Jinliang Shi

Jinliang Shi

Auto-generated profile, based on publicly available information

Jinliang Shi

Beijing University of Posts and Telecommunications, Beijing, China

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

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

Representative Publications

View All Publications
28 February 2025
Verified
Conference Paper
Open access
JSJinliang Shi
Jinliang Shi
SLSHIGANG LI
SHIGANG LI
YXYouxuan Xu
Youxuan Xu
RFRongtian Fu
Rongtian Fu
XWXueying Wang
Xueying Wang
TWTong Wu
Tong Wu
Sparse Matrix-matrix Multiplication (SpMM) and Sampled Dense-dense Matrix Multiplication (SDDMM) are important sparse operators in scientific computing and deep learning. Tensor Core Units (TCUs) enhance modern accelerators with superior computing power, which is promising to boost the performance of matrix operators to a higher level. However, due to the irregularity of unstructured sparse data, it is difficult to deliver practical speedups on TCUs. To this end, we propose FlashSparse, a novel approach to bridge the gap between sparse workloads and the TCU architecture. Specifically, FlashSparse minimizes the sparse granularity for SpMM and SDDMM on TCUs through a novel swap-and-transpose matrix multiplication strategy. Benefiting from the minimum sparse granularity, the computation redundancy is remarkably reduced while the computing power of TCUs is fully utilized. Besides, FlashSparse is equipped with a memory-efficient thread mapping strategy for coalesced data access and a sparse matrix storage format to save memory footprint. Extensive experimental results on H100 and RTX 4090 GPUs show that FlashSparse sets a new state-of-the-art for sparse matrix multiplications (geometric mean 5.5x speedup over DTC-SpMM and 3.22x speedup over RoDe).

28 November 2020
Verified
Conference Paper
Closed access
LXLei Xiao
Lei Xiao
XYXiang Yi
Xiang Yi
YZYuan Zhai
Yuan Zhai
JSJinliang Shi
Jinliang Shi
GLGenli Liu
Genli Liu
ZYZhi Yang
Zhi Yang
A design method of Ku-band high performance frequency synthesizer is introduced in this paper. Based on the traditional in loop mixing phase locking method, the influence of additional phase noise of digital phase detector, frequency divider and other components on the output phase noise performance is reduced through reasonable frequency planning, optimization of bias frequency and frequency division ratio. The developed frequency synthesizer has low spurious and the phase noise performance similar to that of direct frequency synthesis is obtained, which has good engineering application. At the same time, the volume of frequency synthesizer is greatly reduced by adopting 3D assembly technology of RF heterogeneous integration, which provide one method for the miniaturization of frequency synthesizer.

13 July 2026
Verified
Conference Paper
Open access
RFRongtian Fu
Rongtian Fu
SLSHIGANG LI
SHIGANG LI
YXYouxuan Xu
Youxuan Xu
TWTong Wu
Tong Wu
ZMZhi Ma
Zhi Ma
JSJinliang Shi
Jinliang Shi

14 November 2026
Verified
Conference Paper
Open access
YXYouxuan Xu
Youxuan Xu
RFRongtian Fu
Rongtian Fu
JSJinliang Shi
Jinliang Shi
TWTong Wu
Tong Wu
ZMZhi Ma
Zhi Ma
SLSHIGANG LI
SHIGANG LI

1 November 2026 in IEEE Transactions on Parallel and Distributed Systems
Verified
Journal Article
Closed access
JSJinliang Shi
Jinliang Shi
SLSHIGANG LI
SHIGANG LI
RFRongtian Fu
Rongtian Fu
XWXueying Wang
Xueying Wang
YXYouxuan Xu
Youxuan Xu
TWTong Wu
Tong Wu