GFWSVD compresses LLMs by weighting SVD with a Kronecker-factored Fisher information matrix, and reports accuracy gains over FWSVD, ASVD, and SVD-LLM at high compression rates.
A generalized least-square matrix decomposition
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Generalized Fisher-Weighted SVD: Scalable Kronecker-Factored Fisher Approximation for Compressing Large Language Models
GFWSVD compresses LLMs by weighting SVD with a Kronecker-factored Fisher information matrix, and reports accuracy gains over FWSVD, ASVD, and SVD-LLM at high compression rates.