Incoming PhD student at the CLNLP Lab, University of Toronto/AI Privacy & Security Researcher at SysNet
Researcher in Spoken Language Processing | AI Privacy & Security | Tech Enthusiast | Book Lover | Amateur Philosopher
May my knowledge grow as trees, and my thoughts shed as leaves
With which I sail through my years, on the tides of ceaseless change- David Wei
-
Audio Security: Preparing for my upcoming doctoral research by applying circuit-level interpretability to speech foundation models (HuBERT, Wav2Vec 2.0) to stress-test anti-spoofing detectors and audio watermarks against adversarial attacks.
-
Systems Security in LLMs: Discovering novel side-channel vulnerabilities in speculative decoding algorithms, demonstrating how observable metrics like packet size and timing can leak private user queries and retrieval-datastore contents.
-
Voice Privacy: Speaker Biometrics, voice anonymity, attack and defence
My life journey always goes out of plan, but I'm surviving it!!! 🚴♂️💪
Curious about what's the next thing gonna hit me 🧐🔮
Unexpected tasks are opportunities for me to improve, be receptive, and be optimistic! 💡🌱😊
Fun fact: When I'm not coding 💻 or buried in textbooks 📖, you'll find me lost in the world of books 📚 and cooking 🍳.
- AI, gen AI, LLMs, Spoken Language Processing, anything about AI
- AI Privacy & Security (who doesn't want to be a hacker?)
- Embodied Intelligence with native speech/communication capabilities
- Tactical Nuclear GPU
Discord
davidwei01001
