sheaf

A clean PyTorch / PyG implementation library of Sheaf Neural Networks as Message Passing comprising all variants and a benchmark suite with 14+ Datasets.

Stalks & Sections — WebGL explorer of cellular sheaves. Variable-dimension stalks, restriction maps, and sheaf-Laplacian residuals as first-class visual variables.

Sheaf-theoretic gossip reconciliation: H¹ obstruction detection, spectral gap convergence, multi-agent knowledge consistency

Sheaf neural network over the MZV motivic weight filtration for odd Riemann zeta value research, with GPU-accelerated polynomial continued fraction search

[NeurIPS 2026] HetSheaf (Heterogeneous Sheaf Neural Networks) is a framework for learning relation-aware sheaf structures on heterogeneous graphs. HetSheaf learns type- and relation-dependent restriction maps to model structured information transport across heterogeneous nodes and edges.

Contract-driven AI workflow system for deterministic software execution and continuity across sessions.

Last-scattering engine. WMAP ILC + Planck TT + Smith/sheaf Horizon. Finite certificates. Digital Horizon is a criterion on a constructed 12-port sheaf — not a sky measurement.