A stateless, SVD-free regularizer approximates polar factors to induce low-rank weight structure during training, enabling better post-training compression of vision models and LLMs at under 8% overhead.
hub
Speeding up Convolutional Neural Networks with Low Rank Expansions
1 Pith paper cite this work, alongside 1,150 external citations. Polarity classification is still indexing.
1
Pith paper citing it
1,150
external citations · OpenAlex
hub tools
fields
cs.LG 1years
2026 1verdicts
ACCEPT 1representative citing papers
citing papers explorer
-
SLORR: Simple and Efficient In-Training Low-Rank Regularization
A stateless, SVD-free regularizer approximates polar factors to induce low-rank weight structure during training, enabling better post-training compression of vision models and LLMs at under 8% overhead.