SafeRedir: 基于提示嵌入重定向的图像生成模型鲁棒去学习
SafeRedir: Prompt Embedding Redirection for Robust Unlearning in Image Generation Models
AI总结:
SafeRedir通过提示嵌入重定向实现图像生成模型的鲁棒去学习,无需修改模型即可有效消除不安全内容并提升对抗性攻击抵抗力。
摘要:
图像生成模型(IGMs)虽然能够生成令人印象深刻的创造性内容,但往往会记住大量不希望的训练数据中的概念,导致生成不安全内容,如包含不适宜内容的图像和受版权保护的艺术风格。此类行为在现实部署中持续存在安全性和合规性风险,无法通过事后过滤可靠地缓解,因为此类机制的鲁棒性有限且缺乏细粒度的语义控制。最近的去学习方法试图在模型层面擦除有害概念,但存在需要昂贵重新训练、降质良性生成质量或无法抵御提示改写和对抗性攻击的局限性。为了解决这些挑战,我们引入SafeRedir,一种轻量级的推理时间框架,通过提示嵌入重定向实现鲁棒的去学习。不修改底层IGMs,SafeRedir通过嵌入空间中的token级干预将不安全提示路由到安全的语义区域。该框架包含两个核心组件:一个潜意识的多模态安全分类器,用于识别不安全的生成轨迹,以及一个token级的delta生成器,用于精确的语义重定向,配备辅助预测器用于token遮蔽和自适应缩放以局部化和调节干预。在多个代表性的去学习任务中,经验结果表明SafeRedir实现了有效的去学习能力,高语义和感知保留,鲁棒的图像质量以及增强的对抗性攻击抵抗力。此外,SafeRedir在多种扩散后端和现有去学习模型上均能有效泛化,验证了其即插即用的兼容性和广泛的应用性。代码和数据可在https://github.com/ryliu68/SafeRedir上获得。
英文摘要:
Image generation models (IGMs), while capable of producing impressive and creative content, often memorize a wide range of undesirable concepts from their training data, leading to the reproduction of unsafe content such as NSFW imagery and copyrighted artistic styles. Such behaviors pose persistent safety and compliance risks in real-world deployments and cannot be reliably mitigated by post-hoc filtering, owing to the limited robustness of such mechanisms and a lack of fine-grained semantic control. Recent unlearning methods seek to erase harmful concepts at the model level, which exhibit the limitations of requiring costly retraining, degrading the quality of benign generations, or failing to withstand prompt paraphrasing and adversarial attacks. To address these challenges, we introduce SafeRedir, a lightweight inference-time framework for robust unlearning via prompt embedding redirection. Without modifying the underlying IGMs, SafeRedir adaptively routes unsafe prompts toward safe semantic regions through token-level interventions in the embedding space. The framework comprises two core components: a latent-aware multi-modal safety classifier for identifying unsafe generation trajectories, and a token-level delta generator for precise semantic redirection, equipped with auxiliary predictors for token masking and adaptive scaling to localize and regulate the intervention. Empirical results across multiple representative unlearning tasks demonstrate that SafeRedir achieves effective unlearning capability, high semantic and perceptual preservation, robust image quality, and enhanced resistance to adversarial attacks. Furthermore, SafeRedir generalizes effectively across a variety of diffusion backbones and existing unlearned models, validating its plug-and-play compatibility and broad applicability. Code and data are available at https://github.com/ryliu68/SafeRedir.
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。