EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

MGDSL: Multimodal Graph Denoising and Self-Supervised Learning for Multimedia Recommendation

Hongyu Xu, Liye Shi, Pengfei Shao, Yunkai Zhuang

Electronics · 13.06.2026 · 15 (12) · 2616

DOI: 10.3390/electronics15122616 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Electronics, online erschienen am 13.06.2026. Verfasst von Hongyu Xu und 3 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 51 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: the Exploration Project of the Natural Science Foundation of Zhejiang Province, Zhejiang Provincial Natural Science Foundation.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Multimedia recommenders can use behavioral records together with visual and textual item information, but unreliable interactions and sparse histories still make user preference modeling difficult. Most graph-based methods propagate messages over observed user–item edges as if all interactions were equally informative, so incidental or semantically inconsistent behaviors may distort the learned representations. The standard recommendation loss also provides limited context for modeling dependencies within a user’s historical sequence. We propose MGDSL, a MGDSL applies a multimodal-aware topology denoising module to calculate edge reliability weights for historical interactions from collaborative, textual, and visual evidence, and uses these weights for reliability-aware historical aggregation. In parallel, a masked self-supervised auxiliary task reconstructs masked items from sequence context, adding supervision for latent preference learning. Experiments on three benchmark datasets show that MGDSL consistently improves recommendation accuracy over competitive baselines, with particularly clear gains on the sparsest dataset.

Förderung

  • the Exploration Project of the Natural Science Foundation of Zhejiang Province
  • Zhejiang Provincial Natural Science Foundation

Zitieren

APA

Xu, H., Shi, L., Shao, P., & Zhuang, Y. (2026). MGDSL: Multimodal Graph Denoising and Self-Supervised Learning for Multimedia Recommendation. Electronics, 15(12), 2616. https://doi.org/10.3390/electronics15122616

Vancouver

Xu H, Shi L, Shao P, Zhuang Y. MGDSL: Multimodal Graph Denoising and Self-Supervised Learning for Multimedia Recommendation. Electronics. 2026;15(12):2616. doi:10.3390/electronics15122616

BibTeX
@article{Xu2026,
  author = {Xu, Hongyu and Shi, Liye and Shao, Pengfei and Zhuang, Yunkai},
  title = {MGDSL: Multimodal Graph Denoising and Self-Supervised Learning for Multimedia Recommendation},
  journal = {Electronics},
  year = {2026},
  volume = {15},
  number = {12},
  pages = {2616},
  doi = {10.3390/electronics15122616}
}
Kostenlos starten →