通过父概念引导新概念学习，缓解生成模型的遗忘问题。
Forget Less by Learning from Parents Through Hierarchical Relationships
- 在双曲空间中构建概念的父子关系，利用已有知识指导新知识学习。
- 在三个公开数据集和一个合成基准上，显著提升模型持续学习的鲁棒性与泛化能力。
- 适合需要长期增量学习的个性化生成模型研究者使用。
定制化扩散模型（CDMs）在生成建模个性化方面表现优异，但在顺序学习新概念时仍面临灾难性遗忘问题。现有方法多聚焦于减少概念间的干扰，却忽视了正向的概念交互潜力。本文提出一种名为FLLP的新框架，通过在双曲空间中引入父-子概念学习机制来缓解遗忘。该方法将概念表示嵌入洛伦兹流形，天然适合建模树状层级结构，使先前学习的概念作为新概念适应的指导。所提方法不仅有效保留旧知识，还支持新概念的持续集成。我们在三个公开数据集和一个合成基准上验证了FLLP，结果表明其在鲁棒性和泛化能力上均有稳定提升。
原文摘要 · Abstract (English)
Custom Diffusion Models (CDMs) offer impressive capabilities for personalization in generative modeling, yet they remain vulnerable to catastrophic forgetting when learning new concepts sequentially. Existing approaches primarily focus on minimizing interference between concepts, often neglecting the potential for positive inter-concept interactions. In this work, we present Forget Less by Learning from Parents (FLLP), a novel framework that introduces a parent-child inter-concept learning mechanism in hyperbolic space to mitigate forgetting. By embedding concept representations within a Lorentzian manifold, naturally suited to modeling tree-like hierarchies, we define parent-child relationships in which previously learned concepts serve as guidance for adapting to new ones. Our method not only preserves prior knowledge but also supports continual integration of new concepts. We validate FLLP on three public datasets and one synthetic benchmark, showing consistent improvements in both robustness and generalization.
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。