AI Paper Atlas

Research map / Graph learning / GNN architectures

ProGMLP: A Progressive Framework for GNN-to-MLP Knowledge Distillation with Efficient Trade-offs

Weigang Lu, Ziyu Guan, Wei Zhao, Yaming Yang, Yujie Sun, Zheng Liang, Yibing Zhan, Dapeng Tao

AAAI 2026 · Graph learning › GNN architectures · approach: GNN generalization theory · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
distill GCN teacher into ProGMLP instead of the teacher aloneGCN teacher · Cora/Pubmed/CS/Physics/Photo/Computers (transductive)accuracy improvement: +1.50%Improved
use E-GLNN ensemble students with GCN teacherGCN teacher · six transductive datasetsaccuracy improvement: -1.67% (E-GLNN2), -1.36% (E-GLNN4)Hurt
use AdaGMLP ensemble students with GCN teacherGCN teacher · six transductive datasetsaccuracy improvement: -1.23% (AdaGMLP2), -0.56% (AdaGMLP4)Hurt
distill into ProGMLP on ogbn-arxiv (GCN teacher)GCN teacher · ogbn-arxivaccuracy and inference time: 68.42 -> 70.11 accuracy; 84.13 -> 3.95 msImproved
distill into ProGMLP on ogbn-products (GCN teacher)GCN teacher · ogbn-productsaccuracy: 72.94 -> 73.00No clear effect
distill into ProGMLP in the inductive setting (GCN teacher)GCN teacher · Pubmed (inductive)accuracy: 80.06 -> 87.95 (+9.85%)Improved
add Progressive Training Structure (initialize each student from the previous one) (ablation)ProGMLP · Coauthor CSaccuracy: about 4% drop on CS when removedImproved
add Progressive Knowledge Distillation (ablation)ProGMLP · CS, Physics, Photo, ComputersaccuracyImproved
add Progressive Mixup Augmentation (ablation)ProGMLP · CS, Physics, Photo, ComputersaccuracyImproved
vary beta, gamma and early-exit threshold tau_conf (ablation)ProGMLP · hyperparameter sensitivityaccuracyNo clear effect
vary alpha (balance of distillation vs ground-truth loss) (ablation)ProGMLP · hyperparameter sensitivityaccuracyMixed
use ProGMLP with GraphSAGE teacher on Physics versus AdaGMLP4GraphSAGE teacher · Coauthor Physicsaccuracy: 94.20 -> 93.70Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy