Distributionally Robust Wireless Semantic Communication
Robust semantic communication for large AI model deployment over wireless channels.
This project explores semantic communication systems for deploying large AI models over noisy, low-bandwidth, and dynamically changing wireless channels.
My work contributed to modelling and experiments for LLM inference pipelines under dynamic channels, including distributionally robust training simulations and evaluation under constrained communication settings.
Highlights
- Simulated noisy and low-bandwidth wireless conditions for semantic communication.
- Contributed to modelling and experiment design for large AI model inference pipelines.
- Supported the accepted IEEE Journal on Selected Areas in Communications paper (Le et al., 2026).