Track trials, findings, and clinical evidence as it lands.
Search papers, labs, and topics across Lattice.
PANDA, a plug-in architecture for Hypergraph Neural Networks, addresses limitations of fixed structural weights and competitive attention by introducing a dual-path mechanism. The main path balances learned receiver-normalized attention with structural incidence priors, while an auxiliary path allows non-competitive sender contributions. This approach yields significant performance improvements, averaging 5.18 percentage points across diverse HNN tasks and datasets.
Hypergraph attention can be made more robust and expressive by decoupling learned weights from structural priors and allowing non-competitive information flow.
Hypergraph Neural Networks (HNNs), designed to model higher-order relations beyond pairwise interactions, typically rely either on fixed structural weights or on normalised attention. Fixed weights can propagate misleading information in heterophilic settings, while normalisation makes attention coefficients competitive, in the sense that, in a message-passing framework, increasing the weight of one sender must reduce the weight of others in the same aggregation set. Our main contribution is the introduction of PANDA (Prior-guided AttentioNal Dual-path Architecture), a plug-in for two-stage message passing HNNs applicable to both undirected and directed settings. The main path interpolates between learned receiver-normalised attention and structural incidence priors, while a non-competitive (not normalised) auxiliary path lets each sender contribute its own transformed information. We also show that, with fixed structural coefficients, PANDA recovers a certain complex-valued convolution operator adopted in previous works. Across four HNNs spanning spatial and spectral methods in both directed and undirected settings, plugging PANDA in yields an average improvement of 5.18 percentage points across eight real-world benchmark datasets. The code is available at https://github.com/jainris/panda.