mlsanityData to 14 Oct 2026

Layer-wise Swapping for Generalizable Multilingual Safety

arXiv 2601.22620 ↗First posted 30 Jan 2026cs.CLcited by 2

On the map: LLM safety and security → Safety alignment, refusal and guardrails

Abstract (arXiv)

Despite the rapid advancements of Large Language Models (LLMs), safety risks remain a critical challenge for low-resource languages. Existing safety datasets are predominantly English centric, limiting progress in multilingual safety alignment. As a result, low resource expert models, finetuned on their respective instruction datasets, tend to exhibit higher unsafety rates compared to their high resource counterparts. In this work, we propose a safety aware layer swapping method that transfers safety alignment from an English safety expert to low resource language experts without additional training. To further enhance transfer ability, our method adaptively selects or blends modules based on their degree of specialization. Our approach preserves performance on general language understanding tasks while enhancing safety in the target languages. Experimental results show that the proposed method achieves comparable performance to the language expert on general benchmarks such as MMMLU, BELEBELE, and MGSM, while producing more aligned and less harmful responses on the MultiJail safety benchmark.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in Safety alignment, refusal and guardrails, one of 12 directions in LLM safety and security. Builds on 17 of its 30 references that are in the map. First posted on arXiv 30 Jan 2026.

Builds on

17 of 30 references.

References · 17

  1. 2021Training Verifiers to Solve Math Word Problemscited by 10,109
  2. 2023Language Models are Super Mario: Absorbing Abilities from Homologous Models as a Free Lunchcited by 694
  3. 2023Safety-Tuned LLaMAs: Lessons From Improving the Safety of Large Language Models that Follow Instructionscited by 387
  4. 2023The Belebele Benchmark: a Parallel Reading Comprehension Dataset in 122 Language Variantscited by 311
  5. 2024How do Large Language Models Handle Multilingualism?cited by 150
  6. 2024Safety Layers in Aligned Large Language Models: The Key to LLM Securitycited by 134
  7. 2024Representation Surgery for Multi-Task Model Mergingcited by 96
  8. 2024The Language Barrier: Dissecting Safety Challenges of LLMs in Multilingual Contextscited by 94
  9. 2024On the Multilingual Ability of Decoder-based Pre-trained Language Models: Finding and Controlling Language-Specific Neuronscited by 83
  10. 2024Defending Large Language Models Against Jailbreak Attacks via Layer-specific Editingcited by 79
  11. 2024IndicLLMSuite: A Blueprint for Creating Pre-training and Fine-Tuning Datasets for Indian Languagescited by 52
  12. 2024Enhancing Multilingual Capabilities of Large Language Models through Self-Distillation from Resource-Rich Languagescited by 36
5 more
  1. 2024Mitigating Catastrophic Forgetting in Language Transfer via Model Mergingcited by 36
  2. 2023Language and Task Arithmetic with Parameter-Efficient Layers for Zero-Shot Summarizationcited by 28
  3. 2024Layer Swapping for Zero-Shot Cross-Lingual Transfer in Large Language Modelscited by 21
  4. 2024Cross-lingual Transfer of Reward Models in Multilingual Alignmentcited by 17
  5. 2025Leveraging Submodule Linearity Enhances Task Arithmetic Performance in LLMscited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026Towards Truly Multilingual ASR: Generalizing Code-Switching ASR to Unseen Language Pairscited by 0
  2. 2026LLM Safety Alignment in Low-Resource Languages: A Systematic Literature Reviewcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Multilingual Safety Alignment Via Sparse Weight Editing
  2. 2024Towards Comprehensive Post Safety Alignment of Large Language Models via Safety Patching
  3. 2026Safety at One Shot: Patching Fine-Tuned LLMs with A Single Instance
  4. 2026LSSF: Safety Alignment for Large Language Models through Low-Rank Safety Subspace Fusion
  5. 2025SafeMERGE: Preserving Safety Alignment in Fine-Tuned Large Language Models via Selective Layer-Wise Model Merging
  6. 2025Safe at the Margins: A General Approach to Safety Alignment in Low-Resource English Languages -- A Singlish Case Study
  7. 2026Multilingual Safety Signals Are Multi-Layered: Filtering Safety-Degrading Data for Safer LLMs
  8. 2025CREST: Universal Safety Guardrails Through Cluster-Guided Cross-Lingual Transfer