Orthogonal Finetuning Made Scalable
This work addresses scalability issues in parameter-efficient adaptation for machine learning practitioners, enabling broader deployment of OFT methods.
The paper tackled the high runtime and memory demands of orthogonal finetuning (OFT) by proposing OFTv2, which uses an input-centric reformulation and efficient parameterization to achieve up to 10x faster training and 3x lower GPU memory usage without performance loss.
Orthogonal finetuning (OFT) offers highly parameter-efficient adaptation while preventing catastrophic forgetting, but its high runtime and memory demands limit practical deployment. We identify the core computational bottleneck in OFT as its weight-centric implementation, which relies on costly matrix-matrix multiplications with cubic complexity. To overcome this, we propose OFTv2, an input-centric reformulation that instead uses matrix-vector multiplications (i.e., matrix-free computation), reducing the computational cost to quadratic. We further introduce the Cayley-Neumann parameterization, an efficient orthogonal parameterization that approximates the matrix inversion in the Cayley transform via a truncated Neumann series. These modifications allow OFTv2 to achieve up to 10x faster training and 3x lower GPU memory usage without compromising performance. In addition, we extend OFTv2 to support finetuning quantized foundation models and show that it outperforms the popular QLoRA in training stability, efficiency, and memory usage.