Virginia Smith

See "Links and Anonymity" here: icml.cc/Conferences/...

Federated and Collaborative Learning This program aims to develop theoretical foundations for the field of federated and collaborative learning.

🔗 Learn more about the program here: simons.berkeley.edu/programs/fed...
🔗 Apply here (deadline April 1, 2025): simons.berkeley.edu/research-fel...

The Simons Institute Research Fellowship provides a salary & benefits for the duration of the program, along with housing and visa support. The fellowship is intended for junior researchers who received their PhD at most 5 years ago.

ML systems are increasingly powered by data from many sources and stakeholders. Techniques like FL can enable secure, collaborative learning in such scenarios. The program will convene experts in ML, optimization, economics, privacy & security to bridge theory & practice in the field.

Are you a junior researcher interested in federated and collaborative learning? Consider applying to the Simons Institute Research Fellowship (due April 1) to participate in the Spring 2026 FL/CL Program at the Simons Institute at UC Berkeley! details below 👇

#4: Reviewing takes a lot of work and reviews themselves contain highly valuable information. In line with NeurIPS & ICLR, which have made accepted papers’ reviews public for years, reviews and discussions of all *accepted* ICML papers will be made public after the conference.

#3: The use of GenAI tools for reviewing is prohibited, for two key reasons. First, allowing AI generated reviews can damage the credibility of reviewing. Second, submissions and reviews may contain privileged information that should not be inputted into such tools.

#2: Reviewers tend to focus on novelty over correctness. We've updated the review form to emphasize soundness in addition to significance/novelty, and have asked reviewers to provide evidence for their stance on novelty. See the updated review form here: icml.cc/Conferences/...

#1: The discussion period will be simplified (to review→author response→optional reviewer reply→optional author reply). Discussion periods have intensified in recent years, increasing workloads of authors & reviewers alike despite limited evidence that this improves outcomes.

🔗 Blog post on reviewing updates: medium.com/@icml2025pc/...
🔗 ICML author instructions: icml.cc/Conferences/...
🔗 ICML reviewer instructions: icml.cc/Conferences/...
🔗 ICML peer review FAQ: icml.cc/Conferences/...

There are a few updates to the review process at ICML 2025. These updates are all described on the ICML website, but we also released a blog post explaining our decisions (links & summary 🧵 below):

#4: Lay Summaries: Authors of accepted papers will be required to submit a short “lay summary” (non-technical description) of their work with their camera-ready to improve communication of ML research with the public.

#3: Review process: Discussion period will be pared down (limited to review --> rebuttal --> optional reviewer reply --> optional author reply). All reviewers must check a box acknowledging they read the rebuttal and will be prompted to optionally reply/update their score.

#2: Reciprocal Review Requirement: Similar to ICLR, we'll have a reciprocal review requirement for ICML. All submissions must have at least one author register to review, and any author with 4+ submissions must register to review.

New or semi-new updates #1: Based on the success of the track in 2024, we'll be running a position paper track again this year: icml.cc/Conferences/...

ICML Call for Papers is out! icml.cc/Conferences/...

Excited to serve as a program chair this year along with Maryam Fazel, Daniel Hsu, and Simon Lacoste-Julien.

See our blog post highlighting updates this year: medium.com/@icml2025pc/...; also briefly summarized in 🧵 below