# Enhanced Soups for Graph Neural Networks

arXiv: https://arxiv.org/abs/2503.11612 · HTML: https://mlsanity.com/paper/2503.11612/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 31 of its 49 references that are in the map.
- First posted on arXiv 14 Mar 2025.

## Builds on (most cited first, 30 of 31)

- [Semi-Supervised Classification with Graph Convolutional Networks](https://mlsanity.com/paper/1609.02907/) (2016)
- [Graph Attention Networks](https://mlsanity.com/paper/1710.10903/) (2017)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [Inductive Representation Learning on Large Graphs](https://mlsanity.com/paper/1706.02216/) (2017)
- [The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks](https://mlsanity.com/paper/1803.03635/) (2018)
- [Open Graph Benchmark: Datasets for Machine Learning on Graphs](https://mlsanity.com/paper/2005.00687/) (2020)
- [Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time](https://mlsanity.com/paper/2203.05482/) (2022)
- [Averaging Weights Leads to Wider Optima and Better Generalization](https://mlsanity.com/paper/1803.05407/) (2018)
- [OGB-LSC: A Large-Scale Challenge for Machine Learning on Graphs](https://mlsanity.com/paper/2103.09430/) (2021)
- [Personalized Soups: Personalized Large Language Model Alignment via Post-hoc Parameter Merging](https://mlsanity.com/paper/2310.11564/) (2023)
- [GraphCast: Learning skillful medium-range global weather forecasting](https://mlsanity.com/paper/2212.12794/) (2022)
- [Large-Scale Representation Learning on Graphs via Bootstrapping](https://mlsanity.com/paper/2102.06514/) (2021)
- [Layer-Dependent Importance Sampling for Training Deep and Large Graph Convolutional Networks](https://mlsanity.com/paper/1911.07323/) (2019)
- [Training Graph Neural Networks with 1000 Layers](https://mlsanity.com/paper/2106.07476/) (2021)
- [Dirichlet Energy Constrained Learning for Deep Graph Neural Networks](https://mlsanity.com/paper/2107.02392/) (2021)
- [Accelerating Training and Inference of Graph Neural Networks with Fast Sampling and Pipelining](https://mlsanity.com/paper/2110.08450/) (2021)
- [Learn Locally, Correct Globally: A Distributed Algorithm for Training Graph Neural Networks](https://mlsanity.com/paper/2111.08202/) (2021)
- [Seasoning Model Soups for Robustness to Adversarial and Natural Distribution Shifts](https://mlsanity.com/paper/2302.10164/) (2023)
- [Instant Soup: Cheap Pruning Ensembles in A Single Pass Can Draw Lottery Tickets from Large Models](https://mlsanity.com/paper/2306.10460/) (2023)
- [Sparse Model Soups: A Recipe for Improved Pruning via Model Averaging](https://mlsanity.com/paper/2306.16788/) (2023)
- [Layer-Neighbor Sampling -- Defusing Neighborhood Explosion in GNNs](https://mlsanity.com/paper/2210.13339/) (2022)
- [FedSoup: Improving Generalization and Personalization in Federated Learning via Selective Model Interpolation](https://mlsanity.com/paper/2307.10507/) (2023)
- [Old can be Gold: Better Gradient Flow can Make Vanilla-GCNs Great Again](https://mlsanity.com/paper/2210.08122/) (2022)
- [Descriptor and Word Soups: Overcoming the Parameter Efficiency Accuracy Tradeoff for Out-of-Distribution Few-shot Learning](https://mlsanity.com/paper/2311.13612/) (2023)
- [On Graph Neural Network Ensembles for Large-Scale Molecular Property Prediction](https://mlsanity.com/paper/2106.15529/) (2021)
- [GNN-Ensemble: Towards Random Decision Graph Neural Networks](https://mlsanity.com/paper/2303.11376/) (2023)
- [MassiveGNN: Efficient Training via Prefetching for Massively Connected Distributed Graphs](https://mlsanity.com/paper/2410.22697/) (2024)
- [Graph Ladling: Shockingly Simple Parallel GNN Training without Intermediate Communication](https://mlsanity.com/paper/2306.10466/) (2023)
- [BatchGNN: Efficient CPU-Based Distributed GNN Training on Very Large Graphs](https://mlsanity.com/paper/2306.13814/) (2023)
- [Do the Frankenstein, or how to achieve better out-of-distribution performance with manifold mixing model soup](https://mlsanity.com/paper/2309.08610/) (2023)

## Abstract (arXiv)

Graph Neural Networks (GNN) have demonstrated state-of-the-art performance in numerous scientific and high-performance computing (HPC) applications. Recent work suggests that "souping" (combining) individually trained GNNs into a single model can improve performance without increasing compute and memory costs during inference. However, existing souping algorithms are often slow and memory-intensive, which limits their scalability. We introduce Learned Souping for GNNs, a gradient-descent-based souping strategy that substantially reduces time and memory overhead compared to existing methods. Our approach is evaluated across multiple Open Graph Benchmark (OGB) datasets and GNN architectures, achieving up to 1.2% accuracy improvement and 2.1X speedup. Additionally, we propose Partition Learned Souping, a novel partition-based variant of learned souping that significantly reduces memory usage. On the ogbn-products dataset with GraphSAGE, partition learned souping achieves a 24.5X speedup and a 76% memory reduction without compromising accuracy.