CRISP unifies model compression and parameter-efficient fine-tuning by decomposing weights into shared bases and small mixers, reporting 1-5% gains over prior dual-task and specialized methods.
Intra-head pruning for vision transformers via inter-layer dimension relationship modeling.Neural Networks, 190: 107656
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression
CRISP unifies model compression and parameter-efficient fine-tuning by decomposing weights into shared bases and small mixers, reporting 1-5% gains over prior dual-task and specialized methods.