Invariant Graph Representations for Continuous-Time Dynamic Graphs Under Distribution Shifts

Fang, Lanting; Yang, Yulian; Zhang, Yawei; Feng, Shanshan; Feng, Kaiyu; Yuan, Hanning

doi:10.24963/ijcai.2026/286

Invariant Graph Representations for Continuous-Time Dynamic Graphs Under Distribution Shifts

Lanting Fang, Yulian Yang, Yawei Zhang, Shanshan Feng, Kaiyu Feng, Hanning Yuan

Download BibTeX Cite

Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence. Main Track. Pages 2572-2580.

https://doi.org/10.24963/ijcai.2026/286

Keywords

Machine Learning: Representation learning
Knowledge Representation and Reasoning: Learning and reasoning
Data Mining: Mining graphs

Abstract

Continuous-Time Dynamic Graphs (CTDGs) enable fine-grained modeling of evolving relational systems. However, most existing CTDG representation learning methods are tailored to in-distribution settings and exhibit limited robustness under out-of-distribution (OOD) shifts. Although recent causal approaches learn invariant representations via interventions, they are primarily designed for static or discrete-time graphs and become computationally prohibitive for CTDGs due to the combinatorial explosion of structural and temporal variations.

To address these challenges, we propose CIR, a framework grounded in a novel structural causal model termed the ICCM. To avoid exhaustive interventions, we leverage the Normalized Weighted Geometric Mean (NWGM) to efficiently approximate interventional predictions. We further instantiate ICCM within a practical deep learning architecture that jointly captures invariant structural and temporal patterns through dedicated subgraph extractors, and maintains an environment memory bank to model distributional shifts across evolving contexts. Extensive experiments demonstrate that CIR consistently outperforms existing methods under diverse OOD scenarios.

Lanting Fang, Yulian Yang, Yawei Zhang, Shanshan Feng, Kaiyu Feng, and Hanning Yuan. 2026. Invariant Graph Representations for Continuous-Time Dynamic Graphs Under Distribution Shifts. In Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26. IJCAI, 2572–2580. https://doi.org/10.24963/ijcai.2026/286

@inproceedings{ijcai2026p286,
  title     = {Invariant Graph Representations for Continuous-Time Dynamic Graphs Under Distribution Shifts},
  author    = {Fang, Lanting and Yang, Yulian and Zhang, Yawei and Feng, Shanshan and Feng, Kaiyu and Yuan, Hanning},
  booktitle = {Proceedings of the Thirty-Fifth International Joint Conference on
               Artificial Intelligence, {IJCAI-26}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {Diego Calvanese},
  pages     = {2572--2580},
  year      = {2026},
  month     = {8},
  note      = {Main Track},
  doi       = {10.24963/ijcai.2026/286},
  url       = {https://doi.org/10.24963/ijcai.2026/286},
}

TY  - CPAPER
TI  - Invariant Graph Representations for Continuous-Time Dynamic Graphs Under Distribution Shifts
AU  - Fang, Lanting
AU  - Yang, Yulian
AU  - Zhang, Yawei
AU  - Feng, Shanshan
AU  - Feng, Kaiyu
AU  - Yuan, Hanning
T2  - Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2026
SP  - 2572
EP  - 2580
DO  - 10.24963/ijcai.2026/286
UR  - https://doi.org/10.24963/ijcai.2026/286
ER  -