Information Spreading in Diffusion Models from Effective Field Theory
基于有效场论的扩散模型中的信息传播
AI总结 该研究将物理学有效场论应用于带卷积架构的得分匹配扩散模型，通过玩具示例和MNIST实验证实，两点间互信息按布朗运动有效场论预测的方式增长。
Comments This work was previously presented at the Scientific Methods for Understanding Deep Learning Workshop at ICLR 2026. 14 pages + appendices. Code available at: this https URL (https://github.com/NavonilNeogi/MutualInfoEFTDiffusionModels)
