Knowledge Beyond Language: Bridging the Gap in Multilingual Machine Unlearning Evaluation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Hwang, Kyomin, Kim, Hyeonjin, Cho, Sangyeon, Kwak, Nojun
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911684255285248
author Hwang, Kyomin
Kim, Hyeonjin
Cho, Sangyeon
Kwak, Nojun
author_facet Hwang, Kyomin
Kim, Hyeonjin
Cho, Sangyeon
Kwak, Nojun
contents While LLMs are increasingly used in commercial services, they pose privacy risks such as leakage of sensitive personally identifiable information (PII). For LLMs trained on multilingual corpora, Multilingual Machine Unlearning (MMU) aims to remove information across multiple languages. However, prior MMU evaluations fail to capture such cross-linguistic distribution of information, being largely limited to direct extensions of per-language evaluation protocols. To this end, we propose two metrics to evaluate the information spread across languages: the Knowledge Separability Score (KSS) and the Knowledge Persistence Score (KPS). KSS measures the overall unlearning quality across multiple languages, while KPS more specifically aims to assess consistent removal of information among different language pairs. We evaluated various unlearning methods in the multilingual setting with these metrics and conducted comprehensive analyses. Through our investigation, we provide insights into unique phenomena exclusive to MMU and offer a new perspective on MMU evaluation.
format Preprint
id arxiv_https___arxiv_org_abs_2605_14404
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Knowledge Beyond Language: Bridging the Gap in Multilingual Machine Unlearning Evaluation
Hwang, Kyomin
Kim, Hyeonjin
Cho, Sangyeon
Kwak, Nojun
Computation and Language
While LLMs are increasingly used in commercial services, they pose privacy risks such as leakage of sensitive personally identifiable information (PII). For LLMs trained on multilingual corpora, Multilingual Machine Unlearning (MMU) aims to remove information across multiple languages. However, prior MMU evaluations fail to capture such cross-linguistic distribution of information, being largely limited to direct extensions of per-language evaluation protocols. To this end, we propose two metrics to evaluate the information spread across languages: the Knowledge Separability Score (KSS) and the Knowledge Persistence Score (KPS). KSS measures the overall unlearning quality across multiple languages, while KPS more specifically aims to assess consistent removal of information among different language pairs. We evaluated various unlearning methods in the multilingual setting with these metrics and conducted comprehensive analyses. Through our investigation, we provide insights into unique phenomena exclusive to MMU and offer a new perspective on MMU evaluation.
title Knowledge Beyond Language: Bridging the Gap in Multilingual Machine Unlearning Evaluation
topic Computation and Language
url https://arxiv.org/abs/2605.14404