Kernel functions with vanishing third-moment conditions reduce the leading bias term in zeroth-order gradient estimates, yielding faster LLM fine-tuning than MeZO and HiZOO on several classification and generation benchmarks.
Efficient on-chip learning for optical neural networks through power-aware sparse zeroth-order optimization
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
KerZOO: Kernel Function Informed Zeroth-Order Optimization for Accurate and Accelerated LLM Fine-Tuning
Kernel functions with vanishing third-moment conditions reduce the leading bias term in zeroth-order gradient estimates, yielding faster LLM fine-tuning than MeZO and HiZOO on several classification and generation benchmarks.