Keyan Guo

Keyan Guo

I am a PhD candidate in the Department of Computer Science and Engineering at University at Buffalo, supervised by Dr. Hongxin Hu. My background is in AI and cybersecurity, and I am strongly interested in developing AI-driven solutions to address real-world security challenges and enhance AI safety and robustness.

Keyan Guo

I am a final-year Ph.D. candidate in Computer Science and Engineering at the University at Buffalo, SUNY, advised by Dr. Hongxin Hu. My research advances CyberAI, the intersection of AI and cybersecurity, with the goal of building generative AI that protects and can be protected: securing LLMs, AI agents, and AI-generated code against attacks, and harnessing generative AI to protect people, especially children, from online harms.

I am on the 2026–2027 academic job market.

I welcome research collaborations across cybersecurity, AI/ML, software engineering, HCI, and the social sciences.

CV · Publications · Google Scholar · Email

Research Interests

  • AI security and safety: defending LLMs, multimodal models, and AI agents; securing AI-generated code
  • AI for cybersecurity and online safety: detecting and moderating online hate, harmful memes, and unsafe content
  • Human-centered security: understanding how people, especially children, experience online risks

Selected Highlights

Publications

  • NDSS 2027: defending LLM agents against indirect prompt injection
  • ACM CCS 2026: safeguarding multimodal LLMs
  • CHI 2026: rethinking children’s online safety beyond age-based restrictions
  • EMNLP 2026 (Findings): joint security–functionality benchmarking of AI-generated code
  • USENIX Security 2025: defending LLMs against jailbreak attacks
  • NDSS 2025: understanding and detecting harmful memes
  • USENIX Security 2024: moderating illicit promotion of unsafe user-generated content games
  • IEEE S&P 2024: moderating new waves of online hate with chain-of-thought reasoning

Awards

  • NDSS Internet Society Fellowship (2025)
  • UB CSE Best Ph.D. Research Project Award (2024)
  • UB Student Academic Excellence Showcase (2024)
  • UB CSE Best AI Poster Award (2023)
  • UB CSE Best Graduate Teaching Award (2022)

News

  • 10/2026 I am on the 2026–2027 academic job market. Please feel free to reach out!
  • 09/2026 Our paper on how to practically protect AI agents against indirect prompt injection has been accepted by NDSS 2027!
  • 08/2026 Our paper on a joint security-functionality evaluation benchmark for LLMs and coding agents was accepted by EMNLP 2026 (Findings)!
  • 06/2026 I will serve as a Program Committee member for USENIX Security 2027.
  • 05/2026 Our paper on multimodal jailbreak defense was accepted by ACM CCS 2026!
  • 03/2026 I am honored to serve on the IEEE Security and Privacy 2026 Artifact Evaluation Committee.
  • 01/2026 Our work Beyond Age-Based Restrictions: Rethinking Children’s Online Safety Through Comparing Parent–Child Perspectives of Risks in User-Generated Content Games was accepted to CHI 2026.
  • 09/2025 Our research on automatic hate video detection was accepted to EMNLP 2025.
  • 07/2025 I joined the NDSS 2026 Artifact Evaluation Committee.
  • 05/2025 I served on the ASONAM 2025 Program Committee.
  • 02/2025 I presented I know what you MEME! Understanding and Detecting Harmful Memes with Multimodal Large Language Models at NDSS 2025.
  • 02/2025 Our work on detecting and mitigating jailbreak attacks for large language models was accepted to USENIX Security 2025.
  • 01/2025 I was selected as an Internet Society Fellowship recipient for NDSS 2025.
Earlier news
  • 12/2024 I received the CSE Best Research Project (PhD) Award at the University at Buffalo.
  • 11/2024 I joined the artifact technical program committee for the 34th USENIX Security Symposium.
  • 10/2024 Our work on harmful meme detection was accepted to NDSS 2025.
  • 10/2024 Our work on AI-cybersecurity education with cyberharassment detection labs was accepted to CISSE 2024.
  • 10/2024 Our work on an AI-centered social cybersecurity education platform was accepted to CISSE 2024.
  • 10/2024 Our study on understanding cyberbullying images was accepted to ICMLA 2024.
  • 08/2024 I presented Moderating Illicit Online Image Promotion for Unsafe User-Generated Content Games Using Large Vision-Language Models at the 33rd USENIX Security Symposium.
  • 07/2024 I received a USENIX Conference Student Grant.
  • 06/2024 I presented the tutorial Machine Learning Based Online Abuse Defense: Platform, Research, and Hands-on Labs at ICWSM 2024.
  • 05/2024 Our work on unsafe user-generated content was selected as a poster at the 45th IEEE Symposium on Security and Privacy.
  • 04/2024 We had an accepted position statement at a CHI 2024 Workshop.
  • 04/2024 Our ASONAM 2023 paper on COVID-19-related online hate propagation through hateful memes was selected for the ACM Showcase on Kudos.
  • 03/2024 Our paper Moderating Illicit Online Image Promotion for Unsafe User-Generated Content Games Using Large Vision-Language Models was accepted to USENIX Security 2024.
  • 12/2023 I won the 2023 Annual CSE Poster Competition and received the CSE Best AI Poster Award at the University at Buffalo.
  • 12/2023 I presented An Investigation of Large Language Models for Real-World Hate Speech Detection at ICMLA 2023.
  • 10/2023 Our paper Moderating New Waves of Online Hate with Chain-of-Thought Reasoning in Large Language Models was accepted to IEEE S&P 2024.
  • 10/2023 Our paper An Investigation of Large Language Models for Real-World Hate Speech Detection was accepted to ICMLA 2023.
  • 09/2023 I presented Understanding and Measuring Robustness of Vision and Language Multimodal Models at SKM 2023.
  • 09/2023 Our paper on COVID-19-related online hate propagation through hateful memes was accepted to ASONAM 2023.
  • 08/2023 Our paper Understanding and Measuring Robustness of Vision and Language Multimodal Models was accepted to SKM 2023.
  • 08/2023 Our paper Exploring Vulnerabilities in Voice Command Skills for Connected Vehicles was accepted to EAI SmartSP 2023.
  • 04/2023 I presented Mitigating Online Hate in the Evolving Cyber Environment: Rapid Adaptation and Moderation of Emerging Threats at GLSD 2023.
  • 11/2022 Our paper on the generalizability of hateful memes detection models against COVID-19-related hateful memes was accepted to ICMLA 2022.
  • 11/2022 I received the CSE Best Graduate Teaching Award at the University at Buffalo.
  • 11/2022 We presented the demo paper Understanding the Effects of Paint Colors on LiDAR Point Cloud Intensities at AutoSec 2022.

Collaborators

Faculty Collaborators

Dr. Hongxin Hu, Dr. Haipeng Cai, Dr. Ziming Zhao, Dr. Nishant Vishwamitra, Dr. Long Cheng, Dr. Guo Freeman, Dr. Qian Wang, Dr. Juan Wang, Dr. Guofei Gu, Dr. Yongkai Wu, Dr. Xiaohong Yuan, Dr. Chunming Qiao, Dr. Feng Luo, Dr. Jeannette Wade

Ph.D. Student Collaborators

Ebuka Okpala, Feng Wei, Foad Hajiaghajani, Gaoxiang Liu, Md. Armanuzzaman Tomal, Mohammed Aldeen, Qiqing Huang, Ruchi Panchanadikar, Rupam Patir, Shenyi Zhang, Song Liao, Tian Zhang, Wenbo Ding, Xi Tan, Yang Hu, Yiheng Jing, Yong Zhuang, Zheyuan Ma

Sorted alphabetically by first name.

Undergraduate and Graduate Student Collaborators

Amardhruva Narasimha Prabhu, Ayush Utkarsh, Isabelle Ondracek, Pal Dave, Radhika Singh, Shaik Sabiha, Shishir Hebbar

Sorted alphabetically by first name.

High School Student Collaborators

Alexander Hu (now at UCLA), Alicia Gu (now at Stanford University), David Cong (now at Duke University), Helen Qin, Ishan Ajay, Jaden Mu (now at Carnegie Mellon University), Johnson Chen, Wentai Zhao (now at the University of Michigan)

Sorted alphabetically by first name.