paper-with-me

홈 › Papers

DoRAN: Stabilizing Weight-Decomposed Low-Rank Adaptation via Noise Injection and Auxiliary Networks

2025-10-05 · Nghiem T. Diep, Hien Dang, Tuan Truong, Tan Dinh, Huy Nguyen, Nhat Ho arxiv

Parameter-efficient fine-tuning (PEFT) methods have become the standard paradigm for adapting large-scale models. Among these techniques, Weight-Decomposed Low-Rank Adaptation (DoRA) has been shown to improve both the learning capacity and training stability of the Low-Rank Adaptation (LoRA) method by explicitly decomposing pre-trained weights into magnitude and directional components. In this work, we propose DoRAN, a new technique designed to stabilize training and boost the sample efficiency of DoRA. Our framework introduces two key components: (i) the injection of learnable noise into the denominator of DoRA weight decomposition, which serves as an adaptive regularizer to mitigate instabilities and improve the estimation rate of low-rank matrices; and (ii) the replacement of static low-rank matrices with auxiliary networks that generate them dynamically, enabling parameter coupling between the query and value projection matrices, leading to improved sample efficiency both theoretically and empirically. Comprehensive experiments on vision and language benchmarks show that DoRAN consistently outperforms LoRA, DoRA, and other PEFT baselines, underscoring the effectiveness of combining noise-based regularization with network-based parameter generation.

📄 PDF Abstract BibTeX arXiv:2510.04331

Code (0)

등록된 구현이 없습니다.

Tasks

parameter-efficient fine-tuning

Similar Papers 제목 키워드 기반

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

2024-12-09 · Qiushi Wang, Yuchen Fan, Junwei Bao, Hongfei Jiang 외

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)…

parameter-efficient fine-tuning

DoTA: Weight-Decomposed Tensor Adaptation for Large Language Models

2024-12-30 · Xiaolin Hu, Xiang Cheng, Peiyu Liu, Wei Liu 외

Low-rank adaptation (LoRA) reduces the computational and memory demands of fine-tuning large language models (LLMs) by approximating updates with low-rank matrices. However, low-rank approximation in two-dimensional spac…

Arithmetic ReasoningQuantizationTensor Decomposition

EDoRA: Efficient Weight-Decomposed Low-Rank Adaptation via Singular Value Decomposition

2025-01-21 · Hamid Nasiri, Peter Garraghan

Parameter-efficient fine-tuning methods, such as LoRA, reduces the number of trainable parameters. However, they often suffer from scalability issues and differences between their learning pattern and full fine-tuning. T…

parameter-efficient fine-tuning

DoRA: Weight-Decomposed Low-Rank Adaptation

2024-02-14 · Shih-Yang Liu, Chien-Yi Wang, Hongxu Yin, Pavlo Molchanov 외

Among the widely used parameter-efficient fine-tuning (PEFT) methods, LoRA and its variants have gained considerable popularity because of avoiding additional inference costs. However, there still often exists an accurac…

parameter-efficient fine-tuning

HOLa: Zero-Shot HOI Detection with Low-Rank Decomposed VLM Feature Adaptation

2025-07-21 · Qinqian Lei, Bo Wang, Robby T. Tan arxiv

Zero-shot human-object interaction (HOI) detection remains a challenging task, particularly in generalizing to unseen actions. Existing methods address this challenge by tapping Vision-Language Models (VLMs) to access kn…