HodgeCover isolates the harmonic kernel of a simplicial Laplacian on an expert 2-complex to identify irreducible merge cycles and selects experts for aggressive compression, matching or exceeding baselines on open-weight MoE models.
Sim- plicial neural networks.arXiv preprint arXiv:2010.03633
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
fields
cs.LG 3years
2026 3verdicts
UNVERDICTED 3representative citing papers
Hodge Spectral Duality provides a topology-preserving neural operator by isolating unlearnable topological components via Hodge orthogonality and operator splitting.
ModernSASST is the first simplicial complex-based spatiotemporal model that combines random walks on high-dimensional complexes with parallelizable temporal convolutional networks for efficient high-order topology capture.
citing papers explorer
-
HodgeCover: Higher-Order Topological Coverage Drives Compression of Sparse Mixture-of-Experts
HodgeCover isolates the harmonic kernel of a simplicial Laplacian on an expert 2-complex to identify irreducible merge cycles and selects experts for aggressive compression, matching or exceeding baselines on open-weight MoE models.
-
Topology-Preserving Neural Operator Learning via Hodge Decomposition
Hodge Spectral Duality provides a topology-preserving neural operator by isolating unlearnable topological components via Hodge orthogonality and operator splitting.
-
Modern Structure-Aware Simplicial Spatiotemporal Neural Network
ModernSASST is the first simplicial complex-based spatiotemporal model that combines random walks on high-dimensional complexes with parallelizable temporal convolutional networks for efficient high-order topology capture.