Research map / Probabilistic and scientific modeling / Variational and generative sampling
Glass: Geometric Latent Augmentation for Shape Spaces
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use GLASS instead of Vanilla VAE | shape extrapolation · Faust, Centaurs, and Horses sparse sets | surface smoothness: 1.0 -> 0.69; 1.0 -> 0.59; 1.0 -> 0.63; 1.0 -> 0.6; 1.0 -> 0.79; 1.0 -> 0.76; 1.0 -> 0.69; 1.0 -> 0.75; 1.0 -> 0.71; 1.0 | Helped |
| Add interpolation to Vanilla VAE (ablation) | shape extrapolation · Faust, Centaurs, and Horses sparse sets | surface smoothness: 1.0 -> 0.74; 1.0 -> 0.64; 1.0 -> 0.66; 1.0 -> 0.65; 1.0 -> 1.13; 1.0 -> 0.88; 1.0 -> 0.7; 1.0 -> 1.1; 1.0 -> 0.87; 1.0 - | Mixed |
| Add energy to the interpolation variant (ablation) | shape extrapolation · Faust, Centaurs, and Horses sparse sets | surface smoothness: 0.74 -> 0.79; 0.64 -> 0.6; 0.66 -> 0.66; 0.65 -> 0.6; 1.13 -> 1.03; 0.88 -> 0.88; 0.7 -> 0.69; 1.1 -> 1.06; 0.87 -> 0.82 | Mixed |
Closest papers by meaning
- LaGeM: A Large Geometry Model for 3D Representation Learning and Diffusion (ICLR 2025 · similarity 0.87)
- ShapeShifter: 3D Variations Using Multiscale and Sparse Point-Voxel Diffusion (CVPR 2025 · similarity 0.86)
- Flow-Corrected Shape Optimization: Taming Manifold Drift in High-Dimensional 3D Models (NeurIPS 2026 · similarity 0.86)
- GPLD3D: Latent Diffusion of 3D Shape Generative Models by Enforcing Geometric and Physical Priors (CVPR 2024 · similarity 0.86)
- Dora: Sampling and Benchmarking for 3D Shape Variational Auto-Encoders (CVPR 2025 · similarity 0.86)
- The Geometric Wall: Manifold Structure Predicts Layerwise Sparse Autoencoder Scaling Laws (NeurIPS 2026 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.