Results 121 to 130 of about 9,649,421 (248)

HiP-LoRA: Budgeted Spectral Plasticity for Robust Low-Rank Adaptation

Adapting foundation models under resource budgets relies heavily on Parameter-Efficient Fine-Tuning (PEFT), with LoRA being a standard modular solution. However, LoRA suffers from spectral interference. Low-rank updates often concentrate energy on the leading singular directions of pretrained weights, perturbing general capabilities and causing ...
Chen, Lixian, Tan, Jianhong
openaire   +2 more sources

Hypercapnia Induces Mitochondrial Adaptations and Alters Glutamine Metabolism to Drive a Distinct Metabolic Phenotype in Monocytes

Immunology &Cell Biology, EarlyView.
Elevated CO2 (hypercapnia) is a feature of serious lung diseases, for example COPD and contributes to poor patient prognosis through modulation of immune cell function. Here we employ state‐of‐the art metabolomic and isotope tracing approaches to reveal CO2‐dependent mitochondrial alterations linked to suppressed ATP, proline and glutamate levels in ...
Ben Reddan, David E. Phelan, Xiaofei Yin, Lorraine Brennan, Martina Wallace, Eoin P. Cummins   +5 more
wiley   +1 more source

QDyLoRA: Quantized Dynamic Low-Rank Adaptation for Efficient Large Language Model Tuning [PDF]

Finetuning large language models requires huge GPU memory, restricting the choice to acquire Larger models. While the quantized version of the Low-Rank Adaptation technique, named QLoRA, significantly alleviates this issue, finding the efficient LoRA ...
Rezagholizadeh, Mehdi, Ghodsi, Ali, Rajabzadeh, Hossein, Tahaei, Marzieh, Zhu, Tianshu, Kwon, Hyock Ju, Valipour, Mojtaba, Chen, Boxing   +7 more
core   +1 more source

RRLoRA: Refactorized Low-Rank Adaptation With Learning-Rate Restarts for Efficient Fine-Tuning

IEEE Access
Low-rank adaptation (LoRA) has become a standard parameter-efficient fine-tuning technique for adapting large foundation models. However, prior work has shown that LoRA training dynamics and final performance are sensitive to initialization, and state-of-
Mingzhe Yu, Osamu Tatebe
doaj   +1 more source

CL-LoRA: Continual Low-Rank Adaptation for Rehearsal-Free Class-Incremental Learning

2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Class-Incremental Learning (CIL) aims to learn new classes sequentially while retaining the knowledge of previously learned classes. Recently, pre-trained models (PTMs) combined with parameter-efficient fine-tuning (PEFT) have shown remarkable performance in rehearsal-free CIL without requiring exemplars from previous tasks.
Jiangpeng He, Zhihao Duan, Fengqing Zhu 0001   +2 more
openaire   +4 more sources

Artificial Intelligence in Nursing Education: A Scoping Review of Academic Perspectives

Journal of Advanced Nursing, EarlyView.
ABSTRACT Aim To examine nursing academics' perceptions and experiences of artificial intelligence (AI) integration in nursing education. Design Scoping review. Data Sources MEDLINE, CINAHL, ERIC, Scopus, and Web of Science were searched in August 2025. Methods A scoping review using Joanna Briggs Institute methodology.
Natasha Hawkins, Anthea Fagan, Yumiko Coffey, Henri Dohnt   +3 more
wiley   +1 more source

Compress then Merge: From Multiple LoRAs into One Low-Rank Adapter

Accepted to ICML 2026.
He, Zhengbao, Ding, Ruiqi, Huang, Zhehao, Yang, Ruikai, Li, Tao, Huang, Xiaolin   +5 more
openaire   +2 more sources

C-LoRA: Contextual Low-Rank Adaptation for Uncertainty Estimation in Large Language Models

CoRR
Low-Rank Adaptation (LoRA) offers a cost-effective solution for fine-tuning large language models (LLMs), but it often produces overconfident predictions in data-scarce few-shot settings. To address this issue, several classical statistical learning approaches have been repurposed for scalable uncertainty-aware LoRA fine-tuning.
Amir Hossein Rahmati, Sanket R. Jantre, Weifeng Zhang, Yucheng Wang, Byung-Jun Yoon, Nathan M. Urban, Xiaoning Qian   +6 more
openaire   +2 more sources

Understanding culture, memory and trauma in asylum interviews: A mixed‐methods systematic review and meta‐analysis

Legal and Criminological Psychology, EarlyView.
Abstract Purpose Asylum seekers often struggle to recall and report their experiences during asylum interviews. This may occur for several reasons, ranging from communication challenges in high‐context cultures (relying more on indirect and context‐oriented communication) and low‐context cultures (relying more on direct and explicit communication) to ...
Md Yeasir Yunus, Lorraine Hope, Tanja S. van Veldhuizen, Jonathan Koppel   +3 more
wiley   +1 more source

FLoCoRA: Federated learning compression with low-rank adaptation [PDF]

Low-Rank Adaptation (LoRA) methods have gained popularity in efficient parameter fine-tuning of models containing hundreds of billions of parameters. In this work, instead, we demonstrate the application of LoRA methods to train small-vision models in ...
Muller, Guillaume, Leonardon, Mathieu, Ribeiro, Lucas Grativol, Fresse, Virginie, Arzel, Matthieu   +4 more
core   +1 more source