Size Zheng
Scholar

Size Zheng

Google Scholar ID: TMZWR1gAAAAJ
ByteDance Seed
ArchitectureCompilerDeep Learning
Homepage↗Google Scholar↗
Citations & Impact
All-time
Citations
1,419
 
H-index
13
 
i10-index
18
 
Publications
20
 
Co-authors
8
list available
Contact
Emailzheng.size@outlook.comTwitterOpen ↗GitHubOpen ↗LinkedInOpen ↗
Publications
10 items
Teaching and Evaluating LLMs to Reason About Polymer Design Related Tasks
2026
Cited
0
Optimizing Long-context LLM Serving via Fine-grained Sequence Parallelism
2025
Cited
0
SwiftSpec: Ultra-Low Latency LLM Decoding by Scaling Asynchronous Speculative Decoding
2025
Cited
0
MegaScale-MoE: Large-Scale Communication-Efficient Training of Mixture-of-Experts Models in Production
2025
Cited
0
MxMoE: Mixed-precision Quantization for MoE with Accuracy and Performance Co-Design
2025
Cited
0
Triton-distributed: Programming Overlapping Kernels on Distributed AI Systems with the Triton Compiler
2025
Cited
0
MoQa: Rethinking MoE Quantization with Multi-stage Data-model Distribution Awareness
2025
Cited
0
TileLink: Generating Efficient Compute-Communication Overlapping Kernels using Tile-Centric Primitives
2025
Cited
0
Resume (English only)
Academic Achievements
  • - Publications: Multiple papers accepted by ICML 2025, ISCA 2025, MLSys 2025, NeurIPS 2024, etc.
  • - Awards: Outstanding Doctoral Dissertation Award of Peking University (July 2024); Outstanding Ph.D. Graduate of both Beijing and Peking University (July 2024)
Research Experience
  • - ByteDance: Led the Triton-distributed project
  • - University of Washington, SAMPL: Visiting Ph.D. student, September 2023 to January 2024
  • - DeepSeek AI: Short-term research intern
Education
  • - Ph.D.: School of Computer Science, Peking University, Advisor: Prof. Yun Liang
  • - B.S.: Department of Computer Intelligence Science, Peking University
Background
  • - Research Interests: High-performance inference system, AI compiler design, distributed systems
  • - Professional Field: High-performance computing and compiler design
  • - Brief Introduction: Led the Triton-distributed project at ByteDance. Worked as a visiting Ph.D. student with Prof. Luis Ceze on LLM serving and optimization in SAMPL at the University of Washington from September 2023 to January 2024.
Miscellany
  • Personal interests and other information not provided
Co-authors
8 total
Yun (Eric) Liang
Yun (Eric) Liang
Professor of EECS, Peking University, ACM Distinguished Scientist
Renze Chen
Renze Chen
Peking University
Xiuhong Li
Xiuhong Li
Infinigence-AI
Shengen Yan
Shengen Yan
Department of Electronic Engineering, Tsinghua University, China
Liqiang Lu
Liqiang Lu
Zhejiang University
Luis Ceze
Luis Ceze
Professor of Computer Science and Engineering, University of Washington
Co-author 7
Co-author 7
Anjiang Wei
Anjiang Wei
Stanford University