# Who Bridges Safety? Identifying and Targeting Cross-Lingual Shared Safety Pathways

arXiv: https://arxiv.org/abs/2608.09095 · HTML: https://mlsanity.com/paper/2608.09095/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 27 of its 38 references that are in the map.
- First posted on arXiv 10 Aug 2026.

## Builds on (most cited first, 27 of 27)

- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Locating and Editing Factual Associations in GPT](https://mlsanity.com/paper/2202.05262/) (2022)
- [Transformer Feed-Forward Layers Are Key-Value Memories](https://mlsanity.com/paper/2012.14913/) (2020)
- [Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!](https://mlsanity.com/paper/2310.03693/) (2023)
- [Mass-Editing Memory in a Transformer](https://mlsanity.com/paper/2210.07229/) (2022)
- [SimPO: Simple Preference Optimization with a Reference-Free Reward](https://mlsanity.com/paper/2405.14734/) (2024)
- [ORPO: Monolithic Preference Optimization without Reference Model](https://mlsanity.com/paper/2403.07691/) (2024)
- [Language Models are Multilingual Chain-of-Thought Reasoners](https://mlsanity.com/paper/2210.03057/) (2022)
- [Low-Resource Languages Jailbreak GPT-4](https://mlsanity.com/paper/2310.02446/) (2023)
- [Efficient and Effective Text Encoding for Chinese LLaMA and Alpaca](https://mlsanity.com/paper/2304.08177/) (2023)
- [Do Llamas Work in English? On the Latent Language of Multilingual Transformers](https://mlsanity.com/paper/2402.10588/) (2024)
- [KTO: Model Alignment as Prospect Theoretic Optimization](https://mlsanity.com/paper/2402.01306/) (2024)
- [Provably Robust DPO: Aligning Language Models with Noisy Feedback](https://mlsanity.com/paper/2403.00409/) (2024)
- [Aegis2.0: A Diverse AI Safety Dataset and Risks Taxonomy for Alignment of LLM Guardrails](https://mlsanity.com/paper/2501.09004/) (2025)
- [The Language Barrier: Dissecting Safety Challenges of LLMs in Multilingual Contexts](https://mlsanity.com/paper/2401.13136/) (2024)
- [Cognitive Overload: Jailbreaking Large Language Models with Overloaded Logical Thinking](https://mlsanity.com/paper/2311.09827/) (2023)
- [Enhancing Multilingual Capabilities of Large Language Models through Self-Distillation from Resource-Rich Languages](https://mlsanity.com/paper/2402.12204/) (2024)
- [CultureGuard: Towards Culturally-Aware Dataset and Guard Model for Multilingual Safety Applications](https://mlsanity.com/paper/2508.01710/) (2025)
- [Beyond Prompt Engineering: Robust Behavior Control in LLMs via Steering Target Atoms](https://mlsanity.com/paper/2505.20322/) (2025)
- [InstructAlign: High-and-Low Resource Language Alignment via Continual Crosslingual Instruction Tuning](https://mlsanity.com/paper/2305.13627/) (2023)
- [The Emergence of Abstract Thought in Large Language Models Beyond Any Language](https://mlsanity.com/paper/2506.09890/) (2025)
- [LinguaSafe: A Comprehensive Multilingual Safety Benchmark for Large Language Models](https://mlsanity.com/paper/2508.12733/) (2025)
- [Popularity-Aware Alignment and Contrast for Mitigating Popularity Bias](https://mlsanity.com/paper/2405.20718/) (2024)
- [Who Transfers Safety? Identifying and Targeting Cross-Lingual Shared Safety Neurons](https://mlsanity.com/paper/2602.01283/) (2026)
- [Know Your Account: Double Graph Inference-based Account De-anonymization on Ethereum](https://mlsanity.com/paper/2411.18875/) (2024)

## Abstract (arXiv)

Uncovering the internal mechanisms underlying the safety capabilities of large language models (LLMs) is crucial for developing trustworthy artificial intelligence. Currently, mechanistic interpretability studies on multilingual safety are largely confined to local components, such as isolated neurons. However, this static and fragmented perspective overlooks the synergy among components and fails to elucidate how safety signals dynamically propagate within the model to drive safety decisions ultimately. In this work, we move beyond isolated neurons to identify and target the cross-layer functional pathways formed during safety signal propagation, thereby uncovering the mechanisms driving the cross-lingual safety gap. Specifically, we first identify monolingual safety pathways and validate their impact on refusing harmful requests. Subsequent cross-lingual analyses reveal a sparse subset of cross-lingual shared safety pathways, confirming that this intersection acts as the internal bridge transferring safety capabilities from high-resource (HR) languages to non-high-resource (NHR) languages. Building on these mechanistic findings, we propose a pathways-targeted alignment method based on the cross-lingual shared safety pathways. Experimental results show that updating only a small fraction of pathway parameters significantly improves safety in NHR languages while largely preserving the model's general capabilities.