arXiv cs.CL· Siqi Shen, Lajanugen Logeswaran, Moontae Lee, Honglak Lee, Rada Mihalcea·· 12 小时前AI 评分35
当人设变得有害：用全球公民价值观对齐引导 LLM
When Personas Become Harmful: Guiding LLMs with Global Citizenship Values Alignment
AI 导读
研究者构建了 11k 争议话题与基于 Schwartz 价值观维度的人设配对数据集，通过智能体流水线迭代优化话题与人设，从对齐良好的商业 LLM 中诱导出有害回复，成功率超过 47%。团队以 UNESCO 全球公民原则作为区分正当价值表达与有害行为的最小共识，并提出一个可控对齐模型，让用户在推理时平衡人设遵循与安全性，该模型在多数样本上比基线更安全且更贴合人设。
正文
Abstract:LLMs are increasingly being used by users from diverse demographics with pluralistic values and distinct backgrounds. Maintaining safety while respecting value-laden personas is uniquely challenging because persona-conditioned requests are often not overtly malicious, unlike those in existing safety benchmarks. Alignment methods aimed at adversarial requests therefore may miss the harm that emerges from persona use, leaving models to either refuse legitimate value expression or blindly follow unsafe personas. We construct a dataset of 11k controversial topics paired with value-based personas mapping to value dimensions from the Schwartz Value Survey. Through an agentic pipeline that iteratively refines the topic and persona, we surface and collect harmful responses from well-aligned commercial LLMs with a more than 47% success rate. We use UNESCO's global citizenship principles as a minimal common ground for distinguishing valid value expressions from harmful behaviors, and introduce a controllable alignment model that lets users balance persona adherence and safety at inference time. The resulting model is safer and more persona-adherent than baselines on most samples.
| Comments: | Accepted to EMNLP 2026 (Main Conference) |
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.33899 [cs.CL] |
| (or arXiv:2610.33899v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.33899 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Siqi Shen [view email]
[v1]
Sat, 24 Oct 2026 20:18:51 UTC (1,690 KB)
来源：arXiv cs.CL · arxiv.org