H$^2$CL: Heterogeneity-Aware Hypergraph Contrastive Learning for Robust Representation Learning

Hypergraph contrastive learning has shown strong potential for modeling high-order relations, but most existing methods implicitly assume that nodes within the same hyperedge are semantically homogeneous. In real-world hypergraphs, this assumption often breaks down: heterogeneous node--hyperedge interactions may introduce noisy signals during augmentation and message passing, leading to feature contamination and false-positive contrastive alignment. To address this issue, we propose H$^2$CL, a heterogeneity-aware hypergraph contrastive learning framework for robust representation learning. H$^2$CL first estimates node--hyperedge heterogeneity from input features and uses it to guide a Heterogeneity-Aware View Generator, which selectively masks high-heterogeneity incidences and constructs cleaner contrastive views. It further introduces a Heterogeneity-Aware Hypergraph Encoder that dynamically reweights information propagation in both node-to-hyperedge and hyperedge-to-node aggregation, enabling hyperedges to aggregate more homogeneous signals while suppressing heterogeneous noise. We also provide theoretical analysis showing that the encoder corresponds to a coordinate descent step for minimizing a heterogeneity-weighted Dirichlet energy. Extensive experiments on standard benchmarks and larger-scale hypergraphs demonstrate that H$^2$CL achieves competitive or superior performance compared with recent baselines, remains robust under structural noise, and learns reweighting patterns that effectively reduce heterogeneity during message passing. Our code is available at: https://github.com/sxu-yaokx/HHCL

Many real-world data sets are not just made of pairs of connected items. For example, a scientific paper may connect several authors, a purchase record may connect a user with several products, and a recommendation system may connect many related actions. Existing AI methods often assume that all items grouped together are similar, but this can be misleading when one item belongs to a different topic or community. Such mixed connections can pass confusing information through the model and reduce the quality of its predictions. Our work introduces H²CL, a method that first checks how well each item fits the group it belongs to. It then builds cleaner training examples by reducing the influence of confusing connections, and it also adjusts how much information each item is allowed to pass through a group. This helps the model learn from useful group relationships while ignoring misleading signals. Experiments on several standard and larger data sets show that this approach improves prediction accuracy and remains more reliable when extra noisy connections are added. Overall, H²CL provides a more robust way to learn from complex grouped data in applications such as citation analysis, recommendation systems, and other network-based AI tasks.