AI Paper Atlas

Research map / Graph learning / GNN architectures

Certifying Graph Neural Networks Against Label and Structure Poisoning

Lukas Gosch, Xichuan Chen, Yan Scholten, Stephan Günnemann

ICML 2026 · Graph learning › GNN architectures · approach: Graph structure learning

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
add pseudo-labels (co-training and self-training) to each label partition (ablation)partition-based label-flipping certificate · Cora-ML, PubMed (label flipping)certified accuracy: clean accuracy about 40% -> around 80% (Cora-ML, k=80)Improved
add pseudo-edges via link prediction to each edge partition (ablation)partition-based structure certificate · Cora-ML, PubMed (structure perturbation, k=800)certified accuracyImproved
run co-training (label propagation) first, then self-training (ablation)ST-GPA label generation · Cora-ML, k = 80 (label flipping)certified accuracyImproved
replace ParWalks with label propagation for co-training (ablation)ST-GPA · Cora-ML, k = 80certified accuracyNo clear effect
add a second round of pseudo-edge and pseudo-label generation (ablation)ST-GPA · PubMed, LF & SP, k = 200certified accuracyImproved
use GCN, GAT, or APPNP as base classifier (ablation)ST-GPA · Cora-ML, PubMedcertified accuracyNo clear effect
increase number of partitions k with ST-GPA versus naive partitioning (ablation)partition aggregation · Cora-ML, PubMedcertified accuracyMixed
naively partition node labels with many partitionsDPA on graphs · Cora-ML (k = 281)labeled nodes per partition: average of only 4 labeled nodes with 7 classesHurt
naively partition edges with more than 50 partitionsDPA on graphs · Cora-MLsubgraph connectivityHurt

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