通过实体标识符实现动态系统隐空间建模，保持个体可追踪性并提升生成效率。
LaM-SLidE: Latent Space Modeling of Spatial Dynamical Systems via Linked Entities
- 引入实体标识符，从隐空间还原个体属性与组成。
- 在多个领域中实现更快的生成速度与更高的精度和泛化能力。
- 适合需追踪个体行为的复杂系统建模，如分子结构或人群行为。
生成模型正推动深度学习进展，在动态系统轨迹采样方面展现出巨大潜力。然而，尽管隐空间建模已革新图像与视频生成，多数动态系统仍难以适用。这些系统（如化学分子结构或集体人类行为）由实体间相互作用描述，具有连接模式、实体守恒及时间可追踪性等特性。本文提出LaM-SLidE（基于关联实体的隐空间建模），兼顾个体可追踪性与生成效率。核心在于引入标识符表示（IDs），从隐空间中检索实体属性与组成，实现个体追踪。实验表明，该方法在不同领域均具备优异的速度、准确率与泛化性能。代码已公开于https://github.com/ml-jku/LaM-SLidE。
原文摘要 · Abstract (English)
Generative models are spearheading recent progress in deep learning, showcasing strong promise for trajectory sampling in dynamical systems as well. However, whereas latent space modeling paradigms have transformed image and video generation, similar approaches are more difficult for most dynamical systems. Such systems -- from chemical molecule structures to collective human behavior -- are described by interactions of entities, making them inherently linked to connectivity patterns, entity conservation, and the traceability of entities over time. Our approach, LaM-SLidE (Latent Space Modeling of Spatial Dynamical Systems via Linked Entities), bridges the gap between: (1) keeping the traceability of individual entities in a latent system representation, and (2) leveraging the efficiency and scalability of recent advances in image and video generation, where pre-trained encoder and decoder enable generative modeling directly in latent space. The core idea of LaM-SLidE is the introduction of identifier representations (IDs) that enable the retrieval of entity properties and entity composition from latent system representations, thus fostering traceability. Experimentally, across different domains, we show that LaM-SLidE performs favorably in terms of speed, accuracy, and generalizability. Code is available at https://github.com/ml-jku/LaM-SLidE .
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。