Publications

Google Scholar NVIDIA Research papers alphaXiv

2026

Ground-Wing: A Parametric 3D Wing-in-Ground-Effect Dataset for Benchmarking Neural Surrogate Models

A. Cenzato, L. Karch, A. Luckow, Y. Pichon, F. Pasqua, A. Ghasemi, others

arXiv preprint · 2026

A parametric 3D wing-in-ground-effect dataset for benchmarking neural surrogate models of aerodynamics.

arXiv

CANTO: CAD-Native Transformer Operators for AI-Aided Engineering

D. Leibovici, N. B. Kovachki, D. Ahn, R. Ohana, I. J. S. Shokar, A. Ghasemi, others

arXiv preprint · 2026

Neural operators that predict aerodynamics directly from CAD geometry and support differentiable design optimisation.

arXiv
CANTO: CAD-Native Transformer Operators for AI-Aided Engineering

Tessera v2: Scaling pixel-wise earth foundation models

Z. Feng, S. Jaffer, I. Shokar, J. Knezevic, J. Ball, P. Sousa, M. Elvers, M. Lisaius, others

arXiv preprint · 2026

Scaling laws and Matryoshka distillation for pixel-wise Earth observation embeddings, evaluated on 29 downstream tasks.

arXiv
Tessera v2: Scaling pixel-wise earth foundation models

Demystifying data-driven probabilistic medium-range weather forecasting

J. Kossaifi, N. Kovachki, M. Mardani, D. Leibovici, S. Ravuri, I. Shokar, others

arXiv preprint · 2026

Probabilistic medium-range weather forecasting (ATLAS): latent rollouts of probabilistic transformers trained with spatial and spectral CRPS, with over 30x faster inference than diffusion and stochastic-interpolant models and no loss of forecast skill.

arXiv
Demystifying data-driven probabilistic medium-range weather forecasting

2025

Deep Learning Emulation of Chaotic and Stochastic Dynamics for Insights into Regime Transitions and Rare Events

Ira J. S. Shokar

PhD thesis, University of Cambridge · 2025

Emulators of chaotic and stochastic dynamics trained by scoring-rule minimisation: stable rollouts, forecasts of regimes and extreme events absent from training, and generalisation across PDE parameters.

Paper

Deep Learning of the Evolution Operator Enables Forecasting of Out-of-Training Dynamics in Chaotic Systems

Ira J. S. Shokar, Peter H. Haynes, Rich R. Kerswell

arXiv preprint · 2025

Learning the evolution operator lets an emulator forecast dynamics and regimes that never appear in its training data.

arXiv

2024

Stochastic Latent Transformer: Efficient Modelling of Stochastically Forced Zonal Jets

Ira J. S. Shokar, Rich R. Kerswell, Peter H. Haynes

Journal of Advances in Modeling Earth Systems · 2024

A probabilistic latent-space emulator of stochastically forced zonal jets, about five orders of magnitude faster than numerical integration.

Stochastic Latent Transformer: Efficient Modelling of Stochastically Forced Zonal Jets

Extending deep learning emulation across parameter regimes to assess stochastically driven spontaneous transition events

Ira J. S. Shokar, Peter H. Haynes, Rich R. Kerswell

ICLR 2024 Workshop on AI4DifferentialEquations in Science · 2024

Data-efficient fine-tuning across PDE parameters lets emulators generalise to unseen regimes and capture spontaneous transition events.

Extending deep learning emulation across parameter regimes to assess stochastically driven spontaneous transition events

Conditioning deep learning on PDE parameters to generalise emulation of stochastic and chaotic dynamics

Ira Shokar, Peter Haynes, Rich Kerswell

APS Division of Fluid Dynamics Meeting Abstracts · 2024

Conditioning the emulator on PDE parameters (AdaLN) to generalise across stochastic and chaotic regimes.

2023

Learning Stochastic Dynamics with Probabilistic Neural Networks to study Zonal Jets

Ira J. S. Shokar, Rich R. Kerswell, Peter H. Haynes

EGU General Assembly · 2023

Probabilistic neural networks that learn the stochastic dynamics of zonal jets.