MC-HNN: Learning Latent Structural Semantics and High-Rank Representations for Hypergraph Neural Networks

Shuyang Fang, Yuqin Huang, Zelong Yang, Yintao Cai, Xiaoping Min
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:29132-29146, 2026.

Abstract

Hypergraph Neural Networks (HNNs) have emerged as powerful tools for modeling complex high-order correlations. Most existing HNNs adhere to a two-stage message passing paradigm, where node feature propagation is mediated by hyperedges. In this paper, we analyze two structural limitations of this paradigm, which we term rank collapse and hyperedge semantic dependency. To address these challenges, we propose the Multi-Channel Hypergraph Neural Network (MC-HNN). We design a multi-channel message passing mechanism to maintain high-rank representations, while simultaneously introducing a latent hyperedge type encoding mechanism to inject an independent degree of freedom into hyperedge representations. Our analysis and experiments suggest that MC-HNN alleviates these bottlenecks and achieves strong empirical performance.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-fang26g, title = {{MC}-{HNN}: Learning Latent Structural Semantics and High-Rank Representations for Hypergraph Neural Networks}, author = {Fang, Shuyang and Huang, Yuqin and Yang, Zelong and Cai, Yintao and Min, Xiaoping}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {29132--29146}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/fang26g/fang26g.pdf}, url = {https://proceedings.mlr.press/v306/fang26g.html}, abstract = {Hypergraph Neural Networks (HNNs) have emerged as powerful tools for modeling complex high-order correlations. Most existing HNNs adhere to a two-stage message passing paradigm, where node feature propagation is mediated by hyperedges. In this paper, we analyze two structural limitations of this paradigm, which we term rank collapse and hyperedge semantic dependency. To address these challenges, we propose the Multi-Channel Hypergraph Neural Network (MC-HNN). We design a multi-channel message passing mechanism to maintain high-rank representations, while simultaneously introducing a latent hyperedge type encoding mechanism to inject an independent degree of freedom into hyperedge representations. Our analysis and experiments suggest that MC-HNN alleviates these bottlenecks and achieves strong empirical performance.} }
Endnote
%0 Conference Paper %T MC-HNN: Learning Latent Structural Semantics and High-Rank Representations for Hypergraph Neural Networks %A Shuyang Fang %A Yuqin Huang %A Zelong Yang %A Yintao Cai %A Xiaoping Min %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-fang26g %I PMLR %P 29132--29146 %U https://proceedings.mlr.press/v306/fang26g.html %V 306 %X Hypergraph Neural Networks (HNNs) have emerged as powerful tools for modeling complex high-order correlations. Most existing HNNs adhere to a two-stage message passing paradigm, where node feature propagation is mediated by hyperedges. In this paper, we analyze two structural limitations of this paradigm, which we term rank collapse and hyperedge semantic dependency. To address these challenges, we propose the Multi-Channel Hypergraph Neural Network (MC-HNN). We design a multi-channel message passing mechanism to maintain high-rank representations, while simultaneously introducing a latent hyperedge type encoding mechanism to inject an independent degree of freedom into hyperedge representations. Our analysis and experiments suggest that MC-HNN alleviates these bottlenecks and achieves strong empirical performance.
APA
Fang, S., Huang, Y., Yang, Z., Cai, Y. & Min, X.. (2026). MC-HNN: Learning Latent Structural Semantics and High-Rank Representations for Hypergraph Neural Networks. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:29132-29146 Available from https://proceedings.mlr.press/v306/fang26g.html.

Related Material