Xiaoyan Liu

Xiaoyan Liu

Auto-generated profile, based on publicly available information

Xiaoyan Liu

Beihang University, Beijing, China

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

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

Representative Publications

View All Publications
12 August 2024
Verified
Conference Paper
Open access
KZKaige Zhang
Kaige Zhang
XLXiaoyan Liu
Xiaoyan Liu
HYHailong Yang
Hailong Yang
TFTianyu Feng
Tianyu Feng
XYXinyu Yang
Xinyu Yang
DQDepei Qian
Depei Qian
As deep learning models continue to grow larger, model pruning is employed to reduce memory footprint and computation complexity, which generates a large number of sparse matrix-matrix multiplication (SpMM) with unstructured sparsity (e.g., vector sparsity). However, leveraging GPU especially the newly integrated sparse tensor core (SpTC) to accelerate SpMM is quite challenging due to the unstructured sparsity. Unfortunately, existing works fail to fully exploit the SpTC on GPU due to the difficulty of satisfying the stringent requirement for restricted sparsity (e.g., 2:4 sparsity). In this paper, we propose Jigsaw, a novel method to utilize SpTC for accelerating SpMM with vector sparsity. Specifically, we propose the multi-granularity sparsity reorder method to transform the sparse data for satisfying the sparse pattern supported on SpTC. In addition, we propose a reorder-aware storage format for the transformed sparse data to better adapt to the parallelism of SpTC. Moreover, we propose corresponding optimizations to better exploit the SpTC for further accelerating SpMM. The experiment results demonstrate that Jigsaw outperforms state-of-the-art SpMM implementations and achieves promising speedup over cuBLAS.

20 February 2024
Verified
Conference Paper
Open access
XLXiaoyan Liu
Xiaoyan Liu
XZXuegui Zheng
Xuegui Zheng
HYHailong Yang
Hailong Yang
ZLZhongzhi Luan
Zhongzhi Luan
DQDepei Qian
Depei Qian
Convolutional neural networks (CNNs) have achieved remarkable success in various application fields. Although model compression techniques mitigate the ever-increasing resource demands of large CNN models, the compressed models usually exhibit irregular memory access and unstructured sparsity, which are difficult for dominant operators such as sparse convolution to achieve expected performance speedup on popular inference platforms such as GPU. In this paper, we propose Tetris, an efficient sparse convolution approach optimized for GPU. Tetris first fully exploits the input reuse opportunity of sparse convolution to reduce the memory accesses to global memory. It then adopts a stride packed filter (SPF) format and a bank-sensing reorganization scheme to eliminate the irregular memory accesses caused by unstructured sparsity. It also leverages a filter group reorder technique to address load imbalance among threads, and a parameter tuning method to determine the optimal parameters of the sparse convolution implementation. The experiment results show that Tetris outperforms dense/sparse convolution libraries and cutting-edge implementations with promising performance speedup.

8 June 2025
Verified
Conference Paper
Open access
SWSiqi Wang
Siqi Wang
HYHailong Yang
Hailong Yang
PWPengbo Wang
Pengbo Wang
SDShaokang Du
Shaokang Du
YXYufan Xu
Yufan Xu
DQDepei Qian
Depei Qian

8 September 2025
Verified
Conference Paper
Open access
XWXuezhu Wang
Xuezhu Wang
HYHailong Yang
Hailong Yang
XYXin You
Xin You
YXYufan Xu
Yufan Xu
XLXiaoyan Liu
Xiaoyan Liu
DQDepei Qian
Depei Qian