek</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Antonio</namePart>
            <namePart type="family">Toral</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>ELRA Language Resource Association</publisher>
            <place>
                <placeTerm type="text">Palma de Mallorca, Spain</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>The availability of many fine-tuned neural language models for different tasks naturally leads to the question of whether it is worthwhile to combine them, particularly through parameter merging, which is the least resource-intensive option. Among the many existing methods, some focus on parameter alignment before actual merging. In this article, we propose a new method within this research area, based on Procrustes analysis. We evaluate this method for merging fine-tuned models for the same task, derived from the same encoder-based model. Considering nine tasks from the GLUE benchmark, three Named Entity Recognition tasks, and six reference merging methods, we show that our proposal can improve upon existing merging methods in most tested configurations.</abstract>
    <identifier type="citekey">ferret-2026-procrustes</identifier>
    <identifier type="doi">10.63317/3ywqnrqzviyt</identifier>
    <location>
        <url>https://aclanthology.org/2026.lrec-1.783/</url>
    </location>
    <part>
        <date>2026-05</date>
        <extent unit="page">
            <start>9988</start>
            <end>9998</end>
        </extent>
    </part>
</mods>
</modsCollection>

Download as File

%0 Conference Proceedings
%T Procrustes Analysis for Improving Language Model Merging
%A Ferret, Olivier
%Y Piperidis, Stelios
%Y Bel, Núria
%Y van den Heuvel, Henk
%Y Ide, Nancy
%Y Krek, Simon
%Y Toral, Antonio
%S Proceedings of the Fifteenth Language Resources and Evaluation Conference
%D 2026
%8 May
%I ELRA Language Resource Association
%C Palma de Mallorca, Spain
%F ferret-2026-procrustes
%X The availability of many fine-tuned neural language models for different tasks naturally leads to the question of whether it is worthwhile to combine them, particularly through parameter merging, which is the least resource-intensive option. Among the many existing methods, some focus on parameter alignment before actual merging. In this article, we propose a new method within this research area, based on Procrustes analysis. We evaluate this method for merging fine-tuned models for the same task, derived from the same encoder-based model. Considering nine tasks from the GLUE benchmark, three Named Entity Recognition tasks, and six reference merging methods, we show that our proposal can improve upon existing merging methods in most tested configurations.
%R 10.63317/3ywqnrqzviyt
%U https://aclanthology.org/2026.lrec-1.783/
%U https://doi.org/10.63317/3ywqnrqzviyt
%P 9988-9998

Download as File

Markdown (Informal)

[Procrustes Analysis for Improving Language Model Merging](https://aclanthology.org/2026.lrec-1.783/) (Ferret, LREC 2026)

Procrustes Analysis for Improving Language Model Merging (Ferret, LREC 2026)

ACL

Olivier Ferret. 2026. Procrustes Analysis for Improving Language Model Merging. In Proceedings of the Fifteenth Language Resources and Evaluation Conference, pages 9988–9998, Palma de Mallorca, Spain. ELRA Language Resource Association.