7 papers
Rainbow percolation
Peter Gracar, Benjamin Lees
We consider the weight-dependent random connection model on a Poisson point process of intensity on in which the vertices and are joined…
Interpretability-Guided Soft Pruning of Attention Heads in Vision Transformers
Kamil KsiÄ Å¼ek, Piotr SuszyÅski, MichaÅ Jan WÅodarczyk +2
Vision foundation models, such as DINOv2, learn highly expressive representations but rely on massive, opaque architectures that demand substantial computational power and memory.…
Local Intrinsic Dimension Unveils Hallucinations in Diffusion Models
Bartlomiej Sobieski, Matthew Tivnan, Dawid PÅudowski +4
Diffusion models are prone to generating structural hallucinations - samples that match the statistical properties of the training data yet defy underlying structural rules, result…
HOSC: A Periodic Activation with Saturation Control for High-Fidelity Implicit Neural Representations
Michal Jan Wlodarczyk, Danzel Serrano, Przemyslaw Musialski
Periodic activations such as sine preserve high-frequency information in implicit neural representations (INRs) through their oscillatory structure, but often suffer from gradient…
A Finite Difference Approximation of Second Order Regularization of Neural-SDFs
Haotian Yin, Aleksander Plocharski, Michal Jan Wlodarczyk +1
We introduce a finite-difference framework for curvature regularization in neural signed distance field (SDF) learning. Existing approaches enforce curvature priors using full Hess…
FlatCAD: Fast Curvature Regularization of Neural SDFs for CAD Models
Haotian Yin, Aleksander Plocharski, Michal Jan Wlodarczyk +2
Neural signed-distance fields (SDFs) are a versatile backbone for neural geometry representation, but enforcing CAD-style developability usually requires Gaussian-curvature penalti…