DLP-LoRA: Efficient Task-Specific LoRA Fusion with a Dynamic, Lightweight Plugin for Large Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Yuxuan, Li, Ruizhe
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910833677697024
author Zhang, Yuxuan
Li, Ruizhe
author_facet Zhang, Yuxuan
Li, Ruizhe
contents Recent advancements in Large Language Models (LLMs) have achieved robust performance across diverse tasks, but fine-tuning these models for specific domains remains resource-intensive. Parameter-Efficient Fine-Tuning (PEFT) methods like Low-Rank Adaptation (LoRA) address this challenge by fine-tuning a small subset of parameters. However, existing methods for fusing multiple LoRAs lack dynamic fusion based on contextual inputs and often increase inference time due to token-level operations. We propose DLP-LoRA, a Dynamic Lightweight Plugin that employs a mini-MLP module with only 5M parameters to dynamically fuse multiple LoRAs at the sentence level using top-p sampling strategies. This approach reduces inference time to less than twice that of single LoRA inference by leveraging parallel computation. Evaluations across 26 tasks-including multiple-choice questions and question answering-demonstrate that DLP-LoRA achieves an average accuracy of 92.34% on multiple-choice datasets and significant improvements in BLEU and ROUGE scores on QA datasets, outperforming different LLMs backbones under composite task settings. DLP-LoRA effectively balances performance and efficiency, making it a practical solution for dynamic multi-task adaptation in LLMs. Our code is available at https://github.com/MeCuping/DLP-LoRA.
format Preprint
id arxiv_https___arxiv_org_abs_2410_01497
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle DLP-LoRA: Efficient Task-Specific LoRA Fusion with a Dynamic, Lightweight Plugin for Large Language Models
Zhang, Yuxuan
Li, Ruizhe
Computation and Language
Artificial Intelligence
Machine Learning
Recent advancements in Large Language Models (LLMs) have achieved robust performance across diverse tasks, but fine-tuning these models for specific domains remains resource-intensive. Parameter-Efficient Fine-Tuning (PEFT) methods like Low-Rank Adaptation (LoRA) address this challenge by fine-tuning a small subset of parameters. However, existing methods for fusing multiple LoRAs lack dynamic fusion based on contextual inputs and often increase inference time due to token-level operations. We propose DLP-LoRA, a Dynamic Lightweight Plugin that employs a mini-MLP module with only 5M parameters to dynamically fuse multiple LoRAs at the sentence level using top-p sampling strategies. This approach reduces inference time to less than twice that of single LoRA inference by leveraging parallel computation. Evaluations across 26 tasks-including multiple-choice questions and question answering-demonstrate that DLP-LoRA achieves an average accuracy of 92.34% on multiple-choice datasets and significant improvements in BLEU and ROUGE scores on QA datasets, outperforming different LLMs backbones under composite task settings. DLP-LoRA effectively balances performance and efficiency, making it a practical solution for dynamic multi-task adaptation in LLMs. Our code is available at https://github.com/MeCuping/DLP-LoRA.
title DLP-LoRA: Efficient Task-Specific LoRA Fusion with a Dynamic, Lightweight Plugin for Large Language Models
topic Computation and Language
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2410.01497