CaloArt: Large-Patch x-Prediction Diffusion Transformers for High-Granularity Calorimeter Shower Generation

TL;DR

CaloArt introduces a diffusion transformer model with large-patch tokenization and x-prediction for efficient high-granularity calorimeter shower simulation, achieving state-of-the-art quality with low computational cost.

Contribution

The paper presents CaloArt, a novel 3D diffusion transformer with large-patch tokenization and x-prediction, improving efficiency and quality in calorimeter shower generation.

Findings

CaloArt achieves the best FPD and high-level metrics on Dataset 2.

X-prediction outperforms v-prediction on Dataset 3.

Models generate showers in approximately 10 ms per shower on a single GPU.

Abstract

High-granularity calorimeters make ML-based fast shower simulation a high-dimensional generative modeling problem, where voxel-space generators must balance physics fidelity with training and inference cost. This work studies large-patch tokenization with x-prediction, enabling efficient raw voxel generation. We propose CaloArt, a modernized DiT-style backbone with 3D positional encoding and architectural refinements, trained via conditional flow matching with decoupled prediction and loss spaces. On CaloChallenge Dataset 2, where small patch size remains affordable, v-prediction performs well, and CaloArt achieves the best FPD, strongest high-level metrics, and strongest ResNet classifier metrics. On CaloChallenge Dataset 3, the 40500-voxel grid makes large patches necessary; x-prediction improves all reported metrics over v-prediction and places CaloArt on the quality-generation-time Pareto frontier. The final CCD2 and CCD3 models both retain O(10) ms single-GPU generation time, with 9.71 and 11.14 ms per shower. These results support large-patch voxel-space diffusion transformers with x-prediction as a compute-efficient route to high-granularity calorimeter shower synthesis, reducing training and inference cost without a pretrained latent tokenizer.