ABSTRACT
In generative AI, diffusion models are state-of-the-art for high resolution representations of highly diverse complex objects. Recently, the first use of generative conditional diffusion for particle accelerator applications was demonstrated on experimental data at the EuXFEL as a virtual beam diagnostic that maps non-invasive measurements to megapixel resolution longitudinal phase space (z,E) projections of the electron beam [1]. A multi-modal adaptive conditional diffusion approach was also developed to track all 15 unique projections of time-varying beams [2]. This talk gives an overview of generative diffusion and of the results in [1,2] and then discusses a recently developed method of adaptive and physics-constrained superresolution diffusion for noninvasive virtual diagnostics of the six-dimensional (6D) phase space density of charged particle beams [3]. In this approach, an adaptive variational autoencoder is used to embed initial beam condition images and scalar measurements to a low-dimensional latent space from which a 326-pixel 6D tensor representation of the beam’s 6D phase space density is generated. Projecting from a 6D tensor generates physically consistent two-dimensional projections. Physics-guided superresolution diffusion transforms low-resolution images of the 6D density to high resolution 256 × 256 pixel images. Unsupervised adaptive latent space tuning enables tracking of time-varying beams without knowledge of time-varying initial conditions. The method is demonstrated with experimental data and multiparticle simulations at the HiRES UED. The general approach is applicable to a wide range of complex dynamic systems evolving in high-dimensional phase space. The method is shown to be robust to distribution shift without retraining.

[1] Scheinker, Alexander. "Conditional guided generative diffusion for particle accelerator beam diagnostics." Scientific Reports 14.1 (2024): 19210.
[2] Scheinker, Alexander. "cDVAE: VAE-guided diffusion for particle accelerator beam 6D phase space projection diagnostics." Scientific Reports 14.1 (2024): 29303.
[3] Scheinker, Alexander. "Physics-constrained superresolution diffusion for six-dimensional phase space diagnostics." Physical Review Research 7.2 (2025): 023091.