We invite submissions contributing novel research at the intersection of applied mathematics, deep learning, and computational neuroscience, which incorporate symmetry and geometry into the design of artificial neural networks, the mechanistic interpretability of neural systems (biological or artificial), or theories of neural computation. We welcome contributions in the intersection of geometric and topological deep learning, computational and theoretical neuroscience, geometric statistics, and topological data analysis.
The following themes are particularly relevant:
Theory and methods for learning invariant and equivariant representations
Statistical learning theory in the context of topology, geometry, and symmetry
Representational geometry in neural data
Learning and leveraging group structure in data
Equivariant world models for robotics
Dynamics of neural representations
Topological deep learning and topological data analysis
Geometric structure in language
Geometric and topological analysis of generative models
Symmetries, dynamical systems, and learning
We hope to see both theoretical contributions and applied results in domains including vision, motor control, navigation, and language as well as the use of diverse mathematical objects such as quotient spaces, fiber bundles, Lie groups, Riemannian manifolds, graphs, topological domains, and group representations. We are also interested in submissions contributing benchmark datasets or software. This list is intended to provide guidance, but it is far from exhaustive. If you are unsure whether your work is within scope of this workshop, please reach out to the organizers.
There are two tracks for submission to NeurReps. Submissions to both tracks will be featured during the workshop's poster session, and a subset of all submissions will be selected for spotlight talks.
NeurReps Proceedings papers may be up to 9 pages long, excluding references and appendices. Accepted papers are intended to be published in a dedicated workshop proceedings volume. This track is appropriate for self-contained research papers with a high degree of development.
Extended abstracts may be up to 4 pages long, excluding references and appendices. This track creates space for contributions such as early-stage results, insightful negative findings, opinion pieces, or novel datasets. Extended abstracts will not be included in the PMLR volume, but authors may post to arXiv under the NeurReps index.
Dual Submission Policy
Papers in the Proceedings Track will be archivally published. Thus, submissions containing content that has been published or submitted elsewhere must include at least 30% new, unpublished/unsubmitted material. Likewise, to publish a NeurReps paper in another venue down the line, authors must add at least 30% new material. There are no restrictions on Extended Abstract submissions.
Review Process
Reviews are double-blind and conducted through OpenReview. All submissions will receive a minimum of three reviews.
Submission: August 29, 2025 11:59 PM AoE
September 4, 2025 11:59 PM AoE
Author Notification: September 22, 2025 11:59 PM AoE
All submissions must use the NeurReps 2025 LaTeX style files: [.zip].
Authors should select the proceedings or extended abstract format in the pmlr-sample.tex file.
All submissions should be submitted to OpenReview using the link below.
Please reach out to organizers@neurreps.org with any questions.