Saved in:
Bibliographic Details
Main Authors: Zhou, Xiongtao, He, Jie, Ke, Yuhua, Zhu, Guangyao, Gutiérrez-Basulto, Víctor, Pan, Jeff Z.
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2406.05130
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916279679451136
author Zhou, Xiongtao
He, Jie
Ke, Yuhua
Zhu, Guangyao
Gutiérrez-Basulto, Víctor
Pan, Jeff Z.
author_facet Zhou, Xiongtao
He, Jie
Ke, Yuhua
Zhu, Guangyao
Gutiérrez-Basulto, Víctor
Pan, Jeff Z.
contents Multimodal large language models (MLLMs) fine-tuned with multimodal instruction datasets have demonstrated remarkable capabilities in multimodal tasks. However, fine-tuning all parameters of MLLMs has become challenging as they usually contain billions of parameters. To address this issue, we study parameter-efficient fine-tuning (PEFT) methods for MLLMs. We aim to identify effective methods for enhancing the performance of MLLMs in scenarios where only a limited number of parameters are trained. This paper conducts empirical studies using four popular PEFT methods to fine-tune the LLM component of open-source MLLMs. We present a comprehensive analysis that encompasses various aspects, including the impact of PEFT methods on various models, parameters and location of the PEFT module, size of fine-tuning data, model stability based on PEFT methods, MLLM's generalization, and hallucination. We evaluated four PEFT methods on seven datasets from two different categories: unseen and seen datasets. Across all experiments, we show that the adapter is the best-performing PEFT method. At the same time, fine-tuning the connector layers leads to improved performance in most MLLMs. Code and data are available at https://github.com/alenai97/PEFT-MLLM.git.
format Preprint
id arxiv_https___arxiv_org_abs_2406_05130
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle An Empirical Study on Parameter-Efficient Fine-Tuning for MultiModal Large Language Models
Zhou, Xiongtao
He, Jie
Ke, Yuhua
Zhu, Guangyao
Gutiérrez-Basulto, Víctor
Pan, Jeff Z.
Computation and Language
Multimodal large language models (MLLMs) fine-tuned with multimodal instruction datasets have demonstrated remarkable capabilities in multimodal tasks. However, fine-tuning all parameters of MLLMs has become challenging as they usually contain billions of parameters. To address this issue, we study parameter-efficient fine-tuning (PEFT) methods for MLLMs. We aim to identify effective methods for enhancing the performance of MLLMs in scenarios where only a limited number of parameters are trained. This paper conducts empirical studies using four popular PEFT methods to fine-tune the LLM component of open-source MLLMs. We present a comprehensive analysis that encompasses various aspects, including the impact of PEFT methods on various models, parameters and location of the PEFT module, size of fine-tuning data, model stability based on PEFT methods, MLLM's generalization, and hallucination. We evaluated four PEFT methods on seven datasets from two different categories: unseen and seen datasets. Across all experiments, we show that the adapter is the best-performing PEFT method. At the same time, fine-tuning the connector layers leads to improved performance in most MLLMs. Code and data are available at https://github.com/alenai97/PEFT-MLLM.git.
title An Empirical Study on Parameter-Efficient Fine-Tuning for MultiModal Large Language Models
topic Computation and Language
url https://arxiv.org/abs/2406.05130