Keywords
CLIP
poisoned dataset separation
Abstract
Backdoor defense and urban resilience intelligence share a common concern with detecting hidden vulnerabilities before they produce harmful outcomes. CLIP-guided backdoor defense separates poisoned data using entropy-based signals, while universal transferable attacks demonstrate the potential for broad model compromise through shared visual-language representations. Sequential recommendation systems also face vulnerability when user histories or learned representations are manipulated. Biomedical studies of hepatic ischemia-reperfusion and remote cardiac injury show how complex systems can produce indirect downstream harm through inflammatory mechanisms. Flood relocation modeling provides a parallel public-policy environment where hidden risk accumulates before disaster damage occurs. This literature cluster supports intelligent systems that combine anomaly detection, adversarial defense, sequential learning, biomedical risk reasoning, and disaster-oriented resilience analytics.
References
Xu, B., Dai, X., Tang, D., & Zhang, K. (2025, November). One surrogate to fool them all: Universal, transferable, and targeted adversarial attacks with clip. In Proceedings of the 2025 ACM SIGSAC Conference on Computer and Communications Security (pp. 3087-3101).
Xu, B., Yang, F., Dai, X., Tang, D., & Zhang, K. (2025, October). CLIP-Guided Backdoor Defense through Entropy-Based Poisoned Dataset Separation. In Proceedings of the 33rd ACM International Conference on Multimedia (pp. 7415-7423).
Yin, J., Zeng, Z., Li, M., Yan, H., Li, C., Han, W., ... & Wang, S. (2025, April). Unleash llms potential for sequential recommendation by coordinating dual dynamic index mechanism. In Proceedings of the ACM on Web Conference 2025 (pp. 216-227).
Zhang, Z., Duan, Y., Gu, F., Wei, J., Huo, H., Wang, Y., ... & Gao, Z. (2026). Circulating ATP from hepatic ischemia-reperfusion drives remote cardiac injury via macrophage inflammasome activation. Pharmacological Research, 108306.
Zhou, Y. (2022). Pre-disaster relocation and agent-based model for flood disaster [Doctoral dissertation, University of Wisconsin–Madison].
Wang, B., Yao, Y., Shan, S., Li, H., Viswanath, B., Zheng, H., & Zhao, B. Y. (2019). Neural cleanse: Identifying and mitigating backdoor attacks in neural networks. In 2019 IEEE Symposium on Security and Privacy (SP) (pp. 707-723). IEEE.
Carlini, N., & Wagner, D. (2017). Towards evaluating the robustness of neural networks. In 2017 IEEE Symposium on Security and Privacy (SP) (pp. 39-57). IEEE.
Radford, A., Kim, J. W., Hallacy, C., Ramesh, A., Goh, G., Agarwal, S., ... & Sutskever, I. (2021). Learning transferable visual models from natural language supervision. In International conference on machine learning (pp. 8748-8763). PMLR.
Kang, W. C., & McAuley, J. (2018, November). Self-attentive sequential recommendation. In 2018 IEEE International Conference on Data Mining (ICDM) (pp. 197-206). IEEE.
Latz, E., Xiao, T. S., & Stutz, A. (2013). Activation and regulation of the inflammasomes. Nature Reviews Immunology, 13(6), 397-411.