arXiv：cs.LG· Bo Li, Yue Yang, Yujin Huang, Siao Liu, David Lemphers, Sara E Rodriguez Gomez, Luke Borgnolo, Can Wang, Xin Zheng·· 3 小时前
DyG-WM：从时序上下文学习动态图未来演化的世界模型框架
Dynamic Graph World Modelling: Learning Future Dynamics from Temporal Context
AI 导读
研究者提出 Dynamic Graph World Modelling（DyG-WM），首个从时序上下文联合预测未来节点与关系表示的框架，通过历史节点-关系上下文编码器、路径签名编码器和双层联合嵌入目标，回答"什么已演化、如何演化、下一步将演化什么"。在 12 个链接预测和 2 个节点分类数据集上的实验验证了其有效性。
正文
Abstract:Dynamic graph learning models how entities and their relations evolve from historical observations to predict future graph states. Existing methods do not fully address dynamic graph world modelling, which requires learning transitions from temporal context to future latent graph states. This is difficult for three reasons: nodes and relations rely on different historical evidence, structural evolution is path-dependent, and future snapshots provide sparse and realization-specific targets. We propose Dynamic Graph World Modelling (DyG-WM), the first framework to jointly predict future node and relation representations from temporal context by answering three questions: what has evolved, how it evolved, and what will evolve next. To capture what has evolved, a historical node-relation context encoder summarizes node-neighborhood trajectories and pair-specific histories. To characterize how it evolved, a node-relation path-signature encoder captures ordered structural changes. To predict what will evolve next, a dual-level joint-embedding objective predicts target-snapshot node and relation representations without reconstructing the raw graph. Experiments on 12 link-prediction and two node-classification datasets demonstrate the effectiveness of DyG-WM across both tasks.
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.19364 [cs.LG] |
| (or arXiv:2610.19364v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.19364 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Bo Li [view email]
[v1]
Wed, 14 Oct 2026 01:09:47 UTC (629 KB)
来源：arXiv：cs.LG · arxiv.org