← Search

Aneesh Shetty

1 accepted papers

2024

SVFT: Parameter-Efficient Fine-Tuning with Singular Vectors

NeurIPS 2024poster

Popular parameter-efficient fine-tuning (PEFT) methods, such as LoRA and its variants, freeze pre-trained model weights $\(\mathbf{W}\)$ and inject learnable matrices $\(\mathbf{\Delta W}\)$. These $\(\mathbf{\Delta W}\)$ matrices are structured for efficient parameterization, often using techniques…