.png){kind=icon}
.png){kind=icon}
Return
Text2Structure3D: Graph-based generative modeling of equilibrium structures with diffusion transformers
.png){kind=icon}
OA
AI
DOI：10.1016/j.rineng.2026.111375.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Latent diffusion and variational graph auto-encoders can generate structures. • Conditioning models on text enables intuitive natural language-based design processes. • Graph-based data representations enable more general structural design applications. • Residual force optimization of forces and coordinates ensures static equilibrium. • Generated structures can adhere to quantitative and qualitative text descriptions.
Keywords:
Machine learning
Graph machine learning
Diffusion
Structural design
Structural form-finding
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
7.9
Papers:
1.1W
Citations:
1.7W
Organization
View more organizations