Latent Dynamics · 2026-10-15

Dynamic Graph World Modelling: Learning Future Dynamics from Temporal Context

Bo Li, Yue Yang, Yujin Huang, Siao Liu, David Lemphers, Sara E Rodriguez Gomez, Luke Borgnolo, Can Wang, Xin Zheng

arXiv:2610.19364PDFCode

AI-assisted summary based on the paper. Consult the linked source for methods and limitations. Benchmark numbers appear after editor verification.

TL;DR

DyG-WM is a dynamic graph world modelling framework that jointly predicts future latent node and relation representations from temporal context. It uses a historical node-relation context encoder and a node-relation path-signature encoder to capture multi-level evidence and path-dependent evolution. A dual-level joint-embedding predictive objective learns future dynamics in representation space without reconstructing the raw graph. Experiments on 12 link-prediction and two node-classification datasets show effectiveness across both tasks.

Why it matters

The authors frame dynamic graph world modelling as predicting future latent states (nodes and relations) rather than task-specific outputs or raw-graph reconstruction. By explicitly modeling both node- and relation-level history and ordered structural changes, DyG-WM targets the temporal-context-to-future-dynamics setting needed for planning/decision-making in evolving relational systems.

Method

  • Historical node-relation context encoder: summarizes node-neighborhood trajectories (node-level) and pair-specific structural histories (relation-centric) using only preceding snapshots.
  • Node-relation path-signature encoder: uses truncated second-order signatures to capture ordered structural changes at both node and relation levels (path-dependent evolution).
  • Dual-level joint-embedding predictive objective: uses EMA target encoders to predict future latent node and relation representations (JEPA-style), without reconstructing adjacency or raw attributes.

Limitation

The snapshot construction does not generally identify pairwise interaction multiplicity; the authors state that mapping from the raw event stream to snapshot sequence is not injective under their snapshot construction.

Abstract (from arXiv)

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.

Related papers

Publication record

Bibliographic evidence describes where and when a record appeared. It does not validate scientific claims.

First public release: 2026-10-14 · Indexed by Pub-AI: 2026-10-15 UTC

preprint

Dynamic Graph World Modelling: Learning Future Dynamics from Temporal Context

Bo Li, Yue Yang, Yujin Huang, Siao Liu, David Lemphers, Sara E Rodriguez Gomez, Luke Borgnolo, Can Wang, Xin Zheng

Preprint posted: 2026-10-14 · Original record ↗

arxiv
2610.19364
Explicit source versions (1)
  • v1 · 2026-10-14
  1. Preprint available

    Source date: 2026-10-14 · First detected: 2026-10-25 UTC · Historical baseline

    Source evidence (1)
Abstract · arxiv

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.