JI Zhenlan

基本情報

所属

奈良先端科学技術大学院大学 先端科学技術研究科 情報科学領域 助教

学位

博士(2026年1月 香港科技大学)

研究者番号

41045083

ORCID iD

https://orcid.org/0000-0003-3167-0480

J-GLOBAL ID

202501014407434076

researchmap会員ID

R000090518

外部リンク

https://zhenlanji.github.io/

経歴

• 2026年4月 - 現在

  奈良先端科学技術大学院大学, 先端科学技術研究科 情報科学領域, 助教

学歴

• 2021年9月 - 2026年1月

  香港科技大学, コンピュータ理工学部, コンピュータ理工学
• 2017年9月 - 2021年6月

  南京大学

論文

• Beyond Content Safety: Real-Time Monitoring for Reasoning Vulnerabilities in Large Language Models.

  Xunguang Wang, Yuguang Zhou, Qingyue Wang, Zongjie Li, Ruixuan Huang, Zhenlan Ji, Pingchuan Ma 0004, Shuai Wang 0011

  CoRR abs/2603.25412 2026年3月
• CAM: A Causality-based Analysis Framework for Multi-Agent Code Generation Systems.

  Zongyi Lyu, Zhenlan Ji, Songqiang Chen, Liwen Wang, Yuheng Huang 0004, Shuai Wang 0011, Shing-Chi Cheung

  CoRR abs/2602.02138 2026年2月
• Reeq : Testing and Mitigating Ethically Inconsistent Suggestions of Large Language Models with Reflective Equilibrium

  Pingchuan Ma, Zhaoyu Wang, Zongjie Li, Zhenlan Ji, Ao Sun, Juergen Rahmel, Shuai Wang

  ACM Transactions on Software Engineering and Methodology 35(1) 7-27 2026年1月31日
• Privacy-preserving and Verifiable Causal Prescriptive Analytics

  Zhaoyu Wang, Pingchuan Ma, Zhantong Xue, Yanbo Dai, Zhenlan Ji, Shuai Wang

  Proceedings of the ACM on Management of Data 3(6) 1-27 2025年12月4日
• SEAL: Subspace-Anchored Watermarks for LLM Ownership.

  Yanbo Dai, Zongjie Li, Zhenlan Ji, Shuai Wang 0011

  CoRR abs/2511.11356 2025年11月
• Testing and Enhancing Multi-Agent Systems for Robust Code Generation.

  Zongyi Lyu, Songqiang Chen, Zhenlan Ji, Liwen Wang, Shuai Wang 0011, Daoyuan Wu, Wenxuan Wang 0001, Shing-Chi Cheung

  CoRR abs/2510.10460 2025年10月
• G uardrail : Automated Integrity Constraint Synthesis From Noisy Data

  Pingchuan Ma, Zhaoyu Wang, Zhenlan Ji, Zongjie Li, Ao Sun, Shuai Wang

  Proceedings of the ACM on Management of Data 3(4) 248-26 2025年9月22日
• Digging Into the Internal: Causality-Based Analysis of LLM Function Calling.

  Zhenlan Ji, Daoyuan Wu, Wenxuan Wang 0001, Pingchuan Ma 0004, Shuai Wang 0011, Lei Ma 0003

  CoRR abs/2509.16268 2025年9月
• Disabling Self-Correction in Retrieval-Augmented Generation via Stealthy Retriever Poisoning.

  Yanbo Dai, Zhenlan Ji, Zongjie Li, Kuan Li, Shuai Wang 0011

  CoRR abs/2508.20083 2025年8月
• Evaluating LLMs on Sequential API Call Through Automated Test Generation.

  Yuheng Huang 0004, Da Song, Zhenlan Ji, Shuai Wang 0011, Lei Ma 0003

  CoRR abs/2507.09481 2025年7月
• Causality-Aided Evaluation and Explanation of Large Language Model-Based Code Generation

  Zhenlan Ji, Pingchuan Ma, Zongjie Li, Zhaoyu Wang, Shuai Wang

  Proceedings of the ACM on Software Engineering 2(ISSTA) 1374-1397 2025年6月22日
• SoK: Evaluating Jailbreak Guardrails for Large Language Models.

  Xunguang Wang, Zhenlan Ji, Wenxuan Wang 0001, Zongjie Li, Daoyuan Wu, Shuai Wang 0011

  CoRR abs/2506.10597 2025年6月
• IP Leakage Attacks Targeting LLM-Based Multi-Agent Systems.

  Liwen Wang, Wenxuan Wang 0001, Shuai Wang 0011, Zongjie Li, Zhenlan Ji, Zongyi Lyu, Daoyuan Wu, Shing-Chi Cheung

  CoRR abs/2505.12442 2025年5月
• NAMET: Robust Massive Model Editing via Noise-Aware Memory Optimization.

  Yanbo Dai, Zhenlan Ji, Zongjie Li, Shuai Wang 0011

  CoRR abs/2505.11876 2025年5月
• STShield: Single-Token Sentinel for Real-Time Jailbreak Detection in Large Language Models.

  Xunguang Wang, Wenxuan Wang 0001, Zhenlan Ji, Zongjie Li, Pingchuan Ma 0004, Daoyuan Wu, Shuai Wang 0011

  CoRR abs/2503.17932 2025年3月
• SelfDefend: LLMs Can Defend Themselves against Jailbreaking in a Practical Manner.

  Xunguang Wang, Daoyuan Wu, Zhenlan Ji, Zongjie Li, Pingchuan Ma 0004, Shuai Wang 0011, Yingjiu Li, Yang Liu 0003, Ning Liu, Juergen Rahmel

  USENIX Security Symposium 2441-2460 2025年
• Testing and Understanding Deviation Behaviors in FHE-Hardened Machine Learning Models.

  Yiteng Peng, Daoyuan Wu, Zhibo Liu, Dongwei Xiao, Zhenlan Ji, Juergen Rahmel, Shuai Wang 0011

  47th IEEE/ACM International Conference on Software Engineering(ICSE) 2251-2263 2025年
• The Phantom Menace in Crypto-Based PET-Hardened Deep Learning Models: Invisible Configuration-Induced Attacks.

  Yiteng Peng, Dongwei Xiao, Zhibo Liu, Zhenlan Ji, Daoyuan Wu, Shuai Wang 0011, Juergen Rahmel

  CCS 4379-4393 2025年
• SelfDefend: LLMs Can Defend Themselves against Jailbreaking in a Practical Manner.

  Xunguang Wang, Daoyuan Wu, Zhenlan Ji, Zongjie Li, Pingchuan Ma 0004, Shuai Wang 0011, Yingjiu Li, Yang Liu 0003, Ning Liu, Juergen Rahmel

  CoRR abs/2406.05498 2024年
• Testing and Understanding Erroneous Planning in LLM Agents through Synthesized User Inputs.

  Zhenlan Ji, Daoyuan Wu, Pingchuan Ma 0004, Zongjie Li, Shuai Wang 0011

  CoRR abs/2404.17833 2024年