SPQR：良性模型适应下安全对齐的多维基准测试
SPQR: A Multi-Dimensional Benchmark for Safety Alignment under Benign Model Adaptation
- Mohamed bin Zayed University of Artificial Intelligence（穆罕默德·本·扎耶德人工智能大学）
- University of Waterloo（滑铁卢大学）
- Michigan State University（密歇根州立大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
研究文本到图像扩散模型安全对齐在良性微调下的稳定性，引入SPQR基准测试，通过单评分指标提供统一框架，经多方面分析确定安全对齐失败情况，为T2I安全对齐技术提供简洁全面的基准。
AI中文摘要：
文本到图像的扩散模型可能会生成受版权保护、不安全或私密的内容。安全对齐旨在抑制特定概念，但评估很少测试在部署后常规应用的良性下游微调（例如LoRA个性化、风格/域适配器）下安全性是否持续。我们研究了当前安全方法在良性微调下的稳定性，发现频繁出现故障。由于真正的安全对齐必须经受住即使是良性的部署后适应，我们引入了SPQR基准测试（安全、提示遵循、质量和鲁棒性）。SPQR是一个单评分指标，通过报告单个排行榜分数来促进比较，提供一个统一、可重复的框架，以评估安全对齐的扩散模型在良性微调下如何很好地保持安全性、实用性和鲁棒性。我们进行了多语言、特定领域和分布外分析，以及按类别细分，以确定良性微调后安全对齐何时失败，最终展示了SPQR作为T2I模型的T2I安全对齐技术的简洁而全面的基准测试。
英文摘要：
Text-to-image diffusion models can emit copyrighted, unsafe, or private content. Safety alignment aims to suppress specific concepts, yet evaluations seldom test whether safety persists under benign downstream fine-tuning routinely applied after deployment (e.g., LoRA personalization, style/domain adapters). We study the stability of current safety methods under benign fine-tuning and observe frequent breakdowns. As true safety alignment must withstand even benign post-deployment adaptations, we introduce the SPQR benchmark (Safety, Prompt adherence, Quality, and Robustness). SPQR is a single-scored metric that provides a unified, reproducible framework to evaluate how well safety-aligned diffusion models preserve safety, utility, and robustness under benign fine-tuning, by reporting a single leaderboard score to facilitate comparisons. We conduct multilingual, domain-specific, and out-of-distribution analyses, along with category-wise breakdowns, to identify when safety alignment fails after benign fine-tuning, ultimately showcasing SPQR as a concise yet comprehensive benchmark for T2I safety alignment techniques for T2I models.
