Computational Lesions in Multilingual Language Models Separate Shared and Language-specific Brain Alignment
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915933494181888 |
|---|---|
| author | Cui, Yang Sun, Jingyuan Sun, Yizheng Wang, Yifan Zhang, Yunhao Li, Jixing Wang, Shaonan Zhou, Hongpeng Hale, John Zong, Chengqing Nenadic, Goran |
| author_facet | Cui, Yang Sun, Jingyuan Sun, Yizheng Wang, Yifan Zhang, Yunhao Li, Jixing Wang, Shaonan Zhou, Hongpeng Hale, John Zong, Chengqing Nenadic, Goran |
| contents | How the brain supports language across different languages is a basic question in neuroscience and a useful test for multilingual artificial intelligence. Neuroimaging has identified language-responsive brain regions across languages, but it cannot by itself show whether the underlying processing is shared or language-specific. Here we use six multilingual large language models (LLMs) as controllable systems and create targeted ``computational lesions'' by zeroing small parameter sets that are important across languages or especially important for one language. We then compare intact and lesioned models in predicting functional magnetic resonance imaging (fMRI) responses during 100 minutes of naturalistic story listening in native English, Chinese and French (112 participants). Lesioning a compact shared core reduces whole-brain encoding correlation by 60.32% relative to intact models, whereas language-specific lesions preserve cross-language separation in embedding space but selectively weaken brain predictivity for the matched native language. These results support a shared backbone with embedded specializations and provide a causal framework for studying multilingual brain-model alignment. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_10627 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Computational Lesions in Multilingual Language Models Separate Shared and Language-specific Brain Alignment Cui, Yang Sun, Jingyuan Sun, Yizheng Wang, Yifan Zhang, Yunhao Li, Jixing Wang, Shaonan Zhou, Hongpeng Hale, John Zong, Chengqing Nenadic, Goran Computation and Language Artificial Intelligence Computational Engineering, Finance, and Science How the brain supports language across different languages is a basic question in neuroscience and a useful test for multilingual artificial intelligence. Neuroimaging has identified language-responsive brain regions across languages, but it cannot by itself show whether the underlying processing is shared or language-specific. Here we use six multilingual large language models (LLMs) as controllable systems and create targeted ``computational lesions'' by zeroing small parameter sets that are important across languages or especially important for one language. We then compare intact and lesioned models in predicting functional magnetic resonance imaging (fMRI) responses during 100 minutes of naturalistic story listening in native English, Chinese and French (112 participants). Lesioning a compact shared core reduces whole-brain encoding correlation by 60.32% relative to intact models, whereas language-specific lesions preserve cross-language separation in embedding space but selectively weaken brain predictivity for the matched native language. These results support a shared backbone with embedded specializations and provide a causal framework for studying multilingual brain-model alignment. |
| title | Computational Lesions in Multilingual Language Models Separate Shared and Language-specific Brain Alignment |
| topic | Computation and Language Artificial Intelligence Computational Engineering, Finance, and Science |
| url | https://arxiv.org/abs/2604.10627 |