对齐下限：角色定制如何破坏弱对齐大语言模型的安全性
The Alignment Floor: How Persona Customization Breaks Safety in Weakly-Aligned LLMs
AI总结：
通过对比强对齐与弱对齐模型在不同角色条件下的谄媚率变化，定义对齐下限Δ_floor作为评估模型角色定制安全性的审计指标。
摘要：
告诉LLM“要热情”会使轻对齐模型的谄媚率从30%上升到50%，但对强对齐模型没有影响。我们将这一差距定义为对齐下限Δ_floor(m)=max_pS(m,p)-min_pS(m,p)，即模型在不同角色条件下产生的谄媚率范围，并将谄媚视为角色条件属性而非固定模型属性。多元AI依赖于通过角色提示（如“要有创造力”或“要彻底”）进行行为适应，使系统能够尊重不同的用户价值观和沟通风格；安全问题在于给定模型在真实性改变之前能吸收多少定制化。我们进行了一项受控案例研究，对比了强对齐的RLHF+宪法AI模型（Claude Sonnet 4.6）与轻对齐模型（Amazon Nova Lite），涵盖7种角色条件和5个任务，共1800次运行。存在性结果促使进行逐模型审计：至少有一个强对齐模型的Δ_floor=5个百分点（在15%控制率的5个百分点内），至少有一个轻对齐模型的Δ_floor=45个百分点（范围5%-50%）。在轻对齐模型上，所有五种大五人格角色都增加了谄媚率，且反直觉的是，宜人性产生的增幅最小而非最大。研究中最大的单一效果是建设性的：怀疑论者角色使轻对齐模型的谄媚率降低了25个百分点，并且是唯一指示抵制用户主张而非与之互动的角色，这暗示了方向性解释。角色效果的跨模型迁移几乎为零，因此角色-对齐测试必须逐模型进行。我们提出Δ_floor作为部署时的审计指标：在部署角色定制之前，在小规模角色面板上测量该指标。
英文摘要：
Telling an LLM to "be enthusiastic" raises its sycophancy rate from 30\% to 50\% on a lightly-aligned model, but has zero effect on a strongly-aligned one. We define this gap as the alignment floor, $Δ_{\text{floor}}(m)=\max_pS(m,p)-\min_pS(m,p)$, the range of sycophancy rates a model produces across persona conditions, and treat sycophancy as a persona-conditional property rather than a fixed model property. Pluralistic AI relies on behavioral adaptation via persona prompts like "be creative" or "be thorough", which let systems respect diverse user values and communication styles; the safety question is how much customization a given model can absorb before its truthfulness shifts. We present a controlled case study contrasting a strongly-aligned RLHF + Constitutional-AI model (Claude Sonnet 4.6) with a more lightly-aligned model (Amazon Nova Lite), spanning seven persona conditions and five tasks for 1800 total runs. An existence-pair result motivates per-model auditing: there is at least one strongly-aligned model with $Δ_{\text{floor}}=5$pp (within 5pp of the 15\% control rate) and at least one lightly-aligned model with 45pp (5\%--50\% range). On the lightly-aligned model, all five Big Five personas increase sycophancy over control, and counterintuitively Agreeableness produces the smallest increase, not the largest. The single largest effect in the study is constructive: a Skeptic persona reduces sycophancy by 25pp on the lightly-aligned model, and is the only persona that instructs resistance against user claims rather than engagement with them, suggesting a directionality account. Cross-model transfer of persona effects is near-zero, so persona-alignment testing must be per-model. We propose $Δ_{\text{floor}}$ as a deployment-time audit metric: measure it on a small persona panel before deploying persona customization.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
