Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Zhu, Yuke, Xie, Chi, Liang, Shuang, Zheng, Bo, Guo, Sheng
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2411.14228
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917844013285376
author Zhu, Yuke
Xie, Chi
Liang, Shuang
Zheng, Bo
Guo, Sheng
author_facet Zhu, Yuke
Xie, Chi
Liang, Shuang
Zheng, Bo
Guo, Sheng
contents Recent advances on Multi-modal Large Language Models have demonstrated that high-resolution image input is crucial for model capabilities, especially for fine-grained tasks. However, high-resolution images lead to a quadratic increase in the number of visual tokens input into LLMs, resulting in significant computational costs. Current work develop visual token compression methods to achieve efficiency improvements, often at the expense of performance. We argue that removing visual redundancy can simultaneously improve both efficiency and performance. We build a coarse-to-fine visual token compression method, with a vision-guided sampler for compressing redundant regions with low information density, and a text-guided sampler for selecting visual tokens that are strongly correlated with the user instructions.With these two modules, the proposed FocusLLaVA achieves improvements in both efficiency and performance. We validate the effectiveness of our approach on a wide range of evaluation datasets.
format Preprint
id arxiv_https___arxiv_org_abs_2411_14228
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression
Zhu, Yuke
Xie, Chi
Liang, Shuang
Zheng, Bo
Guo, Sheng
Computer Vision and Pattern Recognition
Recent advances on Multi-modal Large Language Models have demonstrated that high-resolution image input is crucial for model capabilities, especially for fine-grained tasks. However, high-resolution images lead to a quadratic increase in the number of visual tokens input into LLMs, resulting in significant computational costs. Current work develop visual token compression methods to achieve efficiency improvements, often at the expense of performance. We argue that removing visual redundancy can simultaneously improve both efficiency and performance. We build a coarse-to-fine visual token compression method, with a vision-guided sampler for compressing redundant regions with low information density, and a text-guided sampler for selecting visual tokens that are strongly correlated with the user instructions.With these two modules, the proposed FocusLLaVA achieves improvements in both efficiency and performance. We validate the effectiveness of our approach on a wide range of evaluation datasets.
title FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2411.14228