He, Y., Tian, J., Zheng, X., Dong, L., Li, Y., & Zhou, J. (2026). Deferred Poisoning: Making the Model More Vulnerable via Hessian Singularization. Proceedings of the AAAI Conference on Artificial Intelligence, 40(26), 21681–21689. https://doi.org/10.1609/aaai.v40i26.39318