Research Papers
Short research papers (four pages or fewer, excluding references) following NeurIPS format, peer-reviewed and ranked for impact; accepted papers are published and presented as posters, with a subset selected for spotlight talks.
AI/CORE Unranked
PAI 2026
Physics and AI
Jun 10 2026Stanford University, Stanford, California,United StatesOfficial conference site Link status unverified
The 2026 Conference on Physics and AI (PAI26) is an interdisciplinary event organized by the Center for Decoding the Universe, the American Physical Society’s Group on Data Science, and the NeurIPS Machine Learning and Physical Sciences Workshop team. It brings together researchers in physics, data science, and AI to advance methodology at their intersection, with a focus on both applying AI to physics problems and using physics insights to improve AI systems.
Key deadlines
Verified September 9, 2026Full paper
May 2, 2026
AoE
Conference timeline
Submission and decisions
Full paperKey deadline
May 2, 2026 · AoE
Paper fit
Contribution paths
A strong submission should clearly identify its contribution and evaluate it appropriately.
Datasets and Benchmarks
Publicly available physics datasets with clear baseline methods, submitted under a single-blind review process.
Perspectives
Compelling and grounded viewpoints on the state and future of AI-physics research, submitted as short papers.
Education and Training Submissions
Descriptions of effective training materials, courses, or programs aimed at building AI and physics literacy.
Research areas in scope
01
Research Tracks
Developing new AI methods for physicsDeveloping new physics methods for AIInnovative applications of AI in physics
02
Datasets and Benchmarks
Well-documented, public physics datasets (real and/or synthetic)Baseline methods presented alongside datasets
03
Perspectives
Viewpoints on past, present, and future status/challenges at the intersection of AI and physics
04
Education and Training
Effective training materials for AI and physics literacyDescriptions of university courses and summer school programs
Policies worth checking twice
- Submissions must be four pages or fewer (excluding references) and follow the NeurIPS format.
- Appendices are not allowed.
- All paper content must be anonymized for double-blind review (except datasets track, which is single-blind).
- Work previously published in archival venues is welcome if modified to fit the venue and properly cited.
- Authors cannot be added after the review begins; only the submitting author needs an OpenReview account.
- One author must be physically present to present a poster if the paper is accepted.
- Submissions are confidential until authors confirm they are not made public.
- AI can be used in any part of the process, but human authors must take full responsibility for the content.
Official sources
Compiled from the official call for papers. The organizers’ pages remain authoritative.
Official conference siteProgramsPostdoc FellowsPhD ScholarsData Science for Social GoodStudent Fellows
Last verified September 9, 2026