Polynomial Neural Sheaf Diffusion: A Spectral Filtering Approach on Cellular Sheaves
Published in arXiv preprint arXiv:2512.00242, 2025
Abstract (snapshot)
This preprint explores polynomial sheaf diffusion paired with equivariant sheaf neural networks to model structured data on graphs. The approach leverages sheaf-theoretic operators to encode consistency and symmetry, providing a principled way to diffuse information while respecting local geometry.
Resources
- 📄 ArXiv: https://arxiv.org/abs/2512.00242
- 🧾 BibTeX: https://arxiv.org/bib/2512.00242
Recommended citation: Borgi, A.; Silvestri F.; Liò P. (2025). "Polynomial Neural Sheaf Diffusion: A Spectral Filtering Approach on Cellular Sheaves.
Download Paper | Download Bibtex