# Temporal graph models fail to capture global temporal dynamics

arXiv: https://arxiv.org/abs/2309.15730 · HTML: https://mlsanity.com/paper/2309.15730/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 14 references that are in the map.
- First posted on arXiv 27 Sep 2023.

## Builds on (most cited first, 11 of 11)

- [Temporal Graph Networks for Deep Learning on Dynamic Graphs](https://mlsanity.com/paper/2006.10637/) (2020)
- [Inductive Representation Learning on Temporal Graphs](https://mlsanity.com/paper/2002.07962/) (2020)
- [Inductive Representation Learning in Temporal Networks via Causal Anonymous Walks](https://mlsanity.com/paper/2101.05974/) (2021)
- [Do We Really Need Complicated Model Architectures For Temporal Networks?](https://mlsanity.com/paper/2302.11636/) (2023)
- [Temporal Graph Benchmark for Machine Learning on Temporal Graphs](https://mlsanity.com/paper/2307.01026/) (2023)
- [Towards Better Evaluation for Dynamic Link Prediction](https://mlsanity.com/paper/2207.10128/) (2022)
- [TCL: Transformer-based Dynamic Graph Modelling via Contrastive Learning](https://mlsanity.com/paper/2105.07944/) (2021)
- [Neighborhood-aware Scalable Temporal Network Representation Learning](https://mlsanity.com/paper/2209.01084/) (2022)
- [An efficient manifold density estimator for all recommendation systems](https://mlsanity.com/paper/2006.01894/) (2020)
- [Cleora: A Simple, Strong and Scalable Graph Embedding Scheme](https://mlsanity.com/paper/2102.02302/) (2021)
- [An Empirical Evaluation of Temporal Graph Benchmark](https://mlsanity.com/paper/2307.12510/) (2023)

## Built on by (most cited first, 5 of 5)

- [Are Large Language Models Good Temporal Graph Learners?](https://mlsanity.com/paper/2506.05393/) (2025)
- [Exploring the Performance of Continuous-Time Dynamic Link Prediction Algorithms](https://mlsanity.com/paper/2405.17182/) (2024)
- [Rolling Forward: Enhancing LightGCN with Causal Graph Convolution for Credit Bond Recommendation](https://mlsanity.com/paper/2503.14213/) (2025)
- [Beat the Counter First: A Baseline for Temporal-Graph Anomaly Detectors](https://mlsanity.com/paper/2608.15965/) (2026)
- [Do Temporal Link Predictors Need Learned Memory? A Smoothed-Count Baseline with a Handful of Parameters](https://mlsanity.com/paper/2609.35364/) (2026)

## Abstract (arXiv)

A recently released Temporal Graph Benchmark is analyzed in the context of Dynamic Link Property Prediction. We outline our observations and propose a trivial optimization-free baseline of "recently popular nodes" outperforming other methods on medium and large-size datasets in the Temporal Graph Benchmark. We propose two measures based on Wasserstein distance which can quantify the strength of short-term and long-term global dynamics of datasets. By analyzing our unexpectedly strong baseline, we show how standard negative sampling evaluation can be unsuitable for datasets with strong temporal dynamics. We also show how simple negative-sampling can lead to model degeneration during training, resulting in impossible to rank, fully saturated predictions of temporal graph networks. We propose improved negative sampling schemes for both training and evaluation and prove their usefulness. We conduct a comparison with a model trained non-contrastively without negative sampling. Our results provide a challenging baseline and indicate that temporal graph network architectures need deep rethinking for usage in problems with significant global dynamics, such as social media, cryptocurrency markets or e-commerce. We open-source the code for baselines, measures and proposed negative sampling schemes.