HALO improves online continual learning under evolving label hierarchies by adaptively combining classification heads regularized with organized learnable prototypes for better adaptation and reduced forgetting.
Title resolution pending
7 Pith papers cite this work. Polarity classification is still indexing.
7
Pith papers citing it
citation-role summary
baseline 1
citation-polarity summary
roles
baseline 1polarities
baseline 1representative citing papers
HC-SOINN with STAR captures topological manifold structure in class features and aligns it to non-linear drift, improving over point-wise NCM when integrated into existing CIL methods.
The paper formalizes Free-Flow Class-Incremental Learning with variable class arrivals and introduces a class-wise mean loss plus targeted adjustments that reduce performance drops seen in standard CIL methods.
A per-class loss reweighting scheme based on distributional robustness allows CLIP models to perform class-incremental and domain-incremental learning with minimal memory while limiting forgetting on CIFAR-100, ImageNet1K, and DomainNet.
DLC inserts lightweight classifier-proximal plugins into distillation-based continual learning to achieve 8% accuracy gains on large benchmarks with only 4% extra backbone parameters.
SoTU merges sparse orthogonal delta parameters learned across streaming tasks to fuse knowledge and mitigate forgetting in pre-trained model continual learning.
citing papers explorer
-
Online Continual Learning with Dynamic Label Hierarchies
HALO improves online continual learning under evolving label hierarchies by adaptively combining classification heads regularized with organized learnable prototypes for better adaptation and reduced forgetting.
-
Beyond Point-wise Neural Collapse: A Topology-Aware Hierarchical Classifier for Class-Incremental Learning
HC-SOINN with STAR captures topological manifold structure in class features and aligns it to non-linear drift, improving over point-wise NCM when integrated into existing CIL methods.
-
Towards Realistic Class-Incremental Learning with Free-Flow Increments
The paper formalizes Free-Flow Class-Incremental Learning with variable class arrivals and introduces a class-wise mean loss plus targeted adjustments that reduce performance drops seen in standard CIL methods.
-
Memory-Efficient Continual Learning with CLIP Models
A per-class loss reweighting scheme based on distributional robustness allows CLIP models to perform class-incremental and domain-incremental learning with minimal memory while limiting forgetting on CIFAR-100, ImageNet1K, and DomainNet.
-
Pushing the Limits of Distillation-Based Continual Learning via Classifier-Proximal Lightweight Plugins
DLC inserts lightweight classifier-proximal plugins into distillation-based continual learning to achieve 8% accuracy gains on large benchmarks with only 4% extra backbone parameters.
-
Sparse Orthogonal Parameters Tuning for Continual Learning
SoTU merges sparse orthogonal delta parameters learned across streaming tasks to fuse knowledge and mitigate forgetting in pre-trained model continual learning.
- Little by Little: Continual Learning via Incremental Mixture of Rank-1 Associative Memory Experts