PulseAugur
LIVE 16:00:25
research · [2 sources] ·
0
research

BRICKS model uses neural Markov kernels for zero-shot radiation-matter simulation

Researchers have developed BRICKS, a novel approach using compositional neural Markov kernels for simulating radiation-matter interactions. This method employs hybrid discrete-continuous transformer models and Riemannian Flow Matching to predict particle behavior and radiation side effects. The system can simulate unseen material distributions in a zero-shot manner and is designed to be differentiable, offering potential for future applications. Additionally, a new dataset of 20 million radiation-matter interaction events has been released to support further research. AI

Summary written by gemini-2.5-flash-lite from 2 sources. How we write summaries →

IMPACT Introduces a new differentiable simulation technique for complex physical interactions, potentially accelerating research in fields like particle physics and medical physics.

RANK_REASON The cluster describes a new scientific paper detailing a novel simulation method.

Read on arXiv cs.LG →

COVERAGE [2]

  1. arXiv cs.LG TIER_1 · Richard Hildebrandt, Evangelos Kourlitis, Baran Hashemi, Manuel B\"unstorf, Thierry Meyer, Nikola Boskov, Michael Kagan, Dan Rosenbaum, Sanmay Ganguly, Lukas Heinrich ·

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

    arXiv:2605.06591v1 Announce Type: new 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 a…

  2. arXiv cs.LG TIER_1 · Lukas Heinrich ·

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

    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, w…