paper-with-me

홈 › Papers

Rank-Accuracy Trade-off for LoRA: A Gradient-Flow Analysis

2026-02-10 · Michael Rushka, Diego Klabjan arxiv

Previous empirical studies have shown that LoRA achieves accuracy comparable to full-parameter methods on downstream fine-tuning tasks, even for rank-1 updates. By contrast, the theoretical underpinnings of the dependence of LoRA's accuracy on update rank remain relatively unexplored. In this work, we compare the accuracy of rank-r LoRA updates against full-parameter updates for fine-tuning tasks from a dynamical systems perspective. We perform gradient flow analysis in both full-rank and low-rank regimes to establish explicit relationships between rank and accuracy for two loss functions under LoRA. While gradient flow equations for LoRA are presented in prior work, we rigorously derive their form and show that they are identical for simultaneous and sequential LoRA parameter updates. We then use the resulting dynamical system equations to obtain closed-form relationships between LoRA rank and accuracy for trace-squared and Frobenius-norm low-rank approximation loss functions.

📄 PDF Abstract BibTeX arXiv:2602.10212

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

MatryoshkaLoRA: Learning Accurate Hierarchical Low-Rank Representations for LLM Fine-Tuning

2026-05-08 · Ionut-Vlad Modoranu, Mher Safaryan, Dan Alistarh arxiv

With the rise in scale for deep learning models to billions of parameters, the computational cost of fine-tuning remains a significant barrier to deployment. While Low-Rank Adaptation (LoRA) has become the standard for p…

parameter-efficient fine-tuning

Parameter-Efficient Domain Adaptation of Physics-Informed Self-Attention based GNNs for AC Power Flow Prediction

2026-02-20 · Redwanul Karim, Changhun Kim, Timon Conrad, Nora Gourmelon 외 arxiv

Accurate AC power flow (AC-PF) prediction under domain shift is critical when models trained on medium-voltage (MV) grids are deployed on high-voltage (HV) networks. Existing physics-informed graph neural network (GNN) s…

Graph Neural NetworkDomain Adaptation

Rethinking LoRA for Privacy-Preserving Federated Learning in Large Models

2026-02-23 · Jin Liu, Yinbin Miao, Ning Xi, Junkang Liu arxiv

Fine-tuning large vision models (LVMs) and large language models (LLMs) under differentially private federated learning (DPFL) is hindered by a fundamental privacy-utility trade-off. Low-Rank Adaptation (LoRA), a promisi…

parameter-efficient fine-tuningFederated Learning

MAMA-FLUX.2: Image-to-Image Synthesis of Post-Contrast Breast DCE-MRI for the MAMA-SYNTH Challenge

2026-08-26 · Kamil Kwarciak, Marek Wodzinski arxiv

Dynamic contrast-enhanced breast MRI is central to cancer diagnosis and monitoring, but requires gadolinium-based contrast agents. In this work, we address pre-to-post contrast breast MRI synthesis for the MAMA-SYNTH cha…

parameter-efficient fine-tuning

When Is Rank-1 Enough? Geometry-Guided Initialization for Parameter-Efficient Fine-Tuning

2026-02-02 · Haoran Zhao, Soyeon Caren Han, Eduard Hovy arxiv

Parameter-efficient fine-tuning (PEFT) is a standard way to adapt multimodal large language models, yet extremely low-rank settings -- especially rank-1 LoRA -- are often unstable. We show that this instability is not so…

parameter-efficient fine-tuning