Forget Less by Learning from Parents Through Hierarchical Relationships
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866909981312286720 |
|---|---|
| author | Kaushik, Arjun Ramesh Devulapally, Naresh Kumar Lokhande, Vishnu Suresh Ratha, Nalini K. Govindaraju, Venu |
| author_facet | Kaushik, Arjun Ramesh Devulapally, Naresh Kumar Lokhande, Vishnu Suresh Ratha, Nalini K. Govindaraju, Venu |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_01892 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Forget Less by Learning from Parents Through Hierarchical Relationships Kaushik, Arjun Ramesh Devulapally, Naresh Kumar Lokhande, Vishnu Suresh Ratha, Nalini K. Govindaraju, Venu Computer Vision and Pattern Recognition Machine Learning 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. |
| title | Forget Less by Learning from Parents Through Hierarchical Relationships |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2601.01892 |