Fuhui Sun

Fuhui Sun

Auto-generated profile, based on publicly available information

Fuhui Sun

Information Technology Service Center of People's Court, Beijing, China

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
22 January 2026 in Pattern Recognition
Verified
Journal Article
Closed access
FHFeihong He
Feihong He
GLGang Li
Gang Li
FSFuhui Sun
Fuhui Sun
MZMengyuan Zhang
Mengyuan Zhang
LSLingyu Si
Lingyu Si
LSLi Shen
Li Shen

2 September 2025 in IEEE Transactions on Computers
Verified
Journal Article
Closed access
GXGuanyu Xu
Guanyu Xu
ZHZhiwei Hao
Zhiwei Hao
LSLi Shen
Li Shen
YLYong Luo
Yong Luo
FSFuhui Sun
Fuhui Sun
YWYonggang Wen
Yonggang Wen
The impressive performance of transformer models has sparked the deployment of intelligent applications on resource-constrained edge devices. However, ensuring high-quality service for real-time edge systems is a significant challenge due to the considerable computational demands and resource requirements of these models. Existing strategies typically either offload transformer computations to other devices or directly deploy compressed models on individual edge devices. These strategies, however, result in either considerable communication overhead or suboptimal trade-offs between accuracy and efficiency. To tackle these challenges, we propose a collaborative inference system for general transformer models, termed CoFormer. The central idea behind CoFormer is to exploit the divisibility and integrability of transformer. An off-the-shelf large transformer can be decomposed into multiple smaller models for distributed inference, and their intermediate results are aggregated to generate the final output. We formulate an optimization problem to minimize both inference latency and accuracy degradation under heterogeneous hardware constraints. DeBo algorithm is proposed to first solve the optimization problem to derive the decomposition policy, and then progressively calibrate decomposed models to restore performance. We demonstrate the capability to support a wide range of transformer models on heterogeneous edge devices, achieving up to 3.1$\times$ inference speedup with large transformer models. Notably, CoFormer enables the efficient inference of GPT2-XL with 1.6 billion parameters on edge devices, reducing memory requirements by 76.3%. CoFormer can also reduce energy consumption by approximately 40% while maintaining satisfactory inference performance.

7 February 2026 in Information Processing & Management
Verified
Journal Article
Closed access
SBSheng Bi
Sheng Bi
KYKai Yu
Kai Yu
JLJiaqi Li
Jiaqi Li
XWXiaoyan Wang
Xiaoyan Wang
FSFuhui Sun
Fuhui Sun
LWLusheng Wang
Lusheng Wang

21 February 2025
Verified
Conference Paper
Open access
XLXiaolong Liu
Xiaolong Liu
FSFuhui Sun
Fuhui Sun
XWXiaoyan Wang
Xiaoyan Wang
TSTanfeng Sun
Tanfeng Sun

14 March 2025
Verified
Conference Paper
Open access
XLXiaolong Liu
Xiaolong Liu
JWJin Wang
Jin Wang
FSFuhui Sun
Fuhui Sun
HSHang Su
Hang Su
ZLZhanhong Lu
Zhanhong Lu
TSTanfeng Sun
Tanfeng Sun

16 January 2026
Verified
Conference Paper
Open access
YGYa Gao
Ya Gao
XLXiaolong Liu
Xiaolong Liu
FSFuhui Sun
Fuhui Sun
XWXiaoyan Wang
Xiaoyan Wang
This paper aims to systematically review domestic and international research on the evaluation system of technological innovation. From the national macro-level perspectives, it examines the progress, shortcomings, and future trends in existing studies, providing a reference for the construction and improvement of scientific and technological innovation evaluation systems across various fields and exploring the inherent connections among different evaluation systems. The article first reviews and analyzes current research on indicator construction in technological innovation evaluation systems both in China and abroad. It then compares the evaluation methods used in different literature, discussing their characteristics and applicable scopes. Furthermore, it focuses on the computational principles of evaluation methods, particularly elaborating on the calculation of indicator weights and evaluation object scores. Taking the China's national macro-level as an example, a research framework is constructed based on two approaches—data modeling and time series analysis—using model-based prediction to analyze the development trends of technological innovation. Finally, the paper concludes with a summary and outlines directions for future research.