Leveraging Peer-Informed Label Consistency for Robust Graph Neural Networks with Noisy Labels
Abstract
Graph Neural Networks (GNNs) excel in many applications but struggle when trained with noisy labels, especially as noise can propagate through the graph structure. Despite recent progress in developing robust GNNs, few methods exploit the intrinsic properties of graph data to filter out noise. In this paper, we introduce ProCon, a novel framework that identifies mislabeled nodes by measuring label consistency among semantically similar peers, which are determined by feature similarity and graph adjacency. Mislabeled nodes typically exhibit lower consistency with these peers, a signal we measure using pseudo-labels derived from representational prototypes. A Gaussian Mixture Model is fitted to the consistency distribution to identify clean samples, which refine prototype quality in an iterative feedback loop. Experiments on multiple datasets demonstrate that ProCon significantly outperforms state-of-the-art methods, effectively mitigating label noise and enhancing GNN robustness.
Cite
Text
Li et al. "Leveraging Peer-Informed Label Consistency for Robust Graph Neural Networks with Noisy Labels." International Joint Conference on Artificial Intelligence, 2025. doi:10.24963/IJCAI.2025/623Markdown
[Li et al. "Leveraging Peer-Informed Label Consistency for Robust Graph Neural Networks with Noisy Labels." International Joint Conference on Artificial Intelligence, 2025.](https://mlanthology.org/ijcai/2025/li2025ijcai-leveraging/) doi:10.24963/IJCAI.2025/623BibTeX
@inproceedings{li2025ijcai-leveraging,
title = {{Leveraging Peer-Informed Label Consistency for Robust Graph Neural Networks with Noisy Labels}},
author = {Li, Kailai and Sun, Jiawei and Lou, Jiong and Feng, Zhanbo and Zhou, Hefeng and Wu, Chentao and Xue, Guangtao and Zhao, Wei and Li, Jie},
booktitle = {International Joint Conference on Artificial Intelligence},
year = {2025},
pages = {5598-5606},
doi = {10.24963/IJCAI.2025/623},
url = {https://mlanthology.org/ijcai/2025/li2025ijcai-leveraging/}
}