Turbo: Informativity-Driven Acceleration Plug-In for Vision-Language Large Models

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Ju, Chen, Wang, Haicheng, Cheng, Haozhe, Chen, Xu, Zhai, Zhonghua, Huang, Weilin, Lan, Jinsong, Xiao, Shuai, Zheng, Bo
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866917723452211200
author Ju, Chen
Wang, Haicheng
Cheng, Haozhe
Chen, Xu
Zhai, Zhonghua
Huang, Weilin
Lan, Jinsong
Xiao, Shuai
Zheng, Bo
author_facet Ju, Chen
Wang, Haicheng
Cheng, Haozhe
Chen, Xu
Zhai, Zhonghua
Huang, Weilin
Lan, Jinsong
Xiao, Shuai
Zheng, Bo
contents Vision-Language Large Models (VLMs) recently become primary backbone of AI, due to the impressive performance. However, their expensive computation costs, i.e., throughput and delay, impede potentials in the real-world scenarios. To achieve acceleration for VLMs, most existing methods focus on the model perspective: pruning, distillation, quantization, but completely overlook the data-perspective redundancy. To fill the overlook, this paper pioneers the severity of data redundancy, and designs one plug-and-play Turbo module guided by information degree to prune inefficient tokens from visual or textual data. In pursuit of efficiency-performance trade-offs, information degree takes two crucial factors into consideration: mutual redundancy and semantic value. Concretely, the former evaluates data duplication between sequential tokens; while the latter evaluates each token by its contribution to the overall semantics. As a result, tokens with high information degree carry less redundancy and stronger semantics. For VLMs' calculation, Turbo works as a user-friendly plug-in that sorts data referring to information degree, utilizing only top-level ones to save costs. Its advantages are multifaceted, e.g., being generally compatible to various VLMs across understanding and generation, simple use without re-training and trivial engineering efforts. On multiple VLMs benchmarks, we fully experiment to demonstrate the good acceleration of Turbo, under negligible performance drop.
format Preprint
id arxiv_https___arxiv_org_abs_2407_11717
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Turbo: Informativity-Driven Acceleration Plug-In for Vision-Language Large Models
Ju, Chen
Wang, Haicheng
Cheng, Haozhe
Chen, Xu
Zhai, Zhonghua
Huang, Weilin
Lan, Jinsong
Xiao, Shuai
Zheng, Bo
Computer Vision and Pattern Recognition
Vision-Language Large Models (VLMs) recently become primary backbone of AI, due to the impressive performance. However, their expensive computation costs, i.e., throughput and delay, impede potentials in the real-world scenarios. To achieve acceleration for VLMs, most existing methods focus on the model perspective: pruning, distillation, quantization, but completely overlook the data-perspective redundancy. To fill the overlook, this paper pioneers the severity of data redundancy, and designs one plug-and-play Turbo module guided by information degree to prune inefficient tokens from visual or textual data. In pursuit of efficiency-performance trade-offs, information degree takes two crucial factors into consideration: mutual redundancy and semantic value. Concretely, the former evaluates data duplication between sequential tokens; while the latter evaluates each token by its contribution to the overall semantics. As a result, tokens with high information degree carry less redundancy and stronger semantics. For VLMs' calculation, Turbo works as a user-friendly plug-in that sorts data referring to information degree, utilizing only top-level ones to save costs. Its advantages are multifaceted, e.g., being generally compatible to various VLMs across understanding and generation, simple use without re-training and trivial engineering efforts. On multiple VLMs benchmarks, we fully experiment to demonstrate the good acceleration of Turbo, under negligible performance drop.
title Turbo: Informativity-Driven Acceleration Plug-In for Vision-Language Large Models
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2407.11717