Zebra: When Temporal Graph Neural Networks Meet Temporal Personalized PageRank
Summary: Links T-GNN temporal message passing to temporal random walks and proposes T-PPR, a parameterized influence score showing a few temporal neighbors suffice for accurate representations. Introduces Zebra: a scalable framework with a single-scan top-k T-PPR algorithm (provable approximation) that aggregates top influencers, delivering up to 100× speedups and often better accuracy. (summarized by gpt-5-mini on Feb 09 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Yiming Li (Hong Kong University of Science and Technology)
- 2. Yanyan Shen (Shanghai Jiao Tong University)
- 3. Lei Chen (Hong Kong University of Science and Technology)
- 4. Mingxuan Yuan (Huawei)
BibTeX Citation
@article{li_vldb23,
title = {{Zebra: When Temporal Graph Neural Networks Meet Temporal Personalized PageRank}},
author = {Li, Yiming and Shen, Yanyan and Chen, Lei and Yuan, Mingxuan},
journal = {PVLDB},
series = {{VLDB} '23},
volume = {16},
number = {6},
pages = {1332--1345},
doi = {10.14778/3583140.3583150},
url = {https://doi.org/10.14778/3583140.3583150},
year = {2023}
}
Incoming Citations (Sorted by Pagerank)
Showing 19 of 19 citing papers.
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 12 of 12 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
Previous
Page 1 / 1
Next