用物理约束融合扩散模型先验，实现无需重训练的精准采样。
Composing diffusion priors with explicit physical context via generative Gibbs sampling
- 通过扩展状态空间建模物理上下文与生成先验的联合分布
- 在双阱系统等场景中准确恢复分布偏移和集体行为
- 适用于科学计算中需结合物理规律的生成建模任务
预训练扩散模型提供强大的学习先验，但在科学采样中，目标分布常依赖于生成模型未充分表示的物理上下文。我们提出无需训练的生成吉布斯采样框架GG-PA，将部分先验与显式物理上下文的组合建模为扩展状态空间中的联合目标分布推断。我们推导出该联合目标的吉布斯采样器，证明其在扩散时间趋近零时渐近精确；在二次相互作用场景下，有限扩散时间仍保持精确。我们进一步引入扩散时间上的复制交换以加速混合。在双阱系统、ϕ⁴格点模型及原子级肽系统上的实验表明，仅使用部分先验，GG-PA无需重训练即可恢复由物理上下文引起的分布偏移与涌现集体行为。这些结果展示了GG-PA作为结合预训练生成先验与显式物理上下文的实用方法。
原文摘要 · Abstract (English)
Pretrained diffusion models provide powerful learned priors, but in scientific sampling the target distribution often depends on physical context that is not fully represented by one generative model. We introduce Generative Gibbs for Physics-Aware Sampling (GG-PA), a training-free framework that formulates the composition of learned partial priors and explicit physical context as inference over a joint target distribution in an augmented state space. We derive a Gibbs sampler for this joint target, show that it is asymptotically exact as the diffusion time approaches zero, and prove that in settings with quadratic interactions it remains exact at finite diffusion times. We further introduce replica exchange over diffusion time to accelerate mixing. Experiments on a double-well system, a $ϕ^4$ lattice model, and atomistic peptide systems show that GG-PA recovers context-induced distribution shifts and emergent collective behavior in interacting systems using partial priors without retraining. These results demonstrate GG-PA as a practical approach for combining pretrained generative priors with explicit physical context.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。