Scholar
Qiusi Zhan
Google Scholar ID: XaYJrgoAAAAJ
University of Illinois Urbana-Champaign
Citations & Impact
Citations
553
H-index
7
i10-index
7
Publications
10
Co-authors
17
list available
Publications
Resume (English only)
Academic Achievements
- Published multiple papers, including 'SafeSearch: Do Not Trade Safety for Utility in LLM Search Agents' and 'Visual Backdoor Attacks on MLLM Embodied Decision Making via Contrastive Trigger Learning'; Recognized as 'Top Reviewer' at NeurIPS 2025; Reviewed for ARR, NeurIPS, and ICLR.
Research Experience
- Applied Scientist Intern at Amazon, Data Scientist Intern at Microsoft, Research Scientist Intern at JD.COM Silicon Valley Research Center, Applied Scientist Intern at ByteDance.
Education
- Ph.D. student at the University of Illinois Urbana-Champaign (UIUC), advised by Prof. Daniel Kang; Master's degree from UIUC, advised by Prof. Heng Ji; Bachelor's degree from Peking University, advised by Prof. Sujian Li.
Background
- Research focuses on developing safe (multimodal) Large Language Models (LLMs) and LLM agents for real-world deployment, with an emphasis on identifying and mitigating safety vulnerabilities. Studied a wide range of safety risks in LLMs and LLM agents, including fine-tuning vulnerabilities, indirect prompt injection attacks, multimodal RAG knowledge poisoning, and backdoor attacks. Explored reinforcement learning approaches to enhance their safety without compromising utility.
Miscellany
- This website is based on a template created by Jon Barron. Last updated: Oct 30, 2025.
