Research

Recent publications

Publications

Google Scholar
2025

GraphQLer: Enhancing GraphQL Security with Context-Aware API Testing

O Tsai, J Li, TT Cheung, L Huang, H Zhu, J Xiao, I Sharafaldin, MA Tayebi

arXiv preprint arXiv:2504.13358

2025

PrediQL: Automated Testing of GraphQL APIs with LLMs

S Liu, S Marefat, O Tsai, Y Chen, Z Deng, J Wang, MA Tayebi

arXiv preprint arXiv:2510.10407

2024

XploitSQL: Advancing Adversarial SQL Injection Attack Generation with Language Models and Reinforcement Learning

D Leung, O Tsai, K Hashemi, B Tayebi, MA Tayebi

Proceedings of the 33rd ACM International Conference on Information and Knowledge Management

Areas I'm actively thinking about. If any of these overlap with your work, I'd love to compare notes.

🔍

API Security & Fuzzing

Automated vulnerability discovery in GraphQL and REST APIs — dependency-aware fuzzing, IDOR detection, and the limits of static vs. dynamic analysis.

🤖

LLM-Assisted Security Testing

Using large language models to improve test coverage, classify endpoints, and reason about access-control chains. Where does the model help, and where does it hallucinate?

🛡️

AI-Powered Cyber Defence

Using AI systems to detect, respond to, and anticipate attacks in real time — from anomaly detection to autonomous incident response and adaptive threat modelling.

📡

Hardware & Embedded Security

Attack surfaces in embedded and IoT devices — firmware analysis, side-channel attacks, and the gap between software security assumptions and hardware reality.

📊

Data-Driven Threat Intelligence

Applying data science and ML to security telemetry — anomaly detection, attribution, and building pipelines that surface signal from noise at scale.

📈

Social Economics

Incentives, information asymmetry, and social dynamics — with intersections in security (bug bounties, ransomware markets) and AI (labour displacement, governance).