RAZOR：面向视觉变换器和扩散模型的比率感知层编辑以实现针对性遗忘
RAZOR: Ratio-Aware Layer Editing for Targeted Unlearning in Vision Transformers and Diffusion Models
AI总结：
RAZOR通过比率感知机制实现视觉变换器和扩散模型的针对性遗忘，通过多层和多头编辑提升模型安全性与合规性，实现高效且稳定的遗忘更新。
摘要：
基于变换器的扩散模型和视觉-语言模型已取得显著成功；然而，高效去除不受欢迎或敏感信息而不重新训练仍是对模型安全性和合规性的核心挑战。我们引入了比率感知零/一步优化保留遗忘（RAZOR），一种轻量、模型无关的遗忘框架，将遗忘更新扩展到变换器骨干中的协调多层和多头编辑。RAZOR通过测量各层和注意力头对遗忘目标数据的贡献程度，识别最重要的层和注意力头。然后，通过精心调节的规则更新这些模型部分，以避免损害整体性能。编辑的组件集逐渐增长，确保精确遗忘，而不会过度编辑或损坏无关能力。我们在CLIP、Stable Diffusion和视觉-语言模型（VLMs）上评估RAZOR，使用广泛采用的遗忘基准测试，涵盖身份、风格和物体擦除任务。我们的结果表明，RAZOR在量化下仍能实现高度准确和稳定的遗忘。该方法相比先前方法具有更强的保留能力和更好的效率。值得注意的是，它比传统方法运行速度显著更快。这些结果表明，RAZOR是视觉模型安全、适应性遗忘的实用且可扩展的解决方案。
英文摘要：
Transformer based diffusion and vision-language models have achieved remarkable success; yet, efficiently removing undesirable or sensitive information without retraining remains a central challenge for model safety and compliance. We introduce Ratio-Aware Zero/One-step Optimized Retentive unlearning (RAZOR), a lightweight, model-agnostic unlearning framework that generalizes forgetting updates to coordinated multi-layer and multi-head edits within transformer backbones. RAZOR identifies the most important layers and attention heads by measuring how much they contribute to forgetting the target data while preserving useful knowledge. Then, it updates these parts of the model using a carefully regularized rule to avoid harming overall performance. The set of edited components grows gradually, ensuring precise unlearning without over-editing or damaging unrelated capabilities. We evaluate RAZOR on CLIP, Stable Diffusion, and vision-language models (VLMs) using widely adopted unlearning benchmarks covering identity, style, and object erasure tasks. Our results show that RAZOR achieves highly accurate and stable forgetting, even under quantization. This approach offers stronger retention and better efficiency than prior methods. Notably, it also operates significant faster than conventional techniques. These results demonstrate that RAZOR is a practical and scalable solution for safe, adaptive unlearning in transformer-based vision models.
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。
