profile photo

Yuntao Du (杜云滔)

PhD Candidate · AI Security & Data Privacy

Department of Computer Science, Purdue University

🔥 I'm on the job market this year!

Email  /  CV  /  Scholar  /  GitHub  /  LinkedIn

I am a PhD candidate in computer science at Purdue University, advised by Prof. Ninghui Li.

I study security and privacy risks in data-driven and AI systems, from discovering new vulnerabilities to auditing and controlling them. My current research focuses on three directions:

🚨 Threat Discovery
Uncovering new privacy and security risks
in data-driven and AI systems, such as privacy leakage from synthetic data, data visualization, and AI misuse.
🔍 Vulnerability Auditing
Quantifying privacy leakage in AI models
through principled membership inference attacks, such as cascading and imitative MIAs, and audits of LLM tokenizers and fine-tuning.
🛡️ AI Control
Control Security and Safety Risks in AI

My research has been recognized and supported by the Ross Fellowship (2023–2027), Presidential Doctoral Excellence Award (2023–2027), and Herbold Scholarship (2023–2024).

News

Selected Publications

All publications →
  1. Manuscript RATION: Towards Quantitative Risk-Based Access Control for AI Agents
    Yuntao Du, Harsh Chaudhari, Yuetian Chen, Ron Heichman, Ashish Jayamohan, Kumar Pallav, Mahim Patel, Kaiyuan Zhang, Kyle Polley, Jerry Ma, Ninghui Li
    Manuscript
    AI Security & SafetyDefense A new security framework for AI agents.
  2. ACL 2026 (Findings) Automated Profile Inference with Language Model Agents
    Yuntao Du, Zitao Li, Bolin Ding, Yaliang Li, Hanshen Xiao, Jingren Zhou, Ninghui Li
    Findings of the Association for Computational Linguistics
    AI Security & SafetyHacking First study showing that LLM agents enable automated doxing at web scale.
    Impact Covered by MIT Technology Review [1] [2].
  1. NDSS 2026 Cascading and Proxy Membership Inference Attacks
    Yuntao Du, Jiacheng Li, Yuetian Chen, Kaiyuan Zhang, Zhizhen Yuan, Hanshen Xiao, Bruno Ribeiro, Ninghui Li
    Network and Distributed System Security Symposium
    Machine Learning Privacy Formulates and categorizes MIAs and first exploits membership dependencies.
  2. USENIX Security 2026 Imitative Membership Inference Attack
    Yuntao Du, Yuetian Chen, Hanshen Xiao, Bruno Ribeiro, Ninghui Li
    USENIX Security Symposium
    Machine Learning Privacy A new shadow training paradigm for MIAs with significantly reduced computation.
  1. ACM CCS 2026 Privacy Leakage from a Thousand Words: Millipixel Location Recovery from Dot Maps
    Yuntao Du⭑, Tanishq Pauskar⭑ , Hao Wang, Jing Su, Ninghui Li
    ACM Conference on Computer and Communications Security
    Data PrivacyHacking Achieve ~1-meter attack accuracy in location recovery from national-scale dot maps.
    Impact This work informed QGIS's privacy guidance. Working with the U.S. CDC to revise its cartographic guidelines.
  2. ACM CCS 2025 Systematic Assessment of Tabular Data Synthesis
    Yuntao Du, Ninghui Li
    ACM Conference on Computer and Communications Security
    Data Privacy Proposes a unified evaluation framework for tabular data synthesis algorithms.

Selected Awards & Honors

Service

  • Program Committee: USENIX Security (2027), VLDB (2027), AsiaCCS (2027), NeurIPS (2026), ICLR (2025-2027), AISTATS (2025-2026), WWW (2026), CODASPY (2026), WSDM (2026), CIKM (2024-2026), SIGIR (2023-2026), IJCNN (2026-2027), TRL (2024-2025), AAAI (2023-2027), SIGIR-AP (2023-2024)
  • Poster Program Committee: IEEE S&P (2026)
  • Journal Reviewers: CSUR, TDSC, TOPS, VLDBJ, TKDE, TORS, TBD