An Empirical Study on Parameter-Efficient Fine-Tuning for MultiModal Large Language Models

TL;DR

This study empirically evaluates parameter-efficient fine-tuning methods for multimodal large language models, demonstrating that adapters generally outperform other methods in enhancing model performance with limited parameter updates.

Contribution

It provides a comprehensive empirical analysis of four PEFT methods across multiple models, datasets, and scenarios, highlighting the effectiveness of adapters and connector layer fine-tuning.

Findings

Adapters are the best-performing PEFT method across experiments.

Fine-tuning connector layers improves performance in most MLLMs.

PEFT methods impact model stability, generalization, and hallucination tendencies.

Abstract

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.