Publications

Students with † are advised by me; Students with * are co-first authors.

2026

  1. SCOPE: Array-Native Nonlinear Acceleration for LLM Inference via Shape-Constrained Neural Approximation
    Chenyu Zhang ,  Chenxi Xu ,  Jiaxiang Zou ,  Yonghao Chen ,  and  Xinyu Chen
    In The 59th IEEE/ACM International Symposium on Microarchitecture (MICRO) , 2026
    to appear
  2. MFNNS: A Multiplier-Free Near-Memory Accelerator for Graph-based ANNS
    Rui Meng ,  Yonghao Chen ,  Jiaxiang Zou ,  Jingyu Guo ,  Chenyu Zhang ,  and  Xinyu Chen
    In The 59th IEEE/ACM International Symposium on Microarchitecture (MICRO) , 2026
    to appear
  3. Ultra-DSP: A Universal Lossless DSP Overpacking Framework for Low-Bit LLM Inference
    Chenyu Zhang ,  Chenxi Xu ,  Jiaxiang Zou ,  Jingyu Guo ,  Rui Meng ,  and  Xinyu Chen
    In The 59th IEEE/ACM International Symposium on Microarchitecture (MICRO) , 2026
    to appear
  4. Sparse by Command: Task-Conditional Compute Skipping for Multi-Task Inference Accelerators
    Afzal Ahmad ,  Gaoyu Mao ,  Shoubo Hu ,  Hui-Ling Zhen ,  Mingxuan Yuan ,  Xinyu Chen ,  and  Wei Zhang
    In The 59th IEEE/ACM International Symposium on Microarchitecture (MICRO) , 2026
    to appear
  5. RidgeBridge: Random Access Optimized Interconnect Architecture for Scalable Graph Random Walks
    Hongshi Tan ,  Yao Chen ,  Xinyu Chen ,  Qizhen Zhang ,  Weng-Fai Wong ,  and  Bingsheng He
    In The 59th IEEE/ACM International Symposium on Microarchitecture (MICRO) , 2026
    to appear
  6. UniCore: A Bit-Width Scalable GEMM Unit for Unified LLM Inference
    Yonghao Chen†* ,  Jiaxiang Zou†* ,  Xingyu Chen† ,  Chenxi Xu† ,  Jingyu Guo† ,  and  Xinyu Chen
    In International Symposium on Computer Architecture (ISCA) , 2026
    PDF Code
  7. MixFP4: Extending NVFP4 to Mixed Micro-Format via Scale-Bit Reuse and Tensor Core Co-design
    Jiaxiang Zou ,  Yonghao Chen ,  Ruilong WU ,  and  Xinyu Chen
    In International Conference on Machine Learning (ICML) , 2026
  8. DSTREE: Data-Driven Synchronous Traversals for Decision Forests on GPUs
    Bi Zeng†* ,  Zhenlin Wu†* ,  Hongyuan Liu ,  and  Xinyu Chen
    In IEEE International Parallel and Distributed Processing Symposium (IPDPS) , 2026
  9. RidgeWalker: Perfectly Pipelined Graph Random Walks on FPGAs
    Hongshi Tan ,  Yao Chen ,  Xinyu Chen ,  Qizhen Zhang ,  Cheng Chen ,  Weng-Fai Wong ,  and  Bingsheng He
    In IEEE International Symposium on High Performance Computer Architecture (HPCA) , 2026

2025

  1. AxCore: A Quantization-Aware Approximate GEMM Unit for LLM Inference
    Jiaxiang Zou†* ,  Yonghao Chen†* ,  Xingyu Chen† ,  Chenxi Xu† ,  and  Xinyu Chen
    In IEEE/ACM International Symposium on Microarchitecture (MICRO) , 2025
    PDF Code
  2. X-SET: An Efficient Graph Pattern Matching Accelerator With Order-Aware Parallel Intersection Units
    Chenxi Xu† ,  Tianhui Shi ,  Shixuan Sun ,  Jidong Zhai ,  and  Xinyu Chen
    In IEEE/ACM International Symposium on Microarchitecture (MICRO) , 2025
    PDF Code
  3. OA-LAMA: An Outlier-Adaptive LLM Inference Accelerator with Memory-Aligned Mixed-Precision Group Quantization
    Huangxu Chen† ,  Yingbo Hao ,  Yi Zou ,  and  Xinyu Chen
    In IEEE/ACM International Conference on Computer-Aided Design (ICCAD) , 2025
    PDF Code
  4. Advancing Matrix Operations for High-Performance and Memory-Efficient Automata Processing on GPUs
    Zhenlin Wu† ,  Tianao Ge ,  Jiajia Li ,  Xinyu Chen ,  and  Hongyuan Liu
    In ACM Transactions on Architecture and Code Optimization (TACO) , 2025
  5. Graphitron: A Domain Specific Language for FPGA-Based Graph Processing Accelerator Generation
    Xinmiao Zhang ,  Zheng Feng ,  Shengwen Liang ,  Xinyu Chen ,  Lei Zhang ,  and  Cheng Liu
    In ACM SIGPLAN/SIGBED International Conference on Languages, Compilers, and Tools for Embedded Systems (LCTES) , 2025
    PDF
  6. Rethinking Dynamic Networks and Heterogeneous Computing with Automatic Parallelization
    Ruilong WU ,  Xinjiao Li ,  Yisu Wang ,  Xinyu Chen ,  and  Dirk Kutscher
    In The 9th Asia-Pacific Workshop on Networking (APNet) , 2025
    PDF
  7. April: Accuracy-Improved Floating-Point Approximation For Neural Network Accelerators
    Yonghao Chen† ,  Jiaxiang Zou† ,  and  Xinyu Chen
    In ACM/IEEE Design Automation Conference (DAC) , 2025
    PDF Code
  8. Clementi: Efficient Load Balancing and Communication Overlap for Multi-FPGA Graph Processing
    Feng Yu ,  Hongshi Tan ,  Xinyu Chen ,  Yao Chen ,  Bingsheng He ,  and  Weng-Fai Wong
    Proceedings of the ACM on Management of Data (SIGMOD), 2025
    PDF Code

2023

  1. LightRW: FPGA Accelerated Graph Dynamic Random Walks
    Hongshi Tan ,  Xinyu Chen ,  Yao Chen ,  Bingsheng He ,  and  Weng-Fai Wong
    Proceedings of the ACM on Management of Data (SIGMOD), 2023
    PDF Code

2022

  1. ReGraph: Scaling Graph Processing on HBM-enabled FPGAs with Heterogeneous Pipelines
    Xinyu Chen ,  Yao Chen ,  Feng Cheng ,  Hongshi Tan ,  Bingsheng He ,  and  Weng-Fai Wong
    In IEEE/ACM International Symposium on Microarchitecture (MICRO) , 2022
    PDF Code
  2. ThunderGP: Resource-efficient graph processing framework on FPGAs with HLS
    Xinyu Chen ,  Feng Cheng ,  Hongshi Tan ,  Yao Chen ,  Bingsheng He ,  Weng-Fai Wong ,  and  Deming Chen
    ACM Transactions on Reconfigurable Technology and Systems (TRETS), 2022
    PDF Code

2021

  1. ThunderGP: HLS-based graph processing framework on FPGAs
    Xinyu Chen ,  Hongshi Tan ,  Yao Chen ,  Bingsheng He ,  Weng-Fai Wong ,  and  Deming Chen
    In ACM/SIGDA International Symposium on Field-Programmable Gate Arrays (FPGA) , 2021
    PDF Code
  2. Skew-oblivious data routing for data intensive applications on FPGAs with HLS
    Xinyu Chen ,  Hongshi Tan ,  Yao Chen ,  Bingsheng He ,  Weng-Fai Wong ,  and  Deming Chen
    In ACM/IEEE Design Automation Conference (DAC) , 2021
    PDF
  3. ThundeRiNG: generating multiple independent random number sequences on FPGAs
    Hongshi Tan ,  Xinyu Chen ,  Yao Chen ,  Bingsheng He ,  and  Weng-Fai Wong
    In ACM International Conference on Supercomputing (SC) , 2021
    PDF Code

2020

  1. Is FPGA useful for hash joins?
    Xinyu Chen ,  Yao Chen ,  Ronak Bajaj ,  Jiong He ,  Bingsheng He ,  Weng-Fai Wong ,  and  Deming Chen
    In Conference on Innovative Data Systems Research (CIDR) , 2020
    PDF Code
  2. G3: when graph neural networks meet parallel graph processing systems on GPUs
    Husong Liu ,  Shengliang Lu ,  Xinyu Chen ,  and  Bingsheng He
    Proceedings of the VLDB Endowment (VLDB), 2020
    PDF Code

2019

  1. Deploying hash tables on die-stacked high bandwidth memory
    Xuntao Cheng ,  Bingsheng He ,  Eric Lo ,  Wei Wang ,  Shengliang Lu ,  and  Xinyu Chen
    In International Conference on Information and Knowledge Management (CIKM) , 2019
    PDF Code
  2. On-the-fly parallel data shuffling for graph processing on OpenCL-based FPGAs
    Xinyu Chen ,  Ronak Bajaj ,  Yao Chen ,  Jiong He ,  Bingsheng He ,  Weng-Fai Wong ,  and  Deming Chen
    In International Conference on Field Programmable Logic and Applications (FPL) , 2019
    PDF Code
  3. A survey on graph processing accelerators: Challenges and opportunities
    Chuang-Yi Gui ,  Long Zheng ,  Bingsheng He ,  Cheng Liu ,  Xinyu Chen ,  Xiao-Fei Liao ,  and  Hai Jin
    Journal of Computer Science and Technology, 2019
    PDF
  4. OBFS: OpenCL based BFS optimizations on software programmable FPGAs
    Cheng Liu ,  Xinyu Chen ,  Bingsheng He ,  Xiaofei Liao ,  Ying Wang ,  and  Lei Zhang
    In International Conference on Field-Programmable Technology (FPT) , 2019
    PDF Code