RepLoRA reparameterizes LoRA's A and B matrices through lightweight MLPs with shared inputs, and claims this changes estimation sample complexity from exponential to polynomial.
emnlp-main.319/
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
-
RepLoRA: Reparameterizing Low-Rank Adaptation via the Perspective of Mixture of Experts
RepLoRA reparameterizes LoRA's A and B matrices through lightweight MLPs with shared inputs, and claims this changes estimation sample complexity from exponential to polynomial.