Comprehensive taxonomy of AI security vulnerabilities, LLM adversarial attacks, prompt injection techniques, and machine learning security research. Covers 71+ attack vectors including model poisoning, agentic AI exploits, and privacy breaches.

Master's students in NCCU SoSLab maintaining a cleaned and restructured version of INCITE (based on PyCT).

Adversarial ML Scanner for threat detection and ML backdoor attcaks

🛡️ Discover and analyze critical vulnerabilities in Meta AI's Instagram Group Chat, ensuring robust security through comprehensive testing and reporting.

Meta AI Bug Bounty focuses on vulnerabilities in Instagram's Group Chat feature, revealing critical security risks. Join us in improving AI safety and explore our findings on GitHub! 🐙✨

Bug bounty report demonstrating prompt injection and command execution vulnerabilities in Meta AI's Instagram Group Chat

Fork for my contributions on Trojans, Comprehensive course materials covering ML safety, robustness, and AI alignment

Placeholder for Annotated Research Papers in Deep learning discussed in Blog Posts. Most of the papers related to Adversarial ML and GAN.

A Gemini-powered explainable AI assistant that analyzes and interprets anomalies in IoT power signals using RAG, few-shot prompting, and embeddings. Built during the 2025 Google GenAI Capstone.