Papers
1
Total Citations
17
H-Index
1
About
Di Huang is a prominent computer architecture researcher whose work sits at the intersection of hardware design, artificial intelligence, and efficient computing systems. His most recognized contribution, "Cambricon-LLM: A Chiplet-Based Hybrid Architecture for On-Device Inference of 70B LLM" (2024), has already garnered 17 citations within a short period — a testament to its timeliness and impact. This work addresses one of the most pressing challenges in modern AI deployment: enabling powerful large language models, such as 70-billion-parameter networks, to run efficiently on resource-constrained edge devices like smartphones and robotics platforms. By leveraging chiplet-based hybrid architectures, Huang's research offers a compelling solution that balances computational performance with the practical demands of on-device inference, including low arithmetic intensity and memory bandwidth constraints. His contributions directly advance the goals of user data privacy, network resilience, and democratized AI access by eliminating reliance on cloud infrastructure. Huang's research sits at a critical frontier where hardware innovation meets the exploding demand for intelligent edge computing, making his work highly relevant to students, engineers, and researchers shaping the next generation of AI systems.
Research Focus
Key Achievements
Top Papers
- 1