Associate Professor

School of Computer Science & Engineering

Beihang University

I am an associate professor in School of Computer Science & Engineering at Beihang University. I received Ph.D degree under supervision of Prof. Yi Liu and Prof. Hailong Yang. My research interests include high performance computing, computer architecture, and deep learning system. My recent research investigates LLM system, GNN system, sparse tensor computation, and deep learning compiler. I have authored over 30 publications in the leading international journals and conferences.

I am awarded with BAAI Young Scholar (北京市智源青年学者), ACM SIGHPC China Doctoral Dissertation Award (ACM SIGHPC中国优博), CCF TCARCH Doctoral Dissertation Award (CCF体系结构优博) and BUAA Doctoral Dissertation Award (北京航空航天大学优博). My papers have been selected as CLUSTER’21 Best Paper Nomination and IEEE Computer’s “Spotlight on Transactions”. I currently serve as executive committee member of CCF TCARCH and TCHPC, as well as reviewers in the journals including TPDS, TC, CSUR, TCC, THPC, SUPE, FCS, JCST, FGCS, 计算机学报, and 计算机工程与科学.

🔥 News

• 🔥 October 2026: One paper (TIDE) is accepted to PPoPP’27, congratulations to Yiqing.
• October 2026: We organize a workshop (第四届多样性算力下的编译、运行时和性能优化论坛) at CCF HPC China 2026.
• October 2026: Invited to give two talks at CCF HPC China 2026 and 中国计算语言学大会.
• 🔥 July 2026: Two papers (SpG-Attention, Grape) are accepted to SC’26, congratulations to Kejie and Siqi.
• 🔥 July 2026: One paper (MAGSAT) is accepted to ICPP’26, congratulations to Haodong.
• June 2026: Invited to give one talk at CCF第一届大模型与AI工程大会.
• April 2026: One paper (ASTRA) is accepted to ICS’26, congratulations to Wenqing and Jianuo.
• March 2026: Invited to give one talk at CCF第一届先进计算与系统结构青年学者研讨会.

📝 Selected Publications

• (Corresponding Author) Exploiting Temporal Heterogeneity in Distributed Temporal Graph Network Training (PPoPP) 2027.
• (Corresponding Author) Towards High-Performance Sparse Graph Attention on Dense Tensor Core Units (SC) 2026.
• Efficient Serving for Agentic LLM Workflows via Micro-Task-Level Parallelism (SC) 2026.
• (Corresponding Author) MAGSAT: Memory-Access-Guided Fusion and Shape-Adaptive Auto-Tuning for Dynamic Tensor Programs (ICPP) 2026.
• (Corresponding Author) Optimizing Streaming Tensor Decomposition on GPU (ICS) 2026.
• (Corresponding Author) Efficient Temporal Graph Network Training via Unified Redundancy Elimination (ASPLOS) 2026.
• (Corresponding Author) APERTURE: Algorithm-System Co-optimization for Temporal Graph Network Inference (PPoPP) 2026.
• GNNPerf: Towards Effective Performance Profiling and Analysis Across GNN Frameworks (IPDPS) 2025.
• MILLION: MasterIng Long-Context LLM Inference Via Outlier-Immunized KV Product QuaNtization (DAC) 2025.
• PISA: Efficient Precision-Slice Framework for LLMs with Adaptive Numerical Type (DAC) 2025.
• Accelerating Complex Stencil Computations with Adaptive Fusion Strategy (ICS) 2025.
• EVASION: Efficient KV CAche CompreSsion vIa PrOduct QuaNtization (DATE) 2025.
• (First Author) 面向GPU平台的通用Stencil自动调优框架 (计算机研究与发展) 2025.
• (Corresponding Author) ScalFrag: Efficient Tiled-MTTKRP with Adaptive Launching on GPUs (CLUSTER) 2024.
• (First Author) Adaptive Auto-Tuning Framework for Global Exploration of Stencil Optimization on GPUs (TPDS) 2024.
• Exploiting Input Tensor Dynamics in Activation Checkpointing for Efficient Training on GPU (IPDPS) 2023.
• (First Author) CoGNN: Efficient Scheduling for Concurrent GNN Training on GPUs (SC) 2022.
• (First Author) QoS-aware Dynamic Resource Allocation with Improved Utilization and Energy Efficiency on GPU (PARCO) 2022.
• (First Author) StencilMART: Predicting Optimization Selection for Stencil Computations across GPUs (IPDPS) 2022.
• (First Author) Input-Aware Sparse Tensor Storage Format Selection for Optimizing MTTKRP (TC) 2021.
• (First Author) csTuner: Scalable Auto-tuning Framework for Complex Stencil Computation on GPUs (CLUSTER) 2021.
• Automatic Code Generation and Optimization of Large-scale Stencil Computation on Many-core Processors (ICPP) 2021.
• An Optimized Tensor Completion Library for Multiple GPUs (ICS) 2021.
• Towards Efficient Canonical Polyadic Decomposition on Sunway Many-core Processor (Information Sciences) 2021.
• The Deep Learning Compiler: A Comprehensive Survey (TPDS) 2021.
• (First Author) SpTFS: Sparse Tensor Format Selection for MTTKRP via Deep Learning (SC) 2020.
• (First Author) SMQoS: Improving Utilization and Energy Efficiency with QoS Awareness on GPUs (CLUSTER) 2019.
• Improving Thread-level Parallelism in GPUs through Expanding Register File to Scratchpad Memory (TACO) 2018.

🎖 Honors and Awards

• BAAI Young Scholar 2026 (北京市智源青年学者).
• BUAA Doctoral Dissertation Award 2024 (北京航空航天大学优秀博士学位论文).
• ACM SIGHPC China Doctoral Dissertation Award 2023 (ACM SIGHPC中国优博奖).
• CCF TCARCH Doctoral Dissertation Award 2023 (CCF体系结构优博奖).
• IEEE Computer’s “Spotlight on Transactions” August 2023.
• Best Paper Award Nomination, IEEE International Conference on Cluster Computing (CLUSTER), 2021. (2 nominations out of 168 submissions).

💬 Teaching

• To be updated.

📖 Educations

• 2017.09 - 2023.06, Computer Architecture, Beihang University.
• 2013.09 - 2017.06, Computer Science and Technology, China University of Petroleum-Beijing.