Publications

Publications

Selected publications — a snapshot of our recent work in SciFM.

1. Wang, Z., Huang, H., Zhao, H., Xu, C., Zhu, S., Janssen, J., Viswanathan, V. DREAMS: Density Functional Theory Based Research Engine for Agentic Materials Simulation. arXiv:2507.14267 (2025).

2. Duraisamy, K. Active Inference AI Systems for Scientific Discovery. arXiv:2506.21329 (2025).

3. Xu, C., Zhu, S., Viswanathan, V. CLOUD: A Scalable and Physics‑Informed Foundation Model for Crystal Representation Learning. arXiv:2506.17345 (2025).

4. Lim, J. Y., Pylorof, D., Garcia, H. E., Duraisamy, K. A Digital Twin Framework for Generation‑IV Reactors with Reinforcement Learning‑Enabled Health‑Aware Supervisory Control. arXiv:2506.17258 (2025).

5. Zhuang, Y., Duraisamy, K. LaDCast: A Latent Diffusion Model for Medium‑Range Ensemble Weather Forecasting. arXiv:2506.09193 (2025).

6. Zhuang, Y., Cheng, S., Duraisamy, K. Spatially‑aware diffusion models with cross‑attention for global field reconstruction with sparse observations. Comput. Methods Appl. Mech. Eng. 435, 117623 (2025).

7. Jacobsen, C., Zhuang, Y. & Duraisamy, K. CoCoGen: Physically Consistent and Conditioned Score‑Based Generative Models for Forward and Inverse Problems. SIAM J. Sci. Comput. 47 (2025). doi:10.1137/24M1636071.

8. Wadell, A., Bhutani, A., Viswanathan, V. Tokenization for Molecular Foundation Models. arXiv:2409.15370 (2024).

9. Wadell, A., Bhutani, A., Viswanathan, V. Scaling Foundation Models for Molecular Chemistry. OpenReview (2024).

10. Madhavan, V., Sebastian, A. S., Ramsundar, B., Viswanathan, V. Self‑supervised Pretraining for Partial Differential Equations. arXiv:2407.06209 (2024).

11. Barwey, S., Shankar, V., Viswanathan, V., Maulik, R. Multiscale graph neural network autoencoders for interpretable scientific machine learning. J. Comput. Phys. 495, 112537 (2023).

Talks & Posters

Selected talks & posters — recent presentations and invited talks where our team shared work and insights.

Venkat Viswanathan — Differentiable Physics, U‑M Knowledge‑Guided Machine Learning (KGML) Workshop, Ann Arbor, MI (2025).

Changwen Xu — Introduction to Scientific Foundation Models, U‑M Knowledge‑Guided Machine Learning (KGML) Workshop, Ann Arbor, MI (2025).

Changwen Xu — CLOUD: A Scalable and Physics‑Informed Foundation Model for Crystal Representation Learning, AI for Chemistry & Materials webinar, Online (2025).

Anoushka Bhutani — Scaling Foundation Models for Molecular Chemistry, AI4X 2025 Conference, Singapore (2025).

Venkat Viswanathan — Building and using Extreme‑scale Molecular Foundation Models, SciFM25: Scientific Discovery in the Age of AI, Ann Arbor, MI (2025).

Karthik Duraisamy — Opening remarks: Vision for AI‑augmented Discovery Engines & Critical Questions for the Conference, SciFM25: Scientific Discovery in the Age of AI, Ann Arbor, MI (2025).

Changwen Xu — CLOUD: A Scalable Scientific Foundation Model for Crystal Representation Learning, NeurIPS 2024 Foundation Models for Science Workshop, Vancouver, BC (2024).

Alexius Wadell - Tokenization for Molecular Foundation Models, FSML Seminar, Ann Arbor, MI (2024).

Anoushka Bhutani - Foundation Model Training and Interpretability for Molecular Design, FSML Seminar, Ann Arbor, MI (2024).

Karthik Duraisamy — Charting the course for SciFMs: Themes & Questions, SciFM24: Scientific Foundation Models, Ann Arbor, MI (2024).

Venkat Viswanathan — Scaling up Materials Foundation Models, SciFM24: Scientific Foundation Models, Ann Arbor, MI (2024).