EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

FAESRec: Frequency Adaptive Encoder for Sequential Recommendation

Wenxian Liu, Yiji Zhao, Haomin Wen, Lei Zhang, Hao Wu

ACM Transactions on Information Systems · 30.11.2026 · 44 (8) · 1-33

DOI: 10.1145/3849380 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACM Transactions on Information Systems, online erschienen am 13.11.2026. Verfasst von Wenxian Liu und 4 weiteren Autor:innen. Die Arbeit nennt 92 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Yunnan Fundamental Research Projects, Open Fund of Beijing Key Laboratory of Traffic Data Mining and Embodied Intelligence.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Sequential recommender systems predominantly focus on time-domain modeling, overlooking rich seasonal and periodic structures revealed in the frequency domain. While recent frequency-aware methods leverage Fourier transforms, their reliance on fixed filters introduces two critical limitations: (i) inadequate sequence encoding due to indiscriminate spectral suppression, causing information degradation, and (ii) semantic-degrading contrastive augmentations that compromise critical frequency components essential for behavioral semantics. To address these challenges, we propose FAESRec, a Frequency Adaptive Encoder with three synergistic innovations: First, a Frequency Adaptive Mixture-of-Experts (FAMoE) layer decomposes sequences into non-overlapping spectral bands with end-to-end learned boundaries, enabling dynamic weighting of heterogeneous frequency drivers. Second, a Filter-Enhanced module (FAFe) employs learnable global/local filters with adaptive thresholds to denoise spectral components while preserving energy-concentrated bands. The filtered spectral features are projected back to the time domain and encoded by a Mamba block. Third, a Frequency Semantic-aware Contrastive Learner (FSaCL) synthesizes semantically coherent views through probabilistic frequency masking, theoretically guaranteeing mutual information conservation between original and augmented sequences. Extensive experiments on six benchmarks from Amazon, Movielens, and LastFM demonstrate FAESRec’s consistent superiority over 13 SOTA baselines. It achieves significant improvements of 2.74–67.69% HR@1 and 3.95–20.38% NDCG@5, particularly under extreme sparsity. Theoretical analysis proves that FSaCL preserves minimal sufficient semantics, establishing a new paradigm for frequency-aware sequential recommendation. Code is available at: https://github.com/0-1user/FAESRec .

Förderung

  • National Natural Science Foundation of China
  • Yunnan Fundamental Research Projects
  • Open Fund of Beijing Key Laboratory of Traffic Data Mining and Embodied Intelligence
  • Open Project Program of Yunnan Key Laboratory of Intelligent Systems and Computing
  • Yunnan University School of ISE Graduate Student Scientific Research Project

Zitieren

APA

Liu, W., Zhao, Y., Wen, H., Zhang, L., & Wu, H. (2026). FAESRec: Frequency Adaptive Encoder for Sequential Recommendation. ACM Transactions on Information Systems, 44(8), 1-33. https://doi.org/10.1145/3849380

Vancouver

Liu W, Zhao Y, Wen H, Zhang L, Wu H. FAESRec: Frequency Adaptive Encoder for Sequential Recommendation. ACM Transactions on Information Systems. 2026;44(8):1-33. doi:10.1145/3849380

BibTeX
@article{Liu2026,
  author = {Liu, Wenxian and Zhao, Yiji and Wen, Haomin and Zhang, Lei and Wu, Hao},
  title = {FAESRec: Frequency Adaptive Encoder for Sequential Recommendation},
  journal = {ACM Transactions on Information Systems},
  year = {2026},
  volume = {44},
  number = {8},
  pages = {1--33},
  doi = {10.1145/3849380}
}
Kostenlos starten →