No article text was recorded for this version.
cross-lingual safety transfer
Version 1 of 125 papers35 conceptsfrozen 2026-10-04
Papers in this version
- Blog: Align Once, Benefit Multilingually: Enforcing Multilingual Consistency for LLM Safety Alignment
- Blog: Code-Switching Curriculum Learning for Multilingual Transfer in LLMs
- Blog: Analyzing LLMs' Knowledge Boundary Cognition Across Languages Through the Lens of Internal Representations
- Blog: Cendol: Open Instruction-tuned Generative Large Language Models for Indonesian Languages
- Blog: From One to Many: Expanding the Scope of Toxicity Mitigation in Language Models
- Blog: DIPLomA: Efficient Adaptation of Instructed LLMs to Low-Resource Languages via Post-Training Delta Merging
- Blog: Safety at One Shot: Patching Fine-Tuned LLMs with A Single Instance
- Blog: Cross-Lingual Transfer of Debiasing and Detoxification in Multilingual LLMs: An Extensive Investigation
- Blog: Aurora-M: Open Source Continual Pre-training for Multilingual Language and Code
- Blog: DIESEL: A Lightweight Inference-Time Safety Enhancement for Language Models
- Blog: Aya Model: An Instruction Finetuned Open-Access Multilingual Language Model
- Blog: Exploring Multilingual Concepts of Human Values in Large Language Models: Is Value Alignment Consistent, Transferable and Controllable across Languages?
- Blog: Multilingual Safety Alignment Via Sparse Weight Editing
- Blog: The Multilingual Alignment Prism: Aligning Global and Local Preferences to Reduce Harm
- Blog: Refuse Whenever You Feel Unsafe: Improving Safety in LLMs via Decoupled Refusal Training
- Blog: Trust The Typical
- Blog: Preference Tuning For Toxicity Mitigation Generalizes Across Languages
- Blog: IndoSafety: Culturally Grounded Safety for LLMs in Indonesian Languages
- Blog: Multilingual Jailbreak Challenges in Large Language Models
- Blog: Governance in Motion: Co-evolution of Constitutions and AI models for Scalable Safety
- Blog: A Generative Approach to LLM Harmfulness Mitigation with Red Flag Tokens
- Blog: Refusal Direction is Universal Across Safety-Aligned Languages
- Blog: MPO: Multilingual Safety Alignment via Reward Gap Optimization
- Blog: Towards Understanding the Fragility of Multilingual LLMs against Fine-Tuning Attacks
- Blog: Language Models are Homer Simpson! Safety Re-Alignment of Fine-tuned Language Models through Task Arithmetic
History & citation
Current · Permanent link
- This direction split from Cross-Lingual Safety Transfer in Large Language Models (76% overlap)
- This direction split from Cross-Lingual Safety Gaps in Large Language Models (52% overlap)
Cite this direction
Latest version (a living page):
@misc{lacuna_cross_lingual_safety_transfer_92e18c39,
title = {cross-lingual safety transfer},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/cross-lingual-safety-transfer/rdr_92e18c3949eb4893863a9b97d26aa716},
note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-21},
}This exact version (version 1, frozen text):
@misc{lacuna_cross_lingual_safety_transfer_92e18c39_v1,
title = {cross-lingual safety transfer},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/cross-lingual-safety-transfer/rdr_92e18c3949eb4893863a9b97d26aa716/v/1},
note = {Version 1, frozen 2026-10-04. Accessed 2026-11-21},
}