提出可持续遗忘的神经网络，解决反复删除数据导致模型能力下降的问题。
Trajectory-Guided Forget-Recover Network for Continual LLM Unlearning
- 通过追踪通道级风险，只清除长期相关但可忽略的计算路径。
- 在保留核心能力的前提下，动态激活低遗忘风险的休眠通道以恢复性能。
- 适合需要频繁处理数据删除请求的工业级大模型场景。
机器遗忘旨在消除敏感数据对模型的影响。现实中，遗忘请求持续到来，带来两个挑战：一是遗忘操作可能重新分布目标相关计算，导致已遗忘知识重现；二是多次遗忘会逐步削弱模型保留有用信息的能力。为此，我们提出轨迹引导的遗忘-恢复网络（TFR-Net）。TFR-Net跟踪跨请求的通道级风险，将持久的目标相关通道与临时热点分离，仅抑制前者。同时，通过重新激活对保留效用贡献大、当前及历史遗忘风险均低的休眠通道来恢复模型容量。恢复操作仅在保留效用下降不超过预设阈值时被接受。在四个数据集上的实验表明，TFR-Net始终在遗忘效果与保留效用之间取得更优平衡，优于代表性基线方法。
原文摘要 · Abstract (English)
Machine unlearning aims to eliminate the influence of sensitive data on a model. In the real world, unlearning requests arrive continually, which gives rise to two challenges. First, an unlearning intervention may redistribute target-related computation across remaining pathways, allowing previously forgotten knowledge to re-emerge. Second, repeated unlearning interventions may progressively reduce the model capacity needed to preserve retained utility. To address these challenges, we propose the Trajectory-guided Forget-Recover Network (TFR-Net). TFR-Net tracks channel-level risk across requests. It separates persistent target-related channels from transient hotspots and suppresses only the persistent ones. TFR-Net also recovers model capacity by reactivating dormant channels. These channels make strong contributions to retained utility and show low current and historical forget risk. The recovery is accepted only when retained-utility degradation remains within a predefined tolerance. Experiments on four datasets show that TFR-Net consistently achieves a more favorable trade-off between unlearning effectiveness and retained utility than representative baselines.
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。