Research
Completed or high-quality work-in-progress original research in ML for the physical sciences, applications of physics methods in ML, or related topics.
Machine Learning/CORE Unranked
ML4PS 2026
Machine Learning and the Physical Sciences 2026
The ML4PS 2026 workshop is an annual gathering at the intersection of machine learning and the physical sciences, fostering research that applies ML to problems in physics, chemistry, astronomy, materials science, and related fields, as well as using physical insights to advance ML methods. It emphasizes in-person participation, community-driven review, and the exploration of emerging topics such as AI's role in reshaping scientific research practices and the interplay between academia and industry in fundamental science.
Paper fit
Contribution paths
A strong submission should clearly identify its contribution and evaluate it appropriately.
Evaluations & Datasets
Contributions that advance evaluative practices in ML and the physical sciences, including development and use of datasets, benchmarks, and resources.
Perspectives
Compelling and thoughtful commentaries on recent directions and open questions at the intersection of ML and the physical sciences.
Research areas in scope
01
Research Tracks
ML for PhysicsPhysics in MLOther areas (e.g., probabilistic methods, deep generative models, scientific foundation models, simulation-based inference, variational inference, causal inference)Evaluations & DatasetsPerspectives
Policies worth checking twice
- Submissions must be short papers up to 4 pages (excluding references) using the NeurIPS 2026 LaTeX template.
- Papers must be fully anonymized for double-blind review, except for the Evaluations & Datasets track which allows single-blind review.
- Code, links, text, and figures must be anonymized; tools like Anonymous Github are encouraged.
- No author additions are permitted after the review process begins.
- Double-submission to other NeurIPS workshops is strictly prohibited and will result in desk rejection.
- Papers with erroneous citations (e.g., mismatched titles or author lists) will not be reviewed.
- Appendices are discouraged and reviewers are not required to read them.
- Authors using generative AI must describe its use and verify the accuracy of generated outputs; unverified AI artifacts may lead to rejection.
Official sources
Compiled from the official call for papers. The organizers’ pages remain authoritative.
Last verified September 9, 2026