Distilling safe refusal behavior from OpenAI o1-mini into Llama-3, Gemma-2, and Qwen3 models via response-based LoRA on multilingual jailbreak data increases jailbreak success rates on MultiJail by up to 16.6 points.
Code-switching red-teaming: Llm evaluation for safety and multilingual understanding.arXiv preprint arXiv:2406.15481
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
fields
cs.CL 3representative citing papers
Phonetic perturbations fragment safety-critical tokens in LLMs, suppressing attribution scores while preserving input understanding and causing safety mechanisms to fail despite good comprehension.
English-only safety alignment fails to transfer cross-lingually, while multilingual DPO training on the new RefusEU dataset improves safety across 12 European languages without degrading Global MMLU performance.
citing papers explorer
-
Response-Based Knowledge Distillation for Multilingual Jailbreak Prevention Unwittingly Compromises Safety
Distilling safe refusal behavior from OpenAI o1-mini into Llama-3, Gemma-2, and Qwen3 models via response-based LoRA on multilingual jailbreak data increases jailbreak success rates on MultiJail by up to 16.6 points.
-
Phonetic Perturbations Reveal Tokenizer-Rooted Safety Gaps in LLMs
Phonetic perturbations fragment safety-critical tokens in LLMs, suppressing attribution scores while preserving input understanding and causing safety mechanisms to fail despite good comprehension.
-
Multilingual Refusal Alignment for Safer Large Language Models
English-only safety alignment fails to transfer cross-lingually, while multilingual DPO training on the new RefusEU dataset improves safety across 12 European languages without degrading Global MMLU performance.