Yage Zhang

prof_pic.jpg

Me and the Shiba

૮₍ ˊᗜˋ₎ა

Hi! I’m Yage Zhang — or you can just call me Laqeey /ˈlɑː.tʃi/.

I am currently a PhD student at CISPA Helmholtz Center for Information Security, jointly supervised by Prof. Yang Zhang and Prof. Dr. Dr. h.c. Michael Backes. Before this, I obtained my Master’s degree at Saarland University.

My research interests focus on

  • Practical AI Ecosystem Security: Identifying and addressing vulnerabilities and safety failures in real-world deployed AI systems.
  • Proactive AI-Society Risk Analysis: Studying emerging risks as AI agents interact with each other and with humans.

news

Apr 06, 2026 Our work DE-CLIP: Few-Shot Anomaly Detection via Difference-Guided Embedding Editing has been accepted to ACL 2026! See you in San Diego!
Aug 01, 2025 I’m excited to share that I officially joined CISPA as a PhD student!
Looking forward to the journey ahead in security & privacy research.
Jul 08, 2025 My work Cutting the Root of Hallucination: Structural Trimming for Vulnerability Mitigation in Code LLMs has been accepted to COLM 2025! See you in Montréal!
Apr 29, 2025 I contributed to the development of a new dataset in collaboration with the Lab for Measurement Technology, Saarland University.
The dataset, titled A Machine Learning Dataset of Artificial Inner Ring Damages on Cylindrical Roller Bearings Measured Under Varying Cross-Influences, is now available on Zenodo. This work was made possible through support from the German Federal Government, funded by the EFRE-Programm (Saarland) within the framework of the VProSaar initiative.
Aug 01, 2024 I joined CISPA as a student research assistant(HiWi)!
Dec 15, 2023 I joined the Lab for Measurement Technology, Saarland University as a student research assistant(HiWi)!

service

  • Conference Reviewer: COLM 2026
  • Journal Reviewer: ACM Transactions on Privacy and Security (TOPS), Transactions on Information Forensics & Security (T-IFS)

selected publications

  1. ACL
    DE_CLIP.png
    DE-CLIP: Few-Shot Anomaly Detection via Difference-Guided Embedding Editing
    Yage Zhang, Yukun Jiang, Michael Backes, and Yang Zhang
    In Annual Meeting of the Association for Computational Linguistics (ACL), 2026
  2. shadowapi.png
    Real Money, Fake Models: Deceptive Model Claims in Shadow APIs
    Yage Zhang, Yukun Jiang, Zeyuan Chen, Michael Backes, Xinyue Shen, and Yang Zhang
    arXiv preprint, 2026
    arXiv PDF
  3. MoE.png
    Sparse Models, Sparse Safety: Unsafe Routes in Mixture-of-Experts LLMs
    Yukun Jiang, Hai Huang, Mingjie Li, Yage Zhang, Michael Backes, and Yang Zhang
    arXiv preprint, 2026
    arXiv PDF
  4. Moltbook_Logo.png
    “Humans welcome to observe”: A First Look at the Agent Social Network Moltbook
    Yukun Jiang*, Yage Zhang*, Xinyue Shen*, Michael Backes, and Yang Zhang†
    arXiv preprint, 2026
    arXiv PDF Website Dataset Coverage: WeChat TechXplore
  5. COLM
    hallu.jpg
    Cutting the Root of Hallucination: Structural Trimming for Vulnerability Mitigation in Code LLMs
    Yage Zhang
    In Conference on Language Modeling (COLM), 2025
    PDF
  6. data-10-00077-g007.png
    A Machine Learning Dataset of Artificial Inner Ring Damages on Cylindrical Roller Bearings Measured Under Varying Cross-Influences
    Schnur Christopher, Goodarzi Payman, Robin Yannick, Schneider Tizian, Schauer Julian, El Moutaouakil Houssam, Morsch Jannis, Ahmad Ali Ali, Yage Zhang, and Schütze Andreas
    2025
    Available from Lab for Measurement Technology, Saarland University
    DOI