Non-Euclidean Foundation Models and Geometric Learning Workshop @ NeurIPS 2025
— Advancing AI Beyond Euclidean Frameworks
The Non-Euclidean Foundation Models and Geometric Learning Workshop will take place at NeurIPS 2025 in San Diego, CA, USA, from December 2–7, 2025. We invite you to join discussions on non-Euclidean representation learning, geometric deep learning, and large foundation models!
News
- 2025-07-22: We are accepting paper for oral and poster presentations. See the Call for Contributions for more details.
- 2025-07-22: If you are interested in being a reviewer for the workshop, please fill out the reviewer nomination form by August 24st.
- Submission site: https://openreview.net/group?id=NeurIPS.cc/2025/Workshop/NEGEL
- Join our
Slack workspace
1. Workshop Introduction
In the era of foundation models and Large Language Models (LLMs), Euclidean space is the de facto geometric setting of our machine learning architectures. However, recent literature has demonstrated that this choice comes with fundamental limitations. Non-Euclidean learning is quickly gaining traction. Non-Euclidean spaces, such as hyperbolic, spherical, and mixed-curvature spaces, have been shown to provide more efficient and effective representations for data with intrinsic geometric properties, like hierarchy, symmetry, and heterogeneity.
Integrating foundation models with non-Euclidean spaces has great potential to enhance their ability to capture and model the underlying structures and relationships in complex real-world data, leading to better performance, generalization, and interpretability. This workshop focuses on the intersection of Non-Euclidean representation learning and Foundation Models, exploring its potential benefits, challenges, and future directions.
2. Call for Contributions (Topics and Scope)
The workshop will include submissions on topics related to the intersection of foundation models and non-Euclidean representation learning, including:
- Theoretical Foundations: Generalization error, representation precision, curvature-dimension tradeoffs, geometric properties (curvature, geodesics, isometries), expressive power of non-Euclidean representations.
- Architectures and Algorithms: Adapting existing foundation models to non-Euclidean spaces, developing new foundation models for non-Euclidean operations, and investigating non-Euclidean architectures in conjunction with foundation models.
- Applications: Graph analysis, text processing, image understanding, biomedical research, and AI for scientific discovery (drug discovery, material science, climate modeling).
- Trustworthiness and Robustness: Adversarial robustness, fairness, interpretability, and privacy in non-Euclidean foundation models.
- Benchmarks and Tools: New datasets, evaluation protocols, and software libraries supporting the integration of foundation models and non-Euclidean representations.
3. Submissions and Timeline
Submissions should be PDFs of ≤ 9 pages (excluding references and appendices) using the NeurIPS 2025 LaTeX style file. The review process is double-blind (submissions must be anonymized) and papers are managed through OpenReview. We welcome work that is original and unpublished, recently published, or work-in-progress. By default, submissions will not have archival proceedings.
Please use the NeurIPS 2025 LaTeX style file; it includes a preprint option for non‑anonymous preprints posted online (see additional formatting details here). Outstanding papers will be selected for lightning talks and a best paper award will be announced at the workshop. All accepted papers will be presented as posters.
Important Dates
- Abstract submission deadline: August 25,2025
- Paper submission deadline: Sept 2,2025
- Author notification: Sept 25,2025
- Camera-ready submission: Sept 30,2025
- NeurIPS early registration deadline: Oct 11, 2025
- NeurIPS financial assistance application deadline: TBA
- Workshop date: December 6 or 7, 2025
- Location: San Diego, CA, USA
- Timezone: Anywhere On Earth (UTC-12)
4. Tentative Schedule
- 8:30–8:50 AM: Poster setup
- 8:50–9:00 AM: Opening remarks
- 9:00–9:50 AM: Invited talk: Philip S. Yu (UIC)
- 9:50–10:40 AM: Invited talk: Pascal Mettes (UvA)
- 10:40–11:30 AM: Invited talk: Smita Krishnaswamy (Yale)
- 11:30–11:50 AM: Discussions and coffee break
- 11:50–12:55 PM: Poster session
- 12:55–1:30 PM: Lunch break
- 1:30–2:20 PM: Invited talk: Min Zhou (Huawei)
- 2:20–3:10 PM: Invited talk: Bo Xiong (Standford)
- 3:10–4:00 PM: Invited talk: Lucas Vinh Tran (JPMorganChase)
- 4:00–4:50 PM: Contributed talks
- 4:50–5:00 PM: Discussions and coffee break
- 5:00–5:30 PM: Panel discussions: Foundation Models Meet Non-Euclidean Space
5. Organizers
The workshop is organized by an international and diverse team of experts in the field:
 Menglin Yang HKUST(GZ) |
 Neil He UIUC/Yale |
 Yifei Zhang NTU |
 Melanie Weber Harvard |
 Rex Ying Yale |
Note: Melanie Weber is an extended organizer.
6.Speakers
The workshop features an international and distinguished lineup of speakers from academia and industry:
Philip S. Yu UIC |
Bo Xiong Stanford |
Min Zhou Huawei |
Pascal Mettes UvA |
Smita Krishnaswamy Yale |
Lucas Vinh Tran JPMorganChase |
7. FAQ
Can I attend virtually?
All talks will be live-streamed, so you can follow them online. However, the poster session is in-person only, and we strongly encourage everyone to attend in person if possible to get the most out of the experience.
What does non-archival mean?
Non-archival means that your submission will not be published in formal proceedings or indexed in academic databases. You still retain the right to publish the same work elsewhere.
Contact
Feel free to contact us at negel2025@outlook.com. We look forward to your participation at NeurIPS 2025!
🔗 Workshop Page: Official NeurIPS Workshop Page