Lune

ICLR2025Top-tier venue

N-ForGOT: Towards Not-forgetting and Generalization of Open Temporal Graph Learning

Liping Wang, Xujia Li, Jingshu Peng, Yue Wang, Chen Zhang, Yan Zhou, Lei Chen

Abstract

Temporal Graph Neural Networks (TGNNs) lay emphasis on capturing node interactions over time but often overlook evolution in node classes and dynamic data distributions triggered by the continuous emergence of new class labels, known as the open-set problem. This problem poses challenges for existing TGNNs in preserving learned classes while rapidly adapting to new, unseen classes. To address this, this paper identifies two primary factors affecting model performance on the open temporal graph, backed by a theoretical guarantee: (1) the forgetting of prior knowledge and (2) distribution discrepancies between successive tasks. Building on theoretical insights, we propose N-ForGOT, which incorporates two plug-in modules into TGNNs to preserve prior knowledge and enhance model generalizability for new classes simultaneously. The first module preserves previously established inter-class connectivity and decision boundaries during the training of new classes to mitigate the forgetting caused by temporal evolutions of class characteristics. The second module introduces an efficient method for measuring distribution discrepancies with designed temporal Weisfeiler-Lehman subtree patterns, effectively addressing both structural and temporal shifts while reducing time complexity. Experimental results on four public datasets demonstrate that our method significantly outperforms state-of-the-art approaches in prediction accuracy, prevention of forgetting, and generalizability 1 .

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

Questions to start from

Lune can draw on

the full paper · related work

Get the answer

Free to start. No credit card required.

Lineage

  • Inductive Representation Learning in Temporal Networks via Causal Anonymous Walks cites Inductive representation learning on temporal graphs
Builds onCited byLeft to right by year; the more cited, the higher.

Builds on16

  1. Inductive representation learning on temporal graphsDa Xu, Chuanwei Ruan, Evren Körpeoglu, Sushant Kumar et al.ICLR 2020 · 901 citations
  2. Inductive Representation Learning in Temporal Networks via Causal Anonymous WalksYanbang Wang, Yen-Yu Chang, Yunyu Liu, Jure Leskovec et al.ICLR 2021 · 326 citations
  3. Overcoming Catastrophic Forgetting in Graph Neural Networks with Experience ReplayFan Zhou, Chengtai CaoAAAI 2021 · 175 citations
  4. Overcoming Catastrophic Forgetting in Graph Neural NetworksHuihui Liu, Yiding Yang, Xinchao WangAAAI 2021 · 166 citations
  5. Shift-Robust GNNs: Overcoming the Limitations of Localized Graph Training dataQi Zhu, Natalia Ponomareva, Jiawei Han, Bryan PerozziNeurIPS 2021 · 152 citations

Related papers

  1. Towards Open Temporal Graph Neural NetworksKaituo Feng, Changsheng Li, Xiaolu Zhang, Jun ZhouICLR 2023 · 4 citations
  2. A Selective Learning Method for Temporal Graph Continual LearningHanmo Liu, Shimin Di, Haoyang Li, Xun Jian et al.ICML 2025
  3. Towards Effective Open-set Graph Class-incremental LearningJiazhen Chen, Zheng Ma, Sichao Fu, Mingbin Feng et al.ACM MM 2025
  4. Class-Domain Incremental Learning on Graphs via Disentangled Knowledge DistillationQin Tian, Chen Zhao, Xintao Wu, Dong Li et al.WWW 2026
  5. FAT-TAG: Mitigating Forgetting in Task-Free Temporal Graph Class Incremental LearningJiyuan Feng, Zhao Liu, Dongyi Zheng, Weihong Han et al.KDD 2026
Dusk over the sea between two cliffs drawn in fine vertical lines