Robust communication
WaSeCom
Wasserstein distributionally robust wireless semantic communication with large AI models.
- Semantic communication
- Distributional robustness
- Wireless AI
Research projects
Research, built into systems.
We turn ideas in distributed intelligence into methods and systems for reliable communication, efficient edge inference, and resource-aware learning.
Efficient inference
GOODSPEED
Speculative decoding for efficient and fair LLM inference at the edge.
- Speculative decoding
- Edge inference
- Fair resource allocation
Research & publications
Follow the ideas behind each system.
Explore the methods, analysis, and published work that support DUAL projects.