GNNCert: Deterministic Certification of Graph Neural Networks against Adversarial Perturbations

Zaishuo Xia, Han Yang, Binghui Wang, Jinyuan Jia
2/12/2026

No official code found. Best implementation: XiaFire/GNNCERT (10 stars, MIT, last commit 2024-04-04).

Semantic Scholar

Code Implementations(1)

XiaFire/GNNCERTCommunity55%
MIT

An official PyTorch implementation of "GNNCert: Deterministic Certification of Graph Neural Networks against Adversarial Perturbations" (ICLR 2024).

104Shell, PythonJan 17, 20242 years agoMIT

Similar Papers

Demystifying Uneven Vulnerability of Link Stealing Attacks against Graph Neural Networks

98% match2026

RE-Bench: Evaluating Frontier AI R&D Capabilities of Language Model Agents against Human Experts

93% match2026

Robust Neural Contextual Bandit against Adversarial Corruptions

91% match2026

GAN You See Me? Enhanced Data Reconstruction Attacks against Split Inference

89% match1 repo142026

Certifiably Robust Policy Learning against Adversarial Multi-Agent Communication

89% match1 repo32026

Defense against Model Extraction Attack by Bayesian Active Watermarking

88% match1 repo62026

Cite this paper

@article{xia2026gnncert,
  title  = {GNNCert: Deterministic Certification of Graph Neural Networks against Adversarial Perturbations},
  author = {Zaishuo Xia and Han Yang and Binghui Wang and Jinyuan Jia},
  year   = {2026},
  url    = {https://api.semanticscholar.org/CorpusID:687abbb274492f95b2c0fe82137c009754456d4c},
  journal = {ICLR 2024 2024}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/687abbb274492f95b2c0fe82137c009754456d4c.svg)](https://codeofpaper.com/paper/687abbb274492f95b2c0fe82137c009754456d4c)

Discussion