OVDU：面向视觉语言模型的开放词表域遗忘协议
Open Vocabulary Domain Unlearning
一篇针对 CLIP 这类视觉语言模型的遗忘论文，指出现有方法只会过拟合见过类别，提出的 4 shots 就能超过 8 shots 基线
arXiv 论文提出 Open-Vocabulary Domain Unlearning（OVDU），指出既有 Approximate Domain Unlearning 方法只在微调时见过的类别上评估遗忘效果，实际只是过拟合了类别-域配对。OVDU 协议要求域遗忘必须迁移到未见过类别，实现类别无关的域擦除。作者用 Fisher Information 掩码隔离域敏感权重，配合 Targeted Manifold Scattering（TMS）目标在局部打散遗忘域的风格几何。在 PACS、OfficeHome、DomainNet 三个基准上，该方法仅用 4 shots 就超过基线 8 shots 的峰值结果。
Open Vocabulary Domain Unlearning
Vision-Language Models (VLMs) exhibit remarkable zero-shot generalization, yet they often encode unwanted or hazardous stylistic domains such as idealized textbook diagrams in medical AI or cartoon vehicles in autonomous driving. Approximate Domain Unlearning (ADU) aims to selectively erase a model's recognition of a target visual domain while preserving accuracy on the remaining domains. However, existing ADU methods operate under a flawed closed-vocabulary assumption: they evaluate unlearning solely on the specific object classes seen during the unlearning fine-tuning phase. Consequently, these methods do not unlearn the domain itself; they merely overfit to seen class-domain pairs, leaving the domain easily recognizable for unseen classes and providing a false sense of removal. We argue that true domain erasure must be class-agnostic. To address this, we formalize Open-Vocabulary Domain Unlearning (OVDU), a rigorous protocol that mandates domain forgetting must transfer to held-out classes. To solve the OVDU challenge, we propose a surgical parameter-editing framework. First, a Fisher Information mask isolates domain-sensitive weights, mathematically protecting foundational zero-shot generalization. Second, our Targeted Manifold Scattering (TMS) objective uses preference-based mining to locally scatter the forget domain's stylistic geometry. Evaluated across PACS, OfficeHome, and DomainNet, our method vastly improves open-vocabulary generalization over existing baselines. Crucially, it delivers exceptional sample efficiency, outperforming peak 8-shot baseline results with only 4 shots.