• I got my PHD from Department of Computer Science at Rutgers University on May 2025, advised by Prof. Shiqing Ma.

• I work at AI for security and security of LLM/Agent systems. I build AI-driven pipelines for auditing, storage, anomaly and threat detection, and forensics across diverse security data (e.g., logs, provenance graphs, and threat intelligence reports). My current focus spans LLM and agent security—including policy enforcement (e.g., OPA), MCP gateway controls, and automated evaluation pipelines for policy creation, adversarial test case generation, and purple-team continuous assessment. I also research model security, covering backdoor and interpretability to diagnose and mitigate unsafe model behaviors.

• I am currently a research scientist at IBM Research. Previously, I was an research intern at IBM Research in Summer 2024, and a machine learning and AI research intern at Bell Labs in Summer 2023. Before my PHD at Rutgers, I earned my B.E. from Central South University in 2020.

News

• [2025-02] I joined IBM Research as a research scientist.
• [2024-08] One paper was accepted by ACSAC 2024.
• [2024-05] I joined IBM Research as a research intern.
• [2024-04] One paper was accepted by ASPLOS 2024.
• [2023-06] I joined Bell Lab as a Machine Learning and AI intern.
• [2023-05] Two papers were accepted to Usenix Security 2023.
• [2022-09] Two papers were accepted to NeurIPS 2022.
• [2022-01] I am a member of Usenix Security 2022 Artifact Evaluation Committee.
• [2021-09] One paper was accepted to Usenix Security 2021.

Publication

• Madeline: Continuous and Low-cost Monitoring with Graph-free Representations to Combat Cyber Threats
  Wenjia Song, Hailun Ding, Na Meng, Peng Gao, Danfeng (Daphne) Yao (ACSAC 2024)

• Merlin: Multi-tier Optimization of eBPF Code for Performance and Compactness
  Jinsong Mao, Hailun Ding, Juan Zhai, Shiqing Ma
  Architectural Support for Programming Languages and Operating Systems 2024 (ASPLOS 2024)

• AirTag: Towards Automated Attack Investigation by Unsupervised Learning with Log Texts
  Hailun Ding, Juan Zhai, Yuhong Nan, Shiqing Ma
  USENIX Security Symposium 2023 (Usenix security 2023)

• The Case for Learned Provenance Graph Storage Systems
  Hailun Ding, Juan Zhai, Dong Deng, Shiqing Ma
  USENIX Security Symposium 2023 (Usenix security 2023)

• Rethinking the Reverse-engineering of Trojan Triggers
  Zhenting Wang, Kai Mei, Hailun Ding, Juan Zhai, Shiqing Ma
  Proceedings of Neural Information Processing Systems 2022 (NeurIPS 2022)

• Training with More Confidence: Mitigating Injected and Natural Backdoors During Training
  Zhenting Wang, Hailun Ding, Juan Zhai, Shiqing Ma
  Proceedings of Neural Information Processing Systems 2022 (NeurIPS 2022)

• ELISE: A Storage Efficient Logging System Powered by Redundancy Reduction and Representation Learning
  Hailun Ding, Shenao Yan, Juan Zhai, Shiqing Ma
  USENIX Security Symposium 2021 (Usenix security 2021)

• Procedural Learning With Robust Visual Features via Low Rank Prior
  Haifeng Li, Li Chen, Hailun Ding, Qi Li, Bingyu Sun, Guohua Wu
  IEEE Access 2019

• A Data-driven Adversarial Examples Recognition Framework via Adversarial Feature Genome
  Li Chen, Hailun Ding, Qi Li, Jiawei Zhu, Jian Peng, Haifeng Li
  Arxiv 2019

• Understanding the Importance of Single Directions via Representative Substitution
  Li Chen, Hailun Ding, Qi Li, Zhuo Li, Jian Peng, Haifeng Li
  AAAI 2019 Workshop on Network Interpretability for Deep Learning**

Service

• Program Committee, AAAI 2025, 2026
• Reviewer, TIFS
• Reviewer, CODASPY 2024
• Reviewer, ICLR 2024, 2025
• Reviewer, NeurIPS Workshop on Backdoors in Deep Learning: The Good, the Bad, and the Ugly (BUGS) 2023
• Reviewer, NeurIPS 2023, 2025
• Reviewer, International Conference on Machine Learning (ICML), 2022, 2024
• Artifact Evaluation Committee, USENIX Security Symposium, 2022 (Fall and Winter cycles)

Award

• National Scholarship (2017, 2019, 0.2% of Chinese undergraduate students)
• First Level Scholarship of Central South University (2017, 2018, 2019, Top 3%)
• Diamond Scholarship (2017, 15 person in 30,000 undergraduate students)
• First Prize in The Hack, Hack Shanghai (2018)