Super–Item Interaction With Contrastive Learning for Structure-Level Cross-Domain Recommendation

Cross-domain recommendation aims to leverage knowledge from multiple domains to mitigate issues of data sparsity and cold–start problems. While traditional cross-domain settings often involve semantic domains (<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">e.g.</i>, movies versus books), recent research has expanded this notion to include structure-level domains that reflect different types of interaction graphs. To this end, we propose <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Super–Item Interaction with Contrastive Learning for Structure-level Cross-domain Recommendation</i> (SI<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">${}^{2}$</tex-math></inline-formula>CL), a framework designed to explore item-level structural associations and transfer knowledge across graph-based domains. Specifically, SI<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">${}^{2}$</tex-math></inline-formula>CL integrates a user–item interaction graph and an item–item graph—each capturing explicit and implicit signals, respectively—while addressing the challenge of noisy or sparse connections via contrastive denoising. Super–item interaction further facilitates knowledge transfer by modeling shared preferences of highly connected items and clusters. By reconstructing an item–item graph and aligning it with user feedback through structure-aware contrastive learning, our approach uncovers latent item relationships and improves recommendation robustness. Extensive experiments validate the effectiveness of SI<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">${}^{2}$</tex-math></inline-formula>CL in enhancing both accuracy and diversity in sparse or cold–start recommendation scenarios.