NeurIPS 2026 workshop · Paris
Self-supervision, transfer learning, sampling, and tokenization for scientific data and physical systems.
Representations map observations into structured embeddings whose geometric structure reveals semantic content. In vision and language, deep learning models trained on large unlabeled datasets have produced representations that transfer broadly across tasks.
As AI progressively permeates scientific research, large-scale datasets together with high-throughput simulation and experimental pipelines, are making it possible to learn general-purpose scientific representations at unprecedented scale.
The scientific setting, however, challenges representation learning in fundamental ways. Data are heterogeneous and structured, and often come as unlabeled streams from a dynamical system. Moreover, the learned embeddings must respect conservation laws, geometry, and causal structure. Ultimately, a useful scientific representation should expose structures that scientists can act upon.
This workshop provides a tightly scoped forum for representation learning in physical systems. For this first edition, we welcome discussions and contributions on topics around Self-Supervision, Transfer Learning, Sampling, and Tokenization, which we believe constitute the most interesting open problems in this space.
Scientific data are often abundant but unlabeled. Self-supervised learning offers a natural route to extracting structure from such data, but standard pretext tasks and augmentations can violate scientific meaning. What self-supervised objectives preserve physical constraints such as conservation laws? What semantic structures do SSL objectives discover when constrained by physical priors?
learning with no labelsScientific models must reliably extrapolate into physically meaningful regimes beyond their training distribution. In the physical sciences, however, verifying a model's out-of-distribution prediction often requires massive computational effort or expensive wet-lab synthesis. This theme investigates the opportunities and limits of scientific transfer to ensure that learned embeddings remain falsifiable and actionable rather than just empirically successful.
generalizationUnlike internet-scale text and image corpora, many scientific domains have access to simulators and experimental loops that can generate new data. Molecular dynamics, DFT, CFD, PDE solvers, and high-throughput experimental platforms make it possible to shape the training distribution itself. This makes adaptive simulation and closed-loop data generation a central opportunity for building better scientific representations.
adaptive data generationThe success of foundation models relies heavily on discrete tokenization, but mapping physical sciences into discrete vocabularies remains a fundamental bottleneck. Physical data are inherently continuous, multi-scale, and often non-Euclidean. Naive grid-based patching or quantization can destroy geometric priors, symmetries, and sub-grid dynamics. This theme focuses on novel tokenization methods for physical systems.
physical modalitiesWe invite interdisciplinary contributions from core machine learning and every area of AI for science.
OpenReview Submission Page | Short PapersFour pages excluding references and appendices. Accepted work will be presented as posters, with selected submissions invited for contributed talks.
Short Papers are limited to four pages of main content; references do not count toward the limit. Appendices are unlimited, but reviewers are not obliged to read them, so keep the main paper self-contained. Unlike the NeurIPS main track, this workshop does not require the NeurIPS paper checklist. Download the workshop LaTeX template , which is based on the NeurIPS style. For styling details, consult the NeurIPS paper-formatting guidance .
Your submission must consist of a single PDF file including the main text, references, and, optionally, an appendix. Additional files are not allowed. You may link to properly anonymized code and/or data repositories.
Short Papers are submitted through OpenReview and reviewed double-blind, following the NeurIPS main-track approach. Concurrent submissions are allowed, but authors are responsible for checking the other venue’s dual-submission policy.
Research Notes have no page limit and will be submitted through a separate channel. Submission details will be announced soon.
Accessible explanations of open problems, paradigms, datasets, or algorithms. New results are not required; clarity and usefulness are.
We welcome expressions of interest from researchers across representation learning, AI for science, computational biology, scientific machine learning, and physics-based simulation—especially early-career researchers with relevant expertise.
Submitting the form is an expression of interest and does not automatically constitute appointment to the committee.
Expect to review up to three submissions and follow the NeurIPS conflict-of-interest policy.
Express your interest in joining the Program Committee Opens a standalone Google Form in a new tabMeet our first confirmed speakers. More coming soon.
Technical University of Munich
Deep learning, partial differential equations, and fluid simulation
Bioptimus
Machine learning for biology, genomics, and precision medicine
University of Cambridge
Statistical ML, uncertainty quantification, and engineeringItalian Institute of Technology
Researcher bridging physics and AI, with a focus on scientific machine learning for dynamical systems and atomistic simulations.
Italian Institute of Technology · UCL
Professor and researcher working on the foundations of machine learning and their applications to scientific modeling.
Télécom Paris · IP Paris
Professor specializing in statistical machine learning for non-Euclidean data, trustworthy AI, and sustainable AI.
Columbia University
Professor working at the intersection of climate science, hydrology, and physics-aware AI for Earth system modeling.
Columbia University
Research scientist developing physics-informed machine learning parameterizations for sub-grid processes in climate models.
University of Stuttgart
Professor researching structured and scientific data through geometric, physics-aware, probabilistic, and generative machine learning.
Interested in sponsoring this workshop? Get in touch ↗