Xilong Wang

Ph.D. Student at Duke University

xilong.wang@duke.edu

Bio

I'm currently a third-year PhD student in the Department of Electrical and Computer Engineering at Duke University, advised by Prof. Neil Gong. I received my Bachelor's degree from the University of Science and Technology of China in 2024.

My research interests lie in LLM agents and AI security, with a particular focus on the following area:

I am looking for research internship opportunities for Summer 2027 — feel free to reach out at xilong.wang@duke.edu.

News

• Jun., 2026    [PAPER] New first-authored paper on securing web agents via security policies!
• May., 2026    [PAPER] New first-authored paper on prompt injection detection for RAG knowledge databases!
• Feb., 2026    [PAPER] New first-authored paper on prompt injection detection and localization for web agents!
• Oct., 2025    [PAPER] New first-authored paper on benchmarking prompt injection detection for web agents!
• Aug., 2025    [PAPER] One first-authored paper has been accepted to EMNLP 2025 main!
• Jan., 2025    [PAPER] One first-authored paper has been accepted to ICLR 2025!
• Jan., 2025    [PAPER] One first-authored paper has been accepted to WWW 2025 (Oral)!
• Aug., 2024    [STATUS] I graduated from USTC and started my PhD journey at Duke!
• Feb., 2023    [PAPER] One first-authored paper has been accepted to ICASSP 2023!
• Aug., 2022    [STATUS] My homepage is on!

Publications

Most recent publications on Google Scholar.
* indicates equal contribution.

Same-Origin Policy for Agentic Browsers

Xilong Wang, Xiaoxing Chen, Patrick Li, Dawn Song, Neil Gong

arXiv Preprint, 2026

Paper Code

CleanBase: Detecting Malicious Documents in RAG Knowledge Databases

Weifei Jin*, Xilong Wang*, Wei Zou, Jinyuan Jia, Neil Gong

arXiv Preprint, 2026

Paper Code

AlignSentinel: Alignment-Aware Detection of Prompt Injection Attacks

Yuqi Jia, Ruiqi Wang, Xilong Wang, Chong Xiang, Neil Gong

arXiv Preprint, 2026

Paper

WebSentinel: Detecting and Localizing Prompt Injection Attacks for Web Agents

Xilong Wang, Yinuo Liu, Zhun Wang, Dawn Song, Neil Gong

arXiv Preprint, 2026

Paper Code

WAInjectBench: Benchmarking Prompt Injection Detections for Web Agents

Yinuo Liu*, Xilong Wang*, Ruohan Xu, Yuqi Jia, Neil Zhenqiang Gong

ACM CCS Workshop on Security, Privacy, and Safety of Agentic AI Systems (Agent-Sec), 2026

Paper Code

WebInject: Prompt Injection Attack to Web Agents

Xilong Wang, John Bloch , Zedian Shao, Yuepeng Hu, Shuyan Zhou, Neil Gong

The 2025 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2025

Paper

StringLLM: Understanding the String Processing Capability of Large Language Models

Xilong Wang, Hao Fu, Jindong Wang, Neil Gong

International Conference on Learning Representations (ICLR), 2025

Paper Code

Provably Robust Federated Reinforcement Learning

Minghong Fang*, Xilong Wang*, Neil Gong

Proceedings of The Web Conference (WWW), 2025, Oral (Top 7.76%)

Paper

ICStega: Image Captioning-based Semantically Controllable Linguistic Steganography

Xilong Wang, Yaofei Wang, Kejiang Chen, Jinyang Ding, Weiming Zhang, Nenghai Yu

IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2023

Paper

Academic Service

• Conference: ICLR 2026, ICML 2026, NeurIPS 2026

• Jornal: IEEE T-IFS

Teaching

Teaching Assistant, Duke University Spring 2026
Generative AI: Foundations, Applications, and Safety, ECE590

Teaching Assistant, USTC Fall 2023
Design and Practice of Information Security II

Vitæ

Full Resume in PDF.

Xilong Wang
xilong.wang@duke.edu

This website is built with Jekyll using a template by Martin Saveski.