Mengyu Yao

I am a Ph.D. student in Computer Science at Peking University, advised by Prof. Yao Guo. I am a member of the Key Laboratory of High Confidence Software Technologies, Ministry of Education and expect to graduate in 2028.

Before joining PKU, I received my B.S. in Computer Science and Technology from Nanjing University in 2023.

Research [Google Scholar]

My research interests lie in security and privacy in AI systems, particularly retrieval-augmented generation, multimodal recommender systems, and AI agents.

Connect the Dots: Knowledge Graph-Guided Crawler Attack on Retrieval-Augmented Generation Systems

Mengyu Yao, Ziqi Zhang, Ning Luo, Shaofei Li, Yifeng Cai, Xiangqun Chen, Yao Guo, and Ding Li.

35th USENIX Security Symposium, 2026

From Zero to Hero: Cross-modal-enhanced Adversarial Item Promotion Attack against Multimodal Recommender Systems

Mengyu Yao, Ziqi Zhang, Yifeng Cai, Junlin Liu, Xinyi Fu, Weiqiang Wang, Xiangqun Chen, Yao Guo, and Ding Li.

35th USENIX Security Symposium, 2026

PredComp: Predicting Compiler Optimization Options with Multi-stage Learning

Bingyu Gao, Ziming Wang, Mengyu Yao, Zhihong Xue, Xiangqun Chen, Ding Li, and Yao Guo.

ACM Transactions on Architecture and Code Optimization (TACO), 2026

I Can Tell Your Secrets: Inferring Privacy Attributes from Mini-app Interaction History in Super-apps

Yifeng Cai, Ziqi Zhang, Mengyu Yao, Junlin Liu, Xiaoke Zhao, Xinyi Fu, Ruoyu Li, Zhe Li, Xiangqun Chen, Yao Guo, and Ding Li.

34th USENIX Security Symposium, 2025

Grouptuner: Efficient Group-Aware Compiler Auto-tuning

Bingyu Gao, Mengyu Yao, Ziming Wang, Dong Liu, Ding Li, Xiangqun Chen, and Yao Guo.

ACM SIGPLAN/SIGBED International Conference on Languages, Compilers, and Tools for Embedded Systems (LCTES), 2025

Not All Exceptions Are Created Equal: Triaging Error Logs in Real-World Enterprises

Junlin Liu, Mengyu Yao, Shaofei Li, Dingyu Yang, Zheshun Wu, Xiaojun Qu, Ziqi Zhang, Ding Li, Yao Guo, and Xiangqun Chen.

ACM Transactions on Software Engineering and Methodology (TOSEM), 2025

TEESlice: Protecting Sensitive Neural Network Models in Trusted Execution Environments when Attackers Have Pre-Trained Models

Ding Li, Ziqi Zhang, Mengyu Yao, Yifeng Cai, Yao Guo, and Xiangqun Chen.

ACM Transactions on Software Engineering and Methodology (TOSEM), 2025

Who Grants the Agent Power? Defending Against Instruction Injection via Task-Centric Access Control

Yifeng Cai, Ziming Wang, Zhaomeng Deng, Mengyu Yao, Junlin Liu, Yutao Hu, Ziqi Zhang, Yao Guo, and Ding Li.

SaTS 2025 (Co-located with ACM CCS 2025)

Who Moved My Transaction? Uncovering Post-Transaction Auditability Vulnerabilities in Modern Super Apps

Junlin Liu, Zhaomeng Deng, Ziming Wang, Mengyu Yao, Yifeng Cai, Yutao Hu, Ziqi Zhang, Yao Guo, and Ding Li.

SaTS 2025 (Co-located with ACM CCS 2025)

Internship Experience

Teaching Experience

Education

Honors and Awards

• Peking University Outstanding Scientific Research AwardDec 2024
• Canon ScholarshipDec 2024
• Outstanding Graduate of Nanjing UniversityJun 2023
• Special Scholarship for Basic Disciplines, Excellence AwardNov 2022
• People's ScholarshipNov 2022
• Special Scholarship for Basic Disciplines, Excellence AwardNov 2021
• Top-notch Undergraduate Program Scholarship, First PrizeNov 2020
• Outstanding Student, Nanjing UniversityNov 2020