About me
Hi, I’m Liya Su (苏莉娅) — currently working as an AI Security R&D Expert at Chaitin Technology, where I explore how LLMs and AI can be applied to cybersecurity (and how to keep them safe in the process).
Before this, I was a Security Algorithm Engineer at JD Cloud. I got my Ph.D. in Cyberspace Security from the Institute of Information Engineering, Chinese Academy of Sciences in 2021. My work sits at the intersection of AI and security — think malicious behavior analysis, graph neural networks, and figuring out creative ways attackers break things.
What’s New
- 2025: 1st Place in Tencent Cloud AI Agent CTF Competition (China’s first AI autonomous penetration challenge).
- 2025: 1st Place in Beijing CAC AI Content Security Protection Competition.
- 2025: Joined Chaitin Technology as AI Security R&D Expert, Oct 2025.
- 2024: Won 1st place in National Cybersecurity AI Capability Competition (guided by CAC).
- 2023: LLM security attack research reported on the front page of The New York Times.
Research Interests
- AI/LLM Security
- Network Data Security & Threat Intelligence
- Web Security & Blockchain Security
- Graph Neural Networks for Security
Experience
- AI Security R&D Expert, Chaitin Technology, Oct 2025 - Present
- Security Algorithm Engineer, JD Cloud Security Product R&D, Mar 2022 - Oct 2025
- DMT (Doctor Management Trainee), JD.com Inc., Jul 2021 - Mar 2022
- Visiting Researcher, CISPA Helmholtz Center for Information Security (Germany), Jan 2020 - Feb 2021, advised by Prof. Yang Zhang
- Visiting Student, Indiana University Bloomington (USA), May 2019 - Jun 2019, advised by Prof. XiaoFeng Wang
Selected Publications
- Liya Su, Xinyue Shen, Xiangyu Du, Xiaojing Liao, XiaoFeng Wang, Luyi Xing, Baoxu Liu. Evil Under the Sun: Understanding and Discovering Attacks on Ethereum Decentralized Applications. USENIX Security 2021. (CCF-A) paper
- Xiaoyi Chen, Siyuan Tang, Rui Zhu, Shijun Yan, Lei Jin, Zihao Wang, Liya Su, XiaoFeng Wang, Haixu Tang. The Janus Interface: How Fine-Tuning in Large Language Models Amplifies the Privacy Risks. arXiv 2023. paper
- Liya Su, Yepeng Yao, Chen Zhang, Zhigang Lu, Baoxu Liu. Marrying Graph Kernel with Deep Neural Network: A Case Study for Network Anomaly Detection. ICCS 2019: 102-115.
- Liya Su, Yepeng Yao, Zhigang Lu, Baoxu Liu. Understanding the Influence of Graph Kernels on Deep Learning Architecture. TrustCom 2019: 312-318. (CCF-C)
- Liya Su, Yepeng Yao, Ning Li, Junrong Liu, Zhigang Lu, Baoxu Liu. Hierarchical Clustering Based Network Traffic Data Reduction for Improving Suspicious Flow Detection. TrustCom 2018: 744-753. (CCF-C)
Full list on Google Scholar.
Awards & Honors
- 2025: 1st Place, Tencent Cloud AI Agent CTF Competition (China’s first AI autonomous penetration challenge)
- 2025: 1st Place, Beijing CAC AI Content Security Protection Competition
- 2024: 1st Place, National Cybersecurity AI Capability Competition (guided by CAC)
- 2023: LLM security attack research reported on NYT front page
- 2023: 3rd Prize, JD Tech AI Innovation Application Contest
- 2022: National HW Sandbox Defense Champion (first-ever defender win)
- 2022: Yizhuang Yiqilin Outstanding Talent; JD Top 100 New Recruits
- 2021: National Scholarship
Education
| Year | Institute | Major | Degree |
|---|---|---|---|
| 2016.09 - 2021.06 | Institute of Information Engineering, CAS | Cyberspace Security | Ph.D. |
| 2012.09 - 2016.06 | University of Science and Technology of China | Information Security | B.S. |
Hi，我是苏莉娅 — 目前在长亭科技担任AI安全研发专家，主要探索大模型和AI技术在网络安全领域怎么用（以及怎么不被人搞）。
来长亭之前，我在京东云安全做安全算法工程师。2021年在中科院信工所拿了网络空间安全的博士学位。日常研究方向是AI与安全的交叉地带 — 恶意行为分析、图神经网络、以及研究攻击者那些花式操作。
最新动态
- 2025年：腾讯云AI智能体CTF夺旗比赛（国内首个AI智能渗透挑战赛）第一名。
- 2025年：北京网信办AI内容安全防护比赛第一名。
- 2025年10月：加入长亭科技，担任AI安全研发专家。
- 2024年：中央网信办指导的网络安全AI能力邀请赛全国第一名。
- 2023年：大模型安全攻击研究被纽约时报首页报道。
研究方向
- AI/大模型安全
- 网络数据安全与威胁情报
- Web安全与区块链安全
- 图神经网络在安全领域的应用
工作经历
- AI安全研发专家，长亭科技，2025年10月 - 至今
- 安全算法工程师，京东科技 京东云安全产品研发部，2022年3月 - 2025年10月
- 博士管培生（DMT），京东集团组织部，2021年7月 - 2022年3月
- 访问学者，德国亥姆霍兹信息安全中心 CISPA，2020年1月 - 2021年2月，合作导师：张阳教授
- 访问学者，美国印第安纳大学伯明顿分校，2019年5月 - 2019年6月，合作导师：王晓峰教授
代表论文
- Liya Su, Xinyue Shen, Xiangyu Du, Xiaojing Liao, XiaoFeng Wang, Luyi Xing, Baoxu Liu. Evil Under the Sun: Understanding and Discovering Attacks on Ethereum Decentralized Applications. USENIX Security 2021. (CCF-A) 论文
- Xiaoyi Chen, Siyuan Tang, Rui Zhu, Shijun Yan, Lei Jin, Zihao Wang, Liya Su, XiaoFeng Wang, Haixu Tang. The Janus Interface: How Fine-Tuning in Large Language Models Amplifies the Privacy Risks. arXiv 2023. 论文
- Liya Su, Yepeng Yao, Chen Zhang, Zhigang Lu, Baoxu Liu. Marrying Graph Kernel with Deep Neural Network: A Case Study for Network Anomaly Detection. ICCS 2019: 102-115.
- Liya Su, Yepeng Yao, Zhigang Lu, Baoxu Liu. Understanding the Influence of Graph Kernels on Deep Learning Architecture. TrustCom 2019: 312-318. (CCF-C)
- Liya Su, Yepeng Yao, Ning Li, Junrong Liu, Zhigang Lu, Baoxu Liu. Hierarchical Clustering Based Network Traffic Data Reduction for Improving Suspicious Flow Detection. TrustCom 2018: 744-753. (CCF-C)
完整列表见 Google Scholar。
荣誉奖项
- 2025年：腾讯云AI智能体CTF夺旗比赛（国内首个AI智能渗透挑战赛）第一名
- 2025年：北京网信办AI内容安全防护比赛第一名
- 2024年：中央网信办指导的网络安全AI能力邀请赛全国第一名
- 2023年：大模型安全攻击研究被纽约时报首页报道
- 2023年：京东科技AI应用创新大赛三等奖
- 2022年：国家HW沙盘推演获得全场唯一防守方获胜奖
- 2022年：北京亦庄亦麒麟优秀人才称号、京东集团校招生百大新人称号
- 2021年：国家奖学金
教育背景
| 年份 | 院校 | 专业 | 学位 |
|---|---|---|---|
| 2016.09 - 2021.06 | 中国科学院信息工程研究所 | 网络空间安全 | 博士 |
| 2012.09 - 2016.06 | 中国科学技术大学 | 信息安全 | 学士 |