Skip to main content
Cornell University
We gratefully acknowledge support from the Simons Foundation, member institutions, and all contributors. Donate
arxiv logo > cs > arXiv:2207.07635

Help | Advanced Search

arXiv logo
Cornell University Logo

quick links

  • Login
  • Help Pages
  • About

Computer Science > Computer Vision and Pattern Recognition

arXiv:2207.07635 (cs)
[Submitted on 15 Jul 2022]

Title:Is a Caption Worth a Thousand Images? A Controlled Study for Representation Learning

Authors:Shibani Santurkar, Yann Dubois, Rohan Taori, Percy Liang, Tatsunori Hashimoto
View a PDF of the paper titled Is a Caption Worth a Thousand Images? A Controlled Study for Representation Learning, by Shibani Santurkar and 3 other authors
View PDF
Abstract:The development of CLIP [Radford et al., 2021] has sparked a debate on whether language supervision can result in vision models with more transferable representations than traditional image-only methods. Our work studies this question through a carefully controlled comparison of two approaches in terms of their ability to learn representations that generalize to downstream classification tasks. We find that when the pre-training dataset meets certain criteria -- it is sufficiently large and contains descriptive captions with low variability -- image-only methods do not match CLIP's transfer performance, even when they are trained with more image data. However, contrary to what one might expect, there are practical settings in which these criteria are not met, wherein added supervision through captions is actually detrimental. Motivated by our findings, we devise simple prescriptions to enable CLIP to better leverage the language information present in existing pre-training datasets.
Subjects: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG); Machine Learning (stat.ML)
Cite as: arXiv:2207.07635 [cs.CV]
  (or arXiv:2207.07635v1 [cs.CV] for this version)
  https://doi.org/10.48550/arXiv.2207.07635
arXiv-issued DOI via DataCite

Submission history

From: Shibani Santurkar [view email]
[v1] Fri, 15 Jul 2022 17:50:51 UTC (8,911 KB)
Full-text links:

Access Paper:

    View a PDF of the paper titled Is a Caption Worth a Thousand Images? A Controlled Study for Representation Learning, by Shibani Santurkar and 3 other authors
  • View PDF
  • TeX Source
  • Other Formats
license icon view license
Current browse context:
cs.CV
< prev   |   next >
new | recent | 2022-07
Change to browse by:
cs
cs.LG
stat
stat.ML

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