# Early-Exit Graph Neural Networks for Link Prediction

arXiv: https://arxiv.org/abs/2606.22167 · HTML: https://mlsanity.com/paper/2606.22167/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 11 references that are in the map.
- First posted on arXiv 20 Jun 2026.

## Builds on (most cited first, 10 of 10)

- [Semi-Supervised Classification with Graph Convolutional Networks](https://mlsanity.com/paper/1609.02907/) (2016)
- [Inductive Representation Learning on Large Graphs](https://mlsanity.com/paper/1706.02216/) (2017)
- [Revisiting Graph Neural Networks: All We Have is Low-Pass Filters](https://mlsanity.com/paper/1905.09550/) (2019)
- [A Survey on Oversmoothing in Graph Neural Networks](https://mlsanity.com/paper/2303.10993/) (2023)
- [Graph-Coupled Oscillator Networks](https://mlsanity.com/paper/2202.02296/) (2022)
- [Why should we add early exits to neural networks?](https://mlsanity.com/paper/2004.12814/) (2020)
- [Evaluating Graph Neural Networks for Link Prediction: Current Pitfalls and New Benchmarking](https://mlsanity.com/paper/2306.10453/) (2023)
- [Adaptive Propagation Graph Convolutional Network](https://mlsanity.com/paper/2002.10306/) (2020)
- [SSFG: Stochastically Scaling Features and Gradients for Regularizing Graph Convolutional Networks](https://mlsanity.com/paper/2102.10338/) (2021)
- [Early-Exit Graph Neural Networks](https://mlsanity.com/paper/2505.18088/) (2025)

## Abstract (arXiv)

Graph Neural Networks are great for link prediction in various network-like structures; however, the question of their speed/quality tradeoff has been barely studied. While in practice the time it takes to do inference matters little for small benchmarks, the latency does limit applicability in large-scale domains. In this work, we explore early-exiting strategies that can be applied to Graph Neural Networks to solve the problem of link-prediction faster. We use no auxiliary losses to enforce early exiting, allowing it to emerge as an implicit property of the architecture. We show that our method enables early exiting in several setups, moving the Pareto frontier on the HeaRT benchmark for GCN and SAS-GNN backbones. Our findings show that inference speed of GNNs on many link-prediction problems can be improved, while losing little, or even winning in terms of prediction quality. The code is available in our repository: https://github.com/knyazer/link_prediction.