# How Narrative Wrapping Affects LLM Refusal: A Cross-Language Benchmark and Defense

**Published:** October 9, 2026
**Author:** Zhankai Ye, Yanning Wang, Yukai Jin, Bo Mei, Fangyi Li, Wei Wang, Shangqian Gao, Xin Liu
**Source:** arXiv — cs.AI
**Language:** en-US
**HTML:** https://aiqazaqstan.com/en/articles/how-narrative-wrapping-affects-llm-refusal-a-cross-language-benchmark-an

Safety-aligned language models often refuse a harmful request stated directly but answer the same request inside a role-play or narrative wrapper. We measure this vulnerability across languages and registers: attack success on Qwen3-1.7B is already 89.4% in English and 93.0% in…

> From the original source: **arXiv — cs.AI**
> [Read full article at arXiv — cs.AI](https://arxiv.org/abs/2610.11005)

**Tags:** #arXiv

---

*AI Qazaqstan — non-commercial informational portal about AI in Kazakhstan. Machine translations and summaries are AI-generated.*