About Me

I am currently a Ph.D. candidate at Pennsylvania State University, advised by Dr. Jinghui Chen, and expect to graduate in August 2026. I have joined ByteDance as a Senior Research Scientist, where I focus on red teaming for AI agents and generative AI (GenAI). I earned my Bachelor’s degree from the University of Electronic Science and Technology of China (UESTC) in 2022. I warmly welcome discussions and collaborations on trustworthy and secure AI!

📝 Publications

• You Can’t Steal Nothing: Mitigating Prompt Leakages in LLMs via System Vectors, Bochuan Cao, Changjiang Li, Yuanpu Cao, Yameng Ge, Ting Wang, Jinghui Chen. In Proceedings of the 32nd ACM Conference on Computer and Communications Security (CCS).
• TruthFlow: Truthful LLM Generation via Representation Flow Correction, Hanyu Wang, Bochuan Cao, Yuanpu Cao, Jinghui Chen. In Proceedings of the 42nd International Conference on Machine Learning (ICML), Vancouver, 2025.
• AdvI2I: Adversarial Image Attack on Image-to-Image Diffusion models, Yaopei Zeng, Yuanpu Cao, Bochuan Cao, Yurui Chang, Jinghui Chen, Lu Lin. In Proceedings of the 42nd International Conference on Machine Learning (ICML), Vancouver, 2025.
• XPrompt: Explaining Large Language Model’s Generation via Joint Prompt Attribution, Yurui Chang, Bochuan Cao, Yujia Wang, Jinghui Chen, Lu Lin. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025 main conference), Vienna, Austria.
• Monitoring Decoding: Mitigating Hallucination via Evaluating the Factuality of Partial Response during Generation, Yurui Chang, Bochuan Cao, Lu Lin. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025 Findings), Vienna, Austria.
• WordGame: Efficient & Effective LLM Jailbreak via Simultaneous Obfuscation in Query and Response, Tianrong Zhang, Bochuan Cao, Yuanpu Cao, Lu Lin, Prasenjit Mitra, Jinghui Chen. In Proceedings of the Annual Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics (NAACL), Albuquerque, New Mexico, 2025.
• Watch the Watcher! Backdoor Attacks on Security-Enhancing Diffusion Models, Changjiang Li, Ren Pang, Bochuan Cao, Jinghui Chen, Fenglong Ma, Shouling Ji, Ting Wang. In Proceedings of the 34th USENIX Security Symposium (USENIX Security), Seattle, WA, USA, 2025.
• Ditect: Lightweight Harmful Content Detector for Text-to-Image Generation, Hangfan Zhang, Bochuan Cao, Jinghui Chen, Lu Lin, Jinyuan Jia, Dinghao Wu. In Proceedings of the ICCV 2025 Workshop on Building Foundation Models You Can Trust.
• Data Free Backdoor Attacks, Bochuan Cao, Jinyuan Jia, Chuxuan Hu, Wenbo Guo, Zhen Xiang, Jinghui Chen, Bo Li, Dawn Song. In Proceedings of the 37th Conference on Neural Information Processing Systems (NeurIPS), Vancouver, 2024.
• Personalized Steering of Large Language Models: Versatile Steering Vectors Through Bi-directional Preference Optimization, Yuanpu Cao, Tianrong Zhang, Bochuan Cao, Ziyi Yin, Lu Lin, Fenglong Ma, Jinghui Chen. In Proceedings of the 37th Conference on Neural Information Processing Systems (NeurIPS), Vancouver, 2024.
• Defending Against Alignment-Breaking Attacks via Robustly Aligned LLM, Bochuan Cao, Yuanpu Cao, Lu Lin, Jinghui Chen. In Proceedings of the The 62nd Annual Meeting of the Association for Computational Linguistics (ACL 2024 main conference), Bangkok, Thailand. [codes]
• On the Safety of Open-Sourced Large Language Models: Does Alignment Really Prevent Them From Being Misused?, Hangfan Zhang, Zhimeng Guo, Huaisheng Zhu, Bochuan Cao, Lu Lin, Jinyuan Jia, Jinghui Chen, Dinghao Wu. In Proceedings of the The 62nd Annual Meeting of the Association for Computational Linguistics (ACL 2024 main conference), Bangkok, Thailand.
• Stealthy and Persistent Unalignment on Large Language Models via Backdoor Injections, Yuanpu Cao, Bochuan Cao, Jinghui Chen. In Proceedings of the 2024 Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL), Mexico City, Mexico, 2024. [codes]
• On the Difficulty of Defending Contrastive Learning against Backdoor Attacks, Changjiang Li, Ren Pang, Bochuan Cao, Zhaohan Xi, Jinghui Chen, Shouling Ji, Ting Wang. In Proceedings of the the 33rd USENIX Security Symposium (USENIX Security ‘24), PHILADELPHIA, PA, USA.
• Is Your Jailbreaking Prompt Truly Effective for Large Language Models?, Bochuan Cao, Tianrong Zhang, Yuanpu Cao, Jinyuan Jia, Lu Lin, Jinghui Chen. In Proceedings of the SECURE AND TRUSTWORTHY LARGE LANGUAGE MODELS (SET LLM @ ICLR 2024).
• On the Intrinsic Self-Correction Capability of LLMs: Uncertainty and Latent Concept, Guangliang Liu, Haitao Mao, Bochuan Cao, Zhiyu Xue, Kristen Johnson, Jiliang Tang, Rongrong Wang. arxiv preprint.
• IMPRESS: Evaluating the Resilience of Imperceptible Perturbations Against Unauthorized Data Usage in Diffusion-Based Generative AI, Bochuan Cao, Changjiang Li, Ting Wang, Jinyuan Jia, Bo Li, Jinghui Chen. In Proceedings of the 37th Conference on Neural Information Processing Systems (NeurIPS), New Orleans, Louisiana, USA, 2023. [codes]
• Wild-Time: A Benchmark of in-the-Wild Distribution Shift over Time, Huaxiu Yao, Caroline Choi, Bochuan Cao, Yoonho Lee, Pang Wei Koh, Chelsea Finn. In Proceedings of the 36th Advances in Neural Information Processing Systems (NeurIPS), New Orleans, Louisiana, USA, 2022.[codes]
• OmniLytics: A Blockchain-based Secure Data Market for Decentralized Machine Learning, Jiacheng Liang, Songze Li, Bochuan Cao, Wensi Jiang, Chaoyang He. In Proceedings of the International Workshop on Federated Learning for User Privacy and Data Confidentiality in Conjunction with ICML 2021.

🎖 Honors and Awards

• 2025.04 Pew Research Center AI Expert Consultation.
• 2024.11 NeurIPS 2024 Top Reviewers.
• 2024.10 IST student travel Award.
• 2023.10 NeurIPS 2023 Scholar Award.

📖 Educations

• 2022.08 - 2026.08 (expected), Pennsylvania State University (Penn State), College of Information Sciences and Technology, Doctor of Philosophy.
• 2018.09 - 2022.06, University of Electronic Science and Technology of China (UESTC), School of Automation Engineering, Bachelor of Engineering

💬 Acadamic Service

Program Committee/Reviewer

• Neural Information Processing Systems (NeurIPS)
• International Conference on Machine Learning (ICML)
• International Conference on Learning Representations (ICLR)
• International Conference on Artificial Intelligence and Statistics (AISTATS)
• Association for the Advancement of Artificial Intelligence (AAAI)
• IEEE International Conference on Big Data (BigData)
• Transactions on Knowledge and Data Engineering (TKDE)
• Transactions on Information Systems (TOIS)

Teaching

• SRA 365: Statistical Analysis for Information Sciences, Teaching Assistant, 2023 Fall and 2024 Spring, Penn State.

💻 Internships

• 2025.05 - 2025.08, Research Intern at Security-Flow, ByteDance, San Jose, USA. Advisor: Dr.Yang Bai and Dr.Xinyu Xing.
• 2024.05 - 2024.08, Research Intern at CR/RSI2-NA Department, Robert Bosch LLC, Sunnyvale, USA. Advisor: Dr.Sashidhar Jakkamsetti.
• 2021.03 - 2022.03, Research Intern at Search Strategy Department, Baidu Inc, Beijing, China. Advisor: Dr.Yiding Liu.