Built with and by Teycir Ben Soltane•
How to Use•FAQ•GitHub•arXiv.org•
Share:
ArXivCSExplorer
☆☆Bookmarks🏆RSSHow to UseFAQ
Home/Authors/Yusen Li

Yusen Li

2 indexed papers

Recent (6 mo)
2
With code
0
Influential cites
0
Benchmarked
0

Publications per year

2
26

Top categories

Algorithms×1Architecture×1

Frequent co-authors

Yinglong Gan1×
Jintao Yu1×
Shenggang Ying1×
Xin Hong1×
Di Mu1×
Tengyuan Jin1×

Research Timeline

2026
ComFuse: Fusing Complex Memory-Intensive Subgraphs with Compute-Intensive Kernels For Modern GPU Architectures

This paper presents ComFuse, an automated GPU compilation system that optimizes heterogeneous deep learning computation graphs by fusing compute-intensive and memory-intensive operators.

XBDD: A Highly Optimized ROBDD with Per-Edge Variable-Flip Maps

This paper proposes XBDD, a highly optimized ROBDD that reduces node count by merging nodes with local polarity differences.

Highlighted terms show continued research focus across papers

Papers

cs.DSEmpiricalRecentSep 29, 2026

XBDD: A Highly Optimized ROBDD with Per-Edge Variable-Flip Maps

Yinglong Gan, Jintao Yu, Shenggang Ying, Yusen Li +1 more

This paper proposes XBDD, a highly optimized ROBDD that reduces node count by merging nodes with local polarity differences.

View →
cs.AREmpiricalRecent
Aug 4, 2026

ComFuse: Fusing Complex Memory-Intensive Subgraphs with Compute-Intensive Kernels For Modern GPU Architectures

Di Mu, Tengyuan Jin, Zhenkun Wang, Jialin Yang +5 more

This paper presents ComFuse, an automated GPU compilation system that optimizes heterogeneous deep learning computation graphs by fusing compute-intensive and memory-intensive operators.

View →