2nd Workshop on Test-Time Adaptation: Putting Updates to the Test! (PUT)#

ICML 2025 workshop

_images/PUT_logo.png

Our second workshop on test-time adaptation will be held at ICML 2025 this summer in Vancouver!

When and Where. The workshop will be held on Jul. 18! Come join us in Meeting 220-222.

Scope. Note the increased scope from the last edition to now encompass test-time updates more broadly. As a workshop at ICML, it is important to host and cross-pollinate work across different learning settings and domains.

Consider joining us to discover and contribute to the latest on updates after training: the test begins now!

Call for Papers#

Topics We will welcome and highlight content on test-time and post-training updates:

  • opportunities and challenges for test-time updates for specific applications or deployments;

  • unsupervised or self-supervised losses for optimization during testing;

  • parameterizations of updates to inputs, models, or outputs at any scale and for white-box and black-box systems (including updates to open or closed foundation model systems);

  • coping with test-time input shifts (domain adaptation), test-time attacks (adversarial defense), or test-time task changes (online continual learning);

  • metrics/datasets/benchmarks to validate and evaluate test-time updates by the changes in performance measures and the computation required;

  • adapting large-scale/foundation models (LLMs, VLMs, etc.) to specialized or personalized domains whether by in-context learning / adaptation / continual learning / editing.

Keywords Adaptation, Continual Learning, Robustness, Personalization

Format We will welcome submission of short papers (= 4 pages content without the references as well as an (optional) Appendix with an unlimited number of pages). Accepted submissions will be selected for poster, lightning talk (= 1 slide in 1 minute), and oral presentation at the workshop. The workshop will not include proceedings.

We welcome the submission of relevant, high-quality, short paper versions of papers accepted to the main conference.

Invited Speakers#


Bonus Track Submission#

Submit one-slide updates to share at our bonus track by Jul. 14 2025 at midnight (AoE).

We invite content for a “bonus track” of our workshop to better broadcast projects, progress, and announcements from and for the community. What could be more in the spirit of test-time updates?

The bonus slides will be auto-played during the poster sessions and breaks plus linked here for virtual attendees.

Paper Submission (Done)#

The deadline is done! It was May 23 2025 at midnight (AoE).

Please see the system on OpenReview: https://openreview.net/group?id=ICML.cc/2025/Workshop/PUT.

  • Decisions: June 9th

  • Camera ready: July 9th

Call for Reviewers (Done)#

Thank you to all of our volunteers who signed up to review! The review period has already begun and the sign-up form is now closed.

We are looking for qualified reviewers to help us select papers for the workshop. All reviewers will be credited for their academic service on the workshop site. If you have published on test-time adaptation, continual learning, model editing, and the other topics of our call then please volunteer.

Organizers#

Program Committee#

We thank all of our reviewrs, area chairs, and program chairs for volunteering their time and expertise!

Reviewers

  • Myungjoon Kim

  • Srishti Gupta

  • Ashwin Vaswani

  • Jihyun Yu

  • Ekagra Ranjan

  • Yingnan Liu

  • Jörg Simon

  • Simran Barnwal

  • Junhyug Noh

  • Amrith Coumaran

  • Olivia Markham

  • Rishit Dholakia

  • Drishti Sharma

  • Damian Sójka

  • Taeckyung Lee

  • Wonho Bae

  • Avi Gupta

  • Issam H. Laradji

  • Tim G. Zhou

  • Yunbei Zhang

  • Nicholas I-Hsien Kuo

  • Shyma Yaser Alhuwaider

  • Enis Simsar

  • Heitor Rapela Medeiros

  • Devavrat Tomar

  • Utsav Garg

  • Vladan Stojnić

  • Hyesu Lim

  • Prabhav Sanga

  • Vineet Jain

  • Alireza Dehghanpour Farashah

  • Neil R. Fernandes

  • Komal K. Teru

  • Gido M van de Ven

  • Shambhavi Mishra

  • Kapilan Balagopalan

  • Zixin Ding

  • Yuran Li

  • Daniele Angioni

  • Tigran Fahradyan

  • Srikar Mutnuri

  • Jiayi Guo

  • Julian Bitterwolf

  • Jason Poulos

  • Stefan Nielsen

  • Siddharth Mansingh

  • Rakesh Chowdary Machineni

  • Antonio Carta

  • Albert Manuel Orozco Camacho

  • Klara Janouskova

  • Laurenz A. Farthofer

  • Trung-Hieu Hoang

  • Sazia Mahfuz

  • Yuto Kojima

  • Zefeng Li

  • Saiyue Lyu

  • Nikhil Pratap Ghanathe

  • Benedikt Tscheschner

  • Michael Cooper

  • Cong Liu

  • Jiahang Li

  • Christoph H. Lampert

  • Christos Ziakas

  • Hicham Eddoubi

  • Jack Lu

  • Gianluca Guglielmo

  • Ali Edalati

  • Shuaicheng Niu

  • Olga Saukh

  • Davis Liang

  • Maura Pintor

  • Qingyang Zhang

  • Akash Gupta

  • Idan Shenfeld

  • Anthony Fuller

  • Bogdan Raducanu

  • Vaclav Voracek

  • Mingxuan Liu

  • Ali Parviz

  • Yuran Li

  • Michael Chang

  • Mikel Menta

  • Gabriel Villalonga Pineda

  • Motasem Alfarra

  • Gabriel Villalonga Pineda

  • Hasan Abed Al Kader Hammoud

Area Chairs

_ Francesco Croce _ Yunhui Guo - Sarthak Kumar Maharana _ Marc Masana _ Evgenia Rusak _ Steffen Schneider _ Evan Shelhamer _ Teresa Yeo

Program Chairs

  • Evgenia Rusak

  • Steffen Schneider

  • Evan Shelhamer

Logistics#

Coming soon

Contact#

Please reach the workshop organizers at tta-icml25@googlegroups.com.