{"type":"ScholarlyArticle","title":"Leveraging Peer-Informed Label Consistency for Robust Graph Neural Networks with Noisy Labels","authors":[{"name":"Kailai Li","slug":"kailai-li","given":"Kailai","family":"Li","affiliation":""},{"name":"Jiawei Sun","slug":"jiawei-sun","given":"Jiawei","family":"Sun","affiliation":""},{"name":"Jiong Lou","slug":"jiong-lou","given":"Jiong","family":"Lou","affiliation":""},{"name":"Zhanbo Feng","slug":"zhanbo-feng","given":"Zhanbo","family":"Feng","affiliation":""},{"name":"Hefeng Zhou","slug":"hefeng-zhou","given":"Hefeng","family":"Zhou","affiliation":""},{"name":"Chentao Wu","slug":"chentao-wu","given":"Chentao","family":"Wu","affiliation":""},{"name":"Guangtao Xue","slug":"guangtao-xue","given":"Guangtao","family":"Xue","affiliation":""},{"name":"Wei Zhao","slug":"wei-zhao","given":"Wei","family":"Zhao","affiliation":""},{"name":"Jie Li","slug":"jie-li","given":"Jie","family":"Li","affiliation":""}],"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.\r\nDespite recent progress in developing robust GNNs, few methods exploit the intrinsic properties of graph data to filter out noise.\r\nIn 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.\r\nMislabeled nodes typically exhibit lower consistency with these peers, a signal we measure using pseudo-labels derived from representational prototypes.\r\nA Gaussian Mixture Model is fitted to the consistency distribution to identify clean samples, which refine prototype quality in an iterative feedback loop.\r\nExperiments on multiple datasets demonstrate that ProCon significantly outperforms state-of-the-art methods, effectively mitigating label noise and enhancing GNN robustness.","doi":"10.24963/ijcai.2025/623","url":"https://ijcai.org/proceedings/2025/623","pdf":"https://ijcai.org/proceedings/2025/0623.pdf","markdown":"https://ijcai.org/proceedings/2025/623.md","bibtex":"https://ijcai.org/proceedings/2025/bibtex/623","volume":{"slug":"2025","title":"Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2025","url":"https://ijcai.org/proceedings/2025/","year":2025,"location":"Montreal, Canada","dates":"16-22 August 2025","editors":"James Kwok","isbn":"978-1-956792-06-5","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2025"},"section":"Machine Learning","track":"Main Track","number":623,"pages":{"first":5598,"last":5606},"keywords":["Machine Learning: ML: Sequence and graph learning","Machine Learning: ML: Weakly supervised learning","Machine Learning: ML: Robustness","Machine Learning: ML: Other"],"videos":[],"references":[],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}