About me
I am a Research Engineer at Character.AI, working on post-training and safety & alignment for large language models. I received my Ph.D. in Computer Science from Purdue University in 2026, advised by Prof. Xiangyu Zhang, where my research focused on AI security. I obtained my Bachelor’s degree in CS from Shanghai Jiao Tong University in 2020, where I was advised by Prof. Haojin Zhu.
Publications
Diagnosing Live Within-Policy Instruction Conflicts in LLM Agents with Witnessed Resolution Profiles Permalink
Lu Yan, Xuan Chen, Xiangyu Zhang
Published in arXiv preprint arXiv:2605.27784, 2026
Who Tests the Testers? Systematic Enumeration and Coverage Audit of LLM Agent Tool Call Safety Permalink
Xuan Chen, Lu Yan, Ruqi Zhang, Xiangyu Zhang
Published in arXiv preprint arXiv:2603.18245, 2026
When the Specification Emerges: Benchmarking Faithfulness Loss in Long-Horizon Coding Agents Permalink
Lu Yan, Xuan Chen, Xiangyu Zhang
Published in arXiv preprint arXiv:2603.17104, 2026
VERA: Variational Inference Framework for Jailbreaking Large Language Models Permalink
Anamika Lochab*, Lu Yan*, Patrick Pynadath*, Xiangyu Zhang, Ruqi Zhang (*Equal contribution; listed in alphabetical order by first name)
Published in Advances in Neural Information Processing Systems (NeurIPS 2025), 2025
Temporal Logic-Based Multi-Vehicle Backdoor Attacks against Offline RL Agents in End-to-end Autonomous Driving Permalink
Xuan Chen, Shiwei Feng, Zikang Xiong, Shengwei An, Yunshu Mao, Lu Yan, Guanhong Tao, Wenbo Guo, Xiangyu Zhang
Published in Advances in Neural Information Processing Systems (NeurIPS 2025), 2025
Feedback-Driven Black-Box Safety Alignment Testing of Large Language Models via Reinforcement Learning Permalink
Xuan Chen, Yuzhou Nie, Lu Yan, Mengxin Zheng, Yunshu Mao, Wenbo Guo, Xiangyu Zhang
Published in Transactions on Machine Learning Research (TMLR), 2025
ASTRA: Autonomous Spatial-Temporal Red-teaming for AI Software Assistants Permalink
Xiangzhe Xu, Guangyu Shen, Zian Su, Siyuan Cheng, Hanxi Guo, Lu Yan, Xuan Chen, Jiasheng Jiang, Xiaolong Jin, Chengpeng Wang, Zhuo Zhang, Xiangyu Zhang
Published in arXiv preprint arXiv:2508.03936, 2025
System Prompt Hijacking via Permutation Triggers in LLM Supply Chains Permalink
Lu Yan, Siyuan Cheng, Xuan Chen, Kaiyuan Zhang, Guangyu Shen, Xiangyu Zhang
Published in Findings of the Association for Computational Linguistics: ACL 2025, 2025
MGC: A Compiler Framework Exploiting Compositional Blindness in Aligned LLMs for Malware Generation Permalink
Lu Yan, Zhuo Zhang, Xiangzhe Xu, Shengwei An, Guangyu Shen, Zhou Xuan, Xuan Chen, Xiangyu Zhang
Published in arXiv preprint arXiv:2507.02057, 2025
BAIT: Large Language Model Backdoor Scanning by Inverting Attack Target Permalink
Guangyu Shen, Siyuan Cheng, Zhuo Zhang, Guanhong Tao, Kaiyuan Zhang, Hanxi Guo, Lu Yan, Xiaolong Jin, Shengwei An, Shiqing Ma, Xiangyu Zhang
Published in 2025 IEEE Symposium on Security and Privacy (S&P 2025), 2025
Rethinking the Invisible Protection against Unauthorized Image Usage in Stable Diffusion Permalink
Shengwei An*, Lu Yan*, Siyuan Cheng, Guangyu Shen, Kaiyuan Zhang, Qiuling Xu, Guanhong Tao, Xiangyu Zhang
Published in Proceedings of the 33rd USENIX Security Symposium (USENIX Security 2024), 2024
Rapid Optimization for Jailbreaking LLMs via Subconscious Exploitation and Echopraxia Permalink
Guangyu Shen, Siyuan Cheng, Kaiyuan Zhang, Guanhong Tao, Shengwei An, Lu Yan, Zhuo Zhang, Shiqing Ma, Xiangyu Zhang
Published in Preprint, 2024
Parafuzz: An interpretability-driven technique for detecting poisoned samples in nlp Permalink
Lu Yan, Zhuo Zhang, Guanhong Tao, Kaiyuan Zhang, Xuan Chen, Guangyu Shen, Xiangyu Zhang
Published in Advances in Neural Information Processing Systems (NeurIPS 2023), 2023
D^3: Detoxing Deep Learning Dataset Permalink
Lu Yan, Siyuan Cheng, Guangyu Shen, Guanhong Tao, Xuan Chen, Kaiyuan Zhang, Yunshu Mao, Xiangyu Zhang
Published in NeurIPS 2023 Workshop on Backdoors in Deep Learning-The Good, the Bad, and the Ugly, 2023
CrossFix: Resolution of GitHub issues via similar bugs recommendation Permalink
Shin Hwei Tan, Ziqiang Li, Lu Yan
Published in Journal of Software: Evolution and Process, 2021
MTFuzz: fuzzing with a multi-task neural network Permalink
Dongdong She, Rahul Krishna, Lu Yan, Suman Jana, Baishakhi Ray
Published in Proceedings of the 28th ACM joint meeting on European software engineering conference and symposium on the foundations of software engineering (ESEC/FSE 2020), 2020
Dynamic traffic feature camouflaging via generative adversarial networks Permalink
Jie Li, Lu Zhou, Huaxin Li, Lu Yan, Haojin Zhu
Published in 2019 IEEE Conference on Communications and Network Security (CNS), 2019