To Preserve or To Compress: An In-Depth Study of Connector Selection in Multimodal Large Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lin, Junyan, Chen, Haoran, Zhu, Dawei, Shen, Xiaoyu
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910641955012608
author Lin, Junyan
Chen, Haoran
Zhu, Dawei
Shen, Xiaoyu
author_facet Lin, Junyan
Chen, Haoran
Zhu, Dawei
Shen, Xiaoyu
contents In recent years, multimodal large language models (MLLMs) have garnered significant attention from both industry and academia. However, there is still considerable debate on constructing MLLM architectures, particularly regarding the selection of appropriate connectors for perception tasks of varying granularities. This paper systematically investigates the impact of connectors on MLLM performance. Specifically, we classify connectors into feature-preserving and feature-compressing types. Utilizing a unified classification standard, we categorize sub-tasks from three comprehensive benchmarks, MMBench, MME, and SEED-Bench, into three task types: coarse-grained perception, fine-grained perception, and reasoning, and evaluate the performance. Our findings reveal that feature-preserving connectors excel in \emph{fine-grained perception} tasks due to their ability to retain detailed visual information. In contrast, feature-compressing connectors, while less effective in fine-grained perception tasks, offer significant speed advantages and perform comparably in \emph{coarse-grained perception} and \emph{reasoning} tasks. These insights are crucial for guiding MLLM architecture design and advancing the optimization of MLLM architectures.
format Preprint
id arxiv_https___arxiv_org_abs_2410_06765
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle To Preserve or To Compress: An In-Depth Study of Connector Selection in Multimodal Large Language Models
Lin, Junyan
Chen, Haoran
Zhu, Dawei
Shen, Xiaoyu
Computation and Language
Computer Vision and Pattern Recognition
In recent years, multimodal large language models (MLLMs) have garnered significant attention from both industry and academia. However, there is still considerable debate on constructing MLLM architectures, particularly regarding the selection of appropriate connectors for perception tasks of varying granularities. This paper systematically investigates the impact of connectors on MLLM performance. Specifically, we classify connectors into feature-preserving and feature-compressing types. Utilizing a unified classification standard, we categorize sub-tasks from three comprehensive benchmarks, MMBench, MME, and SEED-Bench, into three task types: coarse-grained perception, fine-grained perception, and reasoning, and evaluate the performance. Our findings reveal that feature-preserving connectors excel in \emph{fine-grained perception} tasks due to their ability to retain detailed visual information. In contrast, feature-compressing connectors, while less effective in fine-grained perception tasks, offer significant speed advantages and perform comparably in \emph{coarse-grained perception} and \emph{reasoning} tasks. These insights are crucial for guiding MLLM architecture design and advancing the optimization of MLLM architectures.
title To Preserve or To Compress: An In-Depth Study of Connector Selection in Multimodal Large Language Models
topic Computation and Language
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2410.06765