In this work, we establish a direct connection between generative diffusion
models (DMs) and stochastic quantization (SQ). The DM is realized by
approximating the reversal of a stochastic process dictated by the Langevin
equation, generating samples from a prior distribution to effecti
针对量子系统状态的生成建模问题,本文提出了一种基于去噪扩散模型的方法,其关键创新点在于考虑了物理性质中的量子状态约束,通过 Mirror Diffusion Model 和设计出的镜像映射实现了严格保持结构的生成,实验验证了无条件生成和通过无监督分类器引导的条件生成的有效性,后者甚至在未知标签上生成了新的量子状态。