Dear Authors,
Hope you are doing well.
First, thank you for open-sourcing the code for your ICLR 2026 paper, “Poisoning the Inner Prediction Logic of Graph Neural Networks for Clean-Label Backdoor Attacks”. It is an insightful and impactful contribution to the field of GNN security.
I am a master’s student at Guizhou University, currently working on research related to GNN backdoor attacks. I am particularly interested in your BA-LOGIC method and hope to build upon it for my own work.
While the provided node classification implementation has been very helpful for my initial experiments, I noticed that the code for the graph classification tasks (discussed in Section 5.3 and Appendix A.1 of your paper) is not yet included in the repository.
Would it be possible for you to share the graph classification implementation, configuration files, or any brief instructions for this setting? I intend to use these materials solely for academic research and reproduction purposes, and I will properly cite your paper in any work that builds on your method.
Thank you again for your time and for sharing your valuable work with the community.
Best regards,
Yuting Yan
Master Student, Guizhou University
Dear Authors,
Hope you are doing well.
First, thank you for open-sourcing the code for your ICLR 2026 paper, “Poisoning the Inner Prediction Logic of Graph Neural Networks for Clean-Label Backdoor Attacks”. It is an insightful and impactful contribution to the field of GNN security.
I am a master’s student at Guizhou University, currently working on research related to GNN backdoor attacks. I am particularly interested in your BA-LOGIC method and hope to build upon it for my own work.
While the provided node classification implementation has been very helpful for my initial experiments, I noticed that the code for the graph classification tasks (discussed in Section 5.3 and Appendix A.1 of your paper) is not yet included in the repository.
Would it be possible for you to share the graph classification implementation, configuration files, or any brief instructions for this setting? I intend to use these materials solely for academic research and reproduction purposes, and I will properly cite your paper in any work that builds on your method.
Thank you again for your time and for sharing your valuable work with the community.
Best regards,
Yuting Yan
Master Student, Guizhou University