Skip to main content
Cornell University
We gratefully acknowledge support from the Simons Foundation, member institutions, and all contributors. Donate
arxiv logo > cs > arXiv:2412.13180

Help | Advanced Search

arXiv logo
Cornell University Logo

quick links

  • Login
  • Help Pages
  • About

Computer Science > Computer Vision and Pattern Recognition

arXiv:2412.13180 (cs)
[Submitted on 17 Dec 2024]

Title:Feather the Throttle: Revisiting Visual Token Pruning for Vision-Language Model Acceleration

Authors:Mark Endo, Xiaohan Wang, Serena Yeung-Levy
View a PDF of the paper titled Feather the Throttle: Revisiting Visual Token Pruning for Vision-Language Model Acceleration, by Mark Endo and 2 other authors
View PDF HTML (experimental)
Abstract:Recent works on accelerating Vision-Language Models show that strong performance can be maintained across a variety of vision-language tasks despite highly compressing visual information. In this work, we examine the popular acceleration approach of early pruning of visual tokens inside the language model and find that its strong performance across many tasks is not due to an exceptional ability to compress visual information, but rather the benchmarks' limited ability to assess fine-grained visual capabilities. Namely, we demonstrate a core issue with the acceleration approach where most tokens towards the top of the image are pruned away. Yet, this issue is only reflected in performance for a small subset of tasks such as localization. For the other evaluated tasks, strong performance is maintained with the flawed pruning strategy. Noting the limited visual capabilities of the studied acceleration technique, we propose FEATHER (Fast and Effective Acceleration wiTH Ensemble cRiteria), a straightforward approach that (1) resolves the identified issue with early-layer pruning, (2) incorporates uniform sampling to ensure coverage across all image regions, and (3) applies pruning in two stages to allow the criteria to become more effective at a later layer while still achieving significant speedup through early-layer pruning. With comparable computational savings, we find that FEATHER has more than $5\times$ performance improvement on the vision-centric localization benchmarks compared to the original acceleration approach.
Comments: Project page: this https URL
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Cite as: arXiv:2412.13180 [cs.CV]
  (or arXiv:2412.13180v1 [cs.CV] for this version)
  https://doi.org/10.48550/arXiv.2412.13180
arXiv-issued DOI via DataCite

Submission history

From: Mark Endo [view email]
[v1] Tue, 17 Dec 2024 18:56:50 UTC (7,137 KB)
Full-text links:

Access Paper:

    View a PDF of the paper titled Feather the Throttle: Revisiting Visual Token Pruning for Vision-Language Model Acceleration, by Mark Endo and 2 other authors
  • View PDF
  • HTML (experimental)
  • TeX Source
  • Other Formats
view license
Current browse context:
cs.CV
< prev   |   next >
new | recent | 2024-12
Change to browse by:
cs

References & Citations

  • NASA ADS
  • Google Scholar
  • Semantic Scholar
a export BibTeX citation Loading...

BibTeX formatted citation

×
Data provided by:

Bookmark

BibSonomy logo Reddit logo

Bibliographic and Citation Tools

Bibliographic Explorer (What is the Explorer?)
Connected Papers (What is Connected Papers?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)

Code, Data and Media Associated with this Article

alphaXiv (What is alphaXiv?)
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Hugging Face (What is Huggingface?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)

Demos

Replicate (What is Replicate?)
Hugging Face Spaces (What is Spaces?)
TXYZ.AI (What is TXYZ.AI?)

Recommenders and Search Tools

Influence Flower (What are Influence Flowers?)
CORE Recommender (What is CORE?)
  • Author
  • Venue
  • Institution
  • Topic

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)
  • About
  • Help
  • contact arXivClick here to contact arXiv Contact
  • subscribe to arXiv mailingsClick here to subscribe Subscribe
  • Copyright
  • Privacy Policy
  • Web Accessibility Assistance
  • arXiv Operational Status
    Get status notifications via email or slack