Abstract
Undesired information such as harmful content and private data propagates through Multilingual Large Language Models (LLMs) via direct training and indirect cross-linguistic spread. Multilingual Machine Unlearning (MMU) aims to remove such information, yet its evaluation remains underexplored, leaving unclear whether unlearning truly eliminates target knowledge across all languages. To bridge this gap, we introduce $μ^2$-Bench, an MMU benchmark that simulates the full pipeline of memorization, unlearning, and evaluation across diverse languages. It 1) spans a broad set of languages, 2) evaluates on both training and hold-out languages, and 3) assesses knowledge as dispersed across multiple languages. We show that successful MMU requires methods that reflect multilingual characteristics, and conduct analysis to provide deeper insights into MMU.
Keywords
Subject
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Hwang, K., Kim, H., Lee, H., Kim, Y., Song, Y., & Kwak, N. (2026). $μ^2$-Bench: A Multilingual Machine Unlearning Benchmark. https://omanscience.com/en/articles/2-bench-a-multilingual-machine-unlearning-benchmark
MLA 9
Hwang, Kyomin, et al. "$μ^2$-Bench: A Multilingual Machine Unlearning Benchmark." https://omanscience.com/en/articles/2-bench-a-multilingual-machine-unlearning-benchmark.
Chicago (author–date)
Hwang, Kyomin, Hyeonjin Kim, Hyunho Lee, Yearim Kim, Yeji Song, and Nojun Kwak. 2026. "$μ^2$-Bench: A Multilingual Machine Unlearning Benchmark." https://omanscience.com/en/articles/2-bench-a-multilingual-machine-unlearning-benchmark.
Harvard
Hwang, K., Kim, H., Lee, H., Kim, Y., Song, Y. and Kwak, N. (2026) '$μ^2$-Bench: A Multilingual Machine Unlearning Benchmark', Available at: https://omanscience.com/en/articles/2-bench-a-multilingual-machine-unlearning-benchmark.
Vancouver
Hwang K, Kim H, Lee H, Kim Y, Song Y, Kwak N. $μ^2$-Bench: A Multilingual Machine Unlearning Benchmark. https://omanscience.com/en/articles/2-bench-a-multilingual-machine-unlearning-benchmark
IEEE
K. Hwang, H. Kim, H. Lee, Y. Kim, Y. Song, and N. Kwak, "$μ^2$-Bench: A Multilingual Machine Unlearning Benchmark," https://omanscience.com/en/articles/2-bench-a-multilingual-machine-unlearning-benchmark.