当局部规则产生全局秩序：面向潜扩散模型的自组织表征学习
When Local Rules Create Global Order: Self-Organized Representation Learning for Latent Diffusion Models
为潜扩散模型提供了一种通过局部吸引与排斥机制自组织形成兼具局部平滑与全局分散潜空间的方法，可提升重建与生成质量，对设计更好的潜空间正则化有直接启发。
原文依据 SORL promotes these properties through two complementary local mechanisms: local attraction, which encourages coherent reconstructions among nearby latent codes, and local repulsion, which prevents latent codes from collapsing into dense clusters.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论本文研究潜空间结构如何影响潜扩散模型（LDMs）的性能。我们表明，有效的生成需要一个同时具备局部平滑性和全局分散性的潜空间：局部平滑性使重建稳定可靠，全局分散性使模型能够抽取多样且有意义的样本，而不会坍缩到潜空间的狭窄区域。然而，现有方法往往强调平滑性，这可能导致潜区域集中并限制对更广阔空间的探索。为解决这些局限，我们提出自组织表征学习（SORL），一种受复杂系统自组织现象启发的自底向上训练范式，其中全局结构从简单的局部交互中自然涌现。平滑性和最大分散性这两个关键的潜空间性质并非被显式施加，而是通过两种互补的局部机制来促进：局部吸引鼓励邻近潜编码之间的一致重建，局部排斥防止潜编码坍缩成密集簇。通过它们的相互作用，SORL 诱导出一个同时保持局部平滑性和全局分散性的潜流形，从而改善重建和生成。
英文原文摘要 1300 字符
This work studies how latent space structure impacts the performance of Latent Diffusion Models (LDMs). We show that effective generation requires a latent space that is simultaneously locally smooth, enabling stable and reliable reconstruction, and globally dispersive, allowing the model to draw diverse and meaningful samples without collapsing into narrow regions of the latent space. However, existing approaches often emphasize smoothness, which may lead to concentrated latent regions and limited exploration of the broader space. To address these limitations, we propose Self-Organized Representation Learning (SORL), a bottom-up training paradigm inspired by self-organization in complex systems, where global structure emerges naturally from simple local interactions. The critical latent properties of smoothness and maximal dispersity are not explicitly imposed. Instead, SORL promotes these properties through two complementary local mechanisms: local attraction, which encourages coherent reconstructions among nearby latent codes, and local repulsion, which prevents latent codes from collapsing into dense clusters. Through their interaction, SORL induces a latent manifold that maintains both local smoothness and global dispersity, leading to improved reconstruction and generation.