# Enhancing the Expressivity of Temporal Graph Networks through Source-Target Identification

arXiv: https://arxiv.org/abs/2411.03596 · HTML: https://mlsanity.com/paper/2411.03596/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 24 references that are in the map.
- First posted on arXiv 6 Nov 2024.

## Builds on (most cited first, 20 of 20)

- [Inductive Representation Learning on Large Graphs](https://mlsanity.com/paper/1706.02216/) (2017)
- [Learning to Simulate Complex Physics with Graph Networks](https://mlsanity.com/paper/2002.09405/) (2020)
- [Learning Mesh-Based Simulation with Graph Networks](https://mlsanity.com/paper/2010.03409/) (2020)
- [Temporal Graph Networks for Deep Learning on Dynamic Graphs](https://mlsanity.com/paper/2006.10637/) (2020)
- [Masked Label Prediction: Unified Message Passing Model for Semi-Supervised Classification](https://mlsanity.com/paper/2009.03509/) (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)
- [T-GCN: A Temporal Graph ConvolutionalNetwork for Traffic Prediction](https://mlsanity.com/paper/1811.05320/) (2018)
- [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)
- [Foundations and modelling of dynamic networks using Dynamic Graph Neural Networks: A survey](https://mlsanity.com/paper/2005.07496/) (2020)
- [Session-based Social Recommendation via Dynamic Graph Attention Networks](https://mlsanity.com/paper/1902.09362/) (2019)
- [Approximation Ratios of Graph Neural Networks for Combinatorial Problems](https://mlsanity.com/paper/1905.10261/) (2019)
- [TCL: Transformer-based Dynamic Graph Modelling via Contrastive Learning](https://mlsanity.com/paper/2105.07944/) (2021)
- [Dynamic Network Embedding Survey](https://mlsanity.com/paper/2103.15447/) (2021)
- [A Comprehensive Survey of Dynamic Graph Neural Networks: Models, Frameworks, Benchmarks, Experiments and Challenges](https://mlsanity.com/paper/2405.00476/) (2024)
- [Representation Learning over Dynamic Graphs](https://mlsanity.com/paper/1803.04051/) (2018)
- [An Empirical Evaluation of Temporal Graph Benchmark](https://mlsanity.com/paper/2307.12510/) (2023)

## Built on by (most cited first, 6 of 6)

- [Permutation Equivariant Neural Controlled Differential Equations for Dynamic Graph Representation Learning](https://mlsanity.com/paper/2506.20324/) (2025)
- [TGM: a Modular and Efficient Library for Machine Learning on Temporal Graphs](https://mlsanity.com/paper/2510.07586/) (2025)
- [T-GRAB: A Synthetic Diagnostic Benchmark for Learning on Temporal Graphs](https://mlsanity.com/paper/2507.10183/) (2025)
- [Revisiting Node Affinity Prediction in Temporal Graphs](https://mlsanity.com/paper/2510.06940/) (2025)
- [Never Skip a Batch: Dense Learning of Temporal GNNs via Adaptive Pseudo-Supervision](https://mlsanity.com/paper/2505.12526/) (2025)
- [Unifying Graph Neural Networks Through a Common Layer Equation](https://mlsanity.com/paper/2608.16097/) (2026)

## Abstract (arXiv)

Despite the successful application of Temporal Graph Networks (TGNs) for tasks such as dynamic node classification and link prediction, they still perform poorly on the task of dynamic node affinity prediction -- where the goal is to predict 'how much' two nodes will interact in the future. In fact, simple heuristic approaches such as persistent forecasts and moving averages over ground-truth labels significantly and consistently outperform TGNs. Building on this observation, we find that computing heuristics over messages is an equally competitive approach, outperforming TGN and all current temporal graph (TG) models on dynamic node affinity prediction. In this paper, we prove that no formulation of TGN can represent persistent forecasting or moving averages over messages, and propose to enhance the expressivity of TGNs by adding source-target identification to each interaction event message. We show that this modification is required to represent persistent forecasting, moving averages, and the broader class of autoregressive models over messages. Our proposed method, TGNv2, significantly outperforms TGN and all current TG models on all Temporal Graph Benchmark (TGB) dynamic node affinity prediction datasets.