Who Transfers Safety? Identifying and Targeting Cross-Lingual Shared Safety Neurons

TL;DR

This paper identifies a small set of cross-lingual safety neurons in multilingual language models that are crucial for safety transfer, and proposes a neuron-focused training method to improve safety in low-resource languages.

Contribution

The study uncovers shared safety neurons across languages and introduces a neuron-targeted training strategy that outperforms existing methods in enhancing multilingual safety.

Findings

Shared safety neurons regulate safety across languages.

Suppressing these neurons reduces safety in non-high-resource languages.

Reinforcing these neurons improves safety transfer and consistency.

Abstract

Multilingual safety remains significantly imbalanced, leaving non-high-resource (NHR) languages vulnerable compared to robust high-resource (HR) ones. Moreover, the neural mechanisms driving safety alignment remain unclear despite observed cross-lingual representation transfer. In this paper, we find that LLMs contain a set of cross-lingual shared safety neurons (SS-Neurons), a remarkably small yet critical neuronal subset that jointly regulates safety behavior across languages. We first identify monolingual safety neurons (MS-Neurons) and validate their causal role in safety refusal behavior through targeted activation and suppression. Our cross-lingual analyses then identify SS-Neurons as the subset of MS-Neurons shared between HR and NHR languages, serving as a bridge to transfer safety capabilities from HR to NHR domains. We observe that suppressing these neurons causes concurrent safety drops across NHR languages, whereas reinforcing them improves cross-lingual defensive consistency. Building on these insights, we propose a simple neuron-oriented training strategy that targets SS-Neurons based on language resource distribution and model architecture. Experiments demonstrate that fine-tuning this tiny neuronal subset outperforms state-of-the-art methods, significantly enhancing NHR safety while maintaining the model's general capabilities. The code and dataset will be available athttps://github.com/1518630367/SS-Neuron-Expansion.