On the Power of Heuristics in Temporal Graphs

TL;DR

This paper shows that simple heuristics based on recency and popularity can match or outperform complex neural models in temporal graph tasks, highlighting the importance of evaluation schemes.

Contribution

It introduces metrics to quantify recency and popularity effects and demonstrates that heuristics can achieve state-of-the-art results on benchmark datasets.

Findings

Heuristics based on recency and popularity perform competitively with neural models.

Current deep learning methods often fail to capture key temporal patterns.

Refined evaluation schemes are crucial for fair comparison and progress.

Abstract

Dynamic graph datasets often exhibit strong temporal patterns, such as recency, which prioritizes recent interactions, and popularity, which favors frequently occurring nodes. We demonstrate that simple heuristics leveraging only these patterns can perform on par or outperform state-of-the-art neural network models under standard evaluation protocols. To further explore these dynamics, we introduce metrics that quantify the impact of recency and popularity across datasets. Our experiments on BenchTemp and the Temporal Graph Benchmark show that our approaches achieve state-of-the-art performance across all datasets in the latter and secure top ranks on multiple datasets in the former. These results emphasize the importance of refined evaluation schemes to enable fair comparisons and promote the development of more robust temporal graph models. Additionally, they reveal that current deep learning methods often struggle to capture the key patterns underlying predictions in real-world temporal graphs. For reproducibility, we have made our code publicly available.