PC-Droid: Faster diffusion and improved quality for particle cloud generation

TL;DR

PC-Droid introduces a new diffusion model for jet particle cloud generation that significantly improves speed and quality, outperforming previous models and enabling faster, more accurate simulations.

Contribution

The paper presents PC-Droid, a diffusion model with a novel formulation, optimized architectures, and consistency distillation, achieving state-of-the-art performance in jet particle cloud generation.

Findings

Generation time up to 100x faster than PC-JeDi

Outperforms competing models across all metrics

Effective trade-off between speed and quality

Abstract

Building on the success of PC-JeDi we introduce PC-Droid, a substantially improved diffusion model for the generation of jet particle clouds. By leveraging a new diffusion formulation, studying more recent integration solvers, and training on all jet types simultaneously, we are able to achieve state-of-the-art performance for all types of jets across all evaluation metrics. We study the trade-off between generation speed and quality by comparing two attention based architectures, as well as the potential of consistency distillation to reduce the number of diffusion steps. Both the faster architecture and consistency models demonstrate performance surpassing many competing models, with generation time up to two orders of magnitude faster than PC-JeDi and three orders of magnitude faster than Delphes.