Projects

Our projects span various domains of AI and machine learning, from developing novel speculative decoding systems for efficient LLM inference to creating innovative solutions for distributed computing and edge AI. Each project combines rigorous research with practical applications, addressing real-world challenges in modern AI systems.

Below you’ll find our featured projects and ongoing research initiatives that demonstrate our commitment to advancing the state-of-the-art in intelligent systems.

Featured

WaSeCom Wasserstein Distributionally Robust Wireless Semantic Communication with Large AI Models

A research project on enhancing the robustness of wireless semantic communication using large AI models and Wasserstein distributionally robust optimization.

semantic-communication large-ai-models wireless-communication distributionally-robust-optimization uncertainty robustness

GOODSPEED Speculative Decoding for Efficient and Fair LLM Inference at the Edge

A research project exploring speculative decoding techniques for efficient LLM inference in distributed edge environments. Currently in development - focusing on fair resource allocation and optimization strategies for edge computing scenarios.

large-language-models speculative-decoding distributed-systems edge-computing inference-optimization

More