| Graph backdoor Z Xi, R Pang, S Ji, T Wang 30th USENIX Security Symposium (USENIX Security 21), 1523-1540, 2021 | 181 | 2021 |
| TROJANZOO: Towards unified, holistic, and practical evaluation of neural backdoors R Pang, Z Zhang, X Gao, Z Xi, S Ji, P Cheng, X Luo, T Wang 2022 IEEE 7th European Symposium on Security and Privacy (EuroS&P), 684-702, 2022 | 62* | 2022 |
| An Embarrassingly Simple Backdoor Attack on Self-supervised Learning C Li, R Pang, Z Xi, T Du, S Ji, Y Yao, T Wang The 2023 International Conference on Computer Vision (ICCV '23), 2022 | 43* | 2022 |
| Seeing is living? rethinking the security of facial liveness verification in the deepfake era C Li, L Wang, S Ji, X Zhang, Z Xi, S Guo, T Wang 31st USENIX Security Symposium (USENIX Security 22), 2673-2690, 2022 | 37 | 2022 |
| Defending Pre-trained Language Models as Few-shot Learners against Backdoor Attacks Z Xi, T Du, C Li, R Pang, S Ji, J Chen, F Ma, T Wang Neural Information Processing Systems (NeurIPS 2023), 2023 | 26* | 2023 |
| On the security risks of {AutoML} R Pang, Z Xi, S Ji, X Luo, T Wang 31st USENIX Security Symposium (USENIX Security 22), 3953-3970, 2022 | 9 | 2022 |
| Robustifying Safety-Aligned Large Language Models through Clean Data Curation X Liu, J Liang, M Ye, Z Xi arXiv preprint arXiv:2405.19358, 2024 | 8 | 2024 |
| On the Security Risks of Knowledge Graph Reasoning Z Xi, T Du, C Li, R Pang, S Ji, X Luo, X Xiao, F Ma, T Wang Proceedings of the 32nd USENIX Security Symposium (USENIX Security 2023), 2023 | 6 | 2023 |
| The Dark Side of AutoML: Towards Architectural Backdoor Search R Pang, C Li, Z Xi, S Ji, T Wang The 11th International Conference on Learning Representations (ICLR'23), 2022 | 4 | 2022 |
| Towards a secure zero-rating framework with three parties Z Liu, Z Zhang, Y Cao, Z Xi, S Jing, H La Roche 27th USENIX Security Symposium (USENIX Security 18), 711-728, 2018 | 3 | 2018 |
| On the difficulty of defending contrastive learning against backdoor attacks C Li, R Pang, B Cao, Z Xi, J Chen, S Ji, T Wang 33rd USENIX Security Symposium (USENIX Security 24), 2901-2918, 2024 | 2 | 2024 |
| Buckle Up: Robustifying LLMs at Every Customization Stage via Data Curation X Liu, J Liang, L Tang, C You, M Ye, Z Xi arXiv preprint arXiv:2410.02220, 2024 | 1 | 2024 |
| Reasoning over Multi-view Knowledge Graphs Z Xi, R Pang, C Li, T Du, S Ji, F Ma, T Wang arXiv preprint arXiv:2209.13702, 2022 | 1 | 2022 |
| Zodiac: A Cardiologist-Level LLM Framework for Multi-Agent Diagnostics Y Zhou, P Zhang, M Song, A Zheng, Y Lu, Z Liu, Y Chen, Z Xi arXiv preprint arXiv:2410.02026, 2024 | | 2024 |
| PromptFix: Few-shot Backdoor Removal via Adversarial Prompt Tuning T Zhang, Z Xi, T Wang, P Mitra, J Chen arXiv preprint arXiv:2406.04478, 2024 | | 2024 |
| Beyond Images: Robustifying Representation Learning against Trojaning Attacks Z Xi | | 2024 |
| Neural Architectural Backdoors R Pang, C Li, Z Xi, S Ji, T Wang arXiv preprint arXiv:2210.12179, 2022 | | 2022 |
| Towards Robust Reasoning over Knowledge Graphs Z Xi, R Pang, C Li, S Ji, X Luo, X Xiao, T Wang arXiv preprint arXiv:2110.14693, 2021 | | 2021 |
| Poster: Understanding Free-riding Attacks in Internet Zero-rating Services Z Liu, Z Zhang, S Jing, Z Xi, Y Cao | | |
Shouling JiProfessor, Zhejiang University & Georgia Institute of TechnologyVerified email at gatech.edu