AI4Physics
We invite submissions to the 2nd International Conference on AI for Science (AI4Sci 2026), welcoming abstracts on applications of artificial intelligence in and for science. AI4Sci 2026 is part of the AI4Science Week and co-located with the 29th International Conference on Discovery Science (DS 2026). AI4Sci features separate AI4Physics, AI4Chemistry&Materials, AI4LifeSciences, and AI4Humanities&SocialSciences tracks dealing with applications of AI and Machine Learning in a variety of different fields. Contrary to DS 2026, the AI4Sci tracks are non-archival and also suited to ongoing or already published work.
The AI4Physics Track of AI4Sci 2026 invites contributions on topics such as, but not restricted to:
- (innovative) applications of AI in fundamental and applied physics, including (astro)-particle physics, nuclear physics, gravitational wave physics, cosmology, condensed matter physics, quantum computing and related domains
- AI-driven methods for simulation and modeling, such as generative models, surrogate modeling, fast simulation, and AI-enhanced numerical methods
- AI for data analysis and discovery in physics, including anomaly detection, pattern recognition, inverse problems, and automated hypothesis generation
- foundation models and large-scale AI systems in physics, including their adaptation, evaluation, and limitations in scientific contexts
- interpretability, robustness, and reliability of AI methods in physics, including uncertainty quantification, explainable AI, and validation against physical principles
- AI and scientific computing infrastructure, including high-performance computing (HPC), distributed systems, and integration with experimental pipelines
- (innovative) approaches to communication, visualization, and dissemination of physics research using AI
- ethical, epistemological, and methodological implications of AI in physics, including the role of AI in theory building and scientific understanding
The talks of the track will be 15 minutes long, followed by 5 minutes of Q&A.
Submission Procedure
Submitted abstracts must be written in English and should not exceed 500 words, excluding references. The reviewing process will be single-blind. Thus, the submitted abstracts should not be anonymized.
The abstract should be entered in the field "Abstract" of the CMT system directly. Please use the PDF upload for potential supplementary material, for instance, of an already published paper based on which you want to give a talk at the track.
If accepted, we will ask you to submit a two-page PDF of the abstract in Springer LNCS style. A book of abstract will be provided at the conference. Publication of the book of abstracts at, e.g., https://pos.sissa.it/ or https://ceur-ws.org/ is under discussion with the track chairs.
The Microsoft CMT service was used for managing the peer-reviewing process for this conference. This service was provided for free by Microsoft and they bore all expenses, including costs for Azure cloud services as well as for software development and support.
Please submit your contributions at:
https://cmt3.research.microsoft.com/AIforSci2026/.
When submitting, please choose the track "AI4Physics".
The timeline for AI4Sci 2026 is as follows:
- CMT submission system opens: May 8th, 2026
- Abstract submission deadline: June 12th, 2026
- Notification of acceptance or rejection: July 21st, 2026
All deadlines are 23:59 AOE.
Track chairs
Marius Köppel
ETH Zürich
mkoepp@phys.ethz.ch
Thea Aarestad
CERN
thea.aarrestad@cern.ch