Heterophily-Agnostic Hypergraph Neural Networks with Riemannian Local Exchanger

Researchers propose HealHGNN, a novel Hypergraph Neural Network that addresses limitations in traditional networks when dealing with heterophilic hypergraphs. The system uses Riemannian geometry and adaptive local heat exchangers to enable better long-range dependency modeling with linear complexity.

• →Traditional Hypergraph Neural Networks struggle with heterophilic hypergraphs that require long-range dependency modeling.
• →The new approach connects oversquashing and hypergraph bottleneck issues within Riemannian manifold heat flow framework.
• →HealHGNN introduces adaptive local heat exchangers that capture long-range dependencies via Robin condition while preserving representation distinguishability.
• →The system achieves linear complexity in both nodes and hyperedges through bidirectional node-hyperedge architecture.
• →Experimental results show state-of-the-art performance on both homophilic and heterophilic hypergraph cases.