close this message
arXiv smileybones

arXiv Is Hiring a DevOps Engineer

Work on one of the world's most important websites and make an impact on open science.

View Jobs
Skip to main content
Cornell University

arXiv Is Hiring a DevOps Engineer

View Jobs
We gratefully acknowledge support from the Simons Foundation, member institutions, and all contributors. Donate
arxiv logo > cs > arXiv:2109.01903

Help | Advanced Search

arXiv logo
Cornell University Logo

quick links

  • Login
  • Help Pages
  • About

Computer Science > Computer Vision and Pattern Recognition

arXiv:2109.01903 (cs)
[Submitted on 4 Sep 2021 (v1), last revised 21 Jun 2022 (this version, v3)]

Title:Robust fine-tuning of zero-shot models

Authors:Mitchell Wortsman, Gabriel Ilharco, Jong Wook Kim, Mike Li, Simon Kornblith, Rebecca Roelofs, Raphael Gontijo-Lopes, Hannaneh Hajishirzi, Ali Farhadi, Hongseok Namkoong, Ludwig Schmidt
View a PDF of the paper titled Robust fine-tuning of zero-shot models, by Mitchell Wortsman and 10 other authors
View PDF
Abstract:Large pre-trained models such as CLIP or ALIGN offer consistent accuracy across a range of data distributions when performing zero-shot inference (i.e., without fine-tuning on a specific dataset). Although existing fine-tuning methods substantially improve accuracy on a given target distribution, they often reduce robustness to distribution shifts. We address this tension by introducing a simple and effective method for improving robustness while fine-tuning: ensembling the weights of the zero-shot and fine-tuned models (WiSE-FT). Compared to standard fine-tuning, WiSE-FT provides large accuracy improvements under distribution shift, while preserving high accuracy on the target distribution. On ImageNet and five derived distribution shifts, WiSE-FT improves accuracy under distribution shift by 4 to 6 percentage points (pp) over prior work while increasing ImageNet accuracy by 1.6 pp. WiSE-FT achieves similarly large robustness gains (2 to 23 pp) on a diverse set of six further distribution shifts, and accuracy gains of 0.8 to 3.3 pp compared to standard fine-tuning on seven commonly used transfer learning datasets. These improvements come at no additional computational cost during fine-tuning or inference.
Comments: CVPR 2022
Subjects: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
Cite as: arXiv:2109.01903 [cs.CV]
  (or arXiv:2109.01903v3 [cs.CV] for this version)
  https://doi.org/10.48550/arXiv.2109.01903
arXiv-issued DOI via DataCite

Submission history

From: Mitchell Wortsman [view email]
[v1] Sat, 4 Sep 2021 17:11:28 UTC (7,738 KB)
[v2] Fri, 25 Feb 2022 02:29:30 UTC (2,600 KB)
[v3] Tue, 21 Jun 2022 21:50:28 UTC (4,036 KB)
Full-text links:

Access Paper:

    View a PDF of the paper titled Robust fine-tuning of zero-shot models, by Mitchell Wortsman and 10 other authors
  • View PDF
  • TeX Source
  • Other Formats
view license
Current browse context:
cs.CV
< prev   |   next >
new | recent | 2021-09
Change to browse by:
cs
cs.LG

References & Citations

  • NASA ADS
  • Google Scholar
  • Semantic Scholar

DBLP - CS Bibliography

listing | bibtex
Mitchell Wortsman
Mike Li
Jong Wook Kim
Hannaneh Hajishirzi
Ali Farhadi
…
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