Understanding Evolving Graph Structures for Large Discrete-Time Dynamic Graph Representation

Summary: UnderGS targets DTDG representation learning by replacing per-snapshot O(T|V|^2) adjacency storage with a GPU-resident temporal-cohesive neighbor store, maintaining only influential temporal neighbors via a temporal influence score. Lightweight, model-agnostic pipeline (MPNN/non-MPNN) with late-snapshot gradient aggregation; up to 9x faster, +31% accuracy. (summarized by gpt-5.4-mini on Apr 12 2026)

Paper ID: 14566
Venue: VLDB
Year: 2026
Pagerank: 5.093636e-05
Overall Rank: 10,620 | 27.14%
DOI: 10.14778/3796195.3796201

Incoming Non-self Citations Over Time

No non-self incoming citations found for this paper in this database.

Authors

1. Danni Wu (East China Normal University)
2. Yuanyuan Xu (University of New South Wales)
3. Xuemin Lin (Chinese University of Hong Kong)
4. Wenjie Zhang (University of New South Wales)
5. Ying Zhang (Zhejiang Gongshang University)

BibTeX Citation

@article{wu_vldb26,
        title = {{Understanding Evolving Graph Structures for Large Discrete-Time Dynamic Graph Representation}},
        author = {Wu, Danni and Xu, Yuanyuan and Lin, Xuemin and Zhang, Wenjie and Zhang, Ying},
        journal = {PVLDB},
        series = {{VLDB} '26},
        volume = {19},
        number = {5},
        pages = {862--875},
        doi = {10.14778/3796195.3796201},
        url = {https://doi.org/10.14778/3796195.3796201},
        year = {2026}
}

Incoming Citations (Sorted by Pagerank)

Showing 0 of 0 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank

Outgoing Citations (Sorted by Pagerank)

Showing 13 of 13 cited papers.

Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.

Rank	Cited Paper	Year	Venue	Pagerank
223	AliGraph: A Comprehensive Graph Neural Network Platform	2019	VLDB	0.00024182473
1,409	TGL: A General Framework for Temporal GNN Training on Billion-Scale Graphs	2022	VLDB	0.00010854808
3,210	Zebra: When Temporal Graph Neural Networks Meet Temporal Personalized PageRank	2023	VLDB	7.6352864e-05
4,878	NeutronStream: A Dynamic GNN Training Framework with Sliding Window for Graph Streams	2024	VLDB	6.4684388e-05
4,980	Decoupled Graph Neural Networks for Large Dynamic Graphs	2023	VLDB	6.4133375e-05
5,205	ETC: Efficient Training of Temporal Graph Neural Networks over Large-scale Dynamic Graphs	2024	VLDB	6.318626e-05
5,596	Efficient Tree-SVD for Subset Node Embedding over Large Dynamic Graphs	2023	SIGMOD	6.1548101e-05
5,760	DynaHB: A Communication-Avoiding Asynchronous Distributed Framework with Hybrid Batches for Dynamic GNN Training	2024	VLDB	6.0961929e-05
6,261	EARLY: Efficient and Reliable Graph Neural Network for Dynamic Graphs	2023	SIGMOD	5.9366952e-05
6,695	GENTI: GPU-powered Walk-based Subgraph Extraction for Scalable Representation Learning on Dynamic Graphs	2024	VLDB	5.7990641e-05
6,730	SIMPLE: Efficient Temporal Graph Neural Network Training at Scale with Dynamic Data Placement	2024	SIGMOD	5.7895038e-05
8,454	D3-GNN: Dynamic Distributed Dataflow for Streaming Graph Neural Networks	2024	VLDB	5.4226e-05
8,503	Fight Fire with Fire: Towards Robust Graph Neural Networks on Dynamic Graphs via Actively Defense	2024	VLDB	5.4132367e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	5,205	ETC: Efficient Training of Temporal Graph Neural Networks over Large-scale Dynamic Graphs	2024	VLDB
2	6,730	SIMPLE: Efficient Temporal Graph Neural Network Training at Scale with Dynamic Data Placement	2024	SIGMOD
3	11,108	Efficient Graph Embedding Generation and Update for Large-Scale Temporal Graph	2025	VLDB
4	6,261	EARLY: Efficient and Reliable Graph Neural Network for Dynamic Graphs	2023	SIGMOD
5	4,891	DGC: Training Dynamic Graphs with Spatio-Temporal Non-Uniformity using Graph Partitioning by Chunks	2023	SIGMOD
6	1,409	TGL: A General Framework for Temporal GNN Training on Billion-Scale Graphs	2022	VLDB
7	9,349	Temporal SIR-GN: Efficient and Effective Structural Representation Learning for Temporal Graphs	2023	VLDB
8	10,555	FlareDTDG: Harnessing Temporal Recency for Scalable Discrete-Time Dynamic Graph Training	2026	VLDB
9	4,980	Decoupled Graph Neural Networks for Large Dynamic Graphs	2023	VLDB
10	11,110	Towards Ideal Temporal Graph Neural Networks: Evaluations and Conclusions after 10,000 GPU Hours	2025	VLDB