Haoxiang Peng
Papers
2
Total Citations
40
H-Index
2
About
Haoxiang Peng is an emerging researcher at the forefront of efficient AI hardware design, specializing in the deployment of large language models (LLMs) on resource-constrained edge devices. His most notable contribution, EdgeLLM, proposes a highly efficient CPU-FPGA heterogeneous accelerator architecture that tackles one of the field's most pressing challenges: bringing the power of modern LLMs to edge platforms such as robots, where computational resources are severely limited. By leveraging the complementary strengths of CPUs and FPGAs, Peng's work offers a practical and scalable pathway for on-device AI inference without relying on cloud connectivity. This research has rapidly gained traction in the community, accumulating 40 citations across its publications — a remarkable milestone for work this recent, signaling strong interest from both academia and industry. Peng's contributions sit at the intersection of computer architecture, embedded systems, and artificial intelligence, addressing a critical bottleneck in making advanced AI accessible beyond data centers. His work positions him as a promising voice in the growing field of edge AI acceleration, with meaningful implications for robotics, IoT, and real-time intelligent systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2