Is Your Explanation Reliable: Confidence-Aware Explanation on Graph Neural Networks
Published in KDD 2025, 2025
This paper studies reliability of post-hoc GNN explanations and introduces confidence-aware calibration to better handle unknown or out-of-distribution cases.
Recommended citation: Jiaxing Zhang, Xiaoou Liu, Dongsheng Luo, Hua Wei. 2025. Is Your Explanation Reliable: Confidence-Aware Explanation on Graph Neural Networks. KDD 2025.
Recommended citation: Jiaxing Zhang, Xiaoou Liu, Dongsheng Luo, Hua Wei. 2025. Is Your Explanation Reliable: Confidence-Aware Explanation on Graph Neural Networks. Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2025). https://arxiv.org/abs/2506.00437