TsqLoRA: Towards Sensitivity and Quality Low-Rank Adaptation for Efficient Fine-Tuning

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Chen, Yu, Han, Yifei, Zhang, Long, Du, Yue, Li, Bin
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866915508399374336
author Chen, Yu
Han, Yifei
Zhang, Long
Du, Yue
Li, Bin
author_facet Chen, Yu
Han, Yifei
Zhang, Long
Du, Yue
Li, Bin
contents Fine-tuning large pre-trained models for downstream tasks has become a fundamental approach in natural language processing. Fully fine-tuning all model parameters is computationally expensive and memory-intensive, especially in resource-constrained environments. Existing parameter-efficient fine-tuning methods reduce the number of trainable parameters but typically overlook the varying sensitivity of different model layers and the importance of training data. In this work, we propose TsqLoRA, a novel method that integrates data-quality-driven selection with sensitivity-aware low-rank adaptation, consisted of two main components: a quality-aware sampling mechanism for selecting the most informative training data, and a dynamic rank allocation module that adjusts the rank of each layer based on its sensitivity to parameter updates. The experimental results demonstrate that TsqLoRA improves fine-tuning efficiency while maintaining or even improving performance on a variety of NLP tasks. Our code will be available at https://github.com/Benjamin-Ricky/TsqLoRA.
format Preprint
id arxiv_https___arxiv_org_abs_2509_18585
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle TsqLoRA: Towards Sensitivity and Quality Low-Rank Adaptation for Efficient Fine-Tuning
Chen, Yu
Han, Yifei
Zhang, Long
Du, Yue
Li, Bin
Computation and Language
Artificial Intelligence
Fine-tuning large pre-trained models for downstream tasks has become a fundamental approach in natural language processing. Fully fine-tuning all model parameters is computationally expensive and memory-intensive, especially in resource-constrained environments. Existing parameter-efficient fine-tuning methods reduce the number of trainable parameters but typically overlook the varying sensitivity of different model layers and the importance of training data. In this work, we propose TsqLoRA, a novel method that integrates data-quality-driven selection with sensitivity-aware low-rank adaptation, consisted of two main components: a quality-aware sampling mechanism for selecting the most informative training data, and a dynamic rank allocation module that adjusts the rank of each layer based on its sensitivity to parameter updates. The experimental results demonstrate that TsqLoRA improves fine-tuning efficiency while maintaining or even improving performance on a variety of NLP tasks. Our code will be available at https://github.com/Benjamin-Ricky/TsqLoRA.
title TsqLoRA: Towards Sensitivity and Quality Low-Rank Adaptation for Efficient Fine-Tuning
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2509.18585