PDE foundation model-accelerated inverse estimation of system parameters in inertial confinement fusion
Not on the map. Why?
Abstract (arXiv)
PDE foundation models are typically pretrained on large, diverse corpora of PDE datasets and can be adapted to new settings with limited task-specific data. However, most downstream evaluations focus on forward problems, such as autoregressive rollout prediction. In this work, we study an inverse problem in inertial confinement fusion (ICF): estimating system parameters (inputs) from multi-modal, snapshot-style observations (outputs). Using the open JAG benchmark, which provides hyperspectral X-ray images and scalar observables per simulation, we finetune the PDE foundation model and train a lightweight task-specific head to jointly reconstruct hyperspectral images and regress system parameters. The fine-tuned model achieves accurate hyperspectral reconstruction (test MSE 1.2e-3) and strong parameter-estimation performance (up to R^2=0.995). Data-scaling experiments (5%-100% of the training set) show consistent improvements in both reconstruction and regression losses as the amount of training data increases, with the largest marginal gains in the low-data regime. Finally, finetuning from pretrained MORPH weights outperforms training the same architecture from scratch, demonstrating that foundation-model initialization improves sample efficiency for data-limited inverse problems in ICF.
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 24 of its 44 references that are in the map. First posted on arXiv 4 Mar 2026.
Builds on
24 of 44 references.
References · 24
- 2020Language Models are Few-Shot Learnerscited by 32,419
- 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,974
- 2018Neural Ordinary Differential Equationscited by 3,779
- 2020Fourier Neural Operator for Parametric Partial Differential Equationscited by 2,908
- 2021Physics-Informed Neural Operator for Learning Partial Differential Equationscited by 527
- 2022PDEBENCH: An Extensive Benchmark for Scientific Machine Learningcited by 413
- 2024DPOT: Auto-Regressive Denoising Operator Transformer for Large-Scale PDE Pre-Trainingcited by 151
- 2024The Well: a Large-Scale Collection of Diverse Physics Simulations for Machine Learningcited by 145
- 2023Towards Foundation Models for Scientific Machine Learning: Characterizing Scaling and Transfer Behaviorcited by 131
- 2022Deep transfer operator learning for partial differential equations under conditional shiftcited by 83
- 2022Learning two-phase microstructure evolution using neural operators and autoencoder architecturescited by 72
- 2019Improved Surrogates in Inertial Confinement Fusion with Manifold and Cycle Consistenciescited by 14
12 more
- 2025MORPH: PDE Foundation Models with Arbitrary Data Modalitycited by 14
- 2024A conditional latent autoregressive recurrent model for generation and forecasting of beam dynamics in particle acceleratorscited by 8
- 2024Time-inversion of spatiotemporal beam dynamics using uncertainty-aware latent evolution reversalcited by 6
- 2025URSA: The Universal Research and Scientific Agentcited by 6
- 2025Foundation Models for Discovery and Exploration in Chemical Spacecited by 6
- 2022Transfer learning driven design optimization for inertial confinement fusioncited by 5
- 2024Physics consistent machine learning framework for inverse modeling with applications to ICF capsule implosionscited by 4
- 2025Towards Reasoning for PDE Foundation Models: A Reward-Model-Driven Inference-Time-Scaling Algorithmcited by 4
- 2019Parallelizing Training of Deep Generative Models on Massive Scientific Datasetscited by 3
- 2022Inverse characterization of composites using guided waves and convolutional neural networks with dual-branch feature fusioncited by 3
- 2025A Foundation Model for Material Fracture Predictioncited by 2
- 2019Exploring Generative Physics Models with Scientific Priors in Inertial Confinement Fusioncited by 1
Built on by
Citing papers per year: 2026: 2.
2 papers.