发表机构
Zhejiang University（浙江大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
NEMSim通过将事件属性描述编译为可执行转移结构，学习控制条件下的多事件物理动力学，在3D KMC基准上相比最强基线降低Avg. RMSE达58.9%-81.3%，并保持数据效率优势。
AI 中文摘要
控制条件下的多事件物理系统的高保真模拟计算成本高昂，尤其是在广泛的控制空间和长轨迹上。在这些系统中，宏观演化源于局部离散事件，其强度和效果取决于过程控制和演化的局部状态，而可用的系统知识通常以事件属性描述的形式表达。纯数据驱动的替代模型必须从有限的轨迹覆盖中推断这些事件效果，这可能阻碍对未见控制机制的泛化。物理引导的方法则主要基于方程级约束或可微求解器，而非离散事件规则先验。因此，我们提出NEMSim（神经事件-机制模拟器），它将预定义的事件属性描述编译为可执行的转移结构，将控制依赖的事件强度、先验引导的机制归因和状态依赖的响应联系起来。为了利用显式系统知识评估控制条件下的多事件动力学，我们构建了一个基于3D KMC的基准，将高保真轨迹与显式事件规则、标准化划分和评估协议配对。在三种设置下，NEMSim相对于每种设置中最强的基线将平均均方根误差（Avg. RMSE）降低了58.9%-81.3%。在训练数据比例低至10%的数据效率研究中，它仍然保持最佳。机制分析进一步表明，这些收益源于可执行规则集成，而非仅靠先验访问或架构。
英文摘要
High-fidelity simulation of control-conditioned multi-event physical systems is computationally expensive, especially across broad control spaces and long trajectories. In these systems, macroscopic evolution emerges from localized discrete events whose intensities and effects depend on process controls and evolving local states, while the available system knowledge is typically expressed as event-attribute descriptions. Purely data-driven surrogates must infer these event effects from limited trajectory coverage, which can hinder generalization to unseen control regimes. Physics-guided methods instead primarily build on equation-level constraints or differentiable solvers rather than discrete event-rule priors. We therefore propose NEMSim (Neural Event-Mechanism Simulator), which compiles predefined event-attribute descriptions into an executable transition structure linking control-dependent event intensities, prior-guided mechanism attribution, and state-dependent responses. To enable evaluation of control-conditioned multi-event dynamics with explicit system knowledge, we construct a 3D KMC-based benchmark pairing high-fidelity trajectories with explicit event rules, standardized splits, and evaluation protocols. Across three settings, NEMSim reduces Avg. RMSE by 58.9%-81.3% relative to the strongest baseline in each setting. It also remains best in the data-efficiency study with training-data fractions down to 10%. Mechanism analyses further show that these gains arise from executable rule integration rather than prior access or architecture alone.
CommentsMain paper: 9 pages, 6 figures, 2 tables. Supplementary material included
