BRICKS: Compositional Neural Markov Kernels for Zero-Shot Radiation-Matter Simulation

TL;DR

This paper presents a novel, differentiable neural kernel for zero-shot radiation-matter interaction simulation, leveraging compositional transformers and Markov properties for efficient, scalable, and stable predictions.

Contribution

It introduces a new compositional neural kernel based on hybrid transformers that enables zero-shot simulation of complex radiation-matter interactions.

Findings

Achieves significant GPU speed-up over traditional simulators.

Demonstrates stable multi-round autoregressive predictions.

Provides a new large-scale radiation-matter interaction dataset.

Abstract

We introduce a new strategy for compositional neural surrogates for radiation-matter interactions, a key task spanning domains from particle physics through nuclear and space engineering to medical physics. Exploiting the locality and the Markov nature of particle interactions, we create a \emph{next-particle prediction} kernel using hybrid discrete-continuous transformer models based on Riemannian Flow Matching on product manifolds. The model generates variable-sized typed sets of particles and radiation side effects that are the result of the interaction of an incident particle with a material volume. The resulting kernel can be composed to simulate unseen large-scale material distributions in a zero-shot manner. Unlike mechanistic simulators, our model is designed to be differentiable, provides tractable likelihoods for future downstream applications. A significant computational speed-up on GPU compared to CPU-bound mechanistic simulation is observed for single-kernel execution. We evaluate the model at the kernel level and demonstrate predictive stability over multi-round autoregressive rollouts. We additionally release a novel 20M-event radiation-matter interaction dataset for further research.