5 citations · 13 across the 15 of their papers we have counts for
Showing cs.CVShow all
2 papers · 1 filter
cs.CV2026
Structuring Sparsity: Block-Sparse Featurizers Capture Visual Concept Manifolds
Thomas Fel, Matthew Kowal, Mozes Jacobs +22
What is the geometry of a visual percept? The most widely used protocols for decomposing neural network representations into interpretable parts treat concepts as isolated directio…
cs.CV2022★ 2 cited
Do Vision-Language Pretrained Models Learn Composable Primitive Concepts?
Tian Yun, Usha Bhalla, Ellie Pavlick +1
Vision-language (VL) pretrained models have achieved impressive performance on multimodal reasoning and zero-shot recognition tasks. Many of these VL models are pretrained on unlab…