通过竞争机制实现扩散模型中概念遗忘，平衡删除与保留。
Forgetting is Competition: Rethinking Unlearning as Representation Interference in Diffusion Models
- 将遗忘视为干扰竞争，用多样提示生成对抗梯度。
- 在多个数据集上实现更强的删除-保留平衡，减少副作用。
- 适合需要精准内容移除的版权、安全等场景。
已部署的文本到图像扩散模型需在不重新训练的前提下进行后置概念遗忘，以应对版权争议、艺术家退出、安全更新和受保护内容处理。核心挑战是删除与保留之间的失衡：激进更新会削弱目标概念但损害共享能力，而保守或锚点式更新虽保持质量却可能导致通过相关、组合、改写或对抗性提示恢复目标。受逆行干扰启发，我们提出SurgUn，将遗忘视为可控竞争而非直接删除或一对一重分配。SurgUn通过干扰条件梯度竞争实现逆行概念干扰：目标梯度上升削弱目标条件去噪或流匹配行为，而对语义多样的干扰集执行下降操作，在相同提示上下文中引入非目标轨迹竞争。这使输出分布在多个非目标模式之间，而非坍缩为单一代理。为限制共享路径上的附带遗忘，SurgUn引入像素引导的权重空间定位，一种轻量级诊断方法，通过生成图像的删留行为选择注意力模块，利用抑制普遍可实现而保留具有块选择性的不对称性。在UnlearnCanvas、IP角色擦除、整体遗忘、EraseBench和Ring-A-Bell测试集上，SurgUn在Stable Diffusion v1.5、SDXL和SANA-1.5上均优于基线方法。消融实验表明，多样干扰项、对比竞争和定位机制均为实现稳健抑制并保留相关与无关概念所必需。
原文摘要 · Abstract (English)
Deployed text-to-image diffusion models increasingly require post-hoc concept unlearning for copyright claims, artist opt-outs, safety updates, and protected-content mitigation without full retraining. A central challenge is erase-retain imbalance, aggressive updates suppress targets but damage shared capabilities, while conservative or anchor-based updates preserve quality yet leave concepts recoverable through related, compositional, paraphrased, or adversarial prompts. Inspired by retroactive interference, we propose SurgUn, which treats forgetting as controlled competition rather than direct deletion or one-to-one reassignment. SurgUn instantiates retroactive concept interference via distractor-conditioned gradient competition: target-gradient ascent weakens target-conditioned denoising or flow-matching behavior, while descent over a semantically diverse distractor set introduces competing non-target trajectories under the same prompt context. This redistributes outputs across multiple non-target modes instead of collapsing to a single proxy. To limit collateral forgetting through shared pathways, SurgUn adds pixel-grounded weight-space localization, a lightweight diagnostic that selects attention blocks by generated-image erase-retain behavior, exploiting the asymmetry that suppression is broadly achievable whereas retention is block-selective. Across UnlearnCanvas, IP-character erasure, Holistic Unlearning, EraseBench, and Ring-A-Bell on Stable Diffusion v1.5, SDXL, and SANA-1.5, SurgUn achieves a stronger erase-retain balance than baselines. Ablations show that diverse distractors, contrastive competition, and localization are all necessary for robust suppression while preserving related and unrelated concepts.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。