Pith. sign in

BoRA: Bi-dimensional Weight-Decomposed Low-Rank Adaptation

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it
abstract

In recent years, Parameter-Efficient Fine-Tuning (PEFT) methods like Low-Rank Adaptation (LoRA) have significantly enhanced the adaptability of large-scale pre-trained models. Weight-Decomposed Low-Rank Adaptation (DoRA) improves upon LoRA by separating the magnitude and direction components of the weight matrix, leading to superior performance. However, DoRA's improvements are limited to the vertical dimension, resulting in an asymmetrical pattern between horizontal and vertical dimensions. This paper introduces BoRA, an innovative extension of LoRA and DoRA, characterized by symmetrical properties across horizontal and vertical dimensions. Our approach optimizes the weight matrix symmetrically by adjusting both column-wise and row-wise magnitudes. Extensive experiments demonstrate that BoRA surpasses state-of-the-art PEFT methods, including LoRA and DoRA, achieving superior results across various benchmarks.

fields

cs.LG 1

years

2025 1

verdicts

CONDITIONAL 1

representative citing papers

citing papers explorer

Showing 1 of 1 citing paper.

  • MAP: Revisiting Weight Decomposition for Low-Rank Adaptation cs.LG · 2025-05-29 · conditional · none · ref 60 · internal anchor

    MAP decouples a weight matrix's direction and magnitude by normalizing the whole matrix and the low-rank update by their Frobenius norms and scaling each with a learnable scalar.