A per-layer learnable soft-thresholding mechanism for SVD-based low-rank compression reduces language model parameters by 50% with roughly 1% average accuracy loss.
https://pythonhosted.org/nvidia-ml-py/
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
SoftLMs: Efficient Adaptive Low-Rank Approximation of Language Models using Soft-Thresholding Mechanism
A per-layer learnable soft-thresholding mechanism for SVD-based low-rank compression reduces language model parameters by 50% with roughly 1% average accuracy loss.