A recursive least squares framework lets Concept Bottleneck Models add new concepts and classes without gradient updates, reportedly preventing catastrophic forgetting.
Explainable artificial intelligence (xai) 2.0: A manifesto of open challenges and interdisciplinary research directions
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Learning New Concepts, Remembering the Old: Continual Learning for Multimodal Concept Bottleneck Models
A recursive least squares framework lets Concept Bottleneck Models add new concepts and classes without gradient updates, reportedly preventing catastrophic forgetting.