Saved in:
Bibliographic Details
Main Authors: Li, Siwei, Yang, Yifan, Shen, Yifei, Wei, Fangyun, Lu, Zongqing, Qiu, Lili, Yang, Yuqing
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2407.01491
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917709249249280
author Li, Siwei
Yang, Yifan
Shen, Yifei
Wei, Fangyun
Lu, Zongqing
Qiu, Lili
Yang, Yuqing
author_facet Li, Siwei
Yang, Yifan
Shen, Yifei
Wei, Fangyun
Lu, Zongqing
Qiu, Lili
Yang, Yuqing
contents Efficient fine-tuning plays a fundamental role in modern large models, with low-rank adaptation emerging as a particularly promising approach. However, the existing variants of LoRA are hampered by limited expressiveness, a tendency to overfit, and sensitivity to hyperparameter settings. This paper presents LoRA Slow Cascade Learning (LoRASC), an innovative technique designed to enhance LoRA's expressiveness and generalization capabilities while preserving its training efficiency. Our approach augments expressiveness through a cascaded learning strategy that enables a mixture-of-low-rank adaptation, thereby increasing the model's ability to capture complex patterns. Additionally, we introduce a slow-fast update mechanism and cascading noisy tuning to bolster generalization. The extensive experiments on various language and vision datasets, as well as robustness benchmarks, demonstrate that the proposed method not only significantly outperforms existing baselines, but also mitigates overfitting, enhances model stability, and improves OOD robustness. Code will be release in https://github.com/microsoft/LoRASC very soon.
format Preprint
id arxiv_https___arxiv_org_abs_2407_01491
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Expressive and Generalizable Low-rank Adaptation for Large Models via Slow Cascaded Learning
Li, Siwei
Yang, Yifan
Shen, Yifei
Wei, Fangyun
Lu, Zongqing
Qiu, Lili
Yang, Yuqing
Computation and Language
Computer Vision and Pattern Recognition
Efficient fine-tuning plays a fundamental role in modern large models, with low-rank adaptation emerging as a particularly promising approach. However, the existing variants of LoRA are hampered by limited expressiveness, a tendency to overfit, and sensitivity to hyperparameter settings. This paper presents LoRA Slow Cascade Learning (LoRASC), an innovative technique designed to enhance LoRA's expressiveness and generalization capabilities while preserving its training efficiency. Our approach augments expressiveness through a cascaded learning strategy that enables a mixture-of-low-rank adaptation, thereby increasing the model's ability to capture complex patterns. Additionally, we introduce a slow-fast update mechanism and cascading noisy tuning to bolster generalization. The extensive experiments on various language and vision datasets, as well as robustness benchmarks, demonstrate that the proposed method not only significantly outperforms existing baselines, but also mitigates overfitting, enhances model stability, and improves OOD robustness. Code will be release in https://github.com/microsoft/LoRASC very soon.
title Expressive and Generalizable Low-rank Adaptation for Large Models via Slow Cascaded Learning
topic Computation and Language
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2407.01491