A dual-phase framework using self-supervised ViT slots optimizes representations for class identity during training and composes them dynamically at inference to achieve state-of-the-art generalization to unseen concepts with minimal forgetting in continual few-shot learning.
Learning Generative Models with
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Unlocking Compositional Generalization in Continual Few-Shot Learning
A dual-phase framework using self-supervised ViT slots optimizes representations for class identity during training and composes them dynamically at inference to achieve state-of-the-art generalization to unseen concepts with minimal forgetting in continual few-shot learning.