EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

STAGE: LLM-Driven Semantic and Topological Augmented Graph Embedding for Text-Attributed Graphs

Shiwei Huang, Shunxin Xiao, Xu-Yao Zhang, Shunzhi Zhu, Luoqi Liu, Da-Han Wang

Mathematics · 06.05.2026 · 14 (9) · 1568

DOI: 10.3390/math14091568 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Mathematics, online erschienen am 06.05.2026. Verfasst von Shiwei Huang und 5 weiteren Autor:innen. Die Arbeit nennt 44 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Xiamen Municipal Bureau of Science and Technology, State Key Laboratory of Multimodal Artificial Intelligence Systems.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Text-attributed graphs (TAGs) require models to jointly exploit node text and graph structure, yet doing so effectively remains difficult when node text is sparse and the structural context is large. Here, we propose STAGE (Semantic and Topological Augmented Graph Embedding), a two-stage framework for representation learning on TAGs. In Stage I, a frozen large language model is used offline to generate explanatory text that enriches compressed node attributes without introducing online LLM training cost. In Stage II, STAGE performs structure-aware representation learning under a fixed global token budget by combining random-walk-based structural context with graph-conditioned token reduction before PLM encoding. This design preserves informative semantic content while preventing unconstrained sequence expansion. Experiments on seven benchmark datasets show that STAGE consistently outperforms strong baselines under the same evaluation setting and maintains favorable efficiency under bounded input-length constraints.

Förderung

  • National Natural Science Foundation of China
  • Xiamen Municipal Bureau of Science and Technology
  • State Key Laboratory of Multimodal Artificial Intelligence Systems

Zitieren

APA

Huang, S., Xiao, S., Zhang, X., Zhu, S., Liu, L., & Wang, D. (2026). STAGE: LLM-Driven Semantic and Topological Augmented Graph Embedding for Text-Attributed Graphs. Mathematics, 14(9), 1568. https://doi.org/10.3390/math14091568

Vancouver

Huang S, Xiao S, Zhang X, Zhu S, Liu L, Wang D. STAGE: LLM-Driven Semantic and Topological Augmented Graph Embedding for Text-Attributed Graphs. Mathematics. 2026;14(9):1568. doi:10.3390/math14091568

BibTeX
@article{Huang2026,
  author = {Huang, Shiwei and Xiao, Shunxin and Zhang, Xu-Yao and Zhu, Shunzhi and Liu, Luoqi and Wang, Da-Han},
  title = {STAGE: LLM-Driven Semantic and Topological Augmented Graph Embedding for Text-Attributed Graphs},
  journal = {Mathematics},
  year = {2026},
  volume = {14},
  number = {9},
  pages = {1568},
  doi = {10.3390/math14091568}
}
Kostenlos starten →