Generative models have recently demonstrated impressive capabilities in producing high-quality 3D shapes from a variety of user inputs (e.g., text or images). However, generated objects often lack physical integrity. We introduce PhysiOpt, a differentiable physics optimizer designed to improve the physical behavior of 3D generative outputs, enabling them to transition from virtual designs to physically plausible, real-world objects. While most generative models represent geometry as continuous implicit fields, physics-based approaches often rely on the finite element method (FEM), requiring ad hoc mesh extraction to perform shape optimization. In addition, these methods are typically slow, limiting their integration in fast, iterative generative design workflows. Instead, we bridge the representation gap and propose a fast and effective differentiable simulation pipeline that optimizes shapes directly in the latent space of generative models using an intuitive and easy-to-implement differentiable mapping. This approach enables fast optimization while preserving semantic structure, unlike traditional methods relying on local mesh-based adjustments. We demonstrate the versatility of our optimizer across a range of shape priors, from global and part-based latent models to a state-of-the-art large-scale 3D generator, and compare it to a traditional mesh-based shape optimizer. Our method preserves the native representation and capabilities of the underlying generative model while supporting user-specified materials, loads, and boundary conditions. The resulting designs exhibit improved physical behavior, remain faithful to the learned priors, and are suitable for fabrication. We demonstrate the effectiveness of our approach on both virtual and fabricated objects.

There are many applications where it is necessary to find an optimal spanning tree. For several of these, recent research has suggested the use of genetic algorithms (GAs). Historically, the Pruumlfer code has been one of the most popular representations for spanning trees, due largely to the bijective mapping between genotype and phenotype. However, it is has attracted much criticism for its low locality, a primary reason for its poor performance in GAs. Other representations such as direct encoding and network random keys have been shown to be far more effective. In 2001, an alternative called the Blob code was identified and adapted for use in GAs. It was shown to exhibit significantly higher locality than the Pruumlfer code. For a simple test problem, a GA using the Blob code was found to substantially outperform one using the Pruumlfer code. This paper suggests an alternative called the Dandelion code, which is more efficient and exhibits yet higher locality. Although both direct encoding and NetKeys are shown to give better results on the test problems used in this paper, the Dandelion code should be considered as a strong alternative, particularly for very large networks