Hongyang Jia

A heterogeneous microprocessor for energy-scalable sensor inference using genetic programming

2017

We present a heterogeneous microprocessor for IoE sensor-inference applications, which achieves programmability required for feature extraction strictly using application data. Acceleration, though key for energy efficiency, poses substantial programmability challenges. These are overcome by exploiting …

A 28nm 3.14 TFLOP/W BF16 LLM Fine-Tuning Processor with Asymmetric Quantization Computing for AI PC

2025

The powerful capabilities of large language models (LLMs) enable them to function as personal digital assistants. To ensure user privacy, personalized fine-tuning can be conducted locally on memoryconstrained AI PCs using Parameter-Efficient Fine-Tuning (PEFT) algorithms, …