Xilong Wang
Scholar

Xilong Wang

Google Scholar ID: LHTTGy0AAAAJ
Ph.D. student, Duke University
Security and PrivacyAI SafetyMachine Learning
Citations & Impact
All-time
Citations
14
 
H-index
3
 
i10-index
0
 
Publications
6
 
Co-authors
5
list available
Resume (English only)
Academic Achievements
  • - One first-authored paper accepted by EMNLP 2025
  • - One first-authored paper accepted by ICLR 2025
  • - One first-authored paper accepted by WWW 2025 (Oral)
  • - One first-authored paper accepted by ICASSP 2023
  • - WAInjectBench: Benchmarking Prompt Injection Detections for Web Agents (arXiv Preprint, 2025)
  • - WebInject: Prompt Injection Attack to Web Agents (EMNLP 2025)
  • - StringLLM: Understanding the String Processing Capability of Large Language Models (ICLR 2025)
  • - Provably Robust Federated Reinforcement Learning (WWW 2025, Oral)
  • - ICStega: Image Captioning-based Semantically Controllable Linguistic Steganography (ICASSP 2023)
Research Experience
  • - Duke University, Ph.D. Student, Aug. 2024 – Present
  • - Li Auto, Research Intern, Mar. 2024 – Jul. 2024
  • - Duke University, Undergraduate Researcher, Aug. 2022 – Oct. 2023
Education
  • - Duke University, Department of Electrical and Computer Engineering, Ph.D. Student, Aug. 2024 – Present, Advisor: Prof. Neil Gong
  • - University of Science and Technology of China, B.S. Degree, Sept. 2020 – June 2024
Background
  • - Research Interests: AI Security, particularly LLM agent security and federated learning
  • - Field: Electrical and Computer Engineering
  • - Brief Introduction: Currently a second-year Ph.D. student in the Department of Electrical and Computer Engineering at Duke University, supervised by Prof. Neil Gong.
Miscellany
  • - Homepage launched in Aug., 2022
  • - Reviewer for ICLR 2026 Conference
  • - Reviewer for IEEE T-IFS Journal