EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

HMamba : Hyperbolic Mamba for Sequential Recommendation

Qianru Zhang, Honggang Wen, Wei Yuan, Crystal Chen, Menglin Yang, Siu-Ming Yiu, Hongzhi Yin

ACM Transactions on Information Systems · 30.06.2026 · 44 (5) · 1-27

DOI: 10.1145/3811405 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACM Transactions on Information Systems, online erschienen am 23.05.2026. Verfasst von Qianru Zhang und 6 weiteren Autor:innen. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Australian Research Council, Discovery Projects, Linkage Projects.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Sequential recommendation systems require both temporal efficiency to handle long interaction histories and hierarchical representation to model complex user–item relationships. Existing approaches face a fundamental tension: Mamba-based methods offer linear-time efficiency ( \(\mathcal{O}(L)\) ) but operate in Euclidean space, which distorts hierarchical patterns; hyperbolic models capture taxonomies well but suffer quadratic complexity ( \(\mathcal{O}(L^{2})\) ). To solve this dual challenge , we propose Hyperbolic Mamba (HMamba), the first architecture that unifies curvature-aware state spaces with hyperbolic geometry. Our key insight is that hyperbolic curvature \(\kappa\) simultaneously governs: (1) state transition granularity through \(\mathbf{\bar{A}}=\exp(\Delta\mathbf{A}\odot\mathbf{K}(\kappa))\) and (2) hierarchical distance preservation via \(d_{\mathcal{L}}\propto\sqrt{\kappa}\log(\cdot)\) . This enables joint optimization of efficiency and hierarchy—addressing the previously unsolved problem of deep-long modeling . Experiments show HMamba achieves 3–11% accuracy gains while maintaining 3.2 \(\times\) faster training than attention-based models, establishing a new paradigm for hierarchy-aware sequential recommendation. The code and datasets accompanying our paper are publicly available at https://github.com/CoderPowerBeyond/HMamba .

Förderung

  • Australian Research Council
  • Discovery Projects
  • Linkage Projects
  • HKU-SCF FinTech Academy, the Shenzhen-Hong Kong-Macao Science and Technology Plan Project
  • Theme-based Research Scheme of the RGC, Hong Kong

Zitieren

APA

Zhang, Q., Wen, H., Yuan, W., Chen, C., Yang, M., Yiu, S., & Yin, H. (2026). HMamba : Hyperbolic Mamba for Sequential Recommendation. ACM Transactions on Information Systems, 44(5), 1-27. https://doi.org/10.1145/3811405

Vancouver

Zhang Q, Wen H, Yuan W, Chen C, Yang M, Yiu S, et al. HMamba : Hyperbolic Mamba for Sequential Recommendation. ACM Transactions on Information Systems. 2026;44(5):1-27. doi:10.1145/3811405

BibTeX
@article{Zhang2026,
  author = {Zhang, Qianru and Wen, Honggang and Yuan, Wei and Chen, Crystal and Yang, Menglin and Yiu, Siu-Ming and Yin, Hongzhi},
  title = {HMamba : Hyperbolic Mamba for Sequential Recommendation},
  journal = {ACM Transactions on Information Systems},
  year = {2026},
  volume = {44},
  number = {5},
  pages = {1--27},
  doi = {10.1145/3811405}
}
Kostenlos starten →