Test-time GNN Model Evaluation on Dynamic Graphs
Not on the map. Why?
Abstract (arXiv)
Dynamic graph neural networks (DGNNs) have emerged as a leading paradigm for learning from dynamic graphs, which are commonly used to model real-world systems and applications. However, due to the evolving nature of dynamic graph data distributions over time, well-trained DGNNs often face significant performance uncertainty when inferring on unseen and unlabeled test graphs in practical deployment. In this case, evaluating the performance of deployed DGNNs at test time is crucial to determine whether a well-trained DGNN is suited for inference on an unseen dynamic test graph. In this work, we introduce a new research problem: DGNN model evaluation, which aims to assess the performance of a specific DGNN model trained on observed dynamic graphs by estimating its performance on unseen dynamic graphs during test time. Specifically, we propose a Dynamic Graph neural network Evaluator, dubbed DyGEval, to address this new problem. The proposed DyGEval involves a two-stage framework: (1) test-time dynamic graph simulation, which captures the training-test distributional differences as supervision signals and trains an evaluator; and (2) DyGEval development and training, which accurately estimates the performance of the well-trained DGNN model on the test-time dynamic graphs. Extensive experiments demonstrate that the proposed DyGEval serves as an effective evaluator for assessing various DGNN backbones across different dynamic graphs under distribution shifts.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 22 of its 36 references that are in the map. First posted on arXiv 28 Sep 2025.
Builds on
22 of 36 references.
References · 22
- 2023Time-LLM: Time Series Forecasting by Reprogramming Large Language Modelscited by 700
- 2020Temporal Graph Networks for Deep Learning on Dynamic Graphscited by 530
- 2020Inductive Representation Learning on Temporal Graphscited by 462
- 2021Inductive Representation Learning in Temporal Networks via Causal Anonymous Walkscited by 193
- 2021Dynamic Graph Convolutional Recurrent Network for Traffic Prediction: Benchmark and Solutioncited by 157
- 2023Temporal Graph Benchmark for Machine Learning on Temporal Graphscited by 138
- 2023Large Models for Time Series and Spatio-Temporal Data: A Survey and Outlookcited by 110
- 2019Session-based Social Recommendation via Dynamic Graph Attention Networkscited by 76
- 2023Structure-free Graph Condensation: From Large-scale Graphs to Condensed Graph-free Datacited by 71
- 2025Assemble Your Crew: Automatic Multi-agent Communication Topology Design via Autoregressive Graph Generationcited by 58
- 2025Understanding the Information Propagation Effects of Communication Topologies in LLM-based Multi-Agent Systemscited by 43
- 2025BlindGuard: Safeguarding LLM-based Multi-Agent Systems under Unknown Attackscited by 38
10 more
- 2020RiskOracle: A Minute-level Citywide Traffic Accident Forecasting Frameworkcited by 33
- 2023Environment-Aware Dynamic Graph Learning for Out-of-Distribution Generalizationcited by 26
- 2024Spectral Invariant Learning for Dynamic Graphs under Distribution Shiftscited by 19
- 2023GNNEvaluator: Evaluating GNN Performance On Unseen Graphs Without Labelscited by 17
- 2025Graph-Augmented Large Language Model Agents: Current Progress and Future Prospectscited by 14
- 2024Online GNN Evaluation Under Test-time Graph Distribution Shiftscited by 12
- 2023Towards Expressive Spectral-Temporal Graph Neural Networks for Time Series Forecastingcited by 8
- 2023Out-of-Distribution Generalized Dynamic Graph Neural Network for Human Albumin Predictioncited by 5
- 2025THEMIS: Towards Practical Intellectual Property Protection for Post-Deployment On-Device Deep Learning Modelscited by 2
- 2025OASIS: Harnessing Diffusion Adversarial Network for Ocean Salinity Imputation using Sparse Drifter Trajectoriescited by 1
Built on by
Citing papers per year: 2026: 1.
1 paper.