Ziqing Yang

profiles/profile_0.jpg

Shot by Xiao Wu at Istanbul, Turkey

This is Ziqing Yang, a third-year PhD student at CISPA Helmholtz Center for Information Security, co-advised by Prof. Michael Backes and Dr. Yang Zhang. My research focuses on trustworthy machine learning, with a particular interest in exploring the security and safety risks in generative AIs and agents.

Before that, I obtained my Bachelor’s degree from Peking University. Advised by Prof. Ming Zhang, I focused on knowledge representation and graph analysis. During my undergraduate, I was also a research intern at UCLA under the supervision of Prof. Yizhou Sun, working on the combination of knowledge graphs and logical rules.

research interests

  • Trustworthy machine learning (safety, privacy, and security)
  • Generative models and agents
  • Graph analysis

news

Oct 17, 2025 I am delighted to be selected as a NeurIPS 2025 “Reviewer : Top Reviewer”!
May 01, 2025 Our paper titled Comprehensive Assessment of Jailbreak Attacks Against LLMs was accepted by ACL 2025!
Jan 01, 2025 Our paper titled Synthetic Artifact Auditing: Tracing LLM-Generated Synthetic Data Usage in Downstream Applications was accepted by USENIX Security 2025!
Sep 01, 2023 Our paper named SecurityNet: Assessing Machine Learning Vulnerabilities on Public Models was accepted by USENIX Security 2024!
May 01, 2023 Our paper named Data Poisoning Attacks Against Multimodal Encoders was accepted by ICML 2023!

selected publications

2025

  1. USENIX Security
    Synthetic Artifact Auditing: Tracing LLM-Generated Synthetic Data Usage in Downstream Applications
    Yixin Wu, Ziqing Yang, Yun Shen, Michael Backes, and Yang Zhang
    In USENIX Security Symposium (USENIX Security), 2025
    arXiv PDF Code Website
  2. ACL
    JailbreakRadar: Comprehensive Assessment of Jailbreak Attacks Against LLMs
    Junjie Chu, Yugeng Liu, Ziqing Yang, Xinyue Shen, Michael Backes, and Yang Zhang
    In Annual Meeting of the Association for Computational Linguistics (ACL), 2025
    arXiv PDF Code
  3. arxiv
    The Challenge of Identifying the Origin of Black-Box Large Language Models
    Ziqing Yang, Yixin Wu, Yun Shen, Wei Dai, Michael Backes, and Yang Zhang
    CoRR abs/2503.04332, 2025
    arXiv
  4. arxiv
    JADES: A Universal Framework for Jailbreak Assessment via Decompositional Scoring
    Junjie Chu, Mingjie Li, Ziqing Yang, Ye Leng, Chenhao Lin, Chao Shen, Michael Backes, Yun Shen, and Yang Zhang
    CoRR abs/2508.20848, 2025
    arXiv

2024

  1. arxiv
    SOS! Soft Prompt Attack Against Open-Source Large Language Models
    Ziqing Yang, Michael Backes, Yang Zhang, and Ahmed Salem
    CoRR abs/2407.03160, 2024
    arXiv

2023

  1. ICML
    Data Poisoning Attacks Against Multimodal Encoders
    Ziqing Yang, Xinlei He, Zheng Li, Michael Backes, Mathias Humbert, Pascal Berrang, and Yang Zhang
    In International Conference on Machine Learning (ICML), 2023
    arXiv PDF Code

2020

  1. EMNLP
    UniKER: A Unified Framework for Combining Embedding and Definite Horn Rule Reasoning for Knowledge Graph Inference
    Kewei Cheng, Ziqing Yang, Ming Zhang, and Yizhou Sun
    In Conference on Empirical Methods in Natural Language Processing (EMNLP), 2020
    PDF Code