# What Do Temporal Graph Learning Models Learn?

arXiv: https://arxiv.org/abs/2510.09416 · HTML: https://mlsanity.com/paper/2510.09416/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 34 references that are in the map.
- First posted on arXiv 10 Oct 2025.

## Builds on (most cited first, 24 of 24)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Open Graph Benchmark: Datasets for Machine Learning on Graphs](https://mlsanity.com/paper/2005.00687/) (2020)
- [MLP-Mixer: An all-MLP Architecture for Vision](https://mlsanity.com/paper/2105.01601/) (2021)
- [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)
- [Predicting Dynamic Embedding Trajectory in Temporal Interaction Networks](https://mlsanity.com/paper/1908.01207/) (2019)
- [Inductive Representation Learning in Temporal Networks via Causal Anonymous Walks](https://mlsanity.com/paper/2101.05974/) (2021)
- [Towards Better Dynamic Graph Learning: New Architecture and Unified Library](https://mlsanity.com/paper/2303.13047/) (2023)
- [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)
- [Evaluating Graph Neural Networks for Link Prediction: Current Pitfalls and New Benchmarking](https://mlsanity.com/paper/2306.10453/) (2023)
- [TCL: Transformer-based Dynamic Graph Modelling via Contrastive Learning](https://mlsanity.com/paper/2105.07944/) (2021)
- [Position: Graph Learning Will Lose Relevance Due To Poor Benchmarks](https://mlsanity.com/paper/2502.14546/) (2025)
- [TGB 2.0: A Benchmark for Learning on Temporal Knowledge Graphs and Heterogeneous Graphs](https://mlsanity.com/paper/2406.09639/) (2024)
- [Evaluating Link Prediction Methods](https://mlsanity.com/paper/1505.04094/) (2015)
- [A survey of dynamic graph neural networks](https://mlsanity.com/paper/2404.18211/) (2024)
- [Deep learning for dynamic graphs: models and benchmarks](https://mlsanity.com/paper/2307.06104/) (2023)
- [TGB-Seq Benchmark: Challenging Temporal GNNs with Complex Sequential Dynamics](https://mlsanity.com/paper/2502.02975/) (2025)
- [On the Power of Heuristics in Temporal Graphs](https://mlsanity.com/paper/2502.04910/) (2025)
- [UTG: Towards a Unified View of Snapshot and Event Based Models for Temporal Graphs](https://mlsanity.com/paper/2407.12269/) (2024)
- [BenchTemp: A General Benchmark for Evaluating Temporal Graph Neural Networks](https://mlsanity.com/paper/2308.16385/) (2023)
- [From Link Prediction to Forecasting: Addressing Challenges in Batch-based Temporal Graph Learning](https://mlsanity.com/paper/2406.04897/) (2024)
- [Are We Really Measuring Progress? Transferring Insights from Evaluating Recommender Systems to Temporal Link Prediction](https://mlsanity.com/paper/2506.12588/) (2025)

## Built on by (most cited first, 4 of 4)

- [Where Does Neural Advantage Arise in Continuous-Time Dynamic Graph Prediction?](https://mlsanity.com/paper/2608.06765/) (2026)
- [Temporal Knowledge Graph Forecasting under Distribution Shifts: A Synthetic Evaluation](https://mlsanity.com/paper/2607.09232/) (2026)
- [What Softmax Throws Away: Mass-Aware Attention for Evidence Accumulation](https://mlsanity.com/paper/2607.22781/) (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)

Learning on temporal graphs has become a central topic in graph representation learning, with numerous benchmarks indicating the strong performance of state-of-the-art models. However, recent work has raised concerns about the reliability of benchmark results, noting issues with commonly used evaluation protocols and the surprising competitiveness of simple heuristics. This contrast raises the question of which characteristics of the underlying graphs temporal graph learning models actually use to form their predictions. We address this by systematically evaluating eight models on their ability to capture eight fundamental characteristics related to the link structure of temporal graphs. These include structural characteristics such as density, temporal patterns such as recency, and edge formation mechanisms such as homophily. Using both synthetic and real-world datasets, we analyze how well models learn these characteristics. Our findings reveal a mixed picture: models capture some characteristics well but fail to reproduce others. With this, we expose important limitations. Overall, we believe that our results provide practical insights for the application of temporal graph learning models and motivate more interpretability-driven evaluations in graph learning research.