AI News HubLIVE
Original source2 min read

Merging Methods for Multilingual Knowledge Editing for Large Language Models: An Empirical Odyssey

This paper systematically evaluates six vector merging methods for multilingual knowledge editing (MKE). Vector summation with shared covariance is the most reliable strategy, while simple summation performs poorly. TSVM offers limited improvement in reducing multilingual interference. Performance is sensitive to weight scaling and rank compression ratio, with larger scaling and lower rank often yielding better results.

SourcearXiv Computational LinguisticsAuthor: Kunil Lee, Ki-Young Shin, Jong-Hyeok Lee, Young-Joo Suh

[2605.13919] Merging Methods for Multilingual Knowledge Editing for Large Language Models: An Empirical Odyssey

[Submitted on 13 May 2026]

Title:Merging Methods for Multilingual Knowledge Editing for Large Language Models: An Empirical Odyssey

View a PDF of the paper titled Merging Methods for Multilingual Knowledge Editing for Large Language Models: An Empirical Odyssey, by Kunil Lee and 2 other authors

View PDF HTML (experimental)

Abstract:Multilingual knowledge editing (MKE) remains challenging because language-specific edits interfere with one another, even when locate-then-edit methods work well in monolingual settings. This paper focuses on three issues: the effectiveness of vector merging methods for MKE, the extent to which Task Singular Vectors for Merging (TSVM) can reduce multilingual interference, and the influence of the weight scaling factor and rank compression ratio on performance. We evaluate six merging variants with two popular backbone large language models, two base knowledge editing methods, and 12 languages on the MzsRE benchmark under a large-scale batch-editing setting. Our results show that vector summation with shared covariance is the most reliable overall strategy, whereas simple summation without shared covariance performs poorly. TSVM improves performance in some settings, but its ability to mitigate multilingual interference is limited. We also find that performance is sensitive to both weight scale and rank ratio, with larger-than-default scaling and relatively low rank often yielding better results. These findings clarify the practical strengths and limits of current vector merging methods for MKE and provide guidance for future multilingual knowledge editing research.

Subjects:

Computation and Language (cs.CL); Machine Learning (cs.LG)

Cite as: arXiv:2605.13919 [cs.CL]

(or arXiv:2605.13919v1 [cs.CL] for this version)

https://doi.org/10.48550/arXiv.2605.13919

arXiv-issued DOI via DataCite

Submission history

From: Kunil Lee [view email] [v1] Wed, 13 May 2026 13:22:45 UTC (987 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled Merging Methods for Multilingual Knowledge Editing for Large Language Models: An Empirical Odyssey, by Kunil Lee and 2 other authors

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.CL

new | recent | 2026-05

Change to browse by:

cs cs.LG

References & Citations

NASA ADS

Google Scholar

Semantic Scholar

Loading...

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)

Related Papers

Recommenders and Search Tools

Link to Influence Flower

Influence Flower (What are Influence Flowers?)

Core recommender toggle

CORE Recommender (What is CORE?)

Author

Venue

Institution

Topic

About arXivLabs

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)